RESTRICTED

(95-1073)

REPORT OF THE PANEL

Table of Contents Paragraphs

I. INTRODUCTION 1-6

II. FACTUAL BACKGROUND 7-10

III. FINDINGS AND RECOMMENDATIONS REQUESTED 11-14

1. Findings requested 11-12

2. Recommendations requested 13-14

IV. PRELIMINARY OBJECTIONS OF THE EC 15-69

Introduction 15-16

or outside the terms of reference of the Panel 17-46

A. General arguments 17-33

B. Specific arguments 34-46

(1) Introduction 34-35

(2) Calculation of constructed value 36-37

(3) Refunds of anti-dumping duties 38-40

(4) Price undercutting methodology 41-42

(5) Effect of any price increases 43-44

(6) Other factors 45-46

Paragraphs

2. Whether Japan had a legal interest which could give rise

to nullification and impairment 47-69

A. General arguments 47-60

with respect to "asymmetry" and "zeroing" 61-69

(1) Introduction 61-62

(2) "Asymmetry" 63-67

(3) "Zeroing" 68-69

V. ARGUMENTS OF THE PARTIES 70-270

the Anti-Dumping Agreement as exceptions 70-77

2. Calculation of the dumping margin 78-133

A. Introduction 78-79

B. "Asymmetry" 80-102

(1) "level of trade" 103-108

(2) "due allowances" 109-114

C. "Zeroing" of sales at prices above normal value 115-133

(1) Introduction 115-118

(2) Specific arguments 119-133

3. Calculation of the constructed normal value 134-135

A. Allocation of SG&A 136-142

B. Profit 143-145

4. Causation of injury 146-147

A. Specific arguments 148-149

Paragraphs

B. Cumulation of Japanese imports 150-151

(1) Distinct markets 152-157

(2) Inconsistent application of the EC's cumulation

criteria 158-165

C. Volume 166-167

(1) A substantive or a procedural requirement 168-173

(2) The number of elements required 174-177

(3) The meaning of "significant" 178-180

(4) Whether a finding of "significant" volume

increase was required 181-183

(5) Whether the volume of Japanese was

significant 184-197

D. Effect on prices in the EC market 198-201

(1) Price undercutting 202-221

(i) Introduction 202-205

(ii) Margin of undercutting 206-207

(a) Methodology 208-213

(b) Zeroing of overcut prices 214-219

(iii) Volume of undercut products 220-221

E. Price depression/suppression 222-224

(1) Consideration of the effect of the imports 225-239

F. Expenditure on advertising as an effect of dumping 240-247

G. Size of the dumping margins 248-249

Paragraphs

5. Other factors 250-270

A. Introduction 250-252

(1) The effects of dumping 253-257

(2) The burden of proof 258-270

VI. ARGUMENTS OF THIRD PARTIES 271-276

VII. REMEDIES REQUESTED 277-291

1. Revocation 277-284

2. The Basic Regulation 285-288

3. Reimbursement 289-291

VIII. FINDINGS 292-452

1. Preliminary objections 292-346

within the Panel's terms of reference 292-340

(1) General considerations 292-304

(2) Claims not within the Panel's terms of

reference 305-306

(i) EC price undercutting methodology 307-311

(ii) Effect of any price increases 312-316

(iii) EC refund procedures 317-318

(3) Claims not subject to conciliation 319

(i) EC refund procedures 320

(ii) Price undercutting methodology 321-322

(iii) Other factors 323-329

(iv) Calculation of constructed normal value 330-338

(4) Conclusion 389-340

Paragraphs

B. Claims in which Japan had no legal interest 341-346

the existence and extent of dumping 347-403

A. "Zeroing" of sales at prices above normal value 347-366

B. "Asymmetry" 367-388

C. Construction of the normal value: amount for profit 389-396

D. Construction of the normal value: amount for SG&A 397-401

E. Claims under Article 8:3 402-403

3. Determination of the existence of material injury 404-452

A. Cumulation of the effect of dumped imports from Japan

and Korea 406

(1) Claim based on Article 3:4 407-412

(2) Claim based on Article 3:1 413-417

B. Volume and price effects of the dumped imports 418

(1) General considerations 419-422

(2) Volume of the dumped imports 423-427

(3) Price undercutting 428-429

(i) Establishment of "significant price

undercutting" 430-432

(ii) Methodology for calculating margin of

price undercutting 433-439

(4) Price suppression/depression 440-448

(5) Other factors 449-452

IX. CONCLUSIONS AND RECOMMENDATIONS 452-460

X. DISSENTING OPINION OF ONE MEMBER ON ASPECTS

OF THE PANEL'S CONCLUSION RELATING TO ASYMMETRY 461-468

ANNEX Page

100-103

1. This dispute arose out of the imposition of definitive anti-dumping duties on certain imports of audio cassettes from Japan by the European Community ("the EC") in 1991.

2. Japan requested consultations with the EC on 8 July 1991. Consultations were held in July 1991, October 1991, December 1991 and April 1992 under Article 15:2 of the Agreement on Implementation of Article VI of the General Agreement on Tariffs and Trade ("the Agreement"). Japan referred the matter to the Committee for conciliation pursuant to Article 15:3 of the Agreement on 22 May 1992 (ADP/79). A special meeting of the Committee on Anti-Dumping Practices ("the Committee") was held on 9 July 1992 for this purpose (ADP/M/38). The conciliation process did not lead to resolution of the dispute, and Japan requested the establishment of a Panel on 13 October 1992 (ADP/85). On 21 October 1992 Japan provided a reference paper designed to clarify the issues in dispute (ADP/85/Add.1). The Committee established a Panel at its regular meeting held on 26, 27 and 30 October 1992. The United States, Korea and Canada reserved their rights to make submissions as third parties. The Committee authorised the Chairman to conduct informal consultations with the parties to the dispute regarding the terms of reference of the Panel (ADP/M/39).

3. At a meeting of the Committee on 29 April 1993 the Chairman informed the Committee that the parties had agreed on standard terms of reference. They were:

The Chairman noted that certain clarifications regarding the scope of the terms of reference would be provided to the Chairman of the Panel in the form of a letter from Japan. See Annex. The Chairman indicated that the parties had advised him that they considered the clarifications to represent a statement on which the Panel could rely should it need to interpret its terms of reference. In that regard the Chairman also drew to the Committee's attention a communication from the EC (ADP/94).

4. On 25 October 1993 the Chairman informed the Committee (ADP/108) that the composition of the Panel was as follows:

Chairman: Mr. Magnus Lemmel

Members: Mr. Hugh McPhail

Mr. Rudolf Ramsauer

5. The Panel met with the parties to the dispute on 8-9 February and 10 May 1994.

6. On ......... the Report of the Panel was submitted to the parties.

7. In November 1988 the EC Commission received a complaint from the European Council of Chemical Manufacturers' Federation, on behalf of certain EC producers of audio cassettes, alleging that audio cassettes originating in Japan, Hong Kong and the Republic of Korea were dumped and were causing material injury to the complainants. In January 1989 the Commission initiated an anti-dumping enquiry into imports of audio cassettes originating in Japan, Hong Kong and the Republic of Korea.

8. In November 1990, by Commission Regulation 3262/90 (the "Provisional Regulation"), provisional duties were imposed on imports of audio cassettes from Japan, Hong Kong and the Republic of Korea. In May 1991, by Council Regulation 1251/91 (the "Definitive Regulation"), definitive anti-dumping duties were imposed on imports of audio cassettes from Japan and the Republic of Korea.

9. In the Definitive Regulation, in cases where there were no, or insufficient domestic sales, the EC constructed normal values for several Japanese exporters based on cost of production, plus an amount for selling, general and administrative expenses, and an amount for profit (recitals 12-16). The EC also constructed certain export prices, when sales were made to subsidiary companies, by deducting certain expenses and an amount for profit from the price paid by an independent purchaser (recital 17). The EC made certain adjustments for quantity, differences in characteristics, selling expenses and level of trade when comparing normal value and export prices (recital 17). The EC did not make identical deductions or adjustments to the normal values. In the Definitive Regulation, when calculating average dumping margins, the EC treated as undumped those export sales the prices of which exceeded the average or constructed normal values.

10. In the Definitive Regulation, the EC cumulatively analyzed the effect of imports from Korea and Japan (recitals 22-25). In determining the effect of the dumped goods, the EC relied on an increase in volume, price suppression and depression (recitals 26-36). The EC also found that price undercutting of the domestically produced goods occurred during the period under enquiry (recital 26). The EC found a causal link between the injury and the dumped goods, despite the presence of certain other circumstances which some interested parties had asserted could have caused the injury (recital 37-39). The dumping margins determined for Japanese imports ranged from 44.5 to 64.2 per cent (recital 20). In application of the lesser duty rule, definitive duties ranging from 15.2 to 25.5 per cent were imposed on imports of Japanese audio cassettes into the European Community (recital 49).

11. Japan requested the Panel to find that the EC's method of calculating the dumping margin for the Japanese exporters was inconsistent with Articles 2 and 8:3 of the Agreement due to:

(b) the "zeroing" of negative margins in the process of calculating an average dumping margin; and

(c) the failure of the EC to construct properly the normal values of certain models of audio cassettes exported by the largest Japanese exporter.

Japan also requested the Panel to find that the EC's determination that dumped Japanese exports of audio cassettes had caused injury to the EC industry did not comply with the requirements of Article 3:1 through 3:4 of the Agreement.

12. The EC requested that the Panel:

(a) rule that a series of claims put forward by Japan in its submission were inadmissible, and, consequently rule that they be rejected in limine litis (preliminary objections);

and

(b) to the extent that the Panel were not to accept the preliminary objections of the EC, find that the anti-dumping measures taken by the EC were not inconsistent with the provisions of the Agreement;

and therefore reject all the claims made by Japan.

13. Japan requested the Panel to recommend that the Definitive Regulation be revoked, that duties already paid be reimbursed and that the EC bring the relevant provisions of Council Regulation 2423/88 of 11 July 1988 on protection against dumped or subsidized imports from countries not members of the European Economic Community (hereinafter referred to as the "Basic Regulation") and its application into conformity with the Agreement.

14. The EC requested that the Panel find that its measures were in conformity with the Agreement. Were the Panel, however, to uphold any of Japan's arguments, the EC requested the Panel to only recommend that the EC bring its measures into conformity with the Agreement.

15. The EC argued that certain of the claims made by Japan were not properly before the Panel, either because procedural requirements contained in the dispute settlement provisions of the Agreement had not been complied with, or because of lack of a legal interest on the part of Japan, as even a ruling in its favour on certain points would not affect the level of the anti-dumping duties imposed on its imports (due to the application of the lesser duty rule) and consequently that no nullification and impairment had been suffered by Japan. According to the EC the above-mentioned claims should be rejected in limine litis by the Panel.

16. Japan argued that it had complied with the mandatory procedural requirements of the Agreement. In cases of non-compliance, Japan argued that the requirements were of a non-mandatory nature. As a result, Japan argued that all claims were properly before the Panel. Japan also argued that GATT law did not require that an action complained of have an actual impact, and alternatively argued that the Definitive Regulation did have an actual impact on Japan.

17. The EC argued that in GATT dispute settlement there was a distinction between a matter, a claim and an argument. A matter was a sum of claims advanced by a complaining party. A claim was a specific legal assertion (such as an allegation of violation of a provision of a GATT agreement) in relation to certain facts. An argument was legal or factual reasoning advanced in support or rebuttal of a particular claim.

18. The EC argued that Japan had failed to properly raise several of its claims in consultations and conciliation, and therefore that those claims should not be considered by the Panel. The EC also argued that a claim must be raised by the complaining party during consultations or conciliation in order to permit the parties to discuss the claim, and if possible, to reach a mutually satisfactory resolution of the dispute; in order to allow the Committee to 'filter' claims with no reasonable basis; and in order to clarify the facts and issues in dispute. If no solution was found on all or some of the claims, those unresolved claims could then be referred to a Panel. The EC argued that if claims were not raised at the consultation and conciliation stages, the settlement process would be frustrated, and the dispute would be inexorably forced towards arbitration.

19. In support of its argument, the EC noted that paragraph 6 of the 1979 Understanding regarding Notification, Consultation, Dispute Settlement and Surveillance (the "Dispute Settlement Understanding") provided that "... contracting parties should attempt to obtain satisfactory adjustment of the matter in accordance with the provisions of Article XXIII:1 before resorting to Article XXIII:2". This provision, and panel practice, had been designed to ensure that all efforts were taken to reach a mutually acceptable solution to a dispute, prior to resorting to panel proceedings. The Panel on Uruguayan Recourse to Article XXIII (BISD 13S/47, at paragraph 11) had concluded that:

20. Further, Article 15:4 of the Agreement also required parties to "make their best efforts to reach a mutually satisfactory conclusion throughout the period of conciliation". GATT practice therefore required priority be given to the solution of disputes through consultations and conciliation, in order to promote the reaching of "a mutually acceptable solution". The priority of mutually acceptable solutions was made clear by paragraph 18 of the Dispute Settlement Understanding, which left open the possibility to settle matters by mutual agreement at any stage of proceedings.

21. The EC argued that Article 15:5 of the Agreement required that the claims constituting the matter in dispute should be the subject of a "detailed examination" in the Committee during the conciliation stage. The EC argued that footnote 15 to Article 15:3 also supported that argument. The EC relied on the Panel on United States - Imposition of Anti-Dumping Duties on Imports of Fresh and Chilled Atlantic Salmon from Norway, adopted 26 April 1994, ADP/87, paragraphs 335 and 337 (the "Salmon" Panel) as authority for the proposition that unless a matter, and its constituent claims, were referred to the Committee for conciliation a panel could not examine those claims.

22. The EC also argued that the matter before a panel was defined by the terms of reference. In this case the terms of reference of the Panel were to be found in documents ADP/85 and Add.1, it being understood between the parties that the clarifications contained in the "explanatory note" to ADP/85/Add.1 would be used by the Panel to interpret the other documents before it.

23. If claims were either not within the terms of reference, or were not mentioned in consultations and conciliation, the defending party would not have sufficient notice of the case against it, and the claim could not be considered by the Panel. The EC said that the Salmon Panel supported this argument.

24. Japan argued that neither Article 15, in particular Articles 15:3, 15:5, and 15:7 of the Agreement, nor Article XXIII:2 of the GATT, contained express provisions detailing the consequences of a failure to comply with a disclosure requirement. Article 15:5 was the only provision that contained a disclosure requirement. The objectives of any explicit or implicit disclosure requirement were to enable the Committee to conciliate, to permit the Committee to examine whether there were reasonable bases supporting the allegations made, to inform the respondent of the case against it, and to alert other members of the Committee of the dispute in order that they could decide whether they wished to be heard by the panel.

25. Japan argued that in GATT law and practice the Committee conciliation meeting constituted only the commencement phase of any conciliation process. The next phase of the process would have involved the appointment of an impartial conciliator who (in line with GATT and international practice) would thoroughly examine the facts and arguments. However, because of the unwillingness of disputants to be reconciled, the process had never moved beyond the commencement phase (either in this or any other dispute under the Agreement). Japan also argued that in any case, the information in ADP/79 was sufficient to complete this commencement phase.

26. Even on the assumption that the Committee meeting constituted the entire conciliation process, Japan contended that it had made adequate disclosure to the Committee. Japan argued that Article 15:3 referred to cases and not to claims or complaints. Japan agreed that footnote 15 to Article 15:3 required that the Committee be informed of the substance of the dispute, but argued that Article 15:3 did not require that the Committee have an opportunity to express its view on every individual aspect of the complaint. The words "... a solution ..." in Article 15:3 made clear that the Committee's role was to seek a complete settlement of the case. The reason for the requirement to present detailed information was so that the Committee could conciliate. In this case it was not possible to conciliate between the parties. Consequently, a detailed point by point examination of the case was unnecessary, as once it was clear that the parties had elevated the dispute to the step of conciliation, and then could not be reconciled on even one claim, the Committee's conciliation role was exhausted.

27. As regards the second objective, the Committee's role was only to establish that at least some of the claims had a reasonable basis. This was supported by the use of the word "... case ..." in the footnote to Article 15:3 (which meant the sum of the party's case), and because if any reasonable basis existed the Committee could not intervene in the dispute. The ordinary meaning of the language, and the practice of the Committee, made clear that the disclosure requirement was satisfied if the complaining party established that there were reasonable bases for at least some of the claims. Japan argued that as in this case Japan had clearly presented information which indicated that there were reasonable bases to most of its claims, the omission of certain elements should not be considered to be important.

28. The right to a fair hearing, which required that the respondent be informed of the case against it, was supported by Japan. Japan did not support a legalistic approach which required that every issue be raised at every stage of the proceedings. Since the purpose of the requirement was to avoid prejudice to the respondent party, in deciding whether to exclude arguments the Panel should examine the degree of prejudice actually suffered if a particular argument was not raised at every stage of the proceedings. Japan argued that the right to a fair hearing was satisfied by providing sufficient time for the respondent to consider the first written submission of the complainant. Japan recalled that although the EC's claim that certain of Japan's claims had not been properly raised was first raised in its initial submission to the Panel, Japan had not asked the Panel to exclude that claim by the EC.

29. Japan argued that if the EC's arguments were accepted the dispute settlement process would be undermined. A tendency towards formalism and rules of procedure would result in the overwhelming of the Committee with documents. Less developed country Parties to the Agreement who lacked the resources to prepare comprehensive catalogues of their dispute for the Committee, and the Committee itself, would suffer as a result.

30. Japan argued that in the consultation and conciliation discussions and documents it had complained that the EC had not complied with the requirements of the Agreement in its findings of dumping, injury and causal link, and that as the EC had been less than forthcoming in providing certain detailed information during the consultation and conciliation phases, at least part of the lack of disclosure was the fault of the EC.

31. Japan noted that the report of the Salmon Panel was controversial, in particular those parts of it relied on by the EC dealing with Article 15. Japan tendered to the Panel the statement of the Chairman of the Committee made on 27 April 1994 concerning the adoption of that report.

32. Japan argued that in this case the terms of reference of the Panel were defined by the documents ADP/85 and Add.1, the discussions during consultations under Article 15.2, the minutes of the conciliation proceedings under Article 15.3, and the letter of 31 March 1993 to the Chairman of the Committee enclosing a paper clarifying the terms of reference of the Panel.

33. Japan in conclusion argued that any claim in relation to the findings of the EC concerning dumping, injury and causal link were properly before the Panel. In addition, and in the alternative, Japan argued that all its claims were within the terms of reference of the Panel or had been properly raised during the consultation and conciliation processes.

34. The EC argued that the following claims either were not properly raised in consultation and conciliation, or were not covered by the terms of reference: the method of calculation of the constructed normal value, Japan's complaint concerning the EC system for refunds, the methodology used to establish price undercutting, the effect of price increases, and the claim that in determining that injury was caused by Japanese imports the EC had failed to take account of certain factors.

35. Japan provided answers to these points on the premise that the EC was correct regarding the requirements of the disclosure at the conciliation stage. However, Japan argued that its previous arguments (paragraphs 24-32) had shown the premise to be incorrect.

36. The EC argued that this claim was first mentioned in Japan's reference document to the Panel (ADP/85/Add.1) and was not covered by paragraph 19 of Japan's request for conciliation (ADP/79). The EC argued that paragraph 19 of Japan's request for conciliation (ADP/79) dealt only with averaging and zeroing. The EC argued that therefore the Panel could not consider Japan's claim that the amount of costs and profit used in calculation of the constructed normal value breached the requirements of Article 2:4.

37. Japan argued that it had raised this claim in paragraph 19 of document ADP/79. Japan also argued that it was not until the stage of the conciliation meeting that Japan developed concerns about the method of calculation of the constructed normal value because of the comments made by the EC at that meeting. Japan noted that subsequent to the establishment of the Panel, additional consultations were held, resulting in a letter of 31 March 1993 to the Chairman of the Committee. A paper attached to that letter which clarified certain matters dealing with the terms of reference of the Panel specifically included this claim.

38. The EC argued that Japan's allegations concerning EC rules on refunds of anti-dumping duties had not been considered in consultations and conciliation, and were not mentioned in the documents referred to in the terms of reference and, therefore could not be considered by the Panel. The EC later questioned the legal character of the allegations. If it was a claim, the EC objected firstly that no claim concerning the refund process had been previously made, and secondly, that it was speculative as it presupposed (a) a refund application would be made, and (b) that such an application would fulfil the conditions set by the EC authorities for grant of a refund.

39. If however the allegations concerning EC rules on refunds of anti-dumping duties were an argument raised in reply to the EC's objection on the basis of lack of a legal interest to Japan's claims, the EC argued that those allegations by Japan could be characterised as arguments. The EC argued that the sole purpose for which Japan raised this argument was to establish a legal interest. The EC subsequently asked the Panel to treat Japan's allegations as an argument.

40. Japan argued that its comments concerning the refund system were not a claim. Japan argued that it had only raised the subject of the refund mechanism to illustrate that "asymmetry" and "zeroing" had an effect on exporters even when duty was collected by reference to an injury margin.

41. The EC contended that Japan's argument concerning the methodology of selecting models for the analysis of price undercutting was a claim which was not within the terms of reference of the Panel. The EC argued that this claim was not raised in a sufficiently specific manner during conciliation. The EC argued that Japan's later arguments concerning the selection of models of cassettes and averaging of undercut prices of cassettes were separate claims which were not raised in paragraph 21 of document ADP/85/Add.1, and accordingly were not referred to in the documents establishing the Panel's terms of reference.

42. Japan argued that paragraph 24 of ADP/79 and paragraph 21 of document ADP/85/Add.1 had sufficiently alerted the Committee to this claim, and had brought the claim within the Panel's terms of reference.

43. The EC argued that Japan's claim that if the Japanese producers had eliminated any dumping margin EC producers still would not have been able to compete, and would not have been able to raise their prices, had not been properly raised during conciliation or in the request for establishment of a panel in document ADP/85/Add.1. If, however, it was an argument on causation the EC had no procedural objection in relation to it.

44. Japan argued that paragraph 26 of document ADP/79 clearly raised this claim, and alternatively argued that it was one of several arguments interpreting the notion of causation.

45. The EC argued that Japan's claim that the EC failed to take account of certain factors was a claim which had not been raised in conciliation. The EC argued that as it could not be covered by a general reference to causation under Article 3:4 of the Agreement, it could not be said to be included in paragraph 26 of ADP/79. Therefore this claim should not be considered by the Panel.

46. Japan relied on paragraph 26 of document ADP/79. Japan argued that when it challenged the finding "... that dumped imports [were] 'through the effects of dumping causing injury' ..." it was implicit that it was also alleging that the injury being suffered was caused by other factors.

47. The EC argued that to bring a complaint before a Panel, Japan must demonstrate that its benefits under the GATT had been nullified or impaired. The EC acknowledged that when a contracting party established an infringement of a provision of the GATT or of a GATT agreement, paragraph 5 of the Dispute Settlement Understanding provided that the action complained of was considered prima facie to give rise to nullification or impairment. The EC argued that in practice, however, the respondent party could rebut the presumption by showing that no trade effects resulted from the action. The EC argued that in such a case, the complaining party must demonstrate that the action had at least a potential impact on the complaining party. The EC argued that the requests by Japan that the Definitive Determination be revoked and that the EC bring its Basic Regulation and practice into conformity with the Agreement should be ruled inadmissible.

48. Japan argued that, in practice, the presumption of nullification and impairment was irrebuttable. If the Panel was to conclude that the presumption was rebuttable, Japan asked the Panel to find that the EC had failed to rebut the presumption. If the Panel were to find that it had been rebutted, Japan asked the Panel to find that Japan had suffered trade effects.

49. The EC argued that the respondent must be permitted an opportunity to rebut the presumption of nullification or impairment by presenting evidence of no actual nullification or impairment of benefits accruing to the complainant. The EC argued that the doctrine of nullification or impairment of benefits, therefore, had consequences for the legal admissibility of a claim. Once a respondent party provided prima facie evidence that no benefit of the complaining party could be affected by the measures complained of, or that no trade effects resulted from the measures, the complaining party had no legal interest in the case, and the claimant should be determined to have no case unless the complaining party could show that the action complained of had a potential impact.

50. The EC argued that a finding that a provision of Article 2 had been breached resulting in an error in calculation of the dumping margin could have no impact on the amount of duty collected in this case. The EC argued that the methodologies attacked by Japan would only affect one Japanese exporter. In this context, the EC noted that in the case of that one Japanese exporter the duty collected was 49 per cent lower than the margin of dumping due to the application of an injury margin. Therefore the methodologies used had no affect on the duties actually imposed. In such a case a complainant had no legal interest in obtaining a ruling that a dumping margin was incorrectly calculated, as, even if upheld, the claims were incapable of affecting the dumping margin to such an extent that the duty finally imposed ought to have been below what was calculated to be the injury margin.

51. The EC argued that the Report of the Panel in United States - Taxes on Petroleum and Certain Imported Substances, adopted 17 June 1987, BISD 34S/36 (the "Superfund" Panel), and in particular section 5.1 of that Report, did not apply to the current case. The EC noted that section 5.1 of the Superfund Panel Report concluded that an infringement of Article III:2 of the GATT could arise whether or not adverse trade effects occurred. The EC argued that the Agreement, and the Agreement on Interpretation and Application of Articles VI, XVI and XXIII of the GATT (hereinafter referred to as the "Agreement on Subsidies and Countervailing Measures") were distinguishable from Article III:2 of the GATT, because Article III:2 established a prohibition of discriminatory taxation, whereas the Agreement provided primarily procedural guidance to investigating authorities and gave them discretion as to how to conduct investigations. The EC argued that Article 2 of the Agreement set out a number of procedural prescriptions relating to the modalities of determination that dumping was occurring. Article 2 provided discretion to administrative authorities. Article 8:1 of the Agreement requested parties to set a duty at a lower level than the dumping margin. The EC argued that in combination these provisions contemplated a situation where a duty substantially lower than the dumping margin of the exports would be charged.

52. The EC also argued that footnote 26 of the Agreement on Subsidies and Countervailing Measures confirmed that the provisions regarding unfair trade legislation pursuant to the Agreement and the Agreement on Subsidies and Countervailing Measures was intended to be construed differently than GATT provisions, and that a Party was permitted to rebut presumed nullification and impairment in relation to trade effects.

53. The EC argued that it had rebutted the presumption of trade effects by showing that the presence of an injury margin meant that even if Japan's claims were upheld the amount of duty collected may not vary. Consequently, the claims by Japan should be ruled inadmissible by the Panel, on the basis of absence of legal interest.

54. Japan argued that Article 1 of the Agreement required that any anti-dumping measure conform with the Agreement. The reason for this provision was that Article VI, and therefore the Agreement, were exceptions to the central MFN principle at the core of GATT obligations. If the conditions laid down in Article VI and the Agreement were not strictly complied with, an anti-dumping duty could not be imposed. Japan argued that the Panel on Korea - Anti-Dumping Duties on Imports of Polyacetal Resins from the United States, adopted 26 April 1993, ADP/92 (the "Polyacetal Resins" Panel) in the field of anti-dumping duties, and the Panels on Canadian Countervailing Duties on Grain Corn from the United States, adopted 26 March 1992, SCM/140 (the "Grain Corn" Panel) and on United States - Measures affecting Imports of Softwood Lumber from Canada, adopted 27 October 1993, SCM/162 (the "Softwood Lumber" Panel) in the field of countervailing duties, at least implicitly supported this argument.

55. The Superfund Panel made clear that a measure in breach of the General Agreement was presumed to cause nullification and impairment, and that in practice the presumption was irrebuttable. Japan argued that the trade effects of a measure were not relevant to a determination whether nullification and impairment had occurred. Japan argued that it had a legal interest to have the Agreement observed, and that the "benefit" nullified was Japan's right that Parties to the Agreement observe the Agreement's requirements before levying anti-dumping duties against exporters. The Superfund Panel had been concerned with non-mandatory legislation that had permitted the investigating authorities to act inconsistently with the GATT. Japan noted its later elaboration on non-mandatory legislation contained in section VII of this Report.

56. Japan argued that the conclusion of the Superfund Panel (BISD 34S/36) that in practice the presumption of nullification and impairment arising out of a violation of the GATT was irrebuttable should be applied to the Agreement by virtue of the phrase mutatis mutandis in Article 15:7 of the Agreement. The Superfund Panel had concluded that Article III:2 protected expectations of a competitive relationship between imported and domestic goods, breach of which must be regarded as a nullification or impairment of rights. Any mandatory legislation could be challenged as it would destroy those expectations, whereas non-mandatory legislation would not necessarily have a trade effect. When the decision of the Superfund Panel was applied, the result was that any breach of the Agreement, even in the absence of trade effects, would nullify and impair benefits, due to the statement in Article 1 of the Agreement. Japan argued that the EC's attempt to distinguish Article III of the GATT from the Agreement was unpersuasive. Japan also argued that footnote 26 of the Agreement on Subsidies and Countervailing Measures was not relevant to the Agreement, nor to other aspects of the Agreement on Subsidies and Countervailing Measures.

57. Japan also argued that the requirements of Article 2 set the framework for the prohibition of Article 8:3 with the effect that nullification and impairment could be established in the absence of a breach of Article 8:3. Japan argued that a Party did not have discretion about whether to comply with any of the conditions laid down by the Agreement for imposition of an anti-dumping duty. Japan noted that Article 2 of the Agreement used the word "shall" which was indicative of a mandatory obligation, which was distinct from the non-mandatory rules contained in Article 8:1. Therefore, Japan argued, the obligations deriving from Article 2 prevailed over any use of the lesser duty rule.

58. Japan also argued that the Panel on United States - Imposition of Anti-Dumping Duties on Imports of Seamless Stainless Steel Hollow Products from Sweden, ADP/147, paragraph 5.18 (the "Swedish Steel" Panel) also supported an argument that the presence of trade effects was not necessary before a Party could complain that the conditions under the Agreement for imposition of a duty had not been met.

59. Japan alternatively argued that if trade effects were necessary, it had sustained trade effects. This was because the operation of the refund system would deny an importer a refund unless the importer could show that the actual dumping margin was less than the duty collected (using the same methodology as in the original decision). The presence of an injury margin therefore did not alleviate the effects on the importer. Japan also argued that the true margin was below the duty level which also caused trade effects.

60. The EC argued that Japan's arguments concerning the operation of the EC's refund mechanism were incorrect, in so far as Japan's arguments relied on the assertion that in processing a refund application inevitably the same methodology was used. The EC argued that if for instance, at the time of the refund application there was no reason to rely on constructed values, actual values would be used, and cited Commission Decision 92/332/EEC of 3 June 1992 on refunds of duties imposed on ball bearings from Singapore as an example of a situation in which the EC would use a methodology in examination of a refund application that was different to the methodology used in the course of the investigation.

61. The EC argued that Japan's complaints on (a) so called "asymmetry" and (b) so called "zeroing" were inherently incapable of affecting the dumping duty collected. This was because although an allegedly "asymmetrical" approach and "zeroing" were used in calculation of a dumping margin, the actual duties applied were at a lesser duty rate, i.e. at a rate sufficient to remove injury to the domestic industry. Accordingly, even if an "asymmetrical" approach and "zeroing" were not applied, the amount of duty collected would not be affected.

62. Japan argued that because the true dumping margin was less than the amount of duty actually collected, due to the use of an "asymmetrical" comparison and "zeroing", those practices had a direct impact on its exporters. Japan calculated what it considered to be the "true" dumping margin. Japan also argued that there was a direct impact on exporters because the EC had relied on the existence of the resulting high dumping margins in its injury analysis, and because the EC's refund system operated so as to continue to apply an "asymmetrical" comparison and "zeroing" when processing a refund application. In addition, Japan noted that it was challenging the rules on "asymmetry" and "zeroing" being applied by the EC, not simply the specific decision in the audio cassettes case.

63. The EC argued that deduction of certain expenses in construction of an export price, without deduction of similar expenses in construction of the normal value could only result in nullification and impairment if the differential treatment had resulted in higher dumping duties than if equal deductions had been made. Only in the case of one Japanese exporter could equal deductions have resulted in a lower dumping margin, which in any case would have been offset by application of the lesser duty rule. The EC concluded that therefore, Japan had no "legal interest" in a Panel ruling on this point.

64. Japan argued that the application of an asymmetrical methodology produced an inflated dumping margin, which was relied on by the EC in recital 39 of the Definitive Regulation, thereby resulting in an effect on Japanese exporters. Japan argued that the application of this methodology to the refund system continued to affect Japanese exports to the EC, because when an exporter applied for a refund, the exporter had to demonstrate that the amount of duty collected was greater than the dumping margin (which in this case had been established by use of an "asymmetrical" methodology).

65. Japan noted that the EC had initiated the Panel on United States - Definition of industry concerning wine and grape products, adopted 28 April 1992, SCM/71, (the "Wine and Grapes" Panel) because the EC had wanted to contest a rule that would continue to affect other exports to the US market. Japan argued that the purpose behind its challenge in this case was to challenge a consistently applied practice, not simply the terms of the specific decision. Japan also noted, in reply to the EC's argument that there was a difference of 49 per cent between the dumping margin and the anti-dumping duty collected for one Japanese exporter, that Japan had recalculated the dumping margin (in Annex V to its first submission to the Panel) and that on Japan's calculation the actual dumping margin was less than the duty collected. Accordingly, Japan argued that the EC's use of an "asymmetrical" approach had trade effects on its exporters.

66. The EC argued that errors in Annex V to Japan's first submission distorted that Annex to such an extent that Japan's arguments based on Annex V could not be relied on. Specifically, the EC argued that the constructed export price used by Japan was unreliable because Japan had not deducted duties and profits from the sale price in the EC. The EC argued that was contrary to Article 2:6 and distorted Japan's claim concerning the export price. Japan had also only deducted direct expenses, with the result that the price charged to the first independent purchaser in the EC contradicted Article 2:5.

67. The EC also argued that the constructed normal value used in Annex V was flawed for several reasons. The method of construction was mathematically unsound and would understate the normal value. The figures for selling, general and administrative expenses ("SG&A"), profit and research and development were token percentages of gross sales price and were then erroneously applied to an approximate net price (excluding commodity tax and rebates). In addition a profit figure which was based on all sales of TDK Japan, and included indirect exports, sales to related customers and sales of obsolescent models was used.

68. The EC argued that "zeroing" had no impact on the amount of anti-dumping duty collected on Japanese imports and no significant effect on the dumping margin, because the lesser duty rule resulted in the collection of a level of duty below the determined dumping margin, and that therefore its practice of averaging individual dumping margins did not result in nullification and impairment. Accordingly, the EC concluded that Japan could not demonstrate a "legal interest" in a Panel ruling on this point.

69. In relation to the effect that "zeroing" had on Japanese exporters, Japan repeated its arguments concerning "asymmetry". Japan therefore argued that "zeroing" had a trade effect.

70. Japan argued that Articles I, II, III and XI of the GATT contained the fundamental obligations of the GATT. Article VI was a derogation from those fundamental obligations. Accordingly, contracting parties applying Article VI bore the burden of establishing that actions taken under Article VI complied with its requirements. Because Article VI had been viewed as an exception to the GATT, it had been restrictively interpreted by Panels. Japan argued that therefore Article VI and the Agreement should be viewed by the Panel as exceptions to the GATT and should be construed restrictively, with the Panel preferring a restrictive interpretation over a broader construction. Japan argued that the legal status of Article VI as an exception was carried over into the Agreement, through Articles 1 and 16:1 of the Agreement. Japan noted that the need to read exceptions to the GATT restrictively did not mean that a state was incompetent to act unless it had acquired capacity through signature of a treaty. Japan noted that international law limited the sovereignty of a state only as far as was expressly provided in an international agreement.

71. The EC first noted that the terms of reference of the Panel only made reference to the relevant provisions of the Agreement, and did not refer to Article VI. Japan appeared to confuse the theory that Article VI was an exception to the GATT with the consequential argument that anything not specifically authorized by Article VI was therefore prohibited. The EC did not accept that anything that was not specifically authorized by Article VI was prohibited. The EC argued that Japan in effect thereby sought to say that that which a state was not authorized to do was necessarily prohibited. The EC did not agree that international law required a specific attribution of competency. A state's sovereign power to act was unrestricted unless limited by rules of international customary or treaty law. The EC's view was that the Agreement established a limitation on and not an attribution of competence. The EC argued that the fundamental starting point in construction of the Agreement was that it was not drafted to cover comprehensively all aspects of an anti-dumping investigation.

72. The EC accepted that a Party relying on the Agreement had the burden of showing that it had complied with the provisions of the Agreement. The EC argued however that this burden could be discharged by taking an anti-dumping measure which was based on all relevant facts and was properly reasoned. Thereafter, the burden shifted to the complaining party to show that the respondent party had not properly discharged its burden on such bases as not having considered all relevant facts, not having properly explained the basis for the decision or having violated the Agreement. The EC argued that if it had not properly reasoned its decision, the exporters would have challenged the decision of the Commission in the European Court of Justice, and relied on the lack of a challenge as proof of the conformity of the decision with the Agreement.

73. The EC was concerned lest Japan's theory of the restrictive interpretation of the Agreement lead to the conclusion that certain aspects of anti-dumping investigations which are not regulated by the Agreement are prohibited. This conclusion was not supported by general international law. Moreover, Article II:2(b) of GATT did not contain any indication that Article VI or the Agreement had to be interpreted in a restrictive manner. The EC was of the view that Parties are bound by the language of the Agreement, but if it did not provide for a particular obligation and left room for the discretion of administering authorities, Parties are subject to basic panel control for manifest error of fact or of the interpretation of facts or for arbitrariness.

74. Furthermore the EC was of the view that Japan's complaint went beyond the nature of panels' tasks in reviewing findings made by Parties to the Agreement. In this connection the EC argued that the Panel on Brazil - Imposition of provisional and definitive countervailing duties on milk powder and certain types of milk from the European Economic Community, adopted 28 October 1994, SCM/179 and Corr.1 ("Brazilian milk powder" Panel) and the decision of the Polyacetal Resins Panel made clear that the task of a panel was only to review the determination for consistency with the Agreement. The EC argued that in its decision the investigating authority need not set out all the evidence it had relied upon. To meet the Agreement's requirement of "positive evidence", it was sufficient to reference and evaluate the evidence relied upon in the determination.

75. In reply to the EC's argument that the burden of proof could be discharged by taking an anti-dumping measure which was based on all relevant facts and was properly reasoned, Japan argued that a respondent party had to positively establish that the measures taken were consistent with the Agreement to discharge the burden of proof.

76. Japan also argued that because the EC Court of Justice had upheld previously the "asymmetry" and "zeroing" rules, it would have been futile to challenge the decision there. Accordingly the EC could not rely on the exporters' failure to challenge the decision in that Court as evidence of the propriety of the decision.

77. Japan further argued that the issue of restrictively interpreting the provisions of the Agreement was a quite different issue from the issue whether an authority was determined to have acted arbitrarily. The latter issue was concerned with the question whether or not it was proper for panels to conduct de novo reviews of decisions.

78. Japan claimed that the EC had breached the requirements of the Agreement in its calculation of the dumping margin. Japan particularised its claim as follows. First, the EC had not made a fair comparison because it had used an "asymmetrical" approach when comparing the normal values and export prices. Second, the EC had not made a fair comparison when it had "zeroed" the margins of certain sales.

79. The EC argued that it had met the requirements of the Agreement in relation to its comparison of normal values and export prices.

80. Japan argued that in constructing an export price the EC had deducted costs and profits from the price charged to the first arms length purchaser and had not deducted "identical elements" in determining the normal value. Japan argued that in failing to make an adjustment for the corresponding costs and profits in the normal value the EC had conducted an "asymmetrical comparison" and had violated a "fairness" requirement derived from Article 2:1 and 2:6.

81. The EC agreed that it had not made the same deductions to normal values and export prices. This was because normal values and export prices were determined by different rules. The EC also argued that the concept of "fair comparison" was determined by the three elements mentioned in the text of Article 2:6: comparison at the same level of trade, comparison on the basis of sales made at as nearly as possible at the same time, and appropriate due allowances.

82. Japan argued that the mere determination of an appropriate "level of trade" and the making of appropriate "due allowances" did not guarantee a "fair comparison". Japan noted that its complaint in relation to the infringement of Article 2:6 was not a complaint about the use of constructed export prices, provided any such constructed prices were properly compared with the normal value.

83. Japan further argued that when an importer was related to an exporter, the EC would regard the transfer price as unreliable under Article 2:5 of the Agreement. The EC would then construct an export price by making deductions from the price paid by an independent purchaser to arrive at a price as close as possible to the ex-factory level. Japan alleged that the EC deducted direct and indirect selling costs and the profit of related importers in constructing an export price, in order to arrive at a price equivalent to the price for export sales to an independent buyer and only deducted direct selling costs from the normal value, leading to a higher normal value, and ultimately to a higher dumping margin.

84. Japan additionally argued that in some cases if sales on the domestic market were also considered to be unreliable because parties were related, the normal value would also be constructed. Japan argued that to construct a normal value, direct selling costs were deducted from the prices at which the goods were sold to independent buyers. Only those direct costs related to distribution and marketing were deducted.

85. The EC argued that its investigations showed that most of the goods from Japan were sold within the EC by sales subsidiaries of the exporters. The EC argued that accordingly the price at which the imported goods were first resold to an independent buyer had been adjusted by the EC for all expenses (direct and indirect) incurred between import and resale, in order to represent a price comparable with an independent importer (cost, insurance, freight at the EC border).

86. The EC noted that domestic distribution was ordinarily through either a sales division of the producer, or through a sales company economically controlled by the producer. In response to a question by the Panel, the EC argued that in the case of sales made by one economic entity the only way to correctly establish an ex-factory normal value was to take into consideration the first sale to an independent customer. That price was then compared with the export price at the same level of trade, with due allowances being made. In making due allowances, SG&A expenses of the domestic sales departments were not deducted from the normal value, nor were SG&A expenses of the export sales department deducted from the export prices.

87. Japan argued that Articles 2:1 and 2:6 acting together generated a requirement that a comparison be "fair". Japan argued that the definitions of the words "fair comparison" and "fairness" could be ascertained from general English language definitions, and from the use of the term in other GATT agreements. In addition, Japan argued that certain Panel decisions (Italian Discrimination Against Imported Machinery, BISD 7S/60, paragraph 24; Canada - Administration of The Foreign Investment Review Act, BISD 30S/140, para. 6.3; and Panel on Vitamins, BISD 29S/110, para. 22(g)) had interpreted the word "fair" to imply fair and balanced treatment, thereby requiring comparisons that were "fair and equitable". Japan argued therefore that "like should be compared with like". Japan argued that the use of the word "equitable" in the French version of Article 2:6 indicated a similar intention.

88. Japan argued that the EC's argument limiting the scope of the requirement of a "fair comparison" was based on splitting Article 2:6 into two separate subjects. Japan argued that was not warranted. Japan argued that the subjects of the two sentences were the same. The final clause of the second sentence explicitly required that allowances should be made for the "other differences affecting price comparability". If an allowance could not be made to make a comparison fair, a comparison should not be made. Japan argued that the requirements of a "fair comparison" were broad, and alternatively argued that in this case the normal value and export prices were not at the same "level of trade". Japan argued that the phrase in the last part of the second sentence of Article 2:6, "differences affecting price comparability", was intended to be the fundamental criterion for determining allowances and adjustments, and required that the guiding principle for all questions concerning price comparability was "fairness". Japan noted that although no ruling was made on the point, the Salmon Panel had stated (paragraph 481) that it "considered that it was possible to interpret the first sentence [of Article 2:6] to reflect a requirement of a 'fair comparison' which applied generally to any aspect of the comparison of normal values and export prices".

89. Japan also argued that the working papers on the Kennedy Round negotiations and the two reports of the Expert Groups on Anti-Dumping and Countervailing Duties (BISD 8S/145, paragraph 6 and BISD 9S/194, paragraph 13) demonstrated an intention on the part of the negotiators that the phrase "fair comparison" contained in Article 2 of the Agreement be equivalent to the phrase "genuinely comparable" used in certain drafts of the Kennedy Round Anti-Dumping Agreement. Japan inferred that intention on the basis that as the documents mentioned revealed no difference between the Parties over the term, it was reasonable to conclude that the drafters wanted to incorporate the factors usually associated with the notion of fairness.

90. Japan observed, as an example of the unfairness of the EC's practice, that the United States had determined that an "asymmetrical" comparison could not be allowed in its anti-dumping enquiries on the basis that such a comparison was unfair.

91. Japan considered that a "fair comparison" requirement did not only relate to the final stages of calculation of a margin, but to all the stages along the way to it. Japan argued that fairness required that the deductions made to the export price required adjustments for "identical elements" be made to the normal value. Because adjustments necessary to ensure a fair comparison had not been made the EC had infringed Article 2:6.

92. The EC stated that Japan's arguments suggesting a broad definition of "fair comparison" were without merit. The EC argued that Japan had tried to define the word "fair" by reference to other provisions of the GATT which were unrelated to anti-dumping. The panel decisions relied upon by Japan had defined the word "fair" in the context of Article III of the GATT. The references by Japan had no relevance to Article 2:6 of the Agreement, in terms of the principles of interpretation of the context of a treaty contained in Article 31 of the Vienna Convention on the Law of Treaties ("the Vienna Convention"). Japan's conclusion that a "fair comparison" was required was not logically argued.

93. In the context of Article 31 of the Vienna Convention, the EC argued that there was no need to refer to supplementary materials such as travaux préparatoires, as the words and the context of Article 2 were clear, and that therefore no "special meaning" had been intended by the contracting parties. The EC argued that the replacement of the words "genuinely comparable" with "fair comparison" during the last stages of the negotiations of the Kennedy Round's 1967 Anti-Dumping Agreement could not be relied on to demonstrate an intention among the contracting parties that the phrases be equivalent in meaning.

94. The EC argued that the final sentence of Article 2:6 did not add any general requirement of fairness to the textual and contextual interpretation of "fair comparison". The EC also argued that Article 2:1 added nothing to the scope of the requirement contained in Article 2:6. When Japan had argued that "like should be compared with like" to support its conclusion that like deductions or additions should be made to normal values and export prices, it had made a petitio principii. Japan's argument that "like should be compared with like" was not supported by the rules of interpretation of the Vienna Convention, nor by the structure of Article 2:6. The context of the term in the Agreement was quite different to the passages from the panel reports quoted by Japan.

95. The EC also argued that the Salmon Panel had not found it necessary to rule on whether Article 2:6 required a fair comparison which applied generally to any aspect of the comparison of normal values and export prices. The EC argued that this also was adverse to Japan's case.

96. The EC argued that ascertainment of the normal value and export price were governed by different rules (respectively by Article 2:1 to 2:4 and Article 2:5) and reflected different objectives -one was designed to reflect a price for goods destined for consumption in the home market, and the other for goods destined for consumption outside the home market. Both produced a figure that reflected a sale to an independent buyer. There was no requirement that the establishment of normal values and export prices be "symmetrical". A constructed export price was resorted to when the available export sales were to a related company. SG&A expenses need not necessarily be treated in the same way in construction of normal values and export prices. The two methods for determination of normal values and export prices could not be compared as they had different legal bases and were used for different purposes.

97. The EC argued that Article 2:6 only dealt with subsequent comparison of the prices derived pursuant to Article 2:5. Any allowances for costs incurred between importation and resale and for profits accruing, would be made during construction of a price under Article 2:5.

98. The EC argued that the Agreement was concerned with price and not cost comparability, and therefore there was no requirement that adjustments be made to ensure that normal values and export prices were determined by reference to similar levels of costs. Market conditions could not be assumed to be the same in each Party, and therefore it could not be assumed that differences in costs resulted from a difference in the level of trade.

99. The EC argued that following the deduction of all direct and indirect expenses incurred between importation and resale the resulting figure used as an export price was a price equivalent to that charged to an independent buyer at a cost insurance freight EC frontier level. The EC argued that this was the result intended by Article 2:5. If allowance was made for the costs of a related importer, the result could discriminate against unrelated importers, which would breach Article 2:5 as a whole and the first condition of Article 2:6.

100. Japan noted that the unfairness of the EC's approach was reflected in its paradoxical consequences: the larger an exporter's investment in the EC the larger would be the dumping margin. Japan denied that its approach would discriminate against unrelated importers. Such importers had to be persuaded to buy, so sales to them were not directly equivalent to sales to related importers.

101. The EC responded that the assertion of Japan was irrelevant to the approach of anti-dumping law and to its review by the Panel. Decisions related to the vertical integration of companies were matters of commercial strategy, not related to dumping law, which concerned trade in goods, not investment. Investments to improve sales were not normally for the purpose of increasing profit on a rate of return basis (costs would not automatically increase as a proportion of turnover), nor were they made to increase costs.

102. The EC normally allocated SG&A (direct and indirect costs plus costs of importation) and profit of a related importer on a turnover basis. In the normal course of events increased investments in sales operations would lead to increased turnover. Therefore, the cost allocated on a turnover basis could remain the same or may even decrease pro rata.

103. Japan noted that the identification of the same level of trade did not of itself guarantee a fair comparison. The different steps were merely devices by which a fair comparison could be achieved.

104. Japan noted that the term "level of trade" could have three different meanings. The phrase level of trade used in one sense could indicate the nature of the purchaser. An exporter had sought, but had been denied, a "level of trade" adjustment on the basis of differences in the sales made to exclusive distributors and to retailers. Such differences, when they existed, concerned indirect expenses or profits related to sale. Another possible "level of trade" concerned the various costs involved in the terms on which goods were offered for sale. Differences in such terms of trade could ordinarily be eliminated by comparisons at an ex-factory level. A third definition of the term indicated the various stages at which a product moved from manufacturer, via wholesaler to retailer. In contrast to the first meaning, it focused on the nature of the seller rather than that of the purchaser.

105. Japan claimed that the EC in its arguments had used the term "level of trade" in this third sense. Japan claimed that the EC had chosen the level of transfer from the parent company to the sales subsidiary in the EC on the export side, and the level of the sales to the independent purchaser, whether made by a sales department or a sales subsidiary on the domestic side. Japan argued the absence of a sales function in the transfer by the parent company created an "asymmetry" which prevented sales on the export side from being regarded as at the same level of trade as those on the domestic side. Japan argued that, under EC methodology, the comparison was always based upon an "asymmetrical" analysis, especially when export prices were constructed.

106. Japan noted that the EC's argument that it had found the level of trade of the domestic sales department or sales subsidiary to be equivalent to be the export sales department could be regarded as a claim that the EC had chosen to make a comparison between those "levels" only after they had investigated the particular pattern of sales for audio cassettes. However, the EC had not made a finding that both sales were at the same "level of trade". Japan argued that the EC had investigated the sales patterns of some exporters, but only in order to consider whether there were differences in the nature of the purchasers. Japan argued that the EC had never permitted a "level of trade" adjustment based on the nature of the seller.

107. The EC argued that the first question for an investigating authority was whether the "level of trade" was properly identified. The EC argued that the corresponding level of trade in the country of export was the sales department/company of the Japanese producer, adjusted to deduct all direct expenses. The EC argued that a difference in costs between domestic and export sales was not indicative of a difference in levels of trade. Market conditions and distribution networks could not be assumed to be similar in different markets. The EC argued that the investigating authorities had decided in this case that the appropriate level of trade was the domestic sales network or sales subsidiary, which was determined to be equivalent to the export sales department. The reason for that determination was that both agencies sold to distributors and wholesalers with a similar pattern of prices and quantities. The appropriateness of the level of trade was not contested and no adjustments were claimed for the level of trade by the exporter concerned.

108. Japan considered that the EC's reasoning concerning "level of trade" should have resulted in the sales subsidiary in the EC being the comparable level of trade as the only sales made by the export sales department were to a sales subsidiary. Japan submitted that the selection of a comparable level of sale was a different example of a level of trade to the selection of a particular distributor based on similarity of sales and quantity patterns. Japan argued that the EC had failed to select the appropriate level of trade for domestic and export sales, had made no allowance for the resulting unfairness, and therefore had failed to comply with the requirement to effect a"fair comparison".

109. Japan argued that in this case, in addition to not comparing normal values and export prices at the same level of trade, the EC had failed to fulfil the requirement of "fairness" in not making appropriate "due allowances". Japan also argued that in most cases appropriate "due allowances" were necessary to effect a "fair comparison".

110. In this case, allowances should have been made for indirect expenses and profit. Because making allowances for differences affecting price comparability is necessary for a fair comparison, Japan rejected the EC's argument that only variable costs, or factors directly related to the sales could affect price comparability. The EC's deduction of expenses from the export market but leaving them untouched as regards domestic sales was unfair, and ignored the fact that similar expenses were incurred in both the domestic and export markets. Japan did not accept the EC's characterization of the Japanese market as closed to competition.

111. The EC noted that once it had established that the sales took place as nearly as possible at the same time and the prices were at the same level of trade, it made due allowances to offset factors directly related to the sales under consideration which affected price comparability. The EC argued that the phrase "conditions and terms of sale" only referred to variable costs which affected price comparability and which occurred at the time of sale. The EC argued that as dumping was only possible when the market in the exporting country was isolated, costs in the exporting country would nearly always be different from those in the importing country.

112. The EC argued that therefore it was proper to take account of different costs, such as physical characteristics, packing, credit, warranty and other such factors (which may be negotiated by the customer at the point of sale) as due allowances. The EC argued that other costs, including indirect expenses, did not depend on the negotiations between buyer and seller. The EC relied on the words in Article 2:6 of the Agreement "... on its merits ..." to argue that the decision whether to grant due allowances, and the quantum of any such allowances was left to the discretion of the investigating authorities thereby giving administrators the flexibility to modify values according to the circumstances of sale. The EC also noted that the word "due" required that only allowances that achieved "comparability" be made. The EC argued that allowances that made prices less "comparable", i.e., because particular expenses did not affect both prices, were not permitted. Only adjustments for expenses that reflected an actual difference should be made. The EC noted that it had made adjustments to a Korean exporter for indirect expenses and profits that had been shown to have affected the price paid. The EC also noted that if an exporter could establish that the export price was at a different level of trade to the normal value, an adjustment could be made. In the present case TDK had not demonstrated to the satisfaction of the investigating authorities that the different levels of trade resulted in any price differential.

113. The EC argued that the indirect expenses of the foreign producer were irrelevant to determination of a normal value and were only relevant to the construction of the export price. Only factors that directly related to sales of the goods under consideration could be the subject of an allowance. This was because only variable costs could, in the requisite manner, affect prices, and only variable costs should therefore be adjusted. The EC argued that the Japanese market's lack of competition in such areas as advertising also raised costs, and therefore made adjustment of such costs inappropriate. Article 2:6 provided an exclusive statement of both when adjustments could be made, and what type of adjustments could be made. Any distinction between direct and indirect costs was irrelevant.

114. The EC argued that identical allowances were made to both the normal value and export price when dealing with sales to independent importers. When dealing with constructed export prices the EC calculated selling expenses of export sales departments on a turnover basis.

115. Japan claimed that the EC had engaged in a method of averaging it termed "zeroing". According to this method, any sales in which the export price exceeded the normal value were set at "zero". Those "zeroed" transactions were then used in the calculation of an average dumping margin. Japan argued that this resulted in the finding of a higher average margin than would otherwise be found. Japan argued that the practice of "zeroing" was arbitrary because it failed to properly consider export prices above normal value, and unfair, because it always disadvantaged exporters, and therefore was contrary to the requirement in Article 2:1 and 2:6 to effect a fair comparison. Japan also argued that "zeroing" breached the obligation in Article 8:3 that the amount of the anti-dumping duty not exceed the margin of dumping. This was because the dumping margins so determined exceeded the "... margin of dumping as established by Article 2".

116. Japan argued that although in this case the scope for operation of "zeroing" was limited because of the EC's "asymmetrical" comparison of normal values and export prices, "zeroing" still occurred.

117. The EC argued that the definition of "dumping" in the Agreement only required that sales at prices lower than the normal value be examined. The purpose of anti-dumping actions would be nullified if exporters were able to conceal dumping by offsetting dumped sales with sales above the normal value. The EC noted that the terms "zeroing" and "negative margins" used by Japan to describe the methodology of the EC did not appear in the Agreement and were inaccurate: they implied that the Agreement required investigating authorities when determining dumping to consider export sales made at prices above normal value, which it did not.

118. The EC agreed that in averaging the individual dumping margins, it treated transactions where export sales were at prices above normal value as though they had a "zero" dumping margin. The EC argued that its practice did not amount to manipulation of transactions under enquiry. The EC stated that it used the volume of the sales for which no dumping was found to weight the transactions, resulting in the lowering of the dumping margin. If it had not also taken into account and weighted sales at non-dumped prices, the resulting dumping margin would have been higher than under the EC's current practice. The EC argued that it had therefore gone beyond the requirements of the Agreement. The EC admitted that this averaging method was standard practice by its investigating authorities. The EC agreed that duty set on a prospective basis may be applied to undumped transactions, but noted that this would be the case whatever averaging methodology was used. The EC argued that its refund and review mechanism was capable of remedying any situations in which exports were made at undumped prices. The EC argued that almost all of the transactions examined were dumped, and that in this case "zeroing" played an insignificant role.

119. The EC argued that the definition of dumping in Article 2.1 of the Agreement was only concerned with dumping, i.e. with sales in which the normal value exceeded the export price. Thus, a Party was not required to take into account sales in which the export price exceeded the normal value. The EC argued that its legislation and ordinary practice surpassed the requirements of the Agreement, as it was open to it to not examine or average negatively dumped sales at all.

120. Japan argued that there was no support in the Agreement for the EC's position that the dumping margin was defined as the amount by which the normal value exceeded the export price, and that consequently 'negatively dumped sales' need not be taken into account. Japan argued that the absence of a provision in the Agreement dealing with "negative dumping" should not lead the Panel to conclude that instances of negative dumping were not required to be considered by the authorities. In addition, Japan argued that as the Agreement should be interpreted restrictively the EC had not discharged its burden of proof and had not shown that the Agreement did not require investigating authorities to examine negatively dumped sales. The absence of a provision in the Agreement concerning negative dumping should lead the Panel to the conclusion that "zeroing" was not permitted without justification. Japan argued that this negated the EC's argument that because the Agreement was silent concerning how to treat instances of negative dumping it was permissible to "zero" sales.

121. Japan noted that the EC's rules required the EC to weight the individual dumping margins by the value of the corresponding export sale. Japan argued that the result of this was that any sales so "zeroed" would not lower or eliminate a dumping margin.

122. Japan argued that once a dumping margin was assessed and included "zeroed margins" that margin operated to determine the rate of dumping duty applied to all future shipments. The prospectively set duty rate would apply to all transactions and would include undumped transactions. Japan noted that price variations in the export market could generate differences in the dumping margin, possibly leading, on a consignment by consignment basis, to dumping in one instance, and to a "negative dumping margin" on another.

123. Japan argued that the EC's Basic Regulation obliged the investigating authorities to use "zeroing". This was because recital 13 of the Basic Regulation provided that export prices should normally be compared with normal value on a transaction by transaction basis, except where the use of weighted averages "...would not materially affect the results of the investigation...". Because the EC's "zeroing" practice always increased the dumping margin, transaction by transaction comparisons were not conducted. Japan argued that a simple average of "positively" and "negatively" dumped goods would result in no or a lesser dumping margin.

124. Japan argued that zeroing would always inflate the dumping margin. Japan noted that the circumstances of this case were distinguishable from those confronting the Salmon Panel. This Panel had before it evidence of the prejudicial effects of "zeroing", as the EC had admitted that some export prices were above normal values. Therefore the Panel could condemn this practice.

125. In addition to claiming that "zeroing" breached the requirement of a fair comparison discussed above, Japan argued that the operation of the practice of "zeroing" was unfair because it generated higher dumping margins than would otherwise be the case if sales when export prices rose above the normal value were equally considered. It was also unfair when some sales were dumped and others were not, as it operated to always create a dumping margin. Japan also argued that "zeroing" operated to prejudice those exporters who dumped the least.

126. Japan argued that zeroing was inherently unfair because it always adversely effected an exporter. If an exporter made some sales at dumped prices and others at undumped prices "zeroing" would always produce a dumping margin. Whenever prices varied over time an exporter who rigorously maintained equal normal values and export prices would be found to have dumped. This was because the normal value was always an average figure whereas the export price was an individual sales price. Whenever an export price was below the average normal value a dumping margin would be created, whereas a sale above the normal value would not offset the margin. In short, because of its arbitrariness and unfairness, the "zeroing" rule infringed the requirement of a fair comparison.

127. On the definition of "fair comparison" as required by Articles 2:1 and 2:6 of the Agreement the EC recalled its arguments as advanced in the context of the claim that the comparison was unfair (see above, paragraphs 92-95). The EC argued that the notion of a "fair comparison" did not necessitate recourse to some abstract notion of "fairness", as the text of Article 2:6 clearly set out the elements necessary to ensure a fair comparison.

128. The EC argued that zeroing, and comparison of an average normal value with a transaction by transaction average export price, was necessary to detect targeted dumping and avert injury. The EC argued that if Japan's arguments were accepted, negatively dumped sales would be analyzed by means of comparing average normal values with average export prices, with the result that targeted sales at dumped prices into selected sub-markets could not be detected.

129. Injury would indisputably follow from targeted dumping. If sales were not zeroed, a simple average of margins could lead to a result of no dumping overall, thereby denying relief to an industry in that sub-market. This would be contrary to the object and purpose of the Agreement, which should be among the main elements of its interpretation. Further dumping could then follow into other sub-markets. The EC argued that targeting took place in the UK and Germany, and that therefore so called "zeroing" was necessary. In any event the EC stated that almost the totality of the prices for export transactions were below the normal value.

130. Japan said that it did not seek to argue that "zeroing" could never be justified, but objected to the EC's assumption that zeroing was required because of the existence of targeted sales, without having determined whether targeting of sub markets was occurring, and argued that the use of zeroing against targeting would require justification, Japan argued that although Article 2(13) of the Basic Regulation used the word "normally", zeroing was always applied.

131. Japan also argued that targeting was an unrealistic justification for "zeroing". It assumed that the exporter could find purchasers to pay high prices to offset those paying low prices. It also assumed that the local industry could not supply the purchasers prepared to pay the high prices.

132. Japan argued that not all dumping was targeted, and that even if some dumping were shown to be targeted it would be unreasonable to assume that all dumping was targeted. Japan also argued that the EC should be required to establish that "targeted" dumping occurred, and should not be permitted to merely assume it took place as a justification for "zeroing".

133. Japan noted that the EC's argument that pricing in Germany and the UK demonstrated "targeting" did not appear in the Regulations. Japan asserted therefore that this argument by the EC could not be raised, based on the reasoning of the Polyacetal Resins Panel and on the EC's earlier statements during the conciliation and consultation process that the injury determination had not been based on regional injury. In addition, Japan said that the differences in prices between Germany and the UK were explained by differences in the selling activities and levels of trade of the different subsidiaries.

134. Japan objected to (i) the way in which SG&A expenses were allocated, and (ii) the rate of profit selected in the construction of a normal value. Japan claimed that in constructing normal values for certain models of audio cassettes for the Japanese exporter TDK, the EC had acted inconsistently with Article 2:4, as its decision was not "reasonable".

135. The EC denied that Japan's allegations involved questions of interpretation of the Agreement. They were of a purely factual nature and irrelevant unless Japan could demonstrate a manifest error in their appreciation. As to the substance of the factual arguments, the EC said that it had relied entirely on data supplied by the exporter concerned and that the arguments were ill-founded.

136. Japan noted that it did not object to the decision of the EC to construct normal values, only to the manner in which constructed values were calculated for certain TDK products. Japan argued that Article 2:4 of the Agreement required that a constructed normal value be based on the cost of production "... plus a reasonable amount for administrative, selling and any other costs and for profits". Japan argued that the EC had behaved unreasonably in not providing an explanation for its decision to reject a turnover based allocation. Japan argued that the EC had therefore acted in an arbitrary or gratuitously prejudicial manner, and accordingly had not complied with the requirement that the amount for selling, general and administrative costs be "reasonable".

137. Japan also argued that the word "reasonable" in Article 2:4 required that the EC not act inconsistently with its usual practice for allocation of SG&A, and when determining an amount of SG&A that the EC not act in a manner which could prejudice the interests of the exporter.

138. Japan submitted that the result of the method of allocation of SG&A was to increase the amounts of SG&A allocated to the three models under enquiry. This had resulted in an increase in costs, normal values, and ultimately, dumping margins.

139. Japan noted that there were two commonly used methods of allocating costs: allocation proportional to the cost of manufacture of particular models, and allocation proportional to turnover. TDK had requested that the EC allocate costs on the basis of turnover. Article 2(11) of the Basic Regulation provided that, in general, costs should be allocated proportional to turnover. The EC had allocated SG&A expenses on the basis of cost of manufacture. Japan argued that the EC's decision on allocation of SG&A expenses was without explanation, and was therefore unreasoned, and therefore in breach of Article 2:4, as it was perverse and ignored both the exporter's request and the preference contained in the EC's own rules.

140. Japan also argued that the reasoning leading to EC's conclusion that the costing information supplied by TDK was unreliable in recital 29 of the Provisional Regulation was "not convincing". That conclusion of the EC overlapped with the decision to construct a normal value. It was not convincing because a decision to construct a normal value properly only arose when there were problems regarding the prices at which the products were sold in the domestic market. If the turnover basis was rejected whenever such problems existed the information would never be used. This was in obvious conflict with the preference in the Basic Regulation for turnover based allocation. Japan noted that TDK had not objected in principle to construction of normal value, but had objected to the way in which the normal value was constructed.

141. The EC relied on recital 29 of its Provisional Regulation, which stated that:

to show that both the request of the company, and its supporting costing data, had been considered and then rejected as an unsuitable basis for allocation of costs in this case. The EC argued that recital 29 demonstrated that the information provided in support of TDK's request for a turnover based allocation was properly considered and rejected because the information in support of the request was determined to be unreliable.

142. The EC remarked that it had constructed a normal value for TDK because TDK had been unable to provide reliable data for its domestic sales on the goods under enquiry. The EC informed the Panel that TDK had supported the EC's approach.

143. Japan also attacked the rate of profit used in the calculation of the constructed normal value. Japan argued that the EC failed to comply with the obligation in Article 2:4 to base a constructed normal value, on, inter alia, a "reasonable amount" for profit. Japan claimed that the EC's adoption of the profit rate for the like product did not automatically satisfy the requirements of "reasonable profit". Japan noted that the Agreement did not use the term "like product" in the context of the last sentence of Article 2:4. It could not be argued therefore that "goods of the same general category" necessarily corresponded to the definition of "like product". The overriding criterion was that the rate of profit selected be reasonable, and in this case the rate selected was not reasonable. Although the "chrome" and "metal" tapes used in the audio cassettes were of the same general type, the setting of a profit rate for "normal" cassettes based on an average of profit rates for other higher profit cassettes was not a "reasonable amount", because the EC did not adopt the profit rate of the appropriate group of products, although it could have done so.

144. Japan noted that the EC relied on average TDK's profit data for sales of all normal, chrome and metal types sold in Japan to set the rate of profit for normal cassettes. The rate of profit on chrome and metal audio cassettes was much higher than that for normal cassettes, with the consequent result that the average profit rate was raised above that applying to normal cassettes.

145. The EC argued that this was also a factual argument, which would also require a determination by the Panel that the decision of the EC to use a particular profit rate was unreasonable. The EC argued that Article 2:4 required that profit be set at that "normally realized on sales of the same general category". The EC had determined that the profits on the sales of "normal" type of cassettes were unreliable due to sales to a third party. It had therefore relied on arms length sales of goods of the same general type. The EC argued that if it was reasonable to rely on data from sales of products of the same general category, it must be reasonable to rely on data from sales of like goods. Article 2:4 made no distinction between goods of the same general category and like goods. If the profit rate for sales to related parties had been excluded, or profit rates for other sellers had been used, a higher profit rate could have resulted. The EC argued that in using a lower rate than the highest profit rate normally realized on goods of the same general character, it had more than complied with the Agreement's requirements concerning profit.

146. Japan claimed that the EC had improperly decided to cumulate the imports of Japan with those of other countries under investigation. Japan claimed that the EC had failed to establish the existence of the factors identified in Article 3:2 of the Agreement. Japan also claimed that the EC had failed to properly apply the requirements of Article 3:4 particularly in relation to the phrase "the effects of dumping" and claimed that the EC had not properly taken account of "other factors" which may have caused injury to the domestic industry.

147. The EC claimed that it had properly cumulated the imports of Japan with those of other countries under investigation. The EC claimed that it had complied with the requirements of Article 3. The EC argued that it had properly taken account of alleged "other factors".

148. Japan did not contest the EC's finding that the domestic industry was suffering material injury. Rather, Japan claimed that the finding of a causal link between the dumped imports and the material injury suffered by the domestic industry was not properly made. Japan argued that certain parts of the Provisional and Definitive Regulations did not meet the requirements of the Agreement because they either did not contain either sufficient evidence or sufficient reasoning.

149. The EC argued that the requirements of Article 3 were procedural, not of substantive content, and were non-exhaustive. The EC argued that if positive evidence was found of the elements mentioned in Article 3:1 without demonstration of manifest error or arbitrariness, the factors mentioned in Article 3:2 and 3:3 were properly considered and found to be present whether in combination or severally, the investigating authority properly set out the elements on which its findings were based, and the injury found to exist could not be attributed in its entirety to factors other than dumping, the determination of injury could not be impugned by the panel. The EC referred to recitals 55-77 of the Provisional Regulation and recitals 26-32 of the Definitive Regulation to show that all of the elements had been considered and to show that the consideration demonstrated the actual presence of many of the relevant factors and indices.

150. Japan objected to the use of cumulation in this case. Japan complained that the EC's action in cumulating Japan's exports with those of Korea infringed Article 3 of the Agreement. Japan claimed that the EC had failed to comply with the requirements of Article 3:4, as its cumulation of imports denied Japanese exporters the opportunity to demonstrate that any injury caused by them was not through the effects of dumping. Japan also claimed that through its inconsistent approach the EC had failed to comply with the requirement of an "objective examination" provided in Article 3:1.

151. The EC argued that because Article 3 of the Agreement laid down no rules with respect to cumulation, it was permissible to consider the collective effects on the domestic industry of the dumped imports. The decision whether to cumulate was left to the discretion of administering authorities. In the facts of this case, the EC industry had simultaneously to resist the effects of imports from Korea and Japan, which supplied different market sectors. The EC denied that it had deviated from its standard criteria and argued that, even if it had so deviated, it was not obliged under the Agreement always to follow criteria based on a standing administrative practice.

152. Japan argued that Japanese imports supplied the higher-quality segments of the market, whereas Korean imports supplied the lower end of the market. There were substantial differences in prices between the segments of the market. Japan argued that if products from one country were distinguishable from other exports, then cumulation was never justified. Japan argued that Japanese exports were mainly of the "metal" and "chrome" type and competed with the domestic products on a non-price basis, whilst Korean imports were mainly of the "normal" type and competed on price. Japan argued therefore that there were clear and distinct markets for the goods. Japan noted that the EC had found that the Japanese goods competed on non-price elements, such as brand name, marketing features and styling. Therefore, the EC's application of cumulation in this case denied exporters the opportunity of invoking Article 3:4 in order to distinguish themselves from other exporters.

153. Japan noted that there were substantial price differences between the Korean and Japanese goods as well as differences in the increase in volume of the goods in the period 1985 to 1988. Japan argued therefore that if these factors had been considered the EC would not have decided to cumulate Japan's exports with those of Korea. Japan also noted that the cassettes imported into the EC that supplied the upper end of the market identified by the EC were manufactured in Japan and in the EC. Those cassettes were interchangeable with one other in the market. Cassettes manufactured in Korea and sold to the lower priced segment of the market were not interchangeable. Japan argued that this was relevant to the decision to cumulate, and to the examination of injury itself.

154. The EC argued that Article 3 permitted consideration of the collective effects of dumped imports and imposed no limitations on such cumulation. The reference in Article 3:4 to "the dumped imports" confirmed that cumulation was permissible. The EC also argued that the absence of any specific rules in the Agreement on cumulation suggested that it was left to the investigating authorities to determine when it was appropriate to cumulate.

155. The EC argued that the degree of price similarity between products was examined and was determined to be not relevant when the EC industry was forced to compete with products in different market and price sectors. The EC argued that the domestic industry was effectively caught in a "pincer" of low priced Korean imports at the low end of the market, and Japanese imports at the high end of the market. This required the industry to compete in both market segments simultaneously, because leaving one market segment to either of the other parties would have resulted in a speedy demise. The Agreement did not require a differentiation between two injurious effects. Analysis of the effect of dumped imports from all sources on the domestic industry was therefore permitted, and was at the core of an anti-dumping enquiry.

156. The EC argued also that the differences in the price competition of Korean imports, and the non-price competition of Japanese imports did not permit separation of imports from Japan or Korea into different markets. The EC recalled its conclusions in recitals 24 and 25 of the Definitive Regulation that the goods were commercially interchangeable, with the result that Japanese exporters sold in the EC cassettes manufactured in both Japan and Korea, with no difference in the customers perception of the products by reason of their different sources. Although there were some distinguishing characteristics between Korean and Japanese exports, in most cases examined Japanese exporters sold Korean manufactured cassettes, and there was insufficient basis to clearly establish a dividing line separating exports from Japan and Korea. The EC also noted that in the case of TDK, most of its sales were in the "normal" category dominated by Korea.

157. Japan argued that whilst commercial interchangeability existed between the cassettes manufactured by the Japanese exporters and their related Korean producers, it did not exist between those cassettes and the low priced Korean exports. Japan noted also that the high quality image of the Japanese products extended to their "normal" cassettes also, therefore not denying Japan's argument of separate and distinct markets. Japan argued that the largest EC market was biased towards the "chrome" and "metal" cassettes. Japan noted also that during the course of its enquiry the Commission had informed TDK that it did not find TDK to be engaging in price undercutting in Germany. Japan relied on that statement to establish that TDK's majority sales of "normal" cassettes were not competing on price in Germany.

158. Japan also argued that the requirement for an "objective examination" in Article 3:1 obliged the EC to apply the EC's criteria for cumulation in a consistent manner. That requirement would be nullified if a party was free to alter its practices whenever they interfered with the outcome of an investigation. Japan noted that two of the EC's criteria usually taken into account were not taken into account by the EC in this case when deciding whether to cumulate imports from Japan. Those criteria usually were; "... the comparability of the imported products in terms of... [1] the increase in volume of imports from a previous comparable period and [2] the low level of prices attributable to the products of all supplying companies ..." (Hydraulic excavators originating in Japan, Commission Regulation (EEC) no. 595/85 of 7 March 1985, recital 21).

159. The EC argued that there were no provisions in the Agreement concerning cumulation. It argued that the Agreement did not require it to always apply its criteria in the same manner. The EC argued that in any event, it had applied its criteria in a consistent manner.

160. Japan provided to the Panel a detailed outline of Commission decisions in 1991 highlighting the criteria used in decisions whether to cumulate imports. The document analyzed Commission and Council Regulations of 1991. The Regulations were: 129/91 of 11 January 1991, Imposition of provisional measures on Small Screen Colour Television Receivers originating in Hong Kong and the People's Republic of China; 512/91 of 25 July 1991, Acceptance of undertakings on Imports of Artificial Corundum originating in the Soviet Union, Hungary, Poland, Czechoslovakia, and the People's Republic of China; 1386/91 of 23 May 1991, Imposition of provisional duties on Gas-Fuelled non-refillable pocket flint lighters originating in Japan, the People's Republic of China, the Republic of Korea and Thailand; 1391/91 of 27 May 1991, Imposition of definitive duties on Aspartame originating in Japan and the United States of America; 1472/91 of 29 May 1991, Imposition of provisional duties on Oxalic Acid originating in India or China; 2054/91 of 11 July 1991, Imposition of provisional duties on Dihydrostreptomycin originating in the People's Republic of China; 2093/91 of 15 July 1991, Imposition of definitive duties on Small Screen Colour Television Receivers originating in Hong Kong and the People's Republic of China; 2818/91 of 23 September 1991, Imposition of provisional duties on Cotton Yarn originating in Brazil, Egypt and Turkey; 2904/91 of 27 September 1991, Imposition of provisional duty on Certain Polyester Yarns originating in Taiwan, Indonesia, India, the People's Republic of China and Turkey; 3433/91 of 25 November 1991, Imposition of definitive duties on Gas-Fuelled non-refillable pocket flint lighters originating in Japan, the People's Republic of China, the Republic of Korea and Thailand; 3434/91 of 25 November 1991, Imposition of definitive duties on Oxalic Acid originating in India or China. Japan argued that these cases demonstrated that the criterion "similar level of price" was explicitly or implicitly present in virtually every case during that year.

161. Japan argued that as the standard cumulation criteria mentioned above were relevant to the decision to cumulate imports, and had not been taken into account, the decision to cumulate was rendered unreliable. In addition, Japan argued that if those considerations had been taken into account the decision makers would probably have reached a different decision on whether to cumulate imports from Japan. Japan argued that in the one case in which the EC decided to not cumulate imports in 1991, (Dihydrostreptomycin originating in the People's Republic of China) the EC's principal reason for so deciding was the difference in price levels.

162. The EC argued that the standard criteria for cumulation applied in this case only represented a standing administrative practice, and that there was no obligation under the Agreement to follow the same administrative practices. Administrative practices had nothing whatever to do with the phrase "objective examination" in Article 3:1. That phrase only related to the three elements in Article 3:1 that were required to be objectively examined by the investigating authority.

163. The EC argued that the criteria mentioned in Recitals 24 and 25 of the Definitive Regulation were their standard criteria for cumulation, and had been their standard criteria since at least 1988. The additional criteria that Japan had complained were not taken into account were only applied in investigations into commodities where there was a specific reason to take account of prices, because in commodities trade price is a determining factor. Consequently the EC had not applied its standard criteria of cumulation in an inconsistent manner .

164. In reply to the EC's argument that there was no rule in the Agreement concerning cumulation, Japan argued that such a rule must exist by implication, due to the general principle of causation contained in Article 3:4. In reply to the EC's argument that at best the criteria on cumulation were merely standard administrative practice, Japan argued that an unexplained departure from a national rule on cumulation whenever it interfered with the results of an examination would also infringe the Agreement's requirement of an "objective examination", and that whatever rules the Agreement contained on cumulation, they must be applied consistently.

165. Japan argued that as a result, whether or not injury was caused by Japan should have been assessed independently. In relation to volume increase and price effects, Japan argued that because a causal analysis was required for a finding of price suppression or depression an individual assessment of the effect of Japanese exports was required.

166. Japan argued that the EC had failed to find whether or not there was a "significant" increase in the volume of imports from Japan. Although the EC had determined that a volume increase had occurred, in failing to make an explicit finding under Article 3:2 that the volume increase was "significant" the EC had failed to comply with the obligation in Article 3:1 that an injury determination be based on "positive evidence" and involve an "objective examination". Japan also argued that the panel should find that it could not be satisfied from analysis of either the Provisional or Definitive Regulation that the EC had established whether or not a "significant" volume increase had occurred.

167. The EC argued that Article 3 did not require a specific finding of an increase in volume or of its "significance". The indices and factors of paragraphs 2 and 3 of Article 3 must be "considered" and thus shown to be present.

168. Japan argued that Article 3:1 required that a determination of injury be based on "positive evidence", and not on an assumption that any of the Article's elements were satisfied. Japan argued that Article 3:2 gave specific content to the obligation in paragraph 1(a) that an investigating authority conduct an objective examination of "... the volume of the dumped imports and their effect on prices in the domestic market for like products...". Specifically, Article 3:2 required the investigating authority to consider whether there had been a significant increase in dumped imports, either in absolute terms or relative to production or consumption in the importing country. Article 3:1 also required that the determination of injury involve an "objective examination" of volume and price factors. The requirements of positive evidence and an objective examination applied to all factors listed in paragraphs 2, 3 and 4 of Article 3.

169. Japan argued that the requirements of Article 3 must have some substantive content. It noted that Article 3:2 did not provide that "... this list is not exhaustive". Japan argued that the requirements of Article 3 described the types of factors that could properly be considered to cause material injury. Japan argued that the decision of the Polyacetal Resins Panel suggested that at least some of the requirements of Article 3 were of a substantive nature, due to that Panel's finding that Korea had failed to comply with Articles 3:1, 3:3 and 3:4.

170. The EC argued that for the Panel to condemn the EC's findings, Japan was required to show that the EC had violated a provision of the Agreement in assessing the evidence, had manifestly erred, or had acted arbitrarily. The EC argued that the Agreement required an examination of the circumstances of injury as a whole, and that individual injury factors were not determinative of injury. The EC argued that the words "[n]o one or several of these factors can necessarily give decisive guidance" in Article 3:2 confirmed its argument that the task of the investigating authorities was to examine the circumstances of injury as a whole, and also confirmed its argument that the elements contained in Article 3:1 were not conjunctive. The investigating authority was only required to consider each of those elements, but need not find each element was present in every case. In a later argument the EC clarified that the elements must be considered and thus shown to be present.

171. The EC argued that in including import figures (without qualifying them as significant) it had met the procedural requirements of Article 3:2, and the mere absence of an explicit finding of "significant" increase should not lead the Panel to find a failure to comply with Article 8:5 (on which Japan had in any event not relied). Due to the inclusion in the Provisional and Definitive Regulations of the import figures the Panel was in a position to be able to identify whether the determination was based on all relevant facts and whether any manifest error of fact or law had been made. The EC argued that the Agreement did not oblige it to include most or all of the raw evidence relied upon in order to conform with the requirement of "positive evidence". The EC argued that the Polyacetal Resins Panel supported this argument.

172. The EC argued that by arguing that the EC had fallen into error in not determining that the increase in volume was "significant", Japan sought to create a substantive rule from a mere procedural requirement. In reliance upon paragraph 372 of the Softwood Lumber Panel, the EC argued that it need not make a finding concerning the 'significance of the increase of the volume' of imports from Japan, as a condition precedent to the making of an affirmative injury decision.

173. Japan argued that the Softwood Lumber Panel did not support the EC's suggestion that there was only an obligation to consider the presence of significant price effects, and also argued that in any event some combination of the factors must be present and at a significant level.

174. Japan stated that it did not argue that both price and volume criteria must be established to comply with Article 3:2. Japan argued that some combination of the factors mentioned in Article 3:2 must be present, and at a significant level. If the evidence revealed little undercutting but nevertheless a large volume increase, a positive finding under Article 3:2 would be one possible conclusion. Japan noted that in some cases one of the volume factors mentioned in Article 3:2 in isolation could qualify as significant. The circumstances of this case precluded such a conclusion. In any event Japan recalled that the EC had made no determination of significance.

175. The EC noted that in arguing for a substantive link between price effects and volume effects, Japan was again attempting to create from a duty to merely consider whether there has been an increase in dumped imports a substantive rule that where there was no evidence of price effects, there could only be injury if there were a high volume of imports.

176. The EC argued that no particular combination of factors was required, although it recognised at least one of the elements would be required in order to make an injury finding. The EC based this assertion on the last sentence of Article 3:2. Article 3.2 did not require a finding that the level of increase in volume was significant in relation to other factors, nor that any such increase be considered in relation to other factors.

177. The EC argued that Article 3 did not require an authority to consider the impact of the goods on the domestic industry through the volume and the price of the goods conjunctively. The EC argued that Article 3:1 linked price and volume elements by the word "and", which only required that both elements be examined, and did not require that they be examined in relation to one another. Article 3:2 mentioned the volume of dumped imports and the effect on prices. Article 3:3 elaborated further on "consequent impact". Both these paragraphs expressly declared that "no one or several of these factors can necessarily give decisive guidance". Consequently, the Agreement established no relationship between the level of prices and the volume of imports. Japan therefore in this case could not argue because of an absence of price effects that a volume increase of 8 per cent was not significant.

178. Japan argued that volume could be determined to be "significant" on three bases; in absolute terms, relative to production in the EC, and relative to consumption. Consequently, what was "significant" in the context of Article 3:2 could not be absolutely defined. Significance could depend on many factors, and required an analysis of the particular position of the domestic producer. Japan noted that if a producer were faced with low priced imports, the producer would have to decide whether to maintain current prices, thereby risking a loss of volume, or to lower prices. In either case, profits would suffer. If a manufacturer was forced to drop prices by a small amount due to the price of the dumped goods, that could easily amount to "significant" price depression if it was accompanied by a large loss of volume, leading overall to a substantial loss of profits. The volume of sales was crucial to determining whether "significant" price undercutting, depression or suppression had occurred.

179. Japan argued that in considering whether the price factor caused by the dumped imports was sufficiently serious to qualify as "significant", the investigating authorities must consider the seriousness in relation to the volume factor, and vice versa. What could be a significant decline in sales volume when the domestic producer had been obliged to lower prices would not necessarily be significant if the domestic producer had managed to keep the prices constant. Japan noted that the converse was also true, as a price effect was more likely to be considered significant if it was accompanied by a loss of sales volume.

180. The EC argued that the word "significant" used in Article 3:2 meant not insignificant or de minimis, and any increase in volume must not be negligible. The EC did not agree with Japan's argument that the term "significant" should be determined by its context, as this would mean that the term would have a different meaning every time.

181. Japan argued that the absence of a reference in the EC's determinations to a "significant" volume increase was more than a mere oversight. Japan noted that Article 3:2 required that the investigating authorities "consider whether there has been a significant increase in dumped imports, either in absolute terms or relative to production or consumption in the importing country". Japan reiterated that an analysis of volume required two determinations; firstly that there had been a volume increase, and secondly that the increase was "significant". Japan noted that it was insufficient for the EC to make an ex post facto statement that it had considered whether Japanese imports were significant. In a related argument Japan noted that the Polyacetal Resins Panel and the requirements of Article 8.5 supported its argument that a determination of "significant" volume increase must appear in the regulation. Japan argued therefore that the absence of a specific conclusion of "significance" was fatal to that part of the decision.

182. The EC admitted that neither the Provisional nor the Definitive Regulation contained an explicit statement that the volume increase was "significant". The EC argued that a finding of "significant" increase in volume was implicit in the decisions. The EC stated that during the course of the investigation its authorities had accepted that the increase in volume was "significant". The EC argued that a finding, whether implicit or explicit, of a "significant" increase in volume was not necessary. The EC argued that Article 3:2 did not create substantive obligations in that respect. The EC argued that the Agreement only required that an evaluation or consideration of the elements listed should make clear that all elements had been considered and that at least one of them had been established.

183. The EC raised an objection that any complaint concerning the absence of an explicit finding of "significant" volume was properly characterised as an argument about insufficiency of reasoning under Article 8:5. If so, the EC argued that the claim had been insufficiently raised during the consultation and conciliation phases. In the alternative, the EC argued that as its Preliminary and Final Regulations had provided numerical analyses of volume increases, it had met the procedural requirements of Article 3:2 and the requirement for a statement of reasons contained in Article 8:5.

184. Japan argued that in deciding whether an increase in volume under one of the criteria was "significant", account should also be taken of data relating to the other criteria. On this basis the small absolute increase in the present case was not significant. Relative to production in the EC imports from Japan decreased over the enquiry period. Relative to consumption the market share of imports declined significantly during 1985-1988. Therefore two of the three elements upon which a finding of significance could be made were not present.

185. Japan also noted that the Japanese imports' market share of 35 per cent represented a decrease of 7 per cent over the period investigated. Japan argued that although an increase may appear significant in isolation it could be insignificant in context of the other two elements, especially when, as in this case, the other two factors were registering decreases.

186. Japan also noted that although the EC had found that there had been an increase in imports from all countries, it did not make a finding of a significant increase in dumped imports from Japan. In addition, Japan argued that "significance" should be assessed against the background of annual variations in imports between 1985 and 1986, between 1987 and 1988 from Japan and other countries Those annual variations made an absolute increase of 12 million units over the three years 1985-1988 less than "significant". Japan noted that it had only referred to annual variations in imports in order to give meaning to the term "significant" in the context of an absolute increase. Japan also noted that in the present case, there was little or no evidence of price effects, the threshold of "significance" should have been higher. Therefore, Japan argued, the increases in volume of 12 million units could not be considered to be significant.

187. Japan argued that much of the volume change was the result of Japanese manufacturers shifting the sources of supply between their production facilities in various countries. Japan argued that shifting between sources could not be seen to have "exacerbated" injury, at least in a manner relevant to the Agreement. Neither the sale price nor appearance of the cassettes were affected by the change in sourcing.

188. The EC recalled that in recital 26 of the Definitive Regulation it had found that:

189. The EC argued that recital 26 showed a significant volume increase. The EC argued that Japan gave inordinate attention to the question of the increase in volume of dumped imports from Korea alone. It was uncontested that dumped imports cumulated from Japan and Korea had increased in absolute terms and relative to consumption and that there was also an increase in market share, although that increase was of less importance. The EC argued that the volume of cumulated dumped imports from all sources increased in absolute terms from 149 to 205 million units during the enquiry period. Market share of all dumped imports increased from 44 to 47 per cent. Therefore, the elements of paragraph 3:2 were present, as far as the total volume of imports were concerned.

190. The EC also argued that the increase in volume of dumped imports from Japan were significant on any reasonable analysis. Japanese dumped imports increased in volume by 12 million units over the same period while the domestic industry's volume decreased by 8 million units. Japanese market share at 35 per cent represented almost double the EC industry's market share, and was achieved because of dumping margins of between 44 and 66 per cent.

191. The volume increase of Japanese imports of 8 per cent in absolute terms and an amount of 30 per cent relative to EC domestic industry production was definitely "significant". The EC noted that its assertion that Japanese imports had increased relative to EC domestic industry production by 30 per cent was based on numerical analysis of percentage figures contained in the Provisional and Definitive Regulations.

192. Japan argued that the Agreement made a distinction between the definition of industry in Article 4, and the causation analysis required in Article 3:2. Japanese-owned EC producers were relevant to analysis of volume relative to total EC production. Although the producers were excluded from the definition of domestic industry for the purposes of Article 4:1, they were producers of goods for the purposes of Article 3, with the result that their production would 'dilute' the total production of cassettes manufactured within the boundaries of the EC.

193. The EC argued that there was nothing in the Agreement which suggested that indications of volume relative to production in the importing country under Article 3:2 should not be interpreted in the light of the definition of "domestic industry" in Article 4:1. Indications of volume were "effects" according to footnote 4 to Article 3:4, and such "effects" were supposed to cause injury to the "domestic industry". The EC also argued that the Japanese-owned EC producers' cassettes could not be considered to be in competition with the Japanese manufactured products, and also could not be allowed to camouflage the direct effects of the Japanese dumped imports. The presence of the Japanese-owned EC-produced cassettes enabled the Japanese dumped imports to have an undue influence far beyond their volume, because they did not compete with, and instead complemented the dumped imports.

194. The EC argued that Japan's explanation for the increase in volume being due to a shift in supply did not mitigate the injurious dumping, but in fact exacerbated the detrimental effects on the EC industry. The EC also argued that the argument that the exports from Japan had declined in relation to the total Japanese sales in the EC was irrelevant to the question whether the dumped imports caused injury to the EC industry. The EC argued that a table provided to the Panel by Japan could not be relied on to analyse the effects of the Japanese exports to the EC. The percentage of Japanese-produced cassettes could only be judged in relation to the EC domestic industry.

195. The EC argued that the annual variations in the years 1985-1986, and 1987-1988 relied on by Japan to show that the increase of 8 million units over the investigation period could not be determined to be significant, could not be said to impose an additional requirement necessary to be considered in determining "significance". The EC argued that in fact there were no normal annual variations.

196. The EC noted that recital 36 of the Definitive Regulation concluded that:

Thus, the EC argued, even in the absence of cumulation a finding of injury against the Japanese exports would stand. The EC noted however that no such finding had been made in either the Provisional or Definitive Regulation.

197. Japan also argued that the EC had relied on the retention of a large market share by the Japanese exporters as supporting the finding of injury caused by them alone. Japan argued that retention of a large market share was not necessarily important to the finding of injury . Japan argued that retention of a large market share could also illustrate that there was little change in the factors shaping the market. Japan also noted that a volume comparison between Japanese and Korean imports revealed that Japanese exports during the enquiry period rose by 8 per cent, whereas Korean imports rose by 600 per cent. Japan finally noted that there was no evidence of dumping by Japan prior to 1988, and that there was no evidence of continued dumping between 1985-1988, which thereby brought into question the finding of injury.

198. Japan argued that Article 3:1 required a finding of injury be based on positive evidence and involve an objective examination of the broad concepts mentioned therein. Article 3:2 provided a list of alternative matters which it was mandatory for authorities to examine. Japan further argued that paragraphs 1, 2, and 3 of Article 3, when linked with paragraph 4, provided that the specific injury factors and the general principle mentioned in paragraph 4 must be considered. In relation to this Article 3:2 merely posited various possible effects in the domestic market, which were price undercutting, price depression or price suppression.

199. Japan argued that Article 3:2 presented two alternatives for demonstration of price effects. The first method was a demonstration of "significant" price undercutting, and the second was whether the effect of the goods was to cause "significant" price depression or price suppression. Japan argued establishment of the existence of price undercutting was a relatively simple matter. The dumped goods would be found to be sold at a lesser price than the domestically produced goods. Japan argued that price depression and suppression were more difficult to establish, as those situations required demonstration of a causal link between the price at which the dumped goods were sold and the pricing of the domestic industry's goods.

200. The EC argued that it was not a condition precedent for a finding of significant undercutting or price depression or suppression that a significant volume increase be established. The EC added that over the period examined, there was a very clear depression of prices.

201. Japan noted that the EC provided "no evidence" to support the assertion that very clear price depression had been found, and therefore had failed to demonstrate that pricing had received meaningful consideration.

202. Japan claimed that the EC's finding that price undercutting was significant was not reasonable and therefore was inconsistent with Article 3:2. Two elements that it would have been reasonable to consider in an undercutting analysis were (1) the size of the margin of undercutting, and (2) the volume of products which were involved in the undercutting. Japan argued that the EC exaggerated the size of the margin of undercutting by (i) invalid methodology in selecting unrepresentative models and (ii) "zeroing" of average undercutting margins. Japan argued that the volume of imports of the exporters found to be undercutting into the EC market (which were determined to be undercut) was very low. On the basis of these flaws, the finding of "significant price undercutting" was not reasonable, and therefore failed to satisfy Article 3:2. Consequently, the EC had failed to make an "objective examination" on the basis of "positive evidence".

203. The EC argued that although it had established price undercutting, it had not been an important factor in the determination of injury in this case. The EC argued that price undercutting was not a decisive factor in the injury determination, whereas the finding was based on other injury factors. On that basis the EC found price undercutting in one market. The EC noted that price undercutting found in Germany was "significant", but argued that finding was not material to its injury determination. With respect to the margin of undercutting, the EC noted that its methodology for analysis of undercutting had been disclosed to the exporters, and that no complaint had been made. The EC also argued that it was proper to "zero" sales in order to prevent the concealment of price undercutting. With respect to the volume of undercut sales, the EC argued that undercutting was only found in the most important market in the EC.

204. Japan argued that the Provisional and Definitive Regulations suggested that price undercutting in Germany was a major factor in the original decisions, and that even during the consultation phase the EC had relied on price undercutting in that market. Japan argued that only after the event did the EC choose to rely heavily on price suppression and price depression as a justification for the conclusion of price effects.

205. The EC argued that there was simultaneous price undercutting by Korean and Japanese exporters. Large price undercutting by Korean exporters was found for a large number of undertakings. A limited degree of price undercutting was practised by Japanese exporters, but was concentrated in the most important market, Germany.

206. Japan argued that in its Regulations, the EC had devoted most of its discussion of undercutting to the size of the margin. Japan noted that in the Provisional Regulation (recitals 67 and 84) the EC had found undercutting in only the German market, at an average level of 11 per cent up to a maximum of 18 per cent. In the Provisional Regulation the EC had noted an average level of price undercutting of 11 per cent. In the Definitive Regulation the EC had stated that it had found:

Japan argued that further detailed information submitted to Japan by the EC during the course of consultations, and submitted to the Panel by Japan as Annex VI-A to its first submission, when properly recalculated, showed that the undercutting margins were much lower than the 11 per cent quoted in the Provisional Regulation. Only 2 of the 5 exporters were found to be undercutting.

207. The EC noted that the Japanese reassessment of price undercutting included "positive dumping margins" and "overcut sales".

208. Japan argued that recital 66 of the Provisional Regulation set out the EC's methodology for examination of price undercutting. Japan argued that although 30 per cent of all the Japanese sales were included, the methodology selected was arbitrary and that the samples used were not chosen on a statistically valid basis.

209. The EC referred to recital 66 of the Provisional Regulation to show how the undercutting analysis was conducted, and summarised its price methodology as being based on:

- a selection of the comparable and competing models of the exporters,

The EC argued that this methodology was disclosed to the exporters and was generally not contested.

210. Japan argued that the EC's rationale for its undercutting methodology was "not convincing". Japan also argued that the EC had given no justification why the method of selection of models was reasonable. Japan further argued that the method of selection of Sony models was arbitrary, in particular because various models selected were not representative. Japan also argued that invalid sampling methods had been used. Japan argued that because of the failings in the EC's price undercutting methodology the EC had failed to undertake an "objective examination".

211. Japan argued that the detailed information in Annex VI-A to Japan's first submission to the Panel showed that the Sony models selected for the undercutting analysis were not representative of the total sales of that company in the EC. Japan alleged that there were great variations in the undercutting margins on the different Sony products. Japan argued that the EC's determinations amounted to a claim that the weighted average of the group of selected models were the same as all of the sales. Japan argued that the EC had provided no statistical justification of the methodology used in conducting the undercutting analysis. Japan argued that if an undercutting margin had been determined for another product, and an average had been determined for all the products, it was virtually certain that the average figure would have been different from the figure calculated by the EC. Japan argued that the range of models selected for analysis should be wide enough to produce a reliable margin.

212. Japan complained that although 30 per cent of total sales of all models of audio cassettes were analyzed by the EC, the models chosen were arbitrarily selected and were unrepresentative of all sales. In relation to the total sales of Sony, 45 per cent of the sales examined were of the normal category, whereas only 10 per cent of the sales of the chrome type of audio cassettes were examined.

213. The EC argued that the Japanese exporters had been consulted concerning the selection of models that were most comparable to the best selling models of the EC producers. The investigating authorities had selected the most popular and best selling models of the Japanese exporters and had compared their prices to the models selected by the exporters as comparable. The EC argued that the Japanese exporters were thus "in charge" of the selection of models and had at no time raised objection to the selected models. The EC argued therefore that Japan's arguments concerning the models selected only reflected the position of one of the Japanese exporters. The EC argued that Japan's detailed arguments on selection of models were first raised in Japan's second submission and were thus raised too late. The EC argued that Japan therefore had no serious legal interest in the claim.

214. Japan complained that in analysing undercutting, the EC had "zeroed" sales at prices above those of EC products. Japan noted that absent "zeroing", any price undercutting in the German market was offset by overcutting in the other eleven EC markets.

215. Japan argued that the "zeroing" technique would conceal positive "overcutting" margins. Japan argued that in the case of Fuji, one of the Japanese exporters, undercutting margins were found for two out of four products. Japan argued that if the true margins had been used (i.e. for all models) no margin of undercutting might have been determined. Japan also argued that the EC may have used a "zeroing" technique within each product category. If "zeroing" within each product category had been used, the true margin of undercutting would have been much lower than the undercutting margin determined by the EC. Japan argued that the exporters had conducted an analysis of undercutting for 3 major exporters and had determined much lower margins of undercutting than the rates determined by the EC, and only in the case of sales by Sony in the German market.

216. Japan also complained that under a system of "zeroing" the sales at overcutting prices in other EC markets should have been taken into account, albeit at zero-undercutting levels. Those sales would thereby have a limited offsetting effect (although they could never reduce the overall overcutting to zero). By isolating the calculation for the German market the EC had prevented any such attenuating effect. Japan argued that the EC provided no explanation for the use of "zeroing" in the German market.

217. The EC noted that in its analysis of price undercutting, to prevent undercutting being hidden by price manipulation in different markets it had used a methodology described by Japan as "zeroing", in order to ensure that targeted undercutting was not hidden by selected "overcut" pricing. The EC noted that it had cumulated the exports of all models of all producers from each country, which were later cumulated with all data from all countries.

218. The EC argued that the phenomena of price undercutting and price suppression and depression were obviously related in many cases. The EC argued that an absence of price undercutting could be caused by the effects of previous price depression having taken place in the domestic market. The EC argued that "zeroing" was also necessary because price depression of the EC producers' prices may have already occurred in order to defend market share. The EC noted in this regard that a high level of price depression had been found on the EC market. The EC argued that subsequent analysis of the UK and German markets revealed selective price undercutting by the Japanese exporters in Germany.

219. Japan, in relation to the EC's defence of "zeroing" based on previous price depression, argued that a system of weighted averaging was a natural first choice for calculating undercutting and dumping margins, but that other systems may be justified in some cases. In this case, however, the EC's justification was unconvincing, no evidence of targeted undercutting had been provided, and if price depression had already occurred, separate and subsequent findings of price depression caused by dumping were called into question.

220. Japan noted that undercutting was confined to the German market, and that the volume of imports found to be undercutting was small. Japan noted that the two exporters found to be undercutting, Fuji and Sony, held respectively 3 per cent and 15 per cent of the EC market. It was apparent that the German market constituted a disproportionately small part of Sony's exports from Japan. Japan suggested that Sony's sales constituted around 10 per cent of all Japanese sales in Germany. On the whole, Japan argued, the Japanese exports to Germany and the other eleven EC countries were at prices which significantly overcut the EC owned producers, and the EC had not taken this into account. Japan argued that as price undercutting had been found in only one market, it was doubtful whether a sufficiently large volume of goods had undercut the domestic produced goods to make the undercutting "significant".

221. The EC relied on its analysis of undercutting in recital 66 of the Provisional Regulation to rebut Japan's arguments on the volume of undercut sales. The EC also argued that recital 84 of the Provisional Regulation had noted that the only market where Japan was found to have undercut the EC industry was the only market where the EC industry retained a substantial sales base.

222. Japan argued that the EC failed to meet the requirements of Article 3:2. Japan argued that the Basic Regulation was a poor transposition of the Agreement which had contributed to the EC's failure to meet the requirements of the Agreement.

223. Japan argued that the findings in the Provisional and Definitive Regulations did not support the claimed findings of price depression or suppression made to the Panel by the EC. Japan argued that whilst price undercutting could be simply established, the existence of price suppression and depression required an analysis of the causal link between the dumped imports and the reason for the pricing behaviour of the domestic industry. Japan argued that there was absent in the either the Provisional or Definitive Regulations sufficient information to show that the EC had satisfied the requirement in Article 3.2 that they consider "... whether the effect of such [dumped] imports is otherwise to depress prices to a significant degree or prevent price increases, which otherwise would have occurred, to a significant degree."

224. The EC argued that the footnote to Article 3:4 described as the "effects" of dumping the factors listed in Article 3:2 and 3:3. The finding of the presence of those factors meant that those effects were established, and therefore also meant that a causal link between dumping and price suppression and price depression was established. Therefore, the coincidence of dumped imports with "effects" established a causal relationship, unless another factor properly explained all the "effects".

225. Japan argued that when the term "consequent impact" was used in Article 3:1, the factors on which the impact was required to be consequent, were the volume of dumped imports, and "their effect on prices". Those price effects should be primary and not secondary. A finding of the primary effects of the volume of the dumped goods and their effect on prices was required. Article 3.2 required that the EC establish that the dumped imports had a direct price suppressing or depressing effect. It was insufficient to show that the price suppressing or depressing effect was consequential or secondary to the loss of profits caused to the domestic industry due to its lack of marketing activity.

226. Japan set out recital 79 of the EC Provisional Regulation. That recital provided that:

Japan argued that this recital was merely an observation regarding the coincidence of events, which did not amount to a finding that increasing volume of itself caused the price effects. Japan argued that there was a need for a finding that the increasing volume by itself caused the price effects. Japan noted that in the next recital increased volume was dealt with in the context of volume rather than price effects.

227. Japan noted that recital 83 of the Provisional Regulation stated as follows:

228. Japan argued that recital 83 raised interesting factual and theoretical issues. Japan argued that examination of profit levels suggested that there was no significant difference in profit levels on sales in the Japanese and EC markets. Japan also argued that the use of "zeroing" and "asymmetry" would have reduced the profitability of the Japanese exporter's export prices, and made their domestic sales appear more profitable. Japan also argued that there was no evidence for the finding that high profits from the Japanese market had a price effect on EC prices. Japan noted that the EC only analyzed profit levels from 1988, whereas the phenomenon alleged by the EC occurred during earlier years. Japan recalled that the significantly increased volume of sales occurred not among imported products, but in cassettes manufactured within the EC. An increase from 142 million units to 154 million units over the period would not have improved economies of scale. Japan argued that the EC had not demonstrated that any of the domestic sales were below the cost of manufacture, and in reply to the EC's allegation in relation to sales by TDK Electronics Europe ("TEE"), argued that it was irrelevant to the present issue.

229. Japan set out recital 84 of the Provisional Regulation, which provided as follows:

"(c) Effect of the prices of the dumped imports

230. Japan argued that recital 84 implied that if the prices of the Japanese exporters had been different the EC domestic industry would not have been forced to try to undersell them. Japan suggested that if Japanese exporters had raised their prices it was not clear that the EC industry would no longer have been forced to undersell the Japanese exporters. It was not plausible to suggest that the sale price of Japanese cassettes restricted the ability of EC domestic producers to raise the selling prices of their cassettes. Japan argued that in any event, the EC made no attempt to explain the assertion in recital 84, nor to provide evidence to support it. Consequently, Japan argued, the EC had failed to satisfy Article 3:2, and to meet the obligation in 3:1 to make an injury determination on the basis of an objective examination.

231. Japan argued that the fact that the Japanese brands tend to be market leader and price setter in the EC does not itself explain why the prices of EC brands would be lower than otherwise. Japan argued that the EC had improperly relied on the position of the Japanese exporters as market leaders to show that price depression and suppression were caused by the dumped goods.

232. Japan argued that none of the recitals relied on by the EC as constituting determinations of price depression and suppression demonstrated the existence of "positive evidence" or showed that an "objective examination" of the causal link between the dumped imports and the price behaviour had been undertaken.

233. The EC argued that Japan disputed partly the facts and partly the conclusions drawn from those facts by the EC. The EC argued that it had met the requirements of Article 3:2 "to consider whether the effect of such imports is otherwise to depress prices to a significant degree or prevent price increases, which otherwise could have occurred, to a significant degree". Those factors were properly considered in the recitals noted by Japan. The EC objected to Japan raising a complaint of insufficient reasoning under Article 8:5 in relation to the recitals quoted, as the EC argued that such a complaint had not been properly raised.

234. The EC argued that there was no doubt that prices of EC producers were depressed. There had been a general price erosion of 12 per cent between 1985 and 1988 on a weighted average basis (which was substantially higher in Germany). Prices of imports dropped 19 per cent over the same period. Over the same period the Japanese companies were market leaders and the EC industry incurred losses. The EC argued that these were incontrovertible signs of price depression and suppression. Together with the volume increases by Korean exporters and undercutting practised by Korean exporters, the findings made were fully justified, and no breach of the Agreement had been established by Japan.

235. The EC argued that the result of the price effects was that the domestic industry's volume decreased by 8.5 per cent, market share decreased from 27 to 19 per cent, capacity utilisation decreased from 100 to 77 per cent, and price erosion of 12 per cent was suffered. The EC argued that all of these factors resulted in losses and insufficient profitability.

236. In relation to Japan's claim that recital 79 did not amount to a claim that increased volume caused price effects, the EC argued that Japan's statement was irrelevant, as it was one of the elements considered, and that a coincidence of all the elements mentioned in recital 79 was highly relevant.

237. The EC argued that in the following recitals of the Provisional Regulation there was to be found the investigating authority's analysis of price depression and suppression: recital 72 (price erosion for all audio cassettes between 1985 and 1988 by 12 per cent on a weighted average basis) 76, 77, 79 (price erosion and undercutting and price suppression) 84, (price undercutting on the German market up to a maximum of 18.5 per cent) 86, (forced reduction in prices) 87, (price competition from Korean exports) and 107 (necessary price increase and price depression). In the Definitive Regulation there was evidence of consideration of price depression and suppression to a significant degree in recitals 26, (price erosion, price undercutting) 33, (price erosion and undercutting) 38, (attempt by an EC producer to retain its market share against the pressure of dumped imports) and 39 (large volumes of sales forced down the prices of the EC industry, in particular in the EC's most important national market where significant price undercutting was found). The EC asked the panel to read the references in the Regulations in the context of the extensive disclosure conference normally held with exporters by the EC. Such disclosure conferences normally covered all aspects of the case.

238. The EC noted that Japan disputed the facts which were the basis for the findings of price erosion and price suppression in recital 79 of the Provisional Regulation, and that Japan relied on the word "coincides" in that recital to argue that the recital's finding that the increased volume caused the price effects was improperly made. The EC argued that the recital showed that the effects of dumping had been properly considered. The EC also argued that if the Japanese imports had been sold without a dumping margin, the EC industry would have suffered no price depression or loss of market share.

239. In relation to the EC's request that the Panel note that extensive disclosure was given to the exporters, Japan argued that consultations with exporters had revealed no substantial consultations with the exporters in relation to the findings of price effects. Japan noted that the consultations were held in the context of the setting of the lesser duty rate, and that those consultations were not held in order to discharge the EC's obligations under Article 3:2.

240. The EC argued that the Japanese exporters' promotional expenditure, financed by high domestic profits, enabled them to impose their brand image, and thereby to increase sales volume. In addition, that increase in volume enabled more efficient economies of scale, and high profits through high priced domestic sales. The EC noted that the Japanese market for audio cassettes was less than 2 per cent supplied by imports. The EC considered this a point of importance, as in its view, dumping most often occurred when the exporter enjoyed a position of market dominance in the country of export. The EC noted also that Japanese owned companies supplied 65.1 per cent of the EC market, from production in Japan, Korea and from the EC, and that Japan held in effect a near monopoly over the EC market.

241. The EC argued that the Agreement only required it to make findings and did not oblige it to provide evidence that exporters funded high promotional activity through high profits, in light of its findings concerning the state of competition on the Japanese market, and the resulting high levels of profit achieved through the dumped sales. The EC recalled that in constructing the export price independent importers' profits were found to average 5 per cent, whereas profit margins in the Japanese market were in the range of 20 per cent. In this context, the EC also argued that because of the dumping, the EC industry found itself in a position in which it was unable to engage in promotional efforts to the same level as the Japanese exporters.

242. The EC also noted in relation to Japan's comments on recital 83 of the Provisional Regulation, that examination of TDK's responses to the enquiry revealed imports were sold below their cost of production through their German subsidiary TEE. The cassettes sold below cost comprised more than 50 per cent of the dumped sales to the EC which were analyzed.

243. Japan denied that Japanese or Japanese owned sellers of audio cassettes in Japan had a "near-monopoly" position either in Japan or the EC. Japan submitted that Japanese or Japanese owned sellers of audio cassettes vigorously competed with each other both in Japan and in the EC.

244. Japan also criticised the EC for supplying no evidence for the assumptions that underlaid its findings that exporters funded their promotional activity through high levels of profit and sales achieved on the domestic market. Japan also argued that comparing the profit levels of importers, which included selling activities, with those of manufacturers, which included production and selling activities would be a meaningless calculation. It claimed that analysis of profit rates of Japanese manufacturers revealed that profits achieved on sales in the EC and in Japan were similar.

245. Japan argued that the EC's assumption that high profits were used to finance promotion was also flawed, because it contradicted the EC's finding of dumping. If Japanese companies had incurred large expenditure on promoting sales in the EC to increase brand awareness as per the EC's assumption, why should they dump and set low prices which would be at odds with the image of high quality which was central to the promotion campaign.

246. Japan recalled that the EC's finding concerning high profits used to increase brand awareness was flawed by "asymmetry" and "zeroing". In this context, Japan also argued that the levels of promotional expenditure of the German producer BASF and the Japanese exporters were at similar levels during the period under enquiry.

247. Japan further argued that the EC's reliance on the home market profits as providing the means to expend more on advertising, leading in turn to greater market success, was in conflict with research by the exporters suggesting that their success was due to better marketing. Japan argued that recital 86 of the Provisional Regulation had no relationship to the factors contained in Article 3:2, and instead referred to non-price effects. Japan argued that recital 86 of the Provisional Regulation said that due to the weakened state of the industry, price depression and/or suppression were a form of consequential injury.

248. Japan argued that the size of the dumping margins was a factor that should have been taken into account, especially in making the findings contained in recital 39 of the Definitive Regulation, in which the EC found that high profits were used to finance sales expenditure. Japan also noted that the EC had argued that absent the high profit obtained through dumping margins the Japanese exporters would have not caused price depression and not have enjoyed such a large market presence.

249. The EC noted that because an injury margin had been established in this case, the actual dumping margins of the exported goods were and need not have been considered during the injury determination. The EC noted that recital 39 of the Definitive Regulation, and recital 83 of the Provisional Regulation showed that the size of the dumping margin had not been taken into account. The EC argued that those recitals merely showed the effects of the dumping on volumes and sales. The size of the dumping margins properly played no role in the injury determination.

250. Japan argued that the EC had failed to demonstrate that the Japanese exports had, through the effects of dumping, caused injury, as required by Article 3:4. To the contrary, Japan argued that the evidence of the nature of the market and the success of the Japanese producers suggested that any injury suffered by the domestic industry was not caused by reason of dumping by the Japanese exporters. Rather any injury was due to factors within the control of the domestic industry, especially a failure to match the Japanese producers in vital non-price-related aspects of fair competition.

251. Japan complained that although certain other factors were considered by the investigating authority, the conclusion reached following their consideration was so unreasonable that no reasonable authority could have reached such a conclusion.

252. The EC argued that unless all of the material injury found to have occurred could be attributed to causes other than dumping, its determination that the dumped goods caused injury could not be impugned. If the effects of dumping as defined in Articles 3:1 and 3:2 were present, and coincided with the presence of dumped imports, Article 3:4 deemed a causal link to exist. Unless any "other factors" were brought to the attention of the investigating authorities, there was no duty imposed on the investigating authorities to search such other factors out.

253. Japan relied on the first sentence of Article 3:4 to show that there was an obligation to prove a causal link between the dumped imports and material injury, and to ensure that injury caused by "other factors" was not attributed to the dumped imports. Japan argued that in certain cases there could be important differences between the tests contained in Article 3:4 and Article 3:2 and 3:3. Articles 3:2 and 3:3 dealt with the effects of the dumped imports. Article 3:4 dealt with the effects of dumping.

254. Japan argued that the phrase "through the effects of dumping" added an element to the issue of causation which was not found in the earlier paragraphs of Article 3. Such an interpretation followed from a natural reading of the Article, and was consistent with the concept of causation.

255. Japan disagreed with certain findings of the Salmon Panel. Japan argued that the Panel had given too much weight to the footnote to Article 3:4. The Panel had distorted the meaning of the Article by rearranging the balance between the various aspects. The Panel had downplayed the significance of the words "through the effects of dumping" in favour of an interpretation which linked the injury suffered to the dumped imports, and not to the incidents of dumping.

256. The Panel had also not properly applied the principle of "effective interpretation". The Panel's interpretation had led to the conclusion that the first sentence of Article 3:4 had no effect and only repeated the test contained in Article 3:2. The Panel had effectively concluded that the Agreement's negotiators had changed their minds after drafting the first sentence of Article 3:4, but instead of deleting it had decided to add a footnote in order to make the sentence meaningless. Japan argued that such behaviour on the part of the negotiators was implausible.

257. Japan argued that the provision should have been interpreted to conclude that if dumped imports were causing injury, but not through the effects of dumping, i.e. through other factors such as quality, or marketing strategy, they could not be determined to have caused injury within the terms Article 3:4. Japan argued that in this case the EC had gone even beyond the approach of the Salmon Panel by suggesting that causation existed unless there were other factors to which all of the injury could be attributed.

258. Japan argued that the burden of proving that a causal link could be established laid with the EC. Japan argued that it was not incumbent upon Japan to prove that a causal link did not exist. Japan argued that the injury suffered by the EC producers was caused by other factors, such as a failure to mount an effective advertising campaign, a failure to invest sufficient sums of money on marketing, a failure to produce new models or new packaging, whilst Japanese exporters were innovative in all those areas. The EC producers were involved in pollution of the Rhine, and in defective products sold in Germany and Italy. The EC industry lacked a sound distribution policy. All of those factors created a poor image among retailers and wholesalers. Japan submitted an Annex IX to its first submission to the Panel which contained a study of the EC market carried out on behalf of TDK.

259. Japan argued that the survey of the market and of consumers revealed a poor customer image of EC industry products. In addition, ill considered pricing strategies, loss of major contracts (to higher priced Japanese products) and a lack of new models or packaging had greatly diminished profitability. Japan also recalled its earlier criticism of the EC's allegation that profits achieved in Japan had been allocated to the marketing of Japanese produced audio cassettes.

260. Japan also argued that the use of an "asymmetrical" comparison and "zeroing" during establishment of the dumping margin effectively made it impossible for the EC to properly determine whether or not injury was caused through the "effects" of dumping.

261. Japan noted that even if the Panel determined that cumulation of Japanese and Korean imports were appropriate in the present case, it was necessary to revisit the presence of the cheap Korean imports in conducting a causation analysis. Japan argued that as the Japanese owned producers within the EC withstood the large volume of cheaper priced Korean imports, there must be some other explanation for the injury suffered by the domestic industry which the EC had attributed to the dumped goods, such as the EC industry's weak marketing strategies.

262. Japan also argued that in examining causal link, the most useful approach would be to examine whether the EC industry would have been able to raise its prices if the prices of the Japanese exported products had been raised. Japan relied on the Grain Corn Panel (paragraph 5.2.9) as authority for such a test. Such an examination revealed that less than 44 per cent of the cassettes sold on the EC market by the Japanese companies were exported from Japan. The rest of the cassettes sold on the EC market by the Japanese companies were either manufactured in the EC or were exported from a third country at undumped prices. Japan argued that as a consequence it was difficult to conclude that if the prices of the allegedly dumped cassettes were raised to eliminate dumping that injury would not continue, and any such raising of prices would only cause the Japanese companies to source their goods outside Japan.

263. The EC recalled that it did consider other possible factors but determined that the injury found to exist could not all be attributed to factors other than the dumped imports.

264. The EC argued that Article 3 only laid down procedural obligations with respect to causation. The EC argued that it must be demonstrated that the injury was caused "through the effects of dumping". Due to the content of footnote 4, this meant that it would be permissible to determine that a causal link existed if some of the factors and indices mentioned in paragraphs 2 and 3 were found to be present. The EC argued that the Agreement contained no definition of causal link. Injury was only required to be demonstrated through the effects of dumping. From both a theoretic and practical perspective, it was difficult to do more than establish the coincidence of certain phenomena. In an investigation the presence of the dumped imports in the market and some of the "effects" as described in Article 3:2 and 3:3 of the dumped imports was sufficient to establish a causal link to material injury.

265. It was unnecessary for all of the injury factors to be present due to the last sentence of Article 3:3. The EC noted that a causal link could not properly be established if other factors were injuring the industry at the same time, provided that all of the injury could be attributed to those other factors. The EC argued that the Panel could only determine whether in assessing the evidence the EC had violated a provision of the Agreement, made a manifest error, or acted arbitrarily. The EC argued that Japan's arguments did not raise such a claim.

266. The EC argued that it did consider certain "other factors" but found that the injury could not be attributed in its entirety to those factors. The EC argued that it had no obligation under Article 3:4 of the Agreement to actively examine whether there were "other factors" which might have caused the injury suffered by the domestic industry, unless those "other factors" were either mentioned by the interested parties or otherwise emerged during the course of the investigation. The EC argued that requiring investigating authorities to actively search for "other factors" causing injury would frustrate the enquiry process. The EC submitted that the extensive discussion of "other factors" contained in both Regulations made clear that the EC had demonstrated 'due diligence' in searching for and assessing "other factors".

267. The EC said that the argument by Japan that the use of an "asymmetrical" comparison and "zeroing" during establishment of the dumping margin effectively made it impossible for the EC to determine whether or not injury was caused through the effect of dumping, could neither disprove nor deny the existence of injury caused by dumping.

268. The EC considered that it had properly considered the issue of marketing or management failure. To prove that proper consideration had occurred, the EC relied on recitals 88-91 of the Provisional Regulation. It was not proper for the Panel to determine as a matter of fact that the marketing or management failure alleged by Japan had caused the material injury sustained by the domestic industry.

269. In relation to Japan's argument that the EC industry had failed to match the Japanese producers in non-price-related aspects of fair competition, the EC recalled its earlier argument that Japanese exporters had allocated high profits achieved on the domestic market towards payment for marketing and promotional strategies within the EC. The EC argued that Japan's argument actually strengthened the finding of injury, as the EC industry had been forced to cut its expenditure on marketing because of the effects of the dumping. The EC argued that the Japanese owned EC producers were also able to spend more on marketing and other non-price related aspects of competition due to the high profits realised by their parent companies on the Japanese market. As evidence for that finding the EC relied on its study of profit levels of independent importers, and its study of normal value profit rates. The obvious explanation for the Japanese owned EC located producers being able to withstand the impact of low priced Korean imports was their "deep pockets" filled by high prices achieved on their home market. The EC also argued that Japanese exporters were in a position to control the EC market due to their market leadership.

270. In reply to Japan's argument based on the Grain Corn Panel, the EC argued that if the Japanese producers had raised their prices to a level of the normal value the market in the EC would have developed in a totally different manner. The EC distinguished the passage relied on by Japan from the Grain Corn Panel, on the basis that the EC had not engaged in any behaviour that would have distorted the market. The price suppression and depression resulted from dumped Japanese and Korean imports. Contrary to the Grain Corn Panel, in this case there was no factor akin to the effect of the world price for grain corn which was determined by the level of subsidisation paid by the United States.

271. Canada made a written submission as a third party. Canada did not seek an opportunity to make an oral presentation to the panel.

272. Canada argued that the determination of injury required by Article 3 was an important obligation, to which Article 3:4 was central. Article 3:2 required a consideration of whether significant price undercutting or suppression had occurred. Recital 26 of the Definitive Regulation indicated that the investigating authorities had determined that significant price undercutting was only present in the German market. Although recital 84 of the Provisional Regulation did mention price suppression in those markets the Definitive Regulation contained no discussion of price undercutting or suppression in the United Kingdom or French markets.

273. In the absence of express references to consideration of those elements, Canada found it difficult to accept the EC's conclusions that injury had been caused to the domestic industry by Japanese exports.

274. Canada argued that Article 3:1 required "positive evidence" and an "objective examination of...(a) the volume of dumped imports and their effect on prices in the domestic market for like products...". Article 3:4 of the Agreement required that there be evidence that the injury was caused by the dumped imports. Canada noted that the market share of the EC producers fell from 42 per cent to 35 per cent during the period of investigation. There was an increase in absolute volume by 8 per cent, but the EC market grew by 30 per cent over the period of enquiry. Canada argued that it was difficult to accept that such a small increase in such a strongly growing market could cause injury.

275. Canada noted that Japan had argued that material injury could be attributed to dumped Korean imports, or to the production of Japanese owned manufacturers located in the EC. Canada noted that in recital 36 of the Definitive Regulation the EC had concluded Japanese exports could not be isolated from other dumped imports. The EC had noted that despite a decrease in market share for the Japanese exports, Japanese exports had retained a larger market share than the domestic industry. Canada argued that the conclusions of the EC suggested that the mere retention of a large market share by the Japanese exports warranted a finding of material injury, as the domestic producers were presumed to have warranted a larger market share than they were found to have had.

276. Canada accepted that exporters who maintained a large market share and who were price leaders could injure the domestic industry. Canada noted in this regard that in recital 78 of the Provisional Regulation the EC had stated that price was only one competition factor. The EC had therefore found that competition could also be through non-price factors such as superior promotion and marketing. Recital 84 stated that a large market share had been retained in Germany despite price undercutting. Canada argued that the retention of market share supported a conclusion that competition in that market had been due to factors other than price. On the basis that the competition was largely due to factors other than price, Canada doubted that a causal link between dumping and injury had been properly established.

277. Japan requested that the Definitive Regulation be revoked, and that the EC be requested to bring its Basic Regulation and practice into conformity with the Agreement.

278. The EC argued that Japan's requests that the Definitive Regulation be revoked, and that the EC bring the provisions of its Basic regulation and its application into conformity with the Agreement were manifestly inadmissible, as such remedies were not within the competence of the Panel. Alternatively they required factual determinations of the Panel which were not within the Panel's competence.

279. Japan argued that a pre-condition for imposition of an anti-dumping duty was that the principles in Article 1 of the Agreement must be respected. Japan argued that if the Panel found that a finding of injurious dumping had been improperly made, the decision should be revoked, and the duties reimbursed. Japan argued that the reasoning in the Report of the Panel in New Zealand - Imports of electrical transformers from Finland, adopted 18 July 1985, BISD 32S/55, paragraphs 4.10 and 4.11 ("Transformers from Finland" Panel) supported its request for revocation of the order and reimbursement of the duties. Japan noted also that in the Report of the Panel in United States - Countervailing duties on fresh, chilled and frozen pork from Canada, adopted 11 July 1991, BISD 32S/30 ("Pork" Panel) the Panel had supported the decision of the Transformers from Finland Panel to recommend revocation in the circumstances of that case, even though in the circumstances of the Pork Panel the Panel had not explicitly determined that revocation was appropriate. Japan also cited the Softwood Lumber Panel as authority for the proposition that unlawfully taken countervailing duties should be refunded. Japan argued that in this case it would be irrational to refund duties illegally taken and not to recommend the revocation of the illegal decision under which they were collected.

280. Japan noted that the Salmon Panel had found that as it was possible that a reconsideration of the case would result in imposition of a different rate of duty as opposed to a negative finding, the Panel would not recommend revocation. Japan argued that in this case it would be irrational to refund duties illegally taken and not to recommend the revocation of the illegal decision under which they were collected. However, Japan argued that the circumstances of the present case were quite different from those facing the Salmon Panel. In this case the Panel had been presented with argument and evidence that showed that the effect of the EC's zeroing and asymmetrical approach was always to prejudice the exporter. Japan also argued that the effect of the burden of proof was to require the Panel, when deciding whether or not to recommend revocation, to weigh the relative degree of prejudice suffered by the exporter against that suffered by the domestic industry. This was because the burden of proof laid with the party seeking to claim the benefit of an exception to the GATT. Since it was presumed that an exception to the GATT prejudiced an exporting country, once a contravention was established the Panel should be cognisant of the effect on exporters of the continuation of the illegal measure.

281. Japan also argued that the process of dispute settlement would be hampered if no remedies were available to successful complainants. Japan also argued that the lack of specific recommendations such as revocation would reduce the incentives on signatories to comply with GATT rules.

282. The EC argued that a finding that any part of the dumping calculation was conducted in an Agreement-inconsistent manner was an insufficient basis to recommend revocation of the entire Regulation, because recalculation of the amount of duty payable in an Agreement-consistent manner would not necessarily result in a determination of no dumping. Revocation of the entire Regulation was also inappropriate due to the effect of the "lesser duty rule". It would not be proper to recommend revocation of the Regulation if the Panel found fault with the injury determination, or the causal link thereto, as the suitable remedy would be to request speedy reconsideration of the decision in an manner consistent with the Agreement.

283. The EC also argued that the Panel was not in a position to reassess the findings of fact and law of the investigating authority. The EC argued that the Polyacetal Resins Panel made clear that the task of a panel was only to review the determination made by the authority for consistency with the Agreement, on the basis of the findings of fact and conclusions on matters of law contained in the public determination, and to determine whether the party had acted arbitrarily or made a manifest error.

284. The EC also argued that if Japan's claims were upheld, the possible adjustments to be made would not affect the anti-dumping duties to be collected. The EC argued that in the Salmon Panel the Panel had not recommended revocation of measures because it could not be satisfied that alteration of the United States' practices so as to conform with the Panel's recommendations would necessarily result in a finding that no dumping had occurred. The Panel had decided that it would not recommend revocation if the only likely outcome of implementation of the Panel's findings was to alter the amount of duty to be collected. The EC relied also on paragraph 4.11 of the Pork Panel for the same proposition. The EC argued that in any case, in order to provide for a flexibility in implementation the only appropriate measure would be a recommendation that the EC bring its measures into conformity with the Agreement.

285. Japan argued that Article 16:6(a) of the Agreement obliged Parties to ensure that all laws and procedures complied with the Agreement. Japan argued that as "zeroing" and "asymmetry" were applied as a matter of course they should be classified as mandatory, even though it was possible that the authorities would in a particular case be free to not apply those rules. Japan noted that certain cases before the European Court of Justice had upheld application of the rules and that therefore the rules were ordinarily applied.

286. Japan argued that as those rules were invariably applied by the administering authorities they should be considered to be mandatory. In the alternative Japan argued that the Panel could properly make a recommendation that the law and practice of the EC be brought into conformity with the Agreement by the Panel, as the practices of "zeroing" and "asymmetry" infringed the Agreement. Japan argued that the Report of the Panel on European Economic Community - Regulation of imports of parts and components, adopted 16 May 1990, BISD 37S/132, ("Parts and components" Panel) supported this approach, as in that case the Panel had recommended that the EC bring its law and practice into conformity with the Agreement even though the Panel had determined that the legislation was discretionary in character. The Panel had also noted that it would be desirable for the EC to repeal the legislation. Japan argued that the Panel's decision was consistent with an argument that Articles III and XI of the GATT had the objective of protecting expectations in relation to trade. That objective could not be attained unless parties had standing to challenge mandatory trade legislation. Japan therefore requested that the Panel recommend that the EC bring its law and practice into conformity with the Agreement in relation to zeroing and asymmetry.

287. The EC argued that the Superfund Panel (paragraph 5.2.9) was authority for the proposition that the Panel could not review the consistency of the Basic Regulation itself with the Agreement but only the particular application of the Basic Regulation. The EC also argued, consistent with the Parts and components Panel, that if the Panel determined that one of the requirements of the Agreement had been breached, the Panel could not recommend revocation of the Basic Regulation, in so far as it was "non-mandatory" in character. The EC argued that the Parts and components Panel had determined that Article 13:10 of the EC's Basic Regulation was non-mandatory. This was because that provision only gave the relevant authorities the "... possibility to act inconsistently with a GATT provision...". The EC argued that on the whole the Basic Regulation merely established a legal framework for the authorities of the EC, and thus that the Panel could not recommend revocation of the Basic Regulation.

288. The EC argued that the Basic Regulation was not mandatory because the language of Article 7(1) and 12(1), though prescribing certain steps, made those steps conditional on certain findings and thereby on the judgement and discretion of the investigating authorities. The EC also argued that two practices complained of by Japan, so called "asymmetrical" comparison and "zeroing", whilst representing standard practice, were not mandatorily required by the Basic Regulation, and that in an appropriate case a different approach could be followed. The EC argued that the Panel could not order the bringing into conformity with the Agreement of non-mandatory legislation, and that as the practices of so called "zeroing" and so called "asymmetry" were clearly not mandatory, the Panel could not order modification of these practices.

289. Japan argued that there was a well established principle that duties paid pursuant to a measure inconsistent with the Agreement should be reimbursed. Japan relied on the Report of the Panel in United States - Anti-Dumping Duties on gray Portland cement and cement clinker from Mexico, unadopted, ADP/82, paragraph 5.40) as a recent statement of that principle. Japan argued that other panels (Transformers from Finland; Report of the Panel in Canada - Imposition of countervailing duties on imports of manufacturing beef from the EEC, unadopted, SCM/85, paragraph 5.17 ("Manufacturing beef; and Swedish Steel) had taken for granted that reimbursement was an ordinary remedy.

290. Japan argued that if the Panel decided that the margin of dumping had been miscalculated, the reasoning in the Pork Panel also suggested reimbursement was an ordinary remedy. Japan argued that the Panel in that case had determined that the duties were imposed inconsistently with Article VI of the Agreement. The Panel had recommended that the United States reach a proper determination consistent with Article VI, and reimburse the duties determined to be collected above the amount of the countervailable subsidy.

291. The EC argued that it was established GATT law that the Panel could not order specific remedies which were of the same nature as those ordinarily ordered by national Courts. The Panel could only order that the EC bring its practices into conformity with the Agreement. Therefore, the Panel should not recommend reimbursement.

292. The EC had argued that certain claims raised by Japan were not admissible because they had not been raised by Japan during consultations and conciliation and/or were not within the terms of reference of the Panel.

293. The Panel noted that the EC had in certain cases indicated that claims were not raised in the "consultation/conciliation" phase of the proceedings. In other instances, however, the EC had referred to the same claims as not having been raised during conciliation. In light of this ambiguity, and taking into account that the information presented by the EC to the Panel focused on the conciliation phase of the proceeding, the Panel considered the argument of the EC was that certain claims had not been raised in conciliation and/or were not within the terms of reference.

294. The Panel recalled that Article 15 of the Agreement governed the process of consultations, conciliation and dispute settlement under the Agreement. Under Article 15:2, a Party that considered that any benefit accruing to it under the Agreement was being nullified or impaired could request consultations with the Party or Parties in question with a view to achieving a mutually satisfactory resolution of the matter. Under Article 15:3, if the Party considered that the consultations had failed to achieve a mutually satisfactory solution of the matter, it could refer the matter to the Committee on Anti-Dumping Practices ("the Committee") for conciliation. Finally, if no mutually agreed solution was reached after detailed examination by the Committee within three months, the Committee was required at the request of any party to the dispute to establish a panel to examine the matter.

295. The Panel noted that a detailed analysis of the provisions of Article 15 had recently been performed by the Panel on United States - Imposition of anti-dumping duties on imports of fresh and chilled Atlantic salmon from Norway, adopted 26 April 1994, ADP/87, paragraphs 331-346 ("the Salmon Panel"). The Salmon Panel had noted that Article 15 referred with respect to each stage of the dispute to the term "matter." In its view this choice of words reflected a decision to establish a three-step process of settlement of a dispute between Parties concerning a single matter and the individual claims of which that matter was composed, in which panel examination of a matter would be preceded by consultations concerning that matter and conciliation concerning that same matter. Accordingly, that Panel had concluded that, for a claim to be properly before a panel, it had to be within the panel's terms of reference and it had to have been identified during prior stages of the dispute settlement process.

296. In support of its conclusions, the Salmon Panel had focused on the objectives of each stage in the dispute settlement process. With respect to conciliation, it was the view of the Salmon Panel that the obligation to seek a mutually satisfactory solution during conciliation could not be fulfilled unless the individual claims of which a matter were composed were set out in the conciliation phase of the proceeding. In addition, the Committee was required under Article 15:5 to conduct a "detailed examination" of the matter during the conciliation process and was authorized under note 15 to draw the Parties' attention to those cases where there were no reasonable bases supporting the allegations made. These provisions implied that in the conciliation process individual legal claims and their bases would be examined and each Committee member would be able to express its views thereon.

297. With respect to the terms of reference, the Salmon Panel considered that terms of reference served two purposes: definition of the scope of a panel proceeding, and provision of notice to the defending Party and to other Parties that could be affected by the panel decision and the outcome of the dispute. The notice function of terms of reference was particularly important in providing the basis for each Party to determine how its interests might be affected and whether it would wish to exercise its right to participate in a dispute as an interested third party. The Panel had concluded that in light of these objectives a matter, including each claim composing that matter, could not be considered by a panel unless it was within the terms of reference of the panel.

298. The Panel noted that the views of previous panels were not binding on other panels. However, the Panel found the reasoning of the Salmon Panel as outlined above to be convincing.

299. The Panel noted Japan's view that conciliation in GATT and international practice meant the appointment of an impartial conciliator to thoroughly examine the facts and arguments; that, due to the unwillingness of disputants to reconcile their differences, the conciliation meeting held in this and other cases was no more than the commencement phase of the conciliation process; and that for the purposes of that phase the document containing Japan's request for conciliation (ADP/79) was sufficient. The Panel could not preclude that conciliation might include appointment of an impartial conciliator in some cases. However, Article 15 made no reference to any conciliator other than the Committee itself. Thus, the Committee was required to meet to review the matter and to encourage the parties through its good offices to develop a mutually acceptable solution, and was authorized to draw the parties' attention to those cases where, in its view, there were no reasonable bases supporting the allegations made. The Panel therefore considered that it must consider the requirements of Article 15 in light of the role assigned to the Committee by that Article.

300. The Panel further noted Japan's argument that it was not necessary that all claims be raised in conciliation because the role of the Committee in the conciliation process was exhausted once it became clear that reconciliation was not possible on even one major issue. However, the Panel did not consider that the objective of the conciliation process was restricted to resolving disputes in their entirety. To the contrary, a conciliation which, for example, resulted in agreement by a party to the dispute to modify certain practices would be a valuable result which would further the objectives of the conciliation process. Similarly, if certain claims advanced during conciliation lacked a reasonable basis but others did not, the Committee as a whole (or individual members) could point out the weaknesses of certain claims and therefore facilitate a narrowing of the work of a possible future panel. In this respect, the Panel noted that it was not uncommon for parties themselves to decide not to bring issues raised in conciliation before a panel.

301. Finally, the Panel recognized Japan's argument that, when determining the consequences of a failure to raise a claim during a previous stage in the dispute settlement process, a panel should consider the extent to which significant interests had been prejudiced by that failure. However, the Panel did not consider that such an assessment would be either appropriate or feasible. Specifically, the Panel could not understand the basis on which a panel could after the fact consider whether certain claims might have been resolved in previous stages of the dispute settlement process had those claims been raised during those stages of the process. Nor would a panel after the fact have a basis on which to consider whether the rights of third parties to protect their interests through participation in the panel process were jeopardized by the failure of a complainant to raise a claim at the time it requested the establishment of a panel.

302. For the foregoing reasons, the Panel concluded that, for a claim to be properly before a panel, it had to have been identified during the conciliation stage of the dispute settlement process and to be within the panel's terms of reference.

303. The Panel emphasized that its conclusions related to the need to identify "claims." This did not mean that a party to a dispute was required to present its full arguments before a panel had been established. Rather, the Panel considered that, to the extent a Party sought establishment of a panel to adjudicate the consistency of another Party's actions with the Agreement, the complaining Party should at a minimum have identified during conciliation and in its request for establishment of a panel the action or factual situation allegedly giving rise to an inconsistency with the Agreement and the obligation under the Agreement that allegedly was violated.

304. The Panel then turned to the specific issues that the EC alleged could not be considered by the Panel in light of the foregoing considerations.

305. The Panel first considered the argument of the EC that certain claims were not within the terms of reference of the Panel.

306. The Panel noted that its terms of reference were:

The Panel further noted that the parties to the dispute had submitted to the Chairman of the Committee for transmission to the Panel an Explanatory Note to ADP/85/Add.1. See Annex. The parties to the dispute had agreed that the clarifications found in this Explanatory Note represented a statement on which the Panel could rely should it need to interpret its terms of reference.

307. The EC contended that Japan's claim that the EC's methodology for selecting the export models to be used in calculating a margin of price undercutting was inconsistent with Article 3 was not within the terms of reference of the Panel. Japan had responded that this claim was raised in paragraphs 21 and 22 of document ADP/85/Add.1, which stated as follows:

308. The Panel noted that paragraph 22 advanced three specific bases for Japan's assertion that the EC's actions with respect to the consideration of price undercutting were inconsistent with the Agreement. The first such basis, the "zeroing" of overcutting, related not to the selection of models to be used in calculating a margin of price undercutting but to the way the price comparison was performed once the export and domestic models to be compared were selected. With respect to the second such basis, that the price comparison was with the product of only one domestic producer, this related to the selection of the domestic product to be compared, whereas the claim to which the EC had raised its preliminary objection related to the representativeness of the Japanese exporters' models selected for comparison. The third such basis, that no reasonable person could conclude that there was "significant" price undercutting given the facts of this case, related to the conclusions drawn by the EC after it had performed its calculation of undercutting, not to the manner in which the EC determined the margin of undercutting. Thus, the Panel considered that Japan's claim regarding the EC's methodology for the selection of exporters' models was not within its terms of reference by reason of the bases stated in items (a) through (c) of paragraph 22.

309. The Panel next considered whether the claim of Japan could be considered to be within the terms of reference of the Panel by reason of paragraph 21 and the chapeau of paragraph 22. The Panel recalled that paragraph 21 simply stated that the EC's price undercutting methodology was defective because its methodology was not in accordance with the Code, while the chapeau to paragraph 22 stated that the methodology used by the EC to calculate an undercutting margin contained "arbitrary and prejudicial elements". The Panel further noted that the list of specific bases that followed was preceded by the phrase "[i]n particular," suggesting that the list might be illustrative rather than exclusive.

310. The Panel considered that, in determining whether the claim of Japan at issue was sufficiently raised by reason of paragraph 21 and the chapeau to paragraph 22, it should keep in mind the notice function served by the terms of reference. The Panel recalled that the terms of reference, inter alia, provided the basis for each Party to determine how its interests might be affected and whether it would wish to exercise its right to participate in a dispute as an interested third party. This, in turn, required that the terms of reference identify not only the obligation under the Agreement allegedly violated but also the action or factual situation allegedly giving rise to an inconsistency with the Agreement. In the view of the Panel, a statement that the EC's undercutting methodology was "defective" and contained "arbitrary and prejudicial elements," without any identification of the element or elements of the EC's methodology deemed to be inconsistent with the Agreement, did not identify the action or factual situation allegedly giving rise to an inconsistency with the Agreement with sufficient particularity to allow a potential third party to decide whether its interests might be affected such that it would exercise its right to participate in the proceeding.

311. For the foregoing reasons, the Panel concluded that Japan's claim that the EC's methodology for selecting the export models to be used in performing a comparison of price undercutting was inconsistent with Article 3 was not within the terms of reference of the Panel.

312. The EC contended that Japan's assertion that the Panel should consider the effect or lack thereof of an increase in prices by Japanese exporters was an issue that was not within the terms of reference of the Panel.

313. The Panel recalled that Japan had argued in its submissions that even if Japanese exporters had raised their export prices to the normal value, this would not have led to a general raising of prices in the EC market; rather, the principal consequence would have been that Japanese producers would have supplied the EC market from sources other than Japan. Japan cited the Report of the Panel in Canadian countervailing duties on grain corn from the United States, adopted 26 March 1992, SCM/140 ("the Grain Corn Panel") as support for the consideration of this factor "as an aid to resolving the issue of causation."

314. The Panel noted that paragraph 24 of ADP/85/Add.1 stated in relevant part as follows:

315. The Panel considered that paragraph 24 of ADP/85/Add.1 explicitly raised the issue that the EC contended was not within the Panel's terms of reference. The Panel noted that paragraph 24 did not cite the Grain Corn Panel as authority for the relevance of this issue. In the view of the Panel, however, a party to a dispute was not required to identify in its request for the establishment of a panel all legal authority relevant to a particular issue.

316. The Panel therefore concluded that the assertion by Japan that the Panel should consider the effect or lack thereof of an increase in prices by Japanese exporters was within the terms of reference of the Panel.

317. The EC had contended in its initial submission that Japan had sought in its first submission to raise a new claim relating to EC refund procedures which was not within the terms of reference of the Panel. Japan had responded that it was not asking the Panel to condemn the EC's refund methodology. It had raised the EC's refund methodology not as a primary claim but as a response to an anticipated defense by the EC.

318. The Panel noted the statement of Japan that it was not asserting a claim regarding the consistency of EC refund procedures with the Agreement, but was rather anticipating and responding to an argument by the EC. The Panel further noted that the EC had recognized, at the second meeting of the Panel, that Japan had "advance[d] this point as an argument and not as a claim," and had requested only that the Panel "register this and decide accordingly." Under these circumstances, the Panel considered that it need take no further action on the preliminary objection of the EC on this issue.

319. The Panel next examined the argument of the EC that certain claims of Japan were not raised during conciliation and therefore could not be considered by the Panel.

320. The Panel recalled that the EC had argued that Japan's allegations regarding the EC's refund procedure were not within the Panel's terms of reference. The EC had also argued that these allegations by Japan were not raised in the document requesting conciliation (ADP/79). For the reasons set forth by the Panel in the context of its discussion of the Panel's terms of reference (see paragraphs 317-318, supra), the Panel considered that it need take no further action on the preliminary objection of the EC on this issue.

321. The EC contended that Japan's claim that the EC's methodology for selecting the export models to be used in performing a comparison of price undercutting was inconsistent with Article 3 was not raised during the conciliation phase of the proceeding.

322. The Panel recalled that it had determined that this claim was not within the terms of reference of the Panel and could not be considered by the Panel. Consequently, the Panel did not consider it necessary to decide whether this claim had been properly raised during the conciliation phase of the proceeding.

323. The EC contended that Japan's claim on pages 70-73 of its first submission that the EC failed to take account of other factors that were causing injury was not raised by Japan during the conciliation phase of the proceeding. Japan responded that its claim regarding "other factors" was sufficiently raised in the first sentence of paragraph 26 of ADP/79, where it argued that the EC had failed to establish that dumped imports from Japan had, 'through the effects of dumping,' caused injury to the EC industry.

324. The Panel noted that Japan, in its first submission, had included a Part 3.3.2 entitled "Other factors." That Part was composed of three sections, entitled "Source of difficulties of the EC industry", "Success of Japanese companies," and "Effect of price increases." The EC, in its preliminary objection, had indicated that its objection related to the claims of Japan in pages 70-73 of its first submission, that is, to this entire Part. The first section (3.3.2.1) of this Part, "Source of difficulties of EC industry," asserted that any injury suffered by the EC industry was not by reason of dumping by Japanese importers, but was rather "due to factors within the control of the EC industry, especially a failure to match Japanese producers in vital non-price-related aspects of fair competition." The submission set forth a list of such factors, including inadequate investment by the EC industry on marketing, a failure to produce new models and packaging, damage to the EC industry's image caused by polluting the Rhine and producing defective products, a poor image among retailers and distributors, and poor distribution strategies. The second section (3.3.2.2) of this Part, "Success of Japanese companies," contended that Japanese producers "had survived the onslaught of cheap Korean imports unscathed," thus confirming that it was the weak marketing strategies of the EC industry that had caused its lack of success. The third section (3.3.2.3), "Effect of price increase," argued that, even if Japanese exporters had raised their prices to normal value, EC producers would not have been able to raise their prices, as Japanese producers would have supplied the EC market from sources other than Japan.

325. The Panel noted that Japan had relied on paragraph 26 of ADP/79 as its basis for the view that these claims were properly raised during the conciliation process. The Panel observed that paragraph 26 provided as follows:

326. The Panel noted that paragraph 26 of ADP/79 was essentially identical to paragraph 24 of ADP/85/Add.1, which was the basis for the Panel's terms of reference. The Panel further noted that it had considered whether the allegations found in the third section of this Part of Japan's submission ("Effect of price increase") had been raised in paragraph 24 of ADP/85/Add.1 and had concluded that it had been. Accordingly, the Panel concluded that, for the reasons set forth in paragraphs 312-316 above, the allegations raised by Japan with respect to the effect of price increases had been raised in the document requesting conciliation and therefore could be considered by the Panel.

327. The Panel next considered whether the issues raised in sections 1 and 2 of the Part of Japan's submission relating to "Other factors" had been raised by Japan during the conciliation process. In this respect, the Panel first observed that paragraph 26 of ADP/79 did not identify the second sentence of Article 3:4 as relevant to Japan's claim. Nor did Japan in paragraph 26 explicitly allege that any injury to the EC industry was due to factors within the control of the EC industry, much less identify any such factors specifically.

328. Japan argued that it had challenged the EC's finding that dumping by Japanese exporters had caused injury to the EC industry, and that "[i]t is a necessary corollary of this challenge that any injury must have been caused by another factor or factors . . . ." The Panel observed in this respect that the first and second sentences of Article 3:4 were closely related. The requirement to demonstrate that dumped imports were, through the effect of dumping, causing injury, and the requirement that injuries caused by other factors not be attributed to the dumped imports, were two aspects of the process of establishing causality. However, the Panel recalled its view that for a claim to be considered by a Panel, it had to have been identified during the conciliation phase of a dispute. The Panel further recalled its view that the statement of a "claim" required the identification not only of the obligation under the Agreement allegedly violated but also of the action or factual situation allegedly giving rise to an inconsistency with the Agreement. In this case, Japan appeared to argue that an unelaborated reference to the first sentence of Article 3:4 was sufficient to bring before the Panel any issue related to causation of injury, including the issue of other factors which was explicitly addressed in the second sentence of Article 3:4. In light of Japan's failure to identify during conciliation, in even the most general terms, the "other factors" that it now contended were responsible for any injury to the EC industry, the Panel concluded that the allegations raised by Japan in the first two sections of its submission were not claims that could be considered by the Panel.

329. The Panel recalled Japan's argument that the burden of proof was on the EC to prove causation and that Japan therefore was not required to prove the existence of other factors or to present the EC or the Committee with a list of such factors. However, the Panel considered that Japan's argument had confused the issue of the obligations of the EC under Article 3:4 of the Agreement and the requirements placed on a Party pursuing dispute settlement by Article 15 of the Agreement. The fact that a Party was required by Article 3:4 to "demonstrate" prior to the imposition of anti-dumping duties that dumped imports were, through the effects of dumping, causing injury did not mean that it would be sufficient for a Party pursuing dispute settlement pursuant to Article 15 of the Agreement simply to allege during the conciliation process that a Party had failed to satisfy the first sentence of Article 3:4 without identifying the action or factual circumstances allegedly giving rise to the inconsistency.

330. The Panel recalled that Japan had in its first submission raised claims regarding the manner in which the EC had derived the profit and selling, general and administrative costs it had used in constructing the normal value for certain models. The EC contended that Japan had not raised these claims in the conciliation phase of this proceeding. Japan had responded that these claims were raised in paragraph 19 of the document requesting conciliation (ADP/79) and that, in the alternative, these claims were clearly asserted in an Explanatory Note containing clarification of the document serving as the basis for the terms of reference of the Panel.

331. The Panel noted that Part I of ADP/79 was entitled "Calculation of the Dumping Margin." Paragraph 11 of that document, which was the chapeau for Part I, stated that:

The Panel further noted that Part I of ADP/79 was divided into sections A and B. Section A was entitled "Asymmetrical comparison of export price and normal value," while Section B was entitled "Distortion of the dumping margin (Wrong method of averaging)." Paragraph 19 of ADP/79 was the last paragraph both of Part I and of Section B of the document. It stated that:

332. The Panel considered it clear from the structure and text of ADP/79 that Japan in its request for conciliation had raised two claims with respect to the EC's dumping calculation methodology, and that these claims related to the so-called issues of "asymmetry" and "zeroing." The Panel further considered that paragraph 19 of the request for conciliation could not be construed to raise a third claim regarding the EC's methodology for calculating a constructed normal value. Rather, paragraph 19 asserted that the allegedly Code-inconsistent methodologies of "asymmetry" and "zeroing" identified in sections I.A and I.B of the request for conciliation had an impact on the outcome of the audio cassettes investigation.

333. The Panel next considered the consequences for the Panel's work of Japan's failure to raise these claims during conciliation in light of the content of and circumstances surrounding the establishment of the Panel's terms of reference.

334. The Panel noted that ADP/85/Add.1, which was incorporated by reference into the terms of reference of the Panel, stated as follows:

The Panel considered that the claims of Japan regarding defects in the manner in which the EC constructed the normal value fell within the scope of this paragraph. However, it noted that there might be some question as to whether this paragraph was sufficiently specific to provide notice to the EC and third parties regarding the nature of the claims asserted by Japan.

335. The Panel further recalled that the parties to the dispute had submitted to the Panel an Explanatory Note which was intended to "clarify" the issues outlined in ADP/85/Add.1. In that Explanatory Note, Japan stated, with respect to paragraph 17 of ADP/85/Add.1, that:

In light of the fact that the parties to the dispute had agreed that this Explanatory Note could be considered by the Panel in interpreting the scope of its terms of reference, the Panel considered that it could take this Note into account in its examination. The Panel further considered that the Panel's terms of reference, when interpreted in light of paragraph 17 of the Explanatory Note, contained the claims to which the EC had raised a preliminary objection.

336. In light of the foregoing conclusion, the Panel considered that it was required to examine whether it could consider claims which were included in the terms of reference when those claims had not been raised in conciliation. In this respect, the Panel noted that panels under the Agreement typically operated pursuant to standard terms of reference. The complaining party in a dispute drafted the document which served as the basis for these standard terms of reference of the Panel. Thus, it could not be assumed that the Committee by establishing a panel with standard terms of reference had decided that the panel should examine any claim identified in the written statement, regardless of whether that claim had been the subject of conciliation in the Committee. To hold otherwise would place on the Committee the obligation to review the document requesting establishment of a panel in each case in light of the claims presented during conciliation, and to redraft that document in those cases where it included claims not raised during the previous phases of dispute settlement. Similarly, the defending party as a member of the Committee would have no choice but to block the adoption of standard terms of reference in any case where it believed that a claim had not properly been raised in prior phases of the dispute settlement process or abandon its objections to that claim. In short, such an approach would shift to the Committee from the Panel the task of reviewing preliminary objections. In light of the foregoing, and taking into account the Panel's views in paragraphs 295-298 above, the Panel considered that a claim that was not raised during the conciliation phase of a proceeding could not normally be considered by a panel irrespective of whether that claim was thereafter included in the terms of reference of the Panel.

337. The Panel observed, however, that the factual situation in this case differed significantly from that which normally existed. In this case, the EC had opposed in the Committee the approval of standard terms of reference for the panel pending clarification by Japan of the claims it had presented in its request for establishment of a panel. See ADP/M/39, 22 January 1993, paragraphs 108-146. At that time, the EC had specifically identified paragraph 17 of ADP/85/Add.1 as requiring further clarification. As a result, the parties had undertaken lengthy consultations, which resulted in the preparation of an Explanatory Note on ADP/85/Add.1 containing a detailed clarification of Japan's claims regarding constructed normal value. See Annex I. As previously noted, this Explanatory Note was submitted by the parties to the Chairman of the Committee with a request by both parties that it be made available to the Chairman of the Panel once its composition was established. A letter from the EC, which was circulated to the Committee in document ADP/94, referred to the clarifications found in the Explanatory Note and stated that:

338. The Panel considered that, although both the EC and Japan had referred to the terms of reference of this Panel as "standard terms of reference," the EC had in fact negotiated with Japan in this case the Panel's terms of reference, and that the EC had blocked Committee approval of the Panel's terms of reference until the EC was satisfied with the underlying basis for those terms of reference. Thus, the Panel considered that Japan had not, as in the usual case, been the "master" of the terms of reference of the Panel. Rather, it considered that the parties to the dispute had, in effect, negotiated special terms of reference for the Panel, and that these special terms of reference included the specific claims to which the EC now raised preliminary objections. Under these particular circumstances, the Panel considered that the EC by agreeing to include these claims in the Panel's terms of reference had waived its right to object to the consideration of these claims by the Panel on the grounds that they had not been raised during conciliation.

339. For the foregoing reasons, the Panel concluded that:

(a) the claim of Japan that the EC's methodology for selecting the export models to be used in a comparison of price undercutting was inconsistent with Article 3 was not within the terms of reference of the Panel and thus were not properly before the Panel;

(b) the allegations raised by Japan in sections 3.3.2.1 and 3.3.2.2 of its first submission regarding factors within the control of the EC industry that allegedly were responsible for any injury suffered by the industry were not identified during conciliation and thus were not properly before the Panel;

340. the remaining claims and/or arguments to which the EC had raised preliminary objections on the grounds that they had not been raised in the conciliation phase and/or were not within the terms of reference of the Panel could be considered by the Panel.

341. The Panel next proceeded to examine the EC's preliminary objection that Japan had no "legal interest" in its claims related to "zeroing" and "asymmetry" and that the Panel therefore should not consider those claims.

342. The Panel recalled the EC's argument that a panel should not consider claims in which the complainant had no legal interest. In the view of the EC, a complaining party had no legal interest in a claim if the alleged inconsistency with the Agreement could not have given rise to the nullification and impairment of benefits under the Agreement. The EC recognized that an action which was inconsistent with the Agreement would be considered prima facie to constitute a case of nullification or impairment. However, the EC argued that a respondent was entitled, as a preliminary matter, to rebut this presumption by presenting evidence that no benefit could be impaired, or that no adverse effects could arise from, the action complained of. In such a case, the complaining party was required to demonstrate that the action complained of had a potential impact. In this case, the EC contended, it had imposed a duty pursuant to the lesser duty rule which was far lower than the margin of dumping it had determined to exist. Thus, the alleged errors of the EC in calculating a margin of dumping could have had no impact on the level of duties imposed in this case.

343. Japan had argued that in GATT practice the presumption of nullification and impairment arising from an inconsistency with the Agreement was irrebuttable. Japan further argued, in the alternative, that the EC had not rebutted the presumption of nullification and impairment in this case. In support of this view, Japan had submitted calculations regarding one Japanese exporter indicating that, had the EC used calculation methodologies that were consistent with the Agreement, the margin of dumping found for that exporter (representing roughly half of total Japanese exports to the EC) would have been less than the duties imposed. Japan also contended that under EC practice individual exporters could seek refunds where the actual margin of dumping was lower than the duty imposed pursuant to the Definitive Regulation. Because the EC used the same methodologies for calculating a margin when considering any request for a refund of duties as were used during the original investigation, the methodologies complained of by Japan had a trade-inhibiting effect. Finally, Japan argued that it had challenged not only the duty imposed in this case but also the practices which gave rise to the alleged inconsistencies with the Agreement.

344. The Panel considered that the EC had, in effect, raised as a preliminary objection the argument that the actions of which Japan complained could not have resulted in the nullification or impairment of benefits accruing to Japan under the Agreement. The Panel recalled that paragraph 5 of the Annex to the Understanding Regarding Notification, Consultation, Dispute Settlement and Surveillance, which applied mutatis mutandis to disputes under the Agreement, stated that "there is normally a presumption that a breach of the rules has an adverse impact on other contracting parties, and in such cases, it is up to the contracting parties against whom the complaint has been brought to rebut the charge." The Panel observed that previous Panels had considered that the presumption of nullification or impairment of benefits identified in paragraph 5 of the Annex to the Understanding had in fact operated as an irrefutable presumption, and that, assuming arguendo the presumption could be rebutted, the absence of trade effects would not be sufficient rebuttal. United States - Taxes on Petroleum and Certain Imported Substances, adopted 17 June 1987, BISD 34S/36, paragraphs 5.1.7., 5.1.9.

345. The Panel recalled that Japan had submitted calculations to the Panel relating to one exporter in support of its view that the margin of dumping, calculated in a manner consistent with the Agreement, would have been lower than the duty imposed. The EC had argued that the methodology used by Japan was flawed in a number of respects and could not be relied on. However, the EC had not submitted calculations of its own to contradict Japan's calculations. Further, the Panel considered that the debate between the EC and Japan regarding the margin of dumping that would have been found to exist had the EC not used methodologies allegedly inconsistent with the Agreement was one that a panel was particularly ill-equipped to resolve. The task for a Panel was to review the legal and factual bases for the imposition of an anti-dumping duty in light of the obligations of the Agreement, and not to reach its own determination regarding the existence and extent of dumping. Therefore, the Panel concluded that, assuming arguendo the presumption of nullification and impairment arising from an inconsistency with the Agreement was rebuttable, and that a demonstration of the absence of adverse trade effects could constitute sufficient rebuttal, the EC had not rebutted the presumption that the measures claimed by Japan to be inconsistent with the Agreement had caused nullification or impairment in this case.

346. For the foregoing reasons, the Panel concluded that it was not precluded from considering the claims of Japan regarding "zeroing" and "asymmetry" on the grounds that Japan lacked a "legal interest" in those claims.

347. The Panel first proceeded to examine Japan's claim that, by treating export sales made at above the normal value as made at a zero margin of dumping when calculating an average margin of dumping in this case, the EC had acted inconsistently with its obligations under Articles 2:1 and 2:6 of the Agreement.

348. The Panel recalled that the EC in this investigation had first established an average normal value for each model of an exporter (on the basis either of the domestic price for that model or a constructed value) during a prior representative period or "period of investigation". It then compared the export price for individual export transactions during that period to that average normal value. When the export price was less than the average normal value, the EC calculated the amount of dumping with respect to a given transaction by subtracting the export price from the average normal value. When the export price was equal to or greater than the normal value, the EC considered that dumping did not occur. (The EC acknowledged that there were in this case certain export sales at prices in excess of the normal value, although it stated that the number of such sales was extremely small.) The EC then calculated a weighted-average margin of dumping for the exporter by totalling the amount of dumping for the individual export transactions and dividing this total by the total c.i.f. value of all the export sales of that exporter. Under the pre-selection system used by the EC, this weighted-average margin of dumping represented the maximum duty that the EC could impose on imports of the product from that exporter. In this case, the EC imposed a lesser duty which it considered adequate to remove the injury to the domestic industry, as encouraged by Article 8:1 of the Agreement.

349. The Panel noted that the claim of Japan regarding the EC's averaging methodology, although couched in terms of the "zeroing of negative margins", in fact arose out of a combination of two factors in the EC's methodology, the comparison of individual export prices to a weighted-average normal value, and the "zeroing" of any "negative margins" arising from that comparison. In the view of the Panel, the issue of "zeroing" would not arise in cases where the comparison made was between an average normal value and an average export price.

350. The Panel observed that Article 2 did not require the use of averaging. A Party was entitled to determine the existence and amount of dumping and to impose anti-dumping duties on a transaction-to-transaction basis. Under such an approach, a Party would determine the export price with respect to a particular transaction, compare it to the comparable price in a particular transaction in the exporting or a third country or to a constructed value, and levy a duty on the specific export transaction. In this manner, each transaction at dumping prices would be subject to a duty no higher than necessary to offset the dumping and undumped transactions would not be subject to duty. In GATT practice, such a transaction-to-transaction methodology had been considered desirable, albeit difficult to implement. Thus, a Group of Experts on Anti-Dumping and Countervailing Duties, in discussing Article VI of the General Agreement, had observed that "the ideal method of fulfilling these principles [of Article VI] 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." See Second Report of the Group of Experts on Anti-Dumping and Countervailing Duties, adopted 27 May 1960, BISD 9S/194, 195, paragraph 8.

351. The Panel observed that most Parties to the Agreement had considered that the use of a transaction-to-transaction methodology for establishing the existence and amount of dumping and imposing duties was impractical, and as a result had resorted to establishing average margins of dumping. Further, most Parties established an average margin of dumping in a prior representative period and used this level as the maximum level of duty that could be applied pursuant to a pre-selection system. The Panel recalled that Japan had not contended in this case either that the use of averaging techniques in the comparison process or the application of a pre-selection system was inconsistent with the Agreement. Rather, Japan had argued that Articles 2:1 and 2:6, taken together, gave rise to a general obligation for Parties to conduct a fair comparison of the export price and the normal value, and that the particular averaging methodology used by the EC in this investigation was inconsistent with that obligation.

352. The Panel noted that it could be possible to interpret Articles 2:1 and 2:6, taken together, to give rise to a requirement of a "fair comparison" which applied generally to any aspect of the comparison of normal values and export prices, including the use of averaging techniques. The Panel therefore proceeded to examine whether, assuming arguendo that such a generalized fair comparison requirement existed, the EC's averaging methodology as applied in this case could be considered to be inconsistent with that requirement.

353. In the view of the Panel, it followed from the considerations set forth in paragraphs 350-351 that the use of averaging reflected an effort to approximate overall the results that would be obtained if a Party were to determine the existence and extent of dumping and impose anti-dumping duties on a transaction-to-transaction basis. Therefore, an assessment of the compatibility of an averaging methodology with any generalized obligation of fairness under Article 2 logically had to use as a benchmark the results that would have been obtained had the Party determined the existence and extent of dumping and imposed duties on a transaction-to-transaction basis. The Panel therefore proceeded to examine whether the averaging methodology used by the EC in this case produced an unfair result in light of this benchmark.

354. The Panel recalled Japan's view that the EC's methodology "always works to the prejudice of the exporter." However, the Panel had examined the results of the EC methodology and concluded that, while in some cases it might allow the collection of duties in excess of the amount of duties that might be collected on a transaction-to-transaction basis, in many other cases it would not permit the collection of duties in an amount as high as could be collected pursuant to a transaction-to-transaction methodology. Whether the EC's methodology permitted the collection of more or less duties than could be collected pursuant to a transaction-to-transaction methodology would depend upon a series of variables. For example, to the extent that prices had changed over time in both markets, but the price charged by an exporter in the domestic and export markets was the same at any given time, the EC's methodology generated an upward bias by comparison with the transaction-to-transaction method. To the extent, however, that some export prices were above comparable domestic prices in some cases and below those prices in others, the EC's methodology would produce a downward bias in the amount of duty that could be collected by comparison with a transaction-to-transaction approach. In addition, a series of other variables (such as the existence of significant changes in market conditions in both markets) could affect whether the EC's methodology in a given case overestimated or underestimated that actual amount of dumping assessed on a transaction-to-transaction basis. Where the price in the domestic market remained unchanged during the investigation period (or where a single constructed normal value had been established for that period), the EC's methodology was neutral in effect as compared to a transaction-to-transaction methodology.

355. The Panel recalled that, as an example of the unfairness of the EC's methodology, Japan had pointed out that whenever prices varied over time, an exporter that rigorously maintained equal domestic and export prices would nevertheless be found to be dumping. The Panel agreed with Japan that in this particular factual situation the EC's methodology might permit the collection of duties where, under a transaction-to-transaction methodology, no dumping existed and no duties could be collected. Thus, there might be situations where the EC's methodology would produce an outcome which would be inconsistent with the Agreement. However, the Panel noted that Japan had not claimed that the example it had offered to demonstrate the unfairness of the EC's methodology reflected the factual situation in this investigation, nor that the EC's methodology in this particular case had allowed the imposition of duties in excess of those that would have been permissible under a transaction-to-transaction methodology. To the contrary, Japan's calculations indicated that, had all the EC methodologies Japan deemed to be inconsistent with the Agreement been modified, some dumping would still have been found to exist.

356. Japan contended that the EC's averaging methodology was inherently unfair because if an exporter made some sales at dumped prices and others at undumped prices it would always produce a dumping margin. The Panel observed that, if Japan meant that the EC's methodology would inevitably produce a dumping margin wherever an individual export price was below the average normal value, Japan's statement was factually correct (although it observed that because the EC calculated the weighted-average margin for each exporter by totalling the amount of dumping and dividing this total by the total C.I.F. value of all export sales of that exporter, the margin of dumping would decrease as the value of sales above the average normal value increased). However, the Panel noted that the benchmark against which it measured an averaging methodology was a transaction-to-transaction methodology. The Panel observed that, if the existence and extent of dumping and the imposition of duties had been conducted on a transaction-to-transaction basis, the EC would have been entitled to impose a duty with respect to dumped transactions, where injury existed, irrespective of the prices at which other undumped transactions occurred.

357. The Panel recalled the argument of Japan that the methodology applied in this case was unfair because it produced arbitrary results. The Panel noted that an assessment of the arbitrariness of an averaging methodology had to be determined in relation to the results that would be obtained if averaging were not utilized, i.e. if the amount of dumping were determined on a transaction-to-transaction basis. Japan's arguments regarding arbitrariness sought to gauge the EC's practice against the results that would be obtained through the use of another averaging methodology. As the Panel had already observed, however, the average to average benchmark proposed by Japan also failed in some instances accurately to reflect the results that would be obtained if the existence and extent of dumping were determined on a transaction-to-transaction basis. In fact, the Panel was aware of no averaging methodology that would not in some cases produce results that differed from those obtained through the determination of the extent of dumping on a transaction-to-transaction basis. In light of this fact, and taking into account that Japan did not contend that the use of averaging was inconsistent with the Agreement per se, the Panel could not conclude that the EC's methodology as applied in this case was unfair on the grounds of arbitrariness.

358. The Panel noted that Japan's argument appeared to be that, if the EC were to use an averaging methodology, it was required to use an average to average comparison methodology (mathematically, to eliminate "zeroing" of "negative margins" would be equivalent to using an average to average comparison methodology). Thus, Japan in Annex V to its first submission had calculated a dumping margin "under GATT-consistent methodology" for the largest Japanese exporter by comparing average normal values to average export prices. However, in the view of the Panel Article 2 did not require the use of an average to average price comparison methodology. Rather, as stated in paragraph 353, the benchmark against which an averaging methodology was to be gauged was that of a transaction-to-transaction methodology. Thus, the fact that the EC's methodology might in this case have resulted in a higher average margin of dumping than an average to average comparison methodology did not mean that the EC's methodology breached a requirement to conduct a "fair comparison."

359. The Panel therefore concluded that, assuming arguendo there existed a generalized obligation of "fair comparison" derived from Articles 2:1 and 2:6 and that this obligation applied to the use of averaging methodologies in the comparison of export prices and normal values, the information before the Panel did not permit it to find that the application of the EC's methodology in this case had been inconsistent with that obligation.

360. The Panel recalled the argument of Japan that the EC's averaging methodology used in this case infringed the requirement of Article 2:1 to calculate the dumping margin on the basis of "export" prices and "comparable" prices for the like product. Japan contended that by "zeroing negative margins" the EC had not used actual "export" prices but had rather used artificial export prices fixed at the level of the normal value, and that because no equivalent reduction was made to the normal value the two prices were no longer comparable. However, the Panel observed that the EC had conducted its analysis by comparing export prices of the product with the comparable price in the ordinary course of trade for the like product when destined for consumption in the exporting country. The EC then totalled the absolute amount of dumping in cases where dumping had occurred, and calculated a weighted average margin of dumping by dividing this total absolute amount of dumping by the total C.I.F. value of all the export sales. Based on this understanding of the EC's methodology, as it was presented to the Panel by the parties to the dispute, the Panel could not understand in what respect the EC could be considered in this case to have used an "artificial" export price, as alleged by Japan. Thus, the Panel concluded that the application of the EC's averaging methodology in this case was not inconsistent with the requirement of Article 2:1 of the Agreement that a Party calculate the dumping margin on the basis of "export" prices and "comparable" prices for the like product in the market of the exporting country.

361. The Panel recalled Japan's claim that not only the EC's actions in this investigation but also its "law and practice" were inconsistent with the Agreement. The Panel recalled in this respect its view under certain factual circumstances the EC's methodology would permit the collection of duties where, under a transaction-to-transaction methodology, no dumping existed and no duty could be collected, and that this situation could represent a result inconsistent with any fair comparison requirement of the Agreement. However, the Panel observed that non-mandatory legislation could not be found to be inconsistent with the Agreement. Accordingly, the Panel proceeded to examine whether the EC's Basic Regulation was mandatory legislation.

362. The Panel recalled the EC's argument that its anti-dumping legislation was non-mandatory in nature. Specifically, the EC had pointed out that while Articles 7(1) (regarding initiation of investigations) and 12(1) (regarding the imposition of duties) of the Basic Regulation, provided that the EC "shall" do certain things, the taking of these obligatory actions was dependent on certain conditions and whether these conditions were fulfilled. It was not clear to the Panel that legislation was non-mandatory merely because obligations under that legislation were dependent on whether certain factual prerequisites were fulfilled. However, the Panel did not consider in any event that its task in this case was to determine whether the EC's Basic Regulation in its totality was non-mandatory in the sense that the initiation of investigations and imposition of duties were not mandatory functions. Should panels accept this approach, they would be precluded from ever reviewing the content of a Party's anti-dumping legislation, a result that would undermine the requirement of Article 16:6 of the Agreement that Parties bring their legislation, regulations and administrative procedures into conformity with the provisions of the Agreement. Rather, the Panel considered that its task in this case was to determine whether the EC's Basic Regulation was mandatory in the sense that the EC was required by its legislation to use the averaging methodology complained of by Japan.

363. The Panel recalled that the EC's averaging methodology in anti-dumping investigations was set forth in Articles 2(13) of its Basic Regulation. That Article provided that:

The Panel observed that Article 2(13) stated that the normal value shall "normally" be established on a weighted-average basis. The Panel considered that the use of the term "normally" with respect to the normal value indicated that the EC had the discretion not to establish the normal value on a weighted-average basis in particular cases. The Panel further noted that the term "normally" with respect to the comparison of export prices to the normal value on a transaction-by-transaction basis was ambiguous. On the one hand, the exception to the "normal" approach could be limited to the case where the use of weighted averages would not materially affect the results of the investigation. In this case, however, the term "normally" was superfluous in light of the explicit exception. In light of the view of the Panel that the Basic Regulation gave the EC discretion to establish the normal value on bases other than a weighted-average normal value, and that the EC arguably had discretion not to compare export prices to the normal value on a "transaction-by-transaction" basis, the Panel concluded that the EC was not required to use the averaging methodology used in this investigation.

364. For the foregoing reasons, the Panel concluded that the EC's Basic Regulation was not mandatory legislation inconsistent with Articles 2:1 and 2:6 of the Agreement.

365. The Panel recalled that Japan had asked the Panel to determine that, whether or not the EC's averaging methodology was applied pursuant to mandatory legislation, that practice was inconsistent with the Agreement. The Panel recalled, however, that the task of a Panel was to resolve a specific dispute. In this case, the dispute involved the application of a methodology in a particular investigation and the consistency of legislation with the Agreement. Although a Panel might well, in the course of examining the consistency of a Party's actions with the Agreement, reach a conclusion on grounds that would call into question the consistency of a "practice" with the Agreement in other cases, the Panel did not consider that it would be appropriate to reach findings on a "practice" in abstracto when it had determined that the actions taken in a particular investigation were not inconsistent with the Agreement and that the "practice" was not pursuant to mandatory legislation.

366. For the foregoing reasons, the Panel concluded that:

367. The Panel next examined Japan’s claim that the EC in cases where it constructed export prices had calculated the export price on a different basis than it had calculated the normal value, and that its failure to make allowance for this fact was a violation of a requirement derived from Articles 2:1 and 2:6 of the Agreement to conduct a "fair comparison" of the export price and the normal value when determining the existence and extent of dumping.

368. The Panel recalled that Japan did not dispute the right of the EC to construct an export price in this case. To the contrary, Japan recognized that Article 2:5 of the Agreement authorized a Party to construct an export price, where there was no export price or that price was unreliable, on the basis of the price at which the imported products were first resold to an independent buyer. Nor had Japan argued that the EC was not entitled, when constructing an export price, to make allowance for costs, including duties and taxes, incurred between importation and resale, and for profits accruing, as provided by the last sentence of Article 2:6. Rather, Japan argued that by failing to make allowance for the fact that the normal value was calculated on a different basis than the constructed export price before comparing the two and calculating a dumping margin, the EC had acted inconsistently with Article 2 of the Agreement. Specifically, Japan contended that indirect selling expenses and profits of the importer which were associated with sales activities were deducted in order to calculate a constructed export price, while identical cost and profit elements which existed in the exporters' domestic sales operations were not deducted in order to establish a normal value. Japan contended that the failure of the EC to make due allowance for these differences before comparing the two prices resulted in an unfair comparison.

369. The Panel recalled that Japan's claim had referred both to Article 2:1 and 2:6. However, the Panel considered that any review of a claim regarding a failure to make "due allowances" properly should begin with an analysis of Article 2:6, which related directly to the way in which the export price and the normal value were to be compared. Article 2:6 provides as follows:

The Panel considered that the purpose of the allowances provided for in the last sentence of Article 2:6 was to permit the establishment of an export price on a basis which would remove the unreliability arising from the relationship between exporter and importer. The Panel noted, however, that such allowances were additional to those provided for in the second sentence of Article 2:6, which would still be required in comparing normal values, however established, with constructed export prices. Whether such allowances were required would depend on the extent to which the elements referred to in the second sentence (differences in terms and conditions of sale, differences in taxation, the other differences affecting price comparability) might apply to the comparison of the normal value with the constructed export price. This meant, in the view of the Panel, that construction of an export price did not per se require adjustments as provided for in the second sentence, but the need for such adjustments would depend on the fact situation of the particular comparison. Accordingly, the "asymmetry" to which Japan objected appeared to derive not from the EC’s methodology for constructing an export price per se, but more generally from the alleged failure of the EC to make due allowance for differences between the Japanese and EC markets in indirect selling expenses and profits associated with sales activities.

370. The Panel then proceeded to examine whether adjustments for differences in indirect selling expenses could be required in order to satisfy the requirement of Article 2:6 that due allowance be made for differences affecting price comparability. The Panel recalled the EC's argument that "price comparability" within the meaning of Article 2:6 could only be affected by differences in "direct" and not indirect costs. The Panel considered, however, that the term "the other differences affecting price comparability" in Article 2:6 in its ordinary meaning referred simply to differences which would affect the comparability of the export price and normal value that were to be compared. The Panel observed that there might be differences in the amount of both direct and indirect selling expenses between the domestic and export market. The Panel considered that such differences in selling expenses, whether direct or indirect, could affect the comparability of the prices being compared.

371. In arriving at this conclusion, the Panel considered that the term "the other differences affecting price comparability" in Article 2:6 had to be interpreted in the context of Article 2. In the view of the Panel, Article 2 as a whole reflected a view that a price could be affected by all costs and not just "direct" costs. Thus, Article 2:4 provided that where there was no comparable price in the domestic market to the export price, such a price could be constructed based on the cost of production plus a reasonable amount for administrative, general and any other costs and for profits. Similarly, where there was no export price or that price was unreliable, a surrogate export price could be "constructed" pursuant to Article 2:5 and the last sentence of Article 2:6 by starting with the first arms-length price and making allowance for costs incurred and profits accruing between importation and resale. To interpret the term "the other differences affecting price comparability" per se to exclude differences in indirect costs would be to conclude that, in the context of constructed normal values and constructed export prices, prices could or should always reflect indirect costs, while in the context of due allowances the opposite was the case, i.e. that prices would never be affected by indirect costs.

372. The Panel recalled the argument of the EC that the reference in Article 2:6 to due allowances for differences in "conditions and terms of sale" indicated that only differences in variable costs, i.e., those incurred in relation to a particular sale, affected price comparability. However, the Panel recalled that Article 2:6 provided not only for due allowance for differences in conditions and terms of sale but also for "the other differences affecting price comparability." Thus, assuming that "differences in conditions and terms of sale" were limited to differences relating to a particular sale, i.e. to variable costs, this did not mean that "the other differences affecting price comparability" were so limited. The Panel further recalled the argument of the EC that the reference to the making of due allowance "on its merits" allowed investigating authorities a certain discretion in deciding when differences affected price comparability such that due allowances were required under Article 2:6. The Panel concurred that Article 2:6 envisioned a case by case analysis of the normal value and export prices being compared to determine whether adjustments were necessary. However, the Panel did not agree that the term "on its merits" supported the EC's view that in making due allowances for differences affecting price comparability the only adjustments required would be for expenses directly related to a particular sale.

373. In this respect, the Panel noted that the EC could under its Basic Regulation, and did in this case, make allowances for differences in indirect costs in one particular situation, in the form of adjustments for salesmen's salaries which appeared to be "indirect" costs to the extent that they were not tied to the volume or price of sales achieved by individual salesmen. Thus, the EC's Basic Regulation itself recognized that allowances for differences in "indirect" costs could at least in certain circumstances be necessary to ensure the comparability of prices.

374. Finally, the Panel further considered that the term "price comparability" should be interpreted in light of the object and purpose of the Agreement. The Panel considered that the first clause of Article 2:6 indicated that one of the objectives of the comparison rules in that Article was to ensure a fair comparison between the export price and the normal value. Whether or not the provisions of Article 2 gave rise to an independent obligation to conduct a fair comparison (an issue on which the Panel did not consider it necessary to express itself in this case), the Panel therefore considered that in its interpretation of the term "the other differences affecting price comparability" it should take into account this objective. The Panel further considered it evident that to interpret the term "the other differences affecting price comparability" per se to exclude differences in indirect costs could result in an unfair comparison between the export price and the normal value. The Panel noted that such an interpretation could result not only in the failure of a Party to make due allowance for differences in the amount of a particular type of cost in the domestic and export markets, but it could also result in a situation where certain costs were reflected in their totality in one price while the same type of costs were not reflected at all in a comparison price. Thus, for example, all indirect selling expenses in an export market might be incurred by an importer, and thus not be reflected in an export price, while in the domestic market all such selling expenses might be incurred by the producer himself and thus be reflected in their totality in the normal value (since if the producer was not pricing to cover costs, a constructed value reflecting all costs plus profits could be constructed). Thus, the Panel considered that to interpret the term "the differences affecting price comparability" per se to exclude differences in indirect expenses would frustrate the objective of Article 2:6 to ensure a fair comparison.

375. The Panel next considered whether differences in profits associated with selling activities could represent "differences affecting price comparability" within the meaning of Article 2:6. In this respect, the Panel noted that the items specifically identified in the second sentence of Article 2:6 - differences in taxes and in conditions and terms of sale - were differences which could be reflected in different costs. Differences in the level of profits, on the other hand, were a function of the relationship between prices and costs. In a case where costs of production and distribution were identical in the domestic and export markets, or where allowance was made for any difference in costs, any difference between the adjusted prices would result in an identical difference in the amount of profits. In other words, the difference in the amount of profits would be the measure of the existence of price discrimination. Thus, the Panel did not consider that the existence of a difference in the level of profits related to sales activities in the export and domestic markets would of itself necessarily be a difference which affected price comparability and for which due allowance therefore had to be made.

376. The Panel did not consider, however, that due allowance would never be required with respect to profits. The Panel recalled its view that in its ordinary meaning the term "the other differences affecting price comparability" referred simply to differences which could affect the comparability of the prices being compared. The Panel further recalled its view that one of the objectives of the comparison rules in Article 2 was to ensure a fair comparison, and that in its interpretation of the term "the other differences affecting price comparability" the Panel should take into account this objective. In this case, the Panel recalled, Japan had not alleged that the EC had failed to take into account differences in the level of profits related to sales activities in the domestic and export markets, but rather that the EC had in constructing an export price deducted an amount for profits related to sales activities while identical profit elements which existed in the exporters' domestic sales operations were not deducted in order to establish a normal value. It appeared to the Panel that in referring to identical profit elements Japan was referring to profits in the two markets relating to the same functions, and in particular to selling functions. The Panel noted that the process of manufacture and sale of a product involved a series of functions. Each of these functions entailed certain investments and expenses. The Panel considered that where a producer performed additional functions in one market beyond those performed in another market, that this situation could affect the prices at which the products were sold, and thus the comparability of the two prices being compared. Specifically, a producer would in the usual case not perform additional functions with respect to the sale of a product unless it anticipated that the additional costs incurred, plus a return on the additional investment, would be recovered either through higher prices for the product or through higher sales volumes. Contextual elements in Article 2 supported this interpretation. By providing for the inclusion of profits in the construction of export prices, the drafters had indicated their expectation that some measure of profit related to post-production functions related to importation and distribution (which could include selling activities) could be expected to be reflected in the price charged for a product. Thus, the Panel concluded that where the price in one market was based on the price charged by a seller performing particular functions, while the price in the other market was based on a price charged by a seller who performed additional functions entailing additional investments and costs, the comparability of the prices could be affected. In order to restore the comparability of the prices, it might be necessary to make allowance in the form of an amount for profits related to the additional functions performed.

377. The Panel therefore concluded that a difference in indirect selling expenses could constitute a difference affecting price comparability for which due allowance on its merits might be required pursuant to Article 2:6 of the Agreement. The Panel further concluded that where the price in one market was based on the price charged by a seller performing particular functions, while the price in the other market was based on a price charged by a seller who performed additional functions entailing additional investments and costs, a difference affecting price comparability could exist for which due allowance on its merits in the form of an amount for profits related to the differences in the functions performed in the export and domestic markets might be required pursuant to Article 2:6 of the Agreement.

378. The Panel then proceeded to examine the EC's Basic Regulation in light of the requirements of Article 2:6. The Panel recalled that paragraph 9(a) of Article 2 of the EC's Basic Regulation provided, in relevant part, that:

(i) physical characteristics;

(ii) import charges and indirect taxes;

(iii) selling expenses resulting from sales made:

- at different levels of trade, or

- in different quantities, or

- under different conditions and terms of sale."

The Panel further noted that pursuant to paragraph 10 of Article 2 of the EC's Basic Regulation:

..............................

379. The Panel observed that the EC's Basic Regulation on its face set forth an exclusive list of all the differences affecting price comparability. That adjustments for other differences than those listed in paragraph 9(a) were not permitted was indicated by the fact that this list was preceded by the term "i.e.," a Latin abbreviation standing for "id est," which means "that is." The Panel noted that paragraph 9(a) did allow adjustments for differences in selling expenses. The Panel observed, however, that pursuant to paragraph 2(10), any adjustments for differences in selling expenses had to be made pursuant to the rules in that paragraph. The Panel further observed that the specific selling expenses identified in paragraph 10(c) for which adjustments could be made were restricted by such terms as "directly related costs," "direct costs" or "direct selling activities." The Panel further noted that the list of selling expenses identified in paragraph 10(c) for which adjustments could be made also was preceded by the term "i.e." and thus represented an exclusive list. The Panel therefore concluded that under the EC's Basic Regulation the EC was precluded from making adjustments for differences in indirect selling expenses (with the limited exception for the salesmen's salaries identified in Article 2(10)(c)(v) of the Basic Regulation).

380. The Panel next examined whether the EC's Basic Regulation permitted due allowances to be made with respect to profits in the circumstances outlined in paragraph 376. The Panel recalled its view that the Basic Regulation set out an exclusive list of the differences affecting price comparability and the particular adjustments allowed to take account of such differences. The Panel further observed that nowhere in paragraphs 9 or 10 did the Basic Regulation indicate that allowances could be made with respect to profits. The Panel therefore concluded that the EC was precluded by its Basic Regulation from making adjustments with respect to profits related to a difference in the functions performed by the seller in the domestic and export markets.

381. The Panel recalled the contention of the EC that the EC would have been able to make adjustments for differences in indirect expenses and profits had it concluded that export sales were at a different level of trade than domestic sales. The EC had referred to recital 14 of its Provisional Regulation in support of this contention. In that recital, the EC had accepted the claim of a Korean exporter that some of its sales in the EC were on an original equipment manufacturer ("OEM") basis, and had for that reason constructed a normal value to be used in the comparison with those OEM export sales using a rate of profit lower than that used in constructing a normal value for comparison to other export sales of that exporter. The Panel noted, however, that recital 14 appeared in the section of the Provisional Regulation entitled "Normal value based on constructed value." Thus, it appeared that the action taken by the EC in recital 14 related not to the making of due allowances for "the other differences affecting price comparability" as required by the second sentence of Article 2:6, but rather to the manner in which the EC had established the normal value. The Panel further noted that the action taken by the EC in this case related to alleged differences in level of trade. The EC had stated in its submissions that it did not consider a difference in costs between export and domestic sales to be indicative of a difference in levels of trade. Further, there was no indication by the EC in its submission that the EC would in circumstances other than differences in level of trade construct a normal value in such a manner as to take into account differences in indirect selling expenses, or profits related to a difference in the functions performed by the seller in the export and domestic markets.

382. For the foregoing reasons, the Panel concluded that Articles 2(9) and 2(10) of the EC's Basic Regulation were mandatory legislation inconsistent with Article 2:6 of the Agreement because they precluded the making of due allowance, on its merits, for differences in indirect selling expenses and with respect to profits related to a difference in the functions performed by the seller in the domestic and export markets, which differences could affect price comparability.

383. The Panel next reviewed the actions of the EC in the audio cassettes investigation in light of the requirements of Article 2:6 of the Agreement. The Panel observed that Japan had only alleged that the EC's actions were inconsistent with the Agreement in those cases where it had constructed export prices pursuant to Article 2:5 and the last sentence of Article 2:6 of the Agreement. The Panel recalled its view that construction of an export price did not per se require adjustments as provided for in the second sentence, but the need for such adjustments would depend on the fact situation of the particular comparison. The Panel further recalled that it had requested Japan to identify those indirect selling expenses and profits related to sales activities which it alleged had been deducted in the calculation of the export price but had not been deducted from the normal value. Japan had responded in the context of one exporter, TDK. Japan had indicated that all salesmen's salaries, salesmen's travel expenses, other travel expenses associated with sales, rental, light, heating and local taxes associated with sales activities, advertising and promotion, samples and entertainment costs associated with sales, telecommunications and indirect research costs associated with sales, and profits associated with sales were incurred by related sales subsidiaries in the EC and thus were deducted in order to construct an export price. Japan further indicated that none of the identical expenses and profits, except for 50 per cent of salesmen's salaries and 60 per cent of promotion costs, were deducted from the normal value.

384. It appeared from the above list that Japan had alleged, in essence, that importers in the EC incurred all or virtually all the selling expenses related to the EC market; that these indirect expenses, and profits related to these sales functions, thus were not reflected in the export price, while in Japan these selling expenses were incurred by the seller and were thus reflected in the normal value; and that no adjustment was made for these differences. The Panel noted that it did not have the full record before it in this case and that in any event a Panel generally was not well-situated to perform the detailed factual review necessary to assess the existence of the differences in indirect selling expenses and profits in the two markets alleged by Japan. The Panel considered, however, that it did have sufficient information before it with respect to advertising expenses. The Panel noted that where the EC had constructed the normal value the EC had indicated to the Panel that it had included among overhead expenses those expenses related to selling, marketing and promotion in Japan; Japan had further indicated to the Panel that these expenses included expenses for national advertising and sales promotions (spent by sales branches). The Panel further noted that, according to Japan, all advertising expenses related to sales in the EC were incurred in the EC and thus were deducted in order to construct an export price. The EC had not challenged this statement. Further, the Panel noted that pursuant to Article 2(8)(b) of the EC's Basic Regulation, relating to the construction of the export price, the EC "adjusted the export price for advertising costs corresponding to sales made in the Community but paid or reimbursed by the Japanese or Korean exporters associated with these importers . . . ." Provisional Regulation, recital 38. The Panel thus concluded, from the facts before it, that in at least some instances in this investigation the normal value was established on the basis of a price charged by a seller incurring advertising expenses while the export price represented the price charged by an exporter who did not incur advertising expenses (those expenses being incurred by or on behalf of the subsequent purchaser, in this case the related importer, of the product). No allowance was made for these differences.

385. The Panel recalled the argument of the EC that it had considered and rejected the claim of one Japanese exporter that some of its sales in the EC had been to exclusive distributors buying in large quantities at prices lower than average, while domestic sales were made directly to retailers, that these sales were therefore at a different level of trade, and that the EC should make an adjustment for differences in indirect selling expenses related to this difference in level of trade. The Panel recalled its view that Article 2(9) and 2(10) on their face did not allow such adjustments to be made. The Panel further noted that, assuming the EC in fact would have been prepared to make such adjustments irrespective of its legislation, adjustments for indirect selling expenses arising out of sales at different levels of trade would not necessarily be sufficient to address the differences in indirect selling expenses that Japan claimed were required in this case to ensure price comparability, because indirect selling expenses incurred at one "level of trade" in a given market might be incurred at a different level of trade in another market. Thus, the EC had contended in this case that sales by Japanese sales departments and related domestic sales subsidiaries were at the same level of trade as sales by the export sales department in Japan, as they "both sold to distributors and wholesalers with a similar pattern of prices and quantities," but it had not disputed that advertising expenses in the EC were incurred by or attributed to the related importer while in Japan advertising costs were incurred by the sales departments or related sales subsidiaries.

386. Finally, the Panel considered it clear from the information put on the record by the parties to the dispute that the EC in this case had applied paragraphs 9 and 10 of Article 2 of the Basic Regulation, and that it had on that basis declined in at least some cases to make due allowance for differences in indirect selling expenses incurred in the two markets purely on the grounds that such differences were indirect, and irrespective of the merits of any particular allowance. Thus, recital 40 of the EC's Provisional Regulation stated that:

The Panel further noted that recital 42 of the Provisional Regulation provided that:

Finally, the Panel noted that, with respect to one exporter, Japan had submitted a letter from the EC Commission explaining in more detail the adjustments made. This letter, dated 11 October 1989, was sent to counsel for TDK as an explanation of the manner in which the dumping calculation was performed. The letter read, in relevant part, as follows:

387. On the basis of the EC Basic Regulation and its application in this case, as set forth above, the Panel considered that the EC had declined to make adjustments for differences in indirect selling expenses on the grounds that they were indirect. The Panel further considered that Articles 2(9) and 2(10) of the EC's Basic Regulation on their face would not have permitted the EC to make allowances for differences in indirect selling expenses or with respect to profits related to a differences in the functions performed by the seller in the domestic and export markets.

388. For the foregoing reasons, the Panel concluded that:

389. The Panel next examined the claim of Japan that in constructing the normal value for three models of audio cassettes the EC had used an amount for profit which was not reasonable and had thus acted inconsistently with its obligations under Article 2:4 of the Agreement.

390. The Panel recalled that in this investigation the EC had calculated a constructed normal value for three of the seven TDK models used as comparison models to export models. The EC had calculated a constructed normal value for these models because it had determined that reported domestic sales included sales not in the ordinary course of trade and original equipment manufacturer ("OEM") sales which could not be isolated. The three TDK models for which a normal value was constructed were all "normal" type audio cassettes. The amount for profit used in constructing the normal value for these models was based on data related to sales of all TDK audio cassettes in Japan (including "chrome" and "metal" types). During the course of the investigation, TDK had suggested as an alternative that the amount for profit used in constructing the normal values for these models be based on data related to sales of all "normal" type TDK audio cassettes in Japan, a methodology which TDK considered would result in a lower amount for profit. However, the EC had considered the profit on "normal" type tapes to be unreliable because sales of this type of audio cassette in the Japanese market included sales to related customers, to trading houses selling for export and sales at a loss.

391. Japan had claimed that, because in this case there was a major difference between the profit levels, on the one hand, for "normal" type audio cassettes, and on the other hand, for "chrome" and "metal" type audio cassettes, an amount for profit based on data relating to sales of the three types of audio cassettes was not "reasonable" within the meaning of Article 2:4. However, the precise argument advanced by Japan in support of this claim was unclear. On the one hand, Japan's argument could be interpreted to be that whether an amount for profit was "reasonable" would depend upon whether it was based on data drawn from sales of "products of the same general category." This latter term, Japan appeared to argue, should be construed, where possible, to be co-extensive with the type of product for which a normal value was being constructed. Where a Party constructed a normal value because all or most of the sales of a particular product were being made at a loss, Japan observed, the term "products of the same general category" might be construed more broadly to permit a Party to derive an amount for profit from data drawn from sales of a broader range of products. In this investigation, however, Japan considered that the EC could have derived a profit level from sales of all TDK "normal" type cassettes; thus, the EC's resort to data relating to sales of the like product resulted in a higher amount for profit that was not "reasonable." In the alternative, Japan's argument could be that, irrespective of whether an amount for profit was based on profits realized on sales of the same general category of products, that amount was not "reasonable" if the Party could have based the amount for profits on a narrower category of product closer to the category for which a normal value was being constructed.

392. The EC contended that an amount for profit was "reasonable" if it was based on data drawn from sales of "products of the same general category" or from sales of some narrower category of product. The EC's amount for profits was based on sales of the "like product," a category narrower than "products of the same general category," and was thus "reasonable." The EC further contended that it could have limited its analysis to profitable sales to independent customers of all TDK audio cassettes (resulting in a higher amount for profit) and still have acted consistently with the Agreement. Finally, the EC argued that it had considered the profit on sales of "normal" audio cassettes to be unreliable because they included sales to related customers, to trading houses selling for export and sales at a loss; if it had based the amount for profit on sales by TDK of "normal" type audio cassettes other than those to related customers, trading houses and at a loss, the amount for profit would have been higher than that actually applied.

393. The Panel observed that Japan's claim was governed by Article 2:4 of the Agreement. That Article provided that

The Panel considered that Article 2:4 contained two separate criteria relevant to the amount of profit that a Party could use in constructing a normal value. First, the amount for profits had to be a "reasonable" amount. Second, the addition for profit could not, as a general rule, exceed the profit normally realized on sales of products of the same general category in the domestic market of the country of origin. With respect to the first criterion, that the amount for profit be "reasonable," while the breadth of the group of products on which the amount of profit was based might be relevant, the Panel did not consider that an amount for profit was by definition reasonable simply because it did not exceed the profit normally realized on sales of the same general category in the domestic market of the country of origin. With respect to the second criterion, the compatibility of an amount for profit with this criterion related not to whether the group of products on which the amount for profit was based was broader or narrower than that of "the same general category of products" per se but to whether the amount for profit was in excess of the profit normally realized on sales in that category. The Panel therefore proceeded to examine the claims of Japan in light of these considerations.

394. The Panel recalled that Japan had claimed that the EC had acted inconsistently with Article 2:4 because profits on sales of the like product used by the EC to establish an amount for profit were higher than profits on the sub-category of the like product composed of "normal" type tapes. To the extent that Japan's argument was based on the proposition that "normal" type audio cassettes in this case were "products of the same general category" within the meaning of Article 2:4 and that because the amount of profits used by the EC exceeded profits realized on that category of product it was "unreasonable," the Panel considered that Japan's argument was without merit. Without seeking to define the term "products of the same general category," the Panel noted that the term in its ordinary meaning was at least as broad as "like product" (defined in Article 2:1 of the Agreement as "a product which is identical, i.e. alike in all respects to the product under consideration, or in the absence of such a product, another product which, although not alike in all respects, has characteristics closely resembling those of the product under consideration"). The Panel further observed that, read in its context within Article 2:4, the term "products of the same general category" appeared to refer to "products of the same general category as the like product." Accordingly, the Panel concluded that the term "products of the same general category" could not properly be interpreted to refer to a category of "normal" type tapes which were a sub-category of the like product under investigation.

395. The Panel recalled its view that an amount for profit was not necessarily "reasonable" simply because it did not exceed the profit normally realized on sales of products of the same general category in the domestic market of the country of origin. The Panel thus examined whether, irrespective of whether the amount for profit used by the EC in this investigation exceeded the profit normally realized on sales of products of the same general category, Japan had asserted any other basis for its claim that the amount for profit used by the EC was not "reasonable." In this respect, the Panel recalled Japan's view that the EC's amount for profit was not reasonable because the EC could and should have used an amount for profit based on sales of the narrower category of all "normal" type audio cassettes sold by TDK in the Japanese market, as suggested by TDK during the course of the investigation. However, the Panel recalled that the EC had constructed the normal value for the three "normal" type cassettes used as comparison models because reported domestic sales included sales not in the ordinary course of trade and OEM sales which could not be isolated. The Panel further observed that the EC had stated, and Japan had not contested in this proceeding, that the EC had considered the profit on "normal" type cassettes proposed by TDK to be unreliable because sales of this type of tape included, inter alia, sales to related customers, to trading houses selling for export and sales at a loss. Under these circumstances, the Panel considered that the "reasonable" amount for profit criterion expressed in Article 2:4 did not require the EC to base the amount for profit on sales of TDK "normal" type cassettes.

396. The Panel therefore concluded that the EC had not acted inconsistently with Article 2:4 by reason of the fact that it had derived an amount for profit in constructing the normal value for three types of "normal" type cassettes from data relating to all sales of the like product by that exporter in the Japanese market.

397. The Panel next examined the claim of Japan that in constructing the normal value for three models of audio cassettes the EC had used an amount for administrative, selling and any other costs which was not reasonable and had thus acted inconsistently with Article 2:4 of the Agreement.

398. The Panel recalled that in this investigation the EC had calculated a constructed normal value for three of the seven TDK models used as comparison models to export models. With respect to the amount for selling, general and administrative expenses (hereinafter referred to as "SG&A") to be used in constructing these normal values, TDK had provided in its questionnaire response data on SG&A which allocated such expenses both on a turnover and a cost-of-manufacture basis, but had indicated its preference for the turnover-based allocation. The EC, however, had decided to allocate SG&A on a cost-of-manufacture basis, as stated in recital 29 of its Provisional Regulation:

It was undisputed that the Council's methodology for determining SG&A for the purposes of the Definitive Regulation was unchanged in this respect from the methodology expressed in the Provisional Regulation.

399. The Panel recalled that Japan had not claimed that the allocation of SG&A on a cost-of-manufacturing basis would necessarily generate an amount for administrative, selling and any other costs which was not reasonable within the meaning of Article 2:4. Rather, Japan had claimed that the EC had failed to satisfy the requirement of Article 2:4 that a constructed normal value contain no more than a "reasonable amount for administrative, selling and any other costs" because the EC had deviated without adequate explanation from the turnover-based allocation methodology which was normally used by the manufacturer and which was preferred by the EC in anti-dumping investigations. In respect to this latter point, Japan referred to Article 2(11) of the EC's Basic Regulation, titled "ALLOCATION OF COSTS," which provided that "[i]n general, all cost calculations shall be based on available accounting data, normally allocated, where necessary, in proportion to the turnover for each product and market under consideration."

400. The Panel first observed that the task of a Panel was to review the consistency of a Party's actions with the Agreement and not with that Party's domestic laws, regulations or practices. Thus, it was not clear to the Panel that the existence of a preference in the EC's Basic Regulation for a particular allocation methodology necessarily was relevant to the work of the Panel. Assuming arguendo that a Party was required by the Agreement to explain why it had deviated from a methodology preferred in its own practice, the Panel noted that the EC had provided an explanation for its use of a cost-of-manufacture allocation methodology in preference to the methodology suggested by TDK and "preferred" by Article 2(11) of the EC's Basic Regulation. Specifically, the EC had determined that the turnover reported for some of the models for which the normal value was being constructed was not reliable because it involved sales of various other models. The Panel noted that Japan had not challenged the factual accuracy of the EC's explanation for its use of a cost-of-manufacturing allocation methodology. Rather, Japan had contended that this explanation was not "convincing" because the use of constructed normal values only arose where there were problems regarding the prices at which products were sold in the market of the exporting country; if the turnover allocation methodology were rejected whenever such problems existing, it would never be used. However, the Panel observed that Article 2(11) of the EC's Basic Regulation was a general allocation principle the application of which was not restricted to the construction of normal values (as, for example, in the construction of export prices). Thus, even if the EC's reasoning, applied consistently, would uniformly have resulted in the use of a cost-of-manufacture based allocation of SG&A in the construction of normal values, this would not have prevented the application of Article 2(11) in other contexts.

401. In light of the foregoing considerations, the Panel concluded that the EC had not acted inconsistently with its obligations under Article 2:4 of the Agreement by using in the construction of certain normal values an amount for selling, administrative and selling costs based on a cost-of-manufacture allocation methodology.

402. The Panel recalled Japan's claim that because the EC's methodology was inconsistent with Article 2 for the reasons discussed in sections A through D, above, the EC had also acted inconsistently with Article 8:3 of the Agreement. Article 8:3 stated that "[t]he amount of the anti-dumping duty must not exceed the margin of dumping as established under Article 2." In the view of Japan, because the EC's methodology for the calculation of a margin of dumping was inconsistent with Article 2 of the Agreement, it was not a margin of dumping "as established under Article 2." Japan further argued that the margin of dumping as properly calculated was less than the duty actually imposed and that in any event, given the exceptional character of anti-dumping duties, the burden was on the EC to demonstrate that the duty imposed was not in excess of the margin of dumping as properly calculated.

403. The Panel recalled that it had found the EC's methodology for the calculation of the margin of dumping in this case to be inconsistent with Article 2 in one respect and to be consistent with Article 2 in other respects. The Panel further observed that Japan's claims under Article 8:3 flowed from and were dependent on its claims that the EC had acted inconsistently with Article 2. Thus, to the extent that the EC had acted inconsistently with Article 2, those actions would in the view of Japan be inconsistent with Article 8:3, while to the extent that the EC's actions were consistent with Article 2 no inconsistency under Japan's argument pursuant to Article 8:3 could arise. Under these circumstances, a determination by the Panel regarding Japan's claim under Article 8:3 could have no effect on the conclusions of the Panel regarding the consistency of the EC's actions with its obligations under the Agreement. Therefore, the Panel did not consider that it was necessary to reach Japan's claim under Article 8:3 of the Agreement.

404. The Panel next examined whether the imposition by the EC of the anti-dumping duty order on audio cassettes from Japan was inconsistent with the obligations of the EC under the Agreement by reason of the affirmative final determination of material injury by the EC.

405. Japan argued that the determination of the EC was inconsistent with the Agreement on the following grounds. First, the cumulation of the effects of Japanese and Korean exports in this case was inconsistent with Articles 3:1 and 3:4 of the Agreement. Second, the EC had failed to establish the existence of significant increases in the volume of dumped imports from Japan or price undercutting by dumped imports from Japan, and had further failed to establish the existence of any price suppression or depression caused by dumped imports from Japan, as required by Article 3:2. The EC further had failed with respect to these factors to meet the requirements of Article 3:1 that it make an injury determination on the basis of an "objective examination" of "positive evidence." Finally, the EC had failed to demonstrate that any injury suffered by the EC industry was caused by dumped imports from Japan through the effects of dumping, and therefore had acted inconsistently with Article 3:4 of the Agreement.

406. The Panel first considered the claims presented by Japan that the cumulation by the EC of the effects of dumped imports from Japan and Korea was inconsistent with the EC's obligations under Articles 3:1 and 3:4 of the Agreement. The Panel noted that Japan did not in this dispute contest the compatibility of cumulation itself with the Agreement. Rather, Japan claimed that the application of the practice of cumulation on the facts of this case was inconsistent with Articles 3:1 and 3:4 of the Agreement.

407. Japan claimed that the application of the practice of cumulation in this case was inconsistent with Article 3:4 of the Agreement. In support of this claim, Japan argued that there were significant factors in this case which distinguished the exports of Korea from those of Japan. Japanese imports were primarily of the chrome and metal varieties and competed based on non-price elements, while Korean imports were principally of the "normal" type and competed based on price. Thus, Korean and Japanese imports of the like product operated in two distinct markets. Under these circumstances, Japan argued, cumulation denied exporters in Japan the possibility to distinguish themselves from those exporters in Korea who were held by the EC to be causing injury. Specifically, cumulation attributed to allegedly dumped imports from Japan the injury caused by dumped imports from Korea. Further, cumulation prevented Japanese exporters the opportunity to demonstrate that any injury caused by allegedly dumped imports from Japan was not through the effects of dumping.

408. The EC contended that Article 3 allowed consideration of the collective effects of dumped imports and imposed no limitations on such cumulation. The reference to "the dumped imports" in Article 3:4 confirmed that cumulation was permissible, and there were no criteria in the Agreement for the application or non-application of cumulation. The EC had further contended that it had in fact concluded that there were not two distinct markets for audio cassettes in the EC; rather, the EC industry had been forced to compete simultaneously with Korean imports at the low end and Japanese exports at the high end of the market. In support of this contention, it noted that in its Definitive Regulation it had found that there was to a large extent commercial interchangeability between the various models of audio cassettes, and further noted that the majority of the sales of the largest Japanese exporter to the EC were of the same "normal" type of cassettes as those of Korea.

409. The Panel noted that Japan's claim was based on Article 3:4 of the Agreement. Thus, any analysis of that claim had to take as a starting point the language of that provision itself. Article 3:4 provided that

Footnote 4 provided that the effects of dumping were "[a]s set forth in paragraphs 2 and 3 of this Article." Footnote 5 provided an illustrative list of other factors which at the same time could be injuring the industry producing the like product.

410. The Panel noted that there was no explicit reference in Article 3:4, or elsewhere in the Agreement, to the "cumulation of the effects of dumped imports" of a product from different sources. Nor did Article 3:4 of the Agreement set forth any list of criteria that had to be fulfilled in order for a Party to "cumulate the effects of dumped imports" of a product from different sources. Rather, Article 3:4 required that a Party demonstrate that the dumped imports were, through the effects of dumping, causing injury, and that injuries caused by the other factors must not be attributed to the dumped imports. The Panel considered that whether a Party was permitted by the Agreement to conduct a single inquiry relating to injury caused by dumped imports of a product originating in more than one Party or was required to conduct an independent assessment of injury caused by dumped imports originating in each individual Party would depend upon the proper interpretation of the term "the dumped imports." If the term "the dumped imports" was deemed to refer to all imports of a product established to be dumped, taken collectively, then a single assessment of injury caused by dumped imports from more than one Party would be permissible. If, on the other hand, the term "the dumped imports" was interpreted to mean "dumped imports from a particular Party," then an independent assessment of injury caused by dumped imports of a product originating in each individual Party would be required.

411. The Panel recalled, however, that Japan in this dispute did not claim that the Agreement always required a Party to conduct an independent assessment of injury caused by dumped imports from each Party. Rather, Japan claimed that, because in the particular factual circumstances of this case audio cassettes originating in Japan and Korea operated in two "distinct markets," the use of "cumulation" in this case was inconsistent with Article 3:4. As Japan did not challenge the compatibility of cumulation itself with Article 3:4, but only its application in the particular factual circumstances identified above, the Panel considered that it was not called on in this case to determine the meaning of the term "the dumped imports" for the purposes of Article 3:4. Rather, the issue before the Panel was whether Article 3:4 should be interpreted to contain criteria regarding the conditions of competition between dumped imports from different Parties which had to be fulfilled prior to assessing the effects of dumped imports from different Parties on a cumulative basis. In this respect, the Panel recalled its view that the Agreement contained no reference to the concept of "cumulation of the effect of dumped imports" from different Parties, much less specific criteria for the application of such a practice, but referred to an assessment of whether "the dumped imports" were causing injury. The Panel further considered that the meaning of the term "the dumped imports" as used in Article 3:4 of the Agreement could not be construed to depend in each particular case upon the extent of competition between dumped imports of a product originating in different Parties. In this respect, the Panel observed that Japan had identified no basis in the text of Article 3 to support the view that the meaning of the term "the dumped imports" was dependent in each particular case upon the extent of competition between dumped imports originating in different Parties. Nor did the Panel consider that anything in the context of Article 3:4 or the object or purpose of the Agreement would allow it to imply such an interpretation.

412. For the foregoing reasons, the Panel concluded that the cumulation by the EC of the effects of dumped imports from Japan and Korea was not inconsistent with its obligations under Article 3:4 of the Agreement by reason of the fact that imports of the like product from the two countries operated in "distinct markets."

413. The Panel next considered the claim of Japan that the EC, in deciding whether to cumulate the effects of dumped imports from Japan and Korea in this case, had failed to take into account certain criteria which it normally took into account when making such a decision, and that this deviation by the EC from its usual rules regarding cumulation was inconsistent with the requirement of Article 3:1 that a Party conduct an "objective examination."

414. The Panel recalled that Japan had argued that the Article 3:1 obligation to carry out an objective examination "would be nullified if a party were free to alter its rules whenever they interfered with the desired outcome of an investigation." In this case, Japan argued, the EC had failed to take into account two important criteria which it usually took into account in its cumulation decisions, i.e., the comparability in the increase in the volume of imports from a previous comparable period and the low level of prices attributable to the products of all supplying countries. Japan contended that the change in imports from Korea (up 500 per cent) was markedly different from that of Japan (up 8 per cent). Similarly, Korean producers competed at the lower end of the market, while Japanese producers supplied the higher-quality segments of the market, and prices in the two sectors differed by a factor of two or more. Japan contended that the EC likely would have reached a different decision on cumulation had it considered these factors.

415. The EC argued that the Agreement did not set forth any criteria governing cumulation. While the EC had not developed "rules" with respect to cumulation, it had developed a coherent administrative practice, from which it had not departed in this case. In any event, the EC was not obliged to forever follow the same administrative practices. Finally, Article 3:1 required an "objective examination" of the three factors identified therein, and had nothing to do with the consistent application of an administrative practice relating to cumulation.

416. The Panel recalled that Article 3:1 of the Agreement required that a determination of injury "involve an objective examination of both (a) the volume of dumped imports and their effect on prices in the domestic market for like products, and (b) the consequent impact of these imports on domestic producers of such products." Thus, the Panel considered that under Article 3:1 a Party was required to conduct an objective examination of injury and causality in light of the factors which were relevant under Article 3. As the Panel noted in its previous discussion (paragraph 411), however, Article 3:4 of the Agreement did not require that a Party make a determination regarding the conditions of competition among dumped imports from different sources or between dumped imports and the domestic like product as a prerequisite to deciding whether to cumulate the effects of dumped imports from multiple sources. Thus, even if the EC had deviated from its usual administrative practice (an issue on which the Panel was reluctant to pronounce itself), the Panel did not consider that the alleged failure of the EC to consider certain facts and to explain how they supported a determination which the EC was not required by Article 3 to make represented a breach of the objective examination requirement of Article 3:1.

417. For the foregoing reasons, the Panel considered that the EC had not acted inconsistently with its obligations under Article 3:1 of the Agreement by reason of its alleged failure to take into account certain criteria which it normally took into account when deciding whether to cumulate the effects of dumped imports from more than one Party.

418. The Panel then examined the claims of Japan that the EC had failed to establish that there had been a significant increase in the volume of dumped imports, that there had been significant price undercutting by the dumped imports, or that the effect of the dumped imports was otherwise to depress prices or prevent price increases which otherwise would have occurred, as required by Article 3:2, and that the EC further had failed with respect to these factors to meet the requirements of Article 3:1 that it make an injury determination on the basis of an "objective examination" of "positive evidence."

419. The Panel recalled that Japan and the EC at points in this dispute had indicated differing views regarding the nature of the obligation in Article 3:1 and 3:2 to consider price and volume effects prior to determining the existence of injury. Japan had argued that the EC had failed to establish that there had been a significant increase in the volume of dumped imports, that there had been significant price undercutting by the dumped imports, or that the effect of the dumped imports was otherwise to depress prices or prevent price increases which otherwise would have occurred. In the view of Japan, some combination of these three factors had to be present, at a significant level, in order for a Party to make an affirmative finding under Article 3. The EC, however, had argued that the requirement of Article 3:2 that an investigating authority "consider" whether there were significant volume or price effects was purely "procedural" and had no substantive component.

420. The Panel considered, however, that during the course of the proceedings the expressed positions of the parties had converged. Specifically, the EC while continuing to assert that Article 3 imposed purely "procedural" requirements, had stated that:

The EC had further stated that "[t]he indices and factors of paragraphs 2 and 3 of Article 3 must be 'considered' and thus shown to be present; obviously consideration of these elements cannot lead to conclusions that are not supported by the consideration and just as obviously that is not the EC approach."

421. The Panel further noted that the EC and Japan had at certain points in the proceeding expressed differing views regarding Japanese statements that "some combination" of significant volume increases, price undercutting and suppression/depression must exist in order to arrive at an affirmative injury determination. The EC had initially interpreted the Japanese position to be that more than one of these criteria had to be satisfied in order to reach an affirmative determination. The EC had responded that "it is sufficient if one of these criteria [i.e. volume, undercutting or suppression/depression] are satisfied . . . One of these criteria is necessary but also sufficient." However, Japan had later clarified that it "does not claim that both significant volume increase and significant price undercutting / price suppression / price depression must be present before an affirmative finding may be made under Article 3 . . . ."

422. In light of the foregoing, the Panel considered that the parties to the dispute were in agreement that an affirmative finding of injury under Article 3 could not be made unless one or more of the criteria set forth in Article 3:2 (i.e. significant increase in dumped imports, significant price undercutting by the dumped imports, significant price suppression or depression) was present. The parties further were in agreement that not all these criteria needed to be present in order to support an affirmative finding of injury; rather the presence of one of these criteria could be sufficient. The Panel therefore proceeded to examine the claims of Japan in light of this interpretation of the Agreement shared by the parties to the dispute.

423. The Panel recalled that Japan had raised two related claims with respect to the EC's consideration of the volume of dumped imports as it related to its affirmative injury determination. First, Japan had argued that, although the EC had considered the volume of dumped imports from Japan, it had failed to make a finding that the increase in the volume of dumped imports from Japan was "significant," in contravention of the obligation in Article 3:1 that an injury determination be based on positive evidence and an objective examination and that of Article 8:5 that a notice contain findings and conclusions on material issues of law and fact. Second, Japan argued that the data in this case would not support a finding that any increase in the volume of dumped imports from Japan was significant as required by Article 3:2.

424. The EC contended that it had considered whether there was a significant increase in the volume of dumped imports from Japan and had found, at least implicitly, that there was. It further argued that

425. The Panel recalled that the EC in its Definitive Regulation had examined the volume of dumped imports from Japan individually as well as the volume of dumped imports as a whole. Thus, the EC Council in its Definitive Regulation had noted that the EC Commission in its Provisional Regulation had based its findings on, inter alia, the following facts:

The Definitive Regulation had noted that no new facts had been submitted concerning these findings. Definitive Regulation, recitals 26-27. The Definitive Regulation further had noted arguments by Japanese exporters that their exports could not have been responsible for causing injury because the market share of their exports decreased from 42 to 35 per cent between 1985 and 1988. The EC had rejected these arguments on the grounds that:

Definitive Regulation, recitals 35-37.

426. The Panel noted that the claim of Japan that the EC had acted inconsistently with Article 3:4 of the Agreement by cumulating the effects of dumped imports from Japan and Korea and its claim that the EC had failed to establish under Article 3:2 that there had been a significant increase in dumped imports from Japan were integrally linked. Specifically, the Panel considered that Japan's claim with respect to Article 3:2 was predicated on the assumption that the EC was required in this case to assess the existence of injury caused by dumped imports on a country-by-country basis. However, the Panel recalled its conclusion that Japan's claims with respect to the cumulation by the EC of the effects of dumped imports from Japan and Korea were without merit. The Panel further recalled that Japan had at no point during the course of the proceedings advanced arguments regarding the meaning of the term "the dumped imports" as used in Article 3:2 or contended that this term had a different meaning in Article 3:2 than in Article 3:4. Under these circumstances, and taking into account the Panel's disposition of Japan's claim regarding cumulation, the Panel did not consider that it was necessary or appropriate for it to reach Japan's claim that the EC had failed to establish that there had been a significant increase in the volume of dumped imports from Japan.

427. The Panel noted that it had specifically asked Japan in a written question how its claims regarding the volume and price effects of dumped imports from Japan related to its claim regarding cumulation, whether these claims stood independently of the cumulation claim, and if so, how. Japan had responded by referring to its view that injury had to be assessed on a country-by-country basis. It had further contended that, although the EC had given cumulated figures for dumped imports from Japan and Korea, it had not made an examination of "overall" price undercutting. In Japan's view a determination of the significance of volume increases had to take into account price effects and vice versa, and volume and price effects therefore had to be considered "on the same basis." As a result, Japan contended, "the application of the issue of cumulation under Article 3:2" was not properly before the Panel. The Panel considered, however, that Japan's argument was without merit. In effect, Japan was asking the Panel, having disposed of Japan's claims regarding cumulation under Articles 3:4 and 3:1, to assume for the purposes of this proceeding that the EC was required by Article 3:2 to assess the existence of a significant increase in the volume of dumped imports on a country-by-country basis, without advancing any basis under Article 3:2 for such a requirement, solely because the EC allegedly had failed to assess the significance of price undercutting on a cumulative basis. In the view of the Panel, however, the existence of an alleged inconsistency in the manner in which the EC considered volume and price effects provided the Panel with no basis to decide whether those effects should be assessed on a cumulative or country-by-country basis.

428. The Panel recalled that Japan had argued that the EC had failed to establish, as required by Article 3:2 of the Agreement, that exports from Japan had been involved in "significant" price undercutting. Japan had further argued that the EC, by "zeroing" "overcut" sales, had acted inconsistently with the requirement of paragraph 3:1 regarding objective examination and the requirement of Article 3:2 that it "consider" price undercutting.

429. The EC responded that, while it did find in its Definitive Regulation that there was significant price undercutting on the part of Japanese exports in the German market, it was not a decisive factor in its injury determination. The EC considered it significant that the only member state market in which price undercutting by Japanese imports, albeit limited, was found to exist (Germany), was the market where the EC industry retained a substantial sales base.

430. The Panel first considered the claim of Japan that the EC had failed to establish that exports from Japan were involved in "significant price undercutting" in the EC market within the meaning of Article 3:2 of the Agreement. Japan had advanced two bases for this claim. First, it contended that the volume of undercutting by Japanese exporters was inadequate to support a determination of "significant price undercutting" within the meaning of Article 3:2 because price undercutting by Japanese exporters was limited to a single market, Germany, and because the proportion of exports from Japan to the German market which were undercutting the prices of the domestic industry was, at most, not much larger than 10 per cent. Second, it contended that the margins of undercutting were too small to be "significant" within the meaning of Article 3:2. Japan further argued that in assessing whether the EC had established that the margin of undercutting was significant, it had to take into account defects in the manner in which the EC calculated the average margin of undercutting of dumped imports from Japan.

431. The Panel noted that, in its Provisional Regulation, the EC had found that there was weighted average price undercutting by Japanese imports in the Community market of 6 per cent. The EC had further noted that price undercutting in the UK and French markets was "insignificant," while that in the German market was on average 11 per cent (Provisional Regulation, para. 67). In its discussion of causation, the EC had noted that, in Germany, Japanese exporters had "practised a significant price undercutting, up to a maximum of 18.5 per cent" (Provisional Regulation, para. 84). The EC summarized its conclusions in its Definitive Regulation, stating that "for a large number of transactions, large price undercutting was found on the part of Korean exporters and significant price undercutting on the part of Japanese exporters in the German market where the Community industry still retained a large market share" (Definitive Regulation, recital 26). Thus, the Panel concluded that the EC had not determined that there was "significant" price undercutting by Japanese exporters in the Community market as a whole. Rather, the EC had concluded only that there had been significant price undercutting by dumped imports from Japan in the German market. The Panel further noted that the EC had not argued before the Panel that it had implicitly made a finding of significant price undercutting in the EC market as a whole.

432. The Panel noted that the claim of Japan that the EC acted inconsistently with Article 3:4 of the Agreement by cumulating the effects of dumped imports from Japan and Korea and its claim that the EC failed to establish under Article 3:2 that there had been a significant price undercutting by dumped imports from Japan were integrally linked. Specifically, the Panel considered that Japan's claim with respect to Article 3:2 was predicated on the assumption that the EC was required in this case to assess the existence of injury caused by dumped imports on a Party by Party basis. However, the Panel recalled its conclusion that Japan's claims with respect to the cumulation by the EC of the effects of dumped imports from Japan and Korea were without merit. The Panel further recalled that Japan had at no point during the course of the proceedings advanced arguments regarding the meaning of the term "the dumped imports" as used in Article 3:2 or contended that this term had a different meaning in Article 3:2 than in Article 3:4. Under these circumstances, and taking into account the Panel's disposition of Japan's claim regarding cumulation, the Panel concluded that it was neither necessary nor appropriate for it to reach Japan's claim that the EC had failed to establish that there had been significant price undercutting by dumped imports from Japan.

433. The Panel recalled that Japan had argued that the methodology used by the EC for calculating the extent of price undercutting by Japanese exports to the EC market was defective and thus constituted neither an "objective examination" of price undercutting within the meaning of Article 3:1 nor the "consideration" of price undercutting required by Article 3:2 of the Agreement. Specifically, Japan argued that the EC had acted inconsistently with Articles 3:1 and 3:2 because it had "zeroed" "overcut" sales when calculating an average margin of undercutting.

434. The Panel recalled its conclusion (paragraph 432) that it was neither necessary nor appropriate for it to reach Japan's claim that the EC had failed to establish that there had been significant price undercutting by dumped imports from Japan. In the view of the Panel, however, this did not mean that the Panel should not consider claims regarding alleged errors in the manner in which the EC had determined the extent of price undercutting by allegedly dumped imports from Japan. The EC had examined the extent of price undercutting by allegedly dumped imports from Japan, and it was clear from the Provisional and Definitive Regulations that this examination of the extent of undercutting by imports from Japan represented an element in the EC's examination of the existence of price undercutting by dumped imports. Under these circumstances, the Panel considered that it was required to review claims relating to the manner in which the EC had determined the extent of price undercutting.

435. The Panel noted the EC's argument that its determination of the existence of price undercutting by Japanese imports in the German market was not a "decisive factor" in its affirmative injury determination. However, the Panel did not consider it appropriate or feasible to examine, based upon statements made subsequent to the conclusion of the investigation, the degree of importance the EC had placed on this factor. Rather, the Panel concluded based on the Provisional and Definitive Regulations themselves that the EC had considered its determination of price undercutting by Japanese dumped imports in the German market to be relevant to its affirmative injury determination. Under these circumstances, the Panel considered that it was obliged to consider claims by Japan that this determination was based on a flawed methodology for determining the existence and extent of price undercutting.

436. The Panel recalled Japan's claim that the methodology used by the EC to calculate an average margin of undercutting, involving the "zeroing" of sales at "overcutting" prices, was therefore defective and thus constituted neither an "objective examination" within the meaning of Article 3:1 nor the "consideration" of price undercutting within the meaning of Article 3:2. The Panel observed that the EC clearly had considered price undercutting by dumped imports from Japan, in the sense that it had examined the existence and extent of such price undercutting. Thus, it appeared to the Panel that Japan's claim related to the manner in which price undercutting was considered. The Panel therefore turned to Japan's claim that the EC's methodology for the consideration of price undercutting was not consistent with the requirement of Article 3:1 that it conduct an "objective examination" of price undercutting. The Panel further considered that a review of whether a determination of material injury was in conformity with this requirement necessitated an examination of whether the investigating authorities had examined all relevant facts before them (including facts which might detract from an affirmative determination) and whether a reasonable explanation had been provided of how the facts as a whole supported the determination made by the investigating authorities. The Panel therefore considered whether as a result of the averaging methodology contested by Japan the EC had failed to conduct an objective examination with respect to price undercutting.

437. The Panel observed that the consideration of the existence of significant price undercutting as envisioned by Articles 3:1 and 3:2 was not an abstract exercise, but rather related to the process of determining whether dumped imports had, through the effects of dumping, caused material injury to a domestic industry. In the view of the Panel, the extent to which price undercutting would have an impact on a domestic industry would be a function of two variables, the number of sales at undercutting prices, and the extent of the undercutting of such sales. The number of sales at undercutting prices was particularly important, because it would provide an indicator of the likely number of sales lost by the domestic industry. The margin of undercutting of such sales was relevant to the extent that in non-price sensitive products a small margin of undercutting might not play a decisive role in purchasers' decision-making. The Panel further observed that the calculation of an average margin of undercutting for all sales, whether or not at undercutting prices, might not be the most effective manner to assess the impact of price undercutting on a domestic industry, as it limited the ability of the investigating authority independently to examine these two variables. Nevertheless, average margins of undercutting could provide data of utility in considering the existence of significant price undercutting.

438. Japan had not claimed that the calculation of average margins of undercutting was inconsistent with the Agreement. Rather, Japan's claim in this case was that the EC in this case should have used a weighted average to weighted average methodology, which did not "zero" sales at overcutting prices, for determining an average margin of undercutting. Put in the context of Japan's claim regarding the failure of the EC to conduct an "objective examination," Japan's argument could be that the EC failed to consider relevant evidence by disregarding the extent to which some sales were at prices in excess of those charged by the domestic industry. However, the Panel did not find this argument convincing. Specifically, the Panel considered that in the event that certain sales were at undercutting prices, such sales could have an impact on the domestic industry (for example, in terms of lost sales) irrespective of whether other sales might be made at prices above those charged by the domestic industry. Thus, to require an investigating authority to base its analysis of undercutting on weighted average margins of undercutting which offset undercutting prices with "overcutting" prices would require the investigating authority to conclude that no undercutting existed when in fact there might be substantial volumes of sales at undercutting prices which might contribute toward material injury suffered by a domestic industry. In this respect, the Panel noted that Japan had argued that it could "reasonably claim that there was no undercutting in the EC because any undercutting found in the German market is certainly offset by overcutting in the other eleven EC markets." While the Panel was not required for the reasons stated in paragraph 432 above to decide, and did not offer any view, as to whether undercutting in the market of a single Member State by imports from Japan might be sufficient to support an affirmative injury determination, the Panel considered it evident that undercutting in the German market did not cease to occur simply because sales in other markets were in excess of the prices charged by the domestic industry.

439. For the reasons stated above, the Panel concluded that the EC's affirmative injury determination was not inconsistent with Articles 3:1 and 3:2 of the Agreement by reason of the methodology used by the EC to calculate an average margin of price undercutting.

440. The Panel recalled that Japan had argued that the EC had failed to satisfy the requirement of Article 3:2 to consider "whether the effect of such [dumped] imports is otherwise [than through price undercutting] to depress prices to a significant degree or prevent price increases, which otherwise would have occurred, to a significant degree . . . . " because the EC had not established the existence of a causal link between dumped imports and price suppression and depression. Japan further argued that the EC's Provisional and Definitive Regulations did not reflect the existence of positive evidence or the conduct of an objective examination with respect to the consideration of price depression / suppression for the same reason.

441. The Panel noted that Japan's claim regarding price suppression/depression differed from its claims regarding volume effects and price undercutting. With respect to the latter two claims, Japan argued that the EC had failed to establish the "significance" of the increase in the volume of dumped imports or of the undercutting by dumped imports from Japan. With respect to price suppression/depression, on the other hand, Japan claimed that the EC had failed to establish the existence of any causal link between dumped imports from Japan, through other than price undercutting, and price suppression or depression in the EC market. In the view of Japan, while the examination of price undercutting was a "straightforward matter of fact" the examination of price suppression and depression incorporated a notion of causation, i.e. that there had to be a link between the price movements of the imports and those of the domestic like product.

442. The Panel noted that the EC in this case had not made an explicit finding that dumped imports had depressed, through other than price undercutting, prices to a significant degree or had prevented price increases, which otherwise would have occurred, to a significant degree. The Panel further noted that Japan had not claimed that the EC's failure to make a finding on this point was inconsistent with the Agreement. Rather, Japan's claim, although couched in terms of a failure to "consider" price depression and suppression caused by other than price undercutting, in fact appeared first to be that the EC had failed to establish the existence of any price suppression or depression because it had failed to establish based on positive evidence and an objective examination the existence of any causal link between the price movements of dumped imports from Japan, other than through price undercutting, and those of the domestic like product. The Panel considered, however, that before it could review whether a factual determination of a Party was supported by positive evidence and reflected an objective examination under Article 3:2 it had first to examine whether the Party had in fact made any such determination and whether that determination represented a basis on which the Party had relied in reaching an affirmative injury determination.

443. In light of the foregoing, the Panel proceeded to review the Provisional and Definitive Regulations of the EC. The Panel observed that the EC in recital 79 of its Provisional Regulation had stated that:

The Panel considered that the reference to price suppression and price erosion separately from price undercutting indicated that the EC had considered at the time it had prepared its Provisional Regulation that there existed price effects caused by other than price undercutting. In addition, the EC in recital 84 of its Provisional Regulation had found that, in Member States markets other than Germany (where there was evidence of price undercutting),

In the view of the Panel, this statement could only be interpreted to mean that the EC had concluded that dumped imports from Japan in Member States other than Germany, although not made at undercutting prices, had forced EC producers to reduce their prices, i.e. that there was price depression in these markets caused by Japanese dumped imports. Accordingly, the Panel considered that the EC's Provisional Regulation indicated that the EC had examined whether there was price suppression or depression caused by dumped imports from Japan otherwise than through price undercutting, had concluded that there was, and had relied in part on these conclusions as a basis for its affirmative injury determination.

444. The Panel noted, however, that the discussion of this matter in the EC Council's Definitive Regulation differed significantly from the Commission's discussion in the Provisional Regulation. Specifically, recital 33 of the EC's Definitive Regulation, which was the first recital in the portion of the EC's Definitive Regulation titled "Effect of dumped imports," stated that:

The Panel noted that this recital did not refer to the Commission's finding of price suppression as stated in recital 79 of the Provisional Regulation. The Panel further noted that the Definitive Regulation did not repeat the statement in recital 84 of the Provisional Regulation that in Member State markets other than Germany, "Japanese exporters resold their dumped imports at prices which forced the Community industry to undersell in an attempt to retain its market share." Finally, the Panel further observed that the EC had stated, in recital 39 of its Definitive Regulation, that

445. The Panel considered that these differences between the Provisional and Definitive Regulations indicated that the Council had modified the bases for its affirmative injury determination as compared to the bases stated by the Commission in the Provisional Regulation. Specifically, the Panel considered that the EC had not, in its Definitive Regulation, relied on the existence of price suppression or depression caused by dumped imports from Japan through other than price undercutting as a basis for its affirmative determination of injury caused by dumped imports through the effects of dumping. To the contrary, the EC Council in its Definitive Regulation had narrowed its conclusions regarding the price effects of dumped imports from Japan to those arising from price undercutting.

446. The Panel noted that the bases for the definitive affirmative injury determination in this case were as stated in the Definitive Regulation promulgated by the EC Council. The Panel observed that in many cases the Council had adopted the views of the Commission and to the extent it did so the reasoning in the Provisional Regulation was relevant as an explanation of the bases for the imposition by the EC Council of a definitive duty. However, where, as here, the findings of the EC Council deviated from those made by the Commission, the Panel considered that the determination of the EC Council was not based on those findings of the Commission.

447. The Panel recalled Japan's claim that the EC had failed to establish that dumped imports from Japan had, through other than price undercutting, caused price suppression or depression. As the foregoing discussion indicates, however, the EC had not in its Definitive Regulation determined that there was price suppression or depression caused by dumped imports from Japan through other than price undercutting and thus had not relied on such price suppression or depression as a basis for its affirmative injury determination. The Panel therefore considered that it could not review whether such a determination was based on positive evidence and reflected an objective examination pursuant to Article 3:1. The Panel further recalled the agreement of the parties to the dispute that it was not necessary for all the criteria identified in Article 3:2 to be satisfied in order to support an affirmative injury determination; thus, the absence of price suppression or depression as a basis for the EC's affirmative injury determination did not give rise to an inconsistency with Article 3:2.

448. For the foregoing reasons, the Panel concluded that the EC had not acted inconsistently with Articles 3:1 and 3:2 of the Agreement by reason of its failure to establish that dumped imports from Japan had, through other than price undercutting, caused price suppression or depression.

449. The Panel recalled Japan's argument that an increase in the price of imports of audio cassettes from Japan to their normal value would not have led to a general increase in the prices of audio cassettes in the EC market, as Japanese exporters would simply have supplied the EC market from sources in the EC or third countries (which already accounted for 56 per cent of the audio cassettes supplied to the EC market by Japanese companies). Japan stated that this shift in the source of supply by Japanese exporters had in fact occurred after anti-dumping duties on audio cassettes from Japan were put in effect. In the view of Japan, this situation supported its view that the EC had failed to establish that dumped imports from Japan were, through the effects of dumping, causing injury to the EC industry within the meaning of Article 3:4. Japan contended that its argument was supported by the Report of the Panel in Canadian countervailing duties on grain corn from the United States, adopted 26 March 1992, BISD 26 March 1992, which concluded that "the Canadian industry would have been injured whether or not the US exports had been subsidized, and that therefore the US subsidies were not the cause of the injury."

450. The Panel observed that Japan appeared to be advancing a factual argument on the basis of which it asked the Panel to decide that the EC had failed to demonstrate that dumped imports from Japan were causing injury through the effects of dumping. The Panel noted that the task of the Panel was to review the affirmative injury determination of the EC and not to make an independent, de novo assessment regarding the existence of injury and causality. Thus, the Panel could not take into account information regarding events which occurred after the date of the EC's Definitive Regulation. The Panel further noted that Japan had not placed before it information available to the EC at the time of its investigation from which the accuracy of Japan's factual assertion could be confirmed. For example, the Panel had before it no information regarding the ability of EC and third country production facilities to supply the EC market. Assuming arguendo that such information existed and was available to the EC at the time it was conducting its investigation, the argument of Japan did not indicate that dumped imports from Japan were not causing injury to the EC industry. Rather, if Japan's factual assumptions were correct, it would indicate imports from Japan had maintained their market share in the EC as a result of their pricing. Further, Japan's argument did not suggest that dumped imports from Japan were not responsible for injury to the EC industry, but rather that any injury suffered by the EC industry as a result of dumped imports from Japan would simply be replaced by injury caused by audio cassettes from other sources in the event that the prices of imports from Japan were raised to their normal value. In this regard, the Panel considered that the task of a Party under Article 3:4 was to establish whether dumped imports were, through the effects of dumping, causing injury to the domestic industry. A Party was not required to examine whether, in the event duties were imposed sufficient to offset injury caused by dumped imports from a particular Party, injury might continue or recur as a result of an increased supply of the product from other sources.

451. Further, the Panel noted that Japan's reliance on the Grain Corn Panel was misplaced. In the view of the Panel, the Grain Corn Panel had concluded that the findings of injury of the Canadian International Trade Tribunal had focused on the effects of US subsidies and their impact on world market prices rather than on the effects of subsidized imports into Canada. Specifically, the Panel had concluded that Canada had not examined the mandatory elements of Article 6:2 of the Agreement on the Interpretation and Application of Articles VI, XVI and XXIII of the General Agreement on Tariffs and Trade, i.e., the volume and price effects of subsidized imports. In this investigation, on the other hand, the EC clearly had examined the volume and price effects of dumped imports on the EC industry within the meaning of Article 3:2 of the Agreement.

452. For the foregoing reasons, the Panel concluded that the EC had not failed to establish that dumped imports from Japan were, through the effects of dumping, causing injury to the EC industry within the meaning of Article 3:4 by reason of Japan's argument that an increase in the price of imports of audio cassettes from Japan to their normal value would not have led to a general increase in the prices of audio cassettes in the EC market.

IX. CONCLUSIONS AND RECOMMENDATIONS

453. The Panel recalled its conclusions with respect to the preliminary objections of the EC that:

454. The Panel further recalled its conclusions with respect to the averaging methodology used by the EC in the comparison of export prices and normal values, that:

455. The Panel further recalled its conclusions with respect to Japan's claims regarding a so-called "asymmetrical" comparison of the export price and normal value, that:

456. The Panel further recalled its conclusions with respect to the construction by the EC of a normal value, that

457. The Panel further recalled its conclusions with respect to the EC's affirmative final determination of material injury that:

458. With respect to the recommendation to be addressed to the Committee on Anti-Dumping Practices, the Panel recalled that Japan had requested that the Panel recommend that the EC Regulation imposing the definitive duty in this case be revoked, that the duties already paid be reimbursed, and that the EC bring the relevant provisions of its Basic Regulation and its application into conformity with the Agreement.

459. The Panel first considered Japan's request that the Panel recommend that the EC revoke the Regulation imposing a definitive duty and reimburse duties already paid in this case. In the view of the Panel the nature of the recommendation made by the Panel should appropriately reflect the nature of the inconsistencies with the Agreement found to exist. In this case, the Panel had disposed of Japan's claims that the EC had imposed a duty without a proper determination of injury. In addition, the Panel had not found that the EC had imposed a duty in the absence of dumping. Rather, the Panel had found that the methodology used by the EC to calculate the extent of dumping was inconsistent in one respect with the Agreement. Further, Japan had not alleged that dumping did not exist in this case. To the contrary, calculations submitted by Japan to the Panel indicated that, even if all four of its claims regarding the EC's methodology for calculating the extent of dumping had been upheld by the Panel, some dumping had occurred. Taking into account the fact that the EC had imposed a duty in this case that was substantially lower than the margin of dumping it had found to exist, the Panel could not judge whether the inconsistent methodology used by the EC had any effect on the duty actually imposed. Under these circumstances, the Panel did not consider it appropriate to recommend that the Committee request that the EC revoke the Regulation imposing a definitive duty in this case. Rather, the Panel recommends that the Committee request the EC to reconsider its determination in light of its obligations under the Agreement. In the view of the Panel, if that reconsideration results in a determination that the imported product was not dumped, then the EC should revoke its anti-dumping duty and reimburse the duties collected. If it determines that those imports were dumped, but to a lesser extent than the duties actually imposed, it should reimburse the duties collected to the extent of the difference.

460. The Panel further recommended that the Committee request that the EC bring its Basic Regulation into conformity with its obligations under the Agreement.

461. One member of the Panel dissociated himself from a number of aspects of the above conclusions relating to asymmetry and expressed the following dissenting opinion:

462. This member agreed with the majority of the Panel on that part of the conclusion contained in paragraph 455 to the effect that the EC, by failing to make due allowance on its merits for differences in indirect selling expenses, which differences could affect price comparability, had acted inconsistently with its obligations under Article 2:6 of the Agreement; and that Articles 2(9) and 2(10) of the EC’s Basic regulation were mandatory legislation inconsistent with Article 2:6 of the Agreement because they precluded the making of due allowance, on its merits, for differences in indirect selling expenses, which differences could affect price comparability. However, this member did not share the view of the majority with regard to the treatment to be accorded to profits related to differences in the functions performed by the seller in the domestic and export markets.

463. This member agreed with the view of the Panel set out in paragraph 377, that it did not consider the existence of a difference in level of profits related to sales activities in the export and domestic markets would of itself necessarily be a difference which affected price comparability and for which due allowance should be made. However, in the opinion of this member, the circumstances where some provision may need to be made in respect to such profits would arise only in the context of differences in the level of trade. In this regard, the member noted that on its face the first sentence of Article 2:6 imposed an unqualified obligation that "the two prices shall be compared at the same level of trade, normally at the ex-factory level". This member considered that the term "other differences affecting price comparability" in Article 2:6 referred to differences which would affect the comparability of the export price and the normal value which were to be compared, and considered it evident that where the export price and normal value were at different levels of trade the comparability of these prices would be affected. He noted that Japan had not argued in this case that it would not be permissible to fulfil the unqualified obligation through the making of due allowances for differences in level of trade. Thus, the member concluded that adjustments for "the other differences affecting price comparability" could include adjustments necessary to ensure that the obligation to compare prices at the same level of trade was met, but did not preclude other approaches to meeting the obligation to compare prices at the same level of trade.

464. The member did not consider that the circumstance which could give rise to the need for an adjustment in the circumstances outlined above could be considered to exist in relation to the performance of different functions in one market beyond those performed in another market unless the differences in function arose from a difference in the level of trade or could lead to the conclusion that there was a need for an adjustment to ensure that the prices were compared at the same level of trade. Where there was no such distinction, it was not appropriate to seek to allocate profit according to functions, nor was it appropriate to assume that Article 2:6 of the Agreement necessarily required adjustments for differences in the way a firm’s activities were organised in different markets. This member did not share the interpretation in paragraph 376 of the contextual elements in Article 2, and did not agree that the drafters expected that some degree of profit relating to functions performed could be expected to be reflected in the price charged. The adjustments to export prices (established on the basis of the first arm’s length sale) provided for in the last sentence of Article 2:6 were aimed at establishing a reliable measure of a price at the ex-factory level, and were additional to any other adjustments which might be required to ensure price comparability (including adjustments to meet the obligation to compare prices at the same level of trade). The process of constructing normal values set out in Article 2:4 of the Agreement did not make reference to allocating profits on the basis of functions performed.

465. This member recalled that Japan’s argument was that there was a general requirement that the comparison of prices should be "fair"; that the mere determination of an appropriate level of trade did not guarantee a "fair comparison"; and that the EC had failed to select the appropriate level of trade for domestic and export sales, had made no allowance for the resulting unfairness, and therefore had failed to comply with the requirement to effect a "fair comparison". The member also recalled that Japan had identified three different meanings for the term "level of trade", i.e. the nature of the purchaser, the costs involved in the terms on which the goods were offered, and the nature of the seller. This member considered that costs involved in the terms on which goods were offered were covered by the requirement to make due allowance for differences in conditions and terms of sale, and that in any event differences in costs did not lead to a requirement to make adjustments for differences in the level of trade. Article 2:6 required that prices be compared "at the same level of trade, normally at the ex-factory level", which indicated clearly to this member that the intention was that the level of trade should be identified in terms of the seller, in particular since identification in terms of the buyer could create distortions where different markets required different approaches to the distribution of goods. Japan had claimed that by comparing the level of transfer from the parent company to the sales subsidiary in the importing country with the level of sales by a domestic sales subsidiary to an independent purchaser, the EC had not compared prices at the same level of trade. The EC had identified the level of trade in terms of the nature of the seller, and had argued that in this case the investigating authorities had decided that the appropriate level of trade in the country of export was the domestic sales network or sales subsidiary (adjusted to deduct all direct expenses), which was determined to be equivalent to the export sales department, since both agencies sold to distributors and wholesalers with a similar pattern of prices and quantities.

466. The member recalled the contention of the EC that it would have been able to make adjustments for differences in indirect expenses and profits had it concluded that export sales were at a different level of trade from domestic sales. In considering whether the EC’s Basic regulation permitted due allowances to be made in relation to profits arising out of differences in levels of trade, the member noted the conclusions recorded in paragraph 380 to the effect that the EC appeared to be precluded by its Basic regulation from making adjustments with respect to profits related to differences in the functions performed by the seller in different markets, but recalled his own conclusion that such differences in function did not define a level of trade, nor were adjustments for "other differences affecting price comparability" the exclusive approach to ensuring that the obligation to compare prices at the same level of trade was met. This member also noted the points raised in paragraph 381 in relation to the contention of the EC that the EC would have been able to make adjustments for differences in indirect expenses and profits had it concluded that the export sales were at a different level of trade than domestic sales. This member’s conclusions, from the information available, were therefore that with regard to adjustments for profits arising out of differences in levels of trade, the EC appeared to be able to take into account the need to adjust for profits in the context of differences in the level of trade. In the view of this member, in the light of the foregoing, the Panel did not have before it the information relating to the nature of the domestic operations which would permit it to reach a conclusion that the EC had failed to make any necessary adjustments relating to profits to ensure such a comparison.

467. Having reached this conclusion with regard to the identification of the level of trade and any adjustments required in respect of profits, the member noted that Japan had claimed that the EC’s failure to make an allowance for the "unfairness" arising from the selection of the level of trade had meant a failure to comply with a requirement to effect a "fair comparison". The member recalled that Japan had claimed that such a requirement was generated by Article 2:1 and 2:6 acting together. The member noted that Article 2:1 referred to export price being "less than the comparable price", while Article 2:6 began "In order to effect a fair comparison...", and went on to provide that the export price and the domestic price should be compared at the same level of trade, normally at the ex-factory level, and in respect of sales made at as nearly as possible the same time. The paragraph provided for due allowance to be made, on its merits, for differences in conditions and terms of sale, for differences in taxation, and for other differences affecting price comparability, and with additional allowances to be made in the case of constructed export prices. In the view of the member, Article 2:6 clearly stated the requirements for effecting a fair comparison. If these requirements were met, then the comparison could be deemed to be fair. With regard to the issue of profits in the context of ensuring that prices were compared at the same level of trade, it was the conclusion of this member that, on the basis of the information before it, the Panel could not conclude that the EC had not made a fair comparison, or that it was precluded from doing so by its Basic regulation.

468. Accordingly, in contrast with paragraph 455, the member’s conclusion with respect to Japan’s claims regarding a so-called "asymmetrical" comparison of the export price and normal value, was that:

ANNEX

Geneva, 2/4/93

MP/PL/eba/39.93

Dear Mr. Ortega,

I am writing to you in relation to the panel established by the Committee on Anti-Dumping Practices, at the request of Japan, to examine the Community's anti-dumping proceedings on imports of audio tapes in cassettes from Japan.

You will recall that the Community had not been able to agree on the terms of reference for this panel, because we felt that greater precision was needed as to the questions, both of fact and of law, to be asked of the panel, and that the documents tabled by Japan (ADP/85 and ADP/85/Add.1) did not allow us to understand properly these questions.

In connection with this issue, the Community and Japan have held several rounds of bilateral consultations, with a view to clarifying the questions which the panel will be required to examine, and we understand that Japan has sent you a letter concerning this matter. We further understand that you will transmit the clarifications you have received to the panel, once its composition will have been agreed upon.

It is only on the basis, in a spirit of compromise, and in the interest of an expeditious settlement of this dispute, that the Community is ready to set aside its many remaining doubts and criticisms, and can accept standard terms of reference based on the request from Japan contained in the above-mentioned documents.

I must point out, however, that this procedure does not prejudge the position that the Community has taken in the Committee as to the need for extreme clarity in a request for the establishment of a panel, and therefore should not be taken as constituting a precedent in future panel cases.

I take this occasion to thank you for your efforts at conciliation in this matter. We shall continue consultations with Japan in order to determine the composition of the panel in an expeditious manner.

I am copying this letter to the Mission of Japan, and the GATT Secretariat for circulation to the Committee.

Yours sincerely,

TRAN Van-Thinh (Signed)

Ambassador Représentant auprès du GATT

Mr. Armando Ortega

Chairman of the Committee on Anti-Dumping Practices

Centre William Rappard

Rue de Lausanne 154

1211 Geneve 21

Copy: M. Nakatomi, Japanese Mission

M. Woznowski, GATT

Geneva, 31 March 1993

MN/se/D.099

Dear Mr. Ortega,

I would like to take this opportunity to take this opportunity to thank you for all your efforts as the Chairman of the Anti-Dumping Committee to make progress in the panel proceedings on the imposition of anti-dumping duties by the EEC on imports of audio tapes in cassettes from Japan.

As was advised by the Chairman at the Anti-Dumping Committee held on 30 October 1992, the Government of Japan and the EEC had a series of bilateral consultations with respect to the questions which the panel would be expected to examine. I attach a copy of the paper, for your reference, which was informally submitted to the EEC in response to the EEC's request for further clarification on several points.

It is my pleasure to inform you that the EEC is satisfied with our clarifications and has agreed to accept the standard terms of reference on the panel proceedings.

Finally, I would like to draw your attention that this letter and the attachment are not legally required for the panel proceedings under the AD Code. Therefore, they should not be quoted in the terms of reference nor taken as constituting a precedent for the future panel cases.

A copy of this letter will be sent to the Mission of the EEC.

Yours sincerely,

Kazuo ASAKAI (Signed)

Minister

Enc.

c.c. Mr. Patrick Laurent

Mr. Armando Ortega

Chairman

Committee on Anti-Dumping Practices

Permanent Mission of Mexico

10a Avenue de Budé

1202 Geneva

Explanatory Note on ADP/85/Add.1

1. Paragraph 17 of ADP/85/Add.1

In the bilateral consultations regarding the audio cassettes dispute, Japan expressed its concern as to the methodology used by the EC to calculate constructed normal values, and it asked the EC to explain how such values were calculated for the companies involved, in particular regarding the amounts included in SG&A (selling, general and administrative expenses) and for profit. These subjects were covered by recitals 23 to 30 of the Commission Regulation No. 3262/90 regarding provisional determination, and recitals 12 to 16 of the Commission Regulation No. 1251/91 regarding definitive determination on the imposition of the anti-dumping duties in question.

The explanations provided by the EC in the course of the consultations, however, have neither answered to the question asked by Japan nor removed Japan's concerns, and that is why the matter is included in the document presented by Japan to the Committee on Anti-Dumping Practices.

But in this case Japan has decided to focus on the profit rates and SG&A expenses of the largest Japanese exporter. In constructed normal values for the largest Japanese exporter, the EC disregarded the different profit levels associated with different categories of audio cassette. By applying the higher profit level obtained by all the products to cassettes of the certain category to which the relevant models belong, the EC increased their apparent normal values, and consequently the dumping margin. Also with regard to the constructed normal values calculated for that company, the EC wrongly applied a cost of manufacturing basis rather than a turnover basis in allocating SG&A expenses, with the result that normal values were inflated.

These actions by the EC were not consistent with Article 2, and in particular paragraph 4.

2. Paragraph 19

The EC said that Japan's intentions on the issue of cumulation in this case was uncertain.

It is Japan's principal intention to challenge the application of the notion of "cumulation" in the circumstances of the audio cassettes decision rather than to deny that "cumulation" may ever be applied in the context of injury. However, this does not prevent Japan from arguing in the panel proceedings regarding the conformity of the practice itself of cumulation with the provisions of the Anti-Dumping Code.

Cumulation in this case conflicts with Article 3 of the Code, and in particular paragraph 4.

3. Paragraph 25

Another matter which was raised by the EC relates to Japan's complaint regarding injury. Japan considers that its position on this question is sufficiently clarified in the document, as several paragraphs in the document are devoted to demonstrate fully how the EC unjustifiably concluded that the EC's domestic industry was suffering injury, and consequently how it failed to comply with Article 3 of the Code. The EC proceedings failed to recognize that the Community industry fell into two distinct segments, that the segment with which Japanese exporters were competing was not suffering injury.

More specifically speaking according to the clarification paper submitted to Mr. Ortega by the EC, Japan does not dispute that all audio cassettes examined in the investigation are "like products".

Consequently, Japan does not dispute that there is only one audio cassette "domestic industry" in the EC.

Japan's argument can be presented as one concerning "impact" of imports, but it can also be set in the context of estimating the extent of injury. When a number of factors are allegedly contributing to the injury, the analysis can start from the total injury and ask what had caused it, or it can start from an alleged cause - dumping from Japan - and ask whether the result of this amounted to material injury. However, the distinction is academic. In practice, it is impossible to completely separate the issues of causation and injury.

Having said so, Japan's claim is that (a) the market for audio cassettes in the EC was divided in two parts, (b) Japan's exports competed in only one of these (the high quality part), (c) exports from Japan could not have been injuring AGFA (because it principally traded in the other part), (d) the EC did not show that BASF (the only possible significant victim of alleged dumping from Japan) was suffering any injury, (e) even if BASF was suffering injury, the EC did not establish that this injury was the result of dumping by Japan's exporters.

Therefore Japan asserts that the EC's decision was inconsistent with Article 3 of the Code, in particular paragraphs 1, 3 and 4.

4. Chapeau

Finally, Japan's intention to "reserve the right to elaborate on the issues" covered in the document is merely referring to the elaboration which always occurs in the parties submission to panels.