30 November 1992
Report of the Panel adopted by the Committee on Anti-Dumping
Practices on 27 April 1994
(ADP/87)
4.1 Request by Norway for certain data 233-234
4.2 Volume of the allegedly dumped imports (Articles 3:1 and 3:2) 235-250
[Page 2]
4.3 Price effects of the imports under investigation (Articles 3:1 and 3:2) 251-267
4.4 Impact of the imports under investigation on domestic producers of the like products (Articles 3:1 and 3:3) 268-274
4.5 Causal relationship between the allegedly dumped imports and material injury to the domestic industry (Article 3:4) 275-313
4.5.1 Other factors affecting the domestic industry 276-296
4.5.2 Material injury caused by the dumped imports, through the effects of dumping 297-306
4.5.3 Whether the imports under investigation were causing present material injury to the domestic industry 307-313
5. Continued imposition of the anti-dumping duty order (Article 9:1) 314-315
VI. ARGUMENTS PRESENTED BY THE EEC AS AN INTERESTED THIRD PARTY 316-327
VII. FINDINGS 328-589
1. INTRODUCTION 328-330
2. PRELIMINARY OBJECTIONS 331-351
(1) Preliminary objections of the United States regarding matters allegedly not within the Panel's terms of reference or not raised during consultations and conciliation 339-346
(2) Preliminary objections of the United States regarding matters not raised before the investigating authorities 347-351
[Page 3]
3. MERITS 352-589
A. INITIATION OF THE ANTI-DUMPING INVESTIGATION 352-364
B. DETERMINATION OF THE EXISTENCE OF DUMPING 365-486
(1) Procedural issues raised by Norway with respect to the investigation conducted by the Department of Commerce 371-384
(2) Issues raised by Norway regarding the substantive methodology used by the Department of Commerce in its determination of dumping 385-486
C. DETERMINATION OF THE EXISTENCE OF MATERIAL INJURY 487-582
(1) Volume of imports subject to investigation, price effects of the imports and consequent impact of these imports on the domestic industry in the United States 490-541
(2) Causal relationship between the allegedly dumped imports from Norway and material injury to the domestic industry in the United States 542-582
D. CONTINUED IMPOSITION OF THE ANTI-DUMPING DUTY ORDER 583-589
VIII. CONCLUSIONS 590-597
Pages
ANNEXES 219-234
[Page 4]
1. In a communication to the Committee on Anti-Dumping Practices ("the Committee") circulated on 17 June 1991 (ADP/57), Norway informed the Committee that on 2 May 1991 consultations had taken place under Article XXIII:1 of the General Agreement between the United States and Norway on the imposition of anti-dumping duties by the United States on imports of fresh and chilled Atlantic salmon from Norway. This communication stated that it was the understanding of Norway that these consultations were also to be considered as consultations under Article 15:2 of the Agreement on Implementation of Article VI of the General Agreement on Tariffs and Trade (hereinafter referred to as "the Agreement"). This understanding was confirmed by the United States in a communication circulated in document ADP/58, dated 17 June 1991.
2. A request by Norway for conciliation under Article 15:3 of the Agreement in this matter was circulated to the Committee on 11 July 1991 (ADP/61). The Committee held a meeting to examine this matter under Article 15:3 on 19 July 1991 (ADP/M/33).
3. On 24 September 1991, Norway requested that the Committee establish a panel in this dispute under Article 15:5 of the Agreement (ADP/65). On 16 October 1991, Norway supplemented its initial request for the establishment of a panel with a list of issues to be examined by the panel (ADP/65/Add.1).
4. At its regular meeting held on 21 October 1991, the Committee decided to establish a panel in the matter referred to the Committee by Norway in documents ADP/65 and Add.1. The Committee authorized its Chairman to decide, in consultation with the parties to the dispute, on the terms of reference of the Panel, and to decide, after obtaining the agreement of the two parties, on the composition of the Panel. The EEC reserved its right to present its views to the Panel as an interested third party (ADP/M/35).
5. On 6 November 1991, the Committee was informed by the Chairman in document ADP/69 that the terms of reference of the Panel established in this dispute were the following:
"To examine, in the light of the relevant provisions of the Agreement on Implementation of Article VI of the General Agreement, the matter referred to by Norway in documents ADP/65 and Add.1 and to make such findings as will assist the Committee in making recommendations or in giving rulings."
In the same communication, the Chairman informed the Committee that the composition of the Panel was as follows:
Chairman: Mr. Janus Kaczurba
Members: Mr. Peter Gulbransen
Mr. Meinhard Hilf
6. The Panel met with the parties to the dispute on 22 and 24 January, 4 and 6 March and 1 October 1992. The Panel heard the delegation of the EEC on 22 January 1992.
7. The Panel submitted its findings and conclusions to the parties on 23 October 1992.1
[Page 5]
8. The dispute before the Panel concerned the imposition by the United States on 12 April 1991 of an anti-dumping duty order on imports of fresh and chilled Atlantic salmon from Norway. The imposition of this order followed an affirmative final determination of dumping by the United States Department of Commerce and an affirmative final determination of injury by the United States International Trade Commission (USITC) with respect to these imports.
9. The anti-dumping duty investigation which led to the above-noted determinations was initiated by the Department of Commerce on 20 March 1990 after the Department had on 28 February 1990 received a petition for the initiation of an investigation from The Coalition for Fair Atlantic Salmon Trade, comprised of domestic producers of fresh and chilled Atlantic salmon. Also on 20 March 1990 the Department initiated a countervailing duty investigation with respect to these imports.
10. As indicated in the public notice of the initiation of this investigation, the product covered by the investigation was the species Atlantic salmon. All other species of salmon were excluded. The notice explains that "Atlantic salmon is a whole or nearly whole fish, typically (but not necessarily) marketed gutted, bled and cleaned, with the head on. The subject merchandise is typically packed in freshwater ice ("chilled"). Excluded from the subject merchandise are fillets, steaks, and other cuts of Atlantic salmon. Also excluded are frozen, canned, smoked or otherwise processed Atlantic salmon".2
11. This notice also explains that, while the petitioner had requested the Department to examine whether sales of salmon took place at prices below costs of production, at the time of the initiation of the investigation the Department of Commerce found the information provided by petitioner insufficient to warrant the initiation of a sales below costs of production investigation.
12. On 16 April 1990, the USITC issued a preliminary affirmative determination of injury in the anti-dumping duty investigation of imports of fresh and chilled Atlantic salmon from Norway.3 An affirmative preliminary determination of dumping by the Department of Commerce was published on 3 October 1990.4 As a result of this affirmative preliminary determination, the US Customs Service was instructed by the Department of Commerce to suspend liquidation of all entries of fresh and chilled Atlantic salmon from Norway which were entered, or withdrawn from warehouse, for consumption, on or after 3 October 1990 and to require a cash deposit or bond for all entries of this product corresponding to the preliminarily estimated margins of dumping, ranging from 1.90 to 4.76 per cent ad valorem.
13. The Federal Register Notice of the affirmative preliminary determination of dumping explains that, for seven out of the eight investigated exporters, the volume of home market sales was insufficient to constitute a viable basis for the calculation of the normal value. For these exporters, the provisional normal value was established on the basis of export prices to EEC countries.5 The Notice also indicates that, at the time of the preliminary determination, the Department of Commerce was investigating an allegation by the petitioner that the export sales to the EEC markets used as a basis for the calculation of the provisional normal value were made at prices below costs of production and that, for the purpose of its investigation of this allegation, the Department had on 21 August 1990 "delivered cost of production
2 55 Fed.Reg. 28 March 1990, p.11419 ↩
3 Fresh and Chilled Atlantic Salmon from Norway: Determination of the Commission in Investigation No. 731-TA-454 (Preliminary) Under the Tariff Act of 1930, Together with the Information Obtained in the Investigation, USITC Publication 2272, April 1990. ↩
4 55 Fed Reg. 3 October 1990, p.40418 ↩
5 55 Fed.Reg, 3 October 1990, pp.40419-40420 ↩
[Page 6]
questionnaires to eleven fish farmers who reportedly supplied the eight exporters with the subject merchandise during the period of investigation".6
14. An affirmative final determination of dumping ("sales at less than fair value") in this investigation was issued by the Department of Commerce on 25 February 1991.7 The Department found margins of dumping for the eight Norwegian exporters of salmon under investigation, ranging from 15.65 to 31.81 per cent ad valorem, and established an "all others" margin of dumping of 23.80 per cent ad valorem.
15. As explained in the Federal Register Notice of the final affirmative determination of dumping, the normal value used for comparison with export prices was determined as follows. Following the allegation by the petitioner that export sales of Atlantic salmon to EEC markets were made at prices below the costs of production, the Department compared the prices of the third country sales to the costs of production of salmon. These costs of production were calculated as the sum of (1) the simple average of the costs of production of farmers from whom the Department had obtained cost of production information through a sampling procedure, and (2) the exporter's selling general and administrative expenses.8 Generally, where the Department found for an individual exporter that more than 90 per cent of the third country sales were at prices below cost of production, the normal value was established on the basis of a constructed value.9
16. The constructed normal value for salmon sold by each exporter was calculated as the sum of (1) the simple average of the farmers' costs of productionand (2) the exporter's selling, general and administrative expenses, profit and packing. The Notice explains that "for all exporters, profit equal to the statutory minimum 8 per cent of the cost of production was applied" and that "In all cases, for salmon sold on or after January 1, 1990, a five NOK/kg. cost was added to the CV before profit".10
17. A more detailed description of aspects of the process of selecting the respondents and of the method of determining the normal value applied by the Department of Commerce in this investigation is appropriate for a better understanding of the issues in dispute before the Panel with respect to the determination of dumping.
18. On 30 April 1990, the Department of Commerce issued Section A questionnaires to eight exporters of Norwegian salmon who accounted for more than 60 per cent of imports of Atlantic salmon from Norway during the period of investigation. Included in these Section A questionnaires was a request to the exporters "to provide the name and address of each salmon farm from whom you purchased salmon for export to the United States during the POI." (period of investigation) This information was requested in order for the Department of Commerce to be able to quickly identify these farmers as respondents, if it turned out that the farmers rather than the exporters should be the principal respondents in the investigation. Responses from the exporters to the Section A questionnaires, including lists of the farms with which each exporter had dealt during the period of investigation, were received on 16 May 1990. Responses to Sections B and C of the questionnaire, issued to the exporters on 15 June 1990, were received on 27 July 1990.
19. On 30 April 1990, the Department of Commerce also issued a special questionnaire to three entities involved in the production and sale of salmon in Norway for purposes of determining whether salmon
6 55 Fed.Reg., 3 October 1990, p.40418. ↩
7 56 Fed.Reg., 25 February 1991, pp.7661-7678. ↩
8 56 Fed.Reg., 25 February 1991, p.7662. ↩
9 56 Fed.Reg., 25 February 1991, p.7664. ↩
10 56 Fed.Reg., 25 February 1991, pp.7663-7664. ↩
[Page 7]
farmers had knowledge, at the time of their sales to exporters, of the ultimate destination of the salmon. Information about the possible knowledge of farmers of the ultimate destination of the product was considered necessary to determine whether the farmers or the exporters were to be treated as respondents in the investigation. It should be noted in this respect that Norwegian salmon exporters generally do not farm salmon and that Norwegian salmon farmers generally do not export salmon. Responses to this special questionnaire were received on 16 May 1990.
20. In order to obtain further information on the farmers' possible knowledge about the ultimate destination of the salmon sold to the exporters, the Department of Commerce on 12 June 1990 issued a supplemental questionnaire to the Norwegian fish farmers' organization, Fiskeoppdretternes Salgslag (FOS). Based on data provided by the FOS in response to this questionnaire, the Department selected a sample of eight farms which were provided with a modified Section A questionnaire on 25 June 1990. On 25 July 1990, the Department determined, based on a review of the responses by these eight farms and other information collected up to that point in the investigation, that Norwegian salmon farmers did not generally know the ultimate market into which their product was sold. The Department concluded that the farmers should be excused from responding to the remaining sections of the questionnaire and that it should continue to consider the exporters as the proper respondent in this case.
21. On 3 August 1990, the petitioner in the investigation requested the Department of Commerce to determine whether export sales of Norwegian salmon to the EEC were at prices below costs of production. In support of this request, the petitioner alleged that actual Norwegian sales prices (the prices submitted by the exporters in their Section B responses) were below the average costs of production reported by a Norwegian Government study. The Department of Commerce accepted this request on 20 August 1990. Because of the structure of the Norwegian industry and the close interrelationship between the exporters and the farms, the Department decided to investigate the farms' costs of production. For the purpose of the investigation of the farms' costs of production, it was decided to develop a sample of farms for each exporter, each sample to be drawn from a universe of farms with which the exporter had actually dealt during the period of investigation. The farms were selected for each exporter from the list of farms submitted by the exporters in their responses to Section A of the questionnaire issued on 30 April 1990. Eleven firms had thus been selected.
22. On 30 August 1990, counsel for the Norwegian respondents reported that several of the farms selected by the Department for purposes of its costs of production investigation had not sold any salmon during the period of investigation to the exporters to which they had been linked. The Department determined that it could not develop new samples of farms for each exporter because it was not possible to determine from the lists provided by the exporters in response to the Section A questionnaire issued on 30 April 1990 which farms actually had sold salmon to the individual exporters during the period of investigation and there was insufficient time left in the investigation to develop a new sample, present questionnaires to new farms and analyze and verify the responses. The Department instead decided to proceed to collect costs of production information from the remaining fish farms selected for the survey, i.e. the seven farms which actually had supplied salmon to the exporters during the period of investigation, and to develop an average cost of production from these remaining farms. A Department of Commerce memorandum dated 13 February 1991 dealt with the question of whether this average costs of production figure should be calculated as a simple, or a weighted average of the costs of production of these famrs (i.e. whether the individual costs of production figures should be assigned a weight proportional to the share in total production volume in Norway of the different size categories of the farms in the sample). The Department found that no basis to weight costs of farms of different sizes and accordingly decided to use a simple average of the costs of production figures of the farms in the sample.
23. A discussion of several aspects of the methodology used by the Department of Commerce to calculate normal value in this investigation appears in the "Interested Party Comments" Section of the Federal
[Page 8]
Register Notice of the affirmative final determination of dumping.11 Particularly relevant to the matter before the Panel are the interested party comments, and responses by the Department of Commerce to such comments, with respect to the inclusion of a "freezing charge" in the cost of production of the farmers; the use of the highest verified cost of production of any farmer for one farmer, Nordsvalaks, which allegedly had failed to provide certain information, the methodology used to calculate cost of production of the farmers on the basis of the sample used by the Department of Commerce, the issue of whether Atlantic salmon should be treated as a perishable product for purposes of cost of production analysis and for purposes of comparing the normal value to the export prices, and the determination of the amount of certain processing costs.
24. On 2 April 1991, the USITC issued one final determination for the purpose of both the anti-dumping and countervailing duty investigations of imports of fresh and chilled Atlantic salmon from Norway12, in which it concluded that an industry in the United States was materially injured by reason of imports from Norway of fresh and chilled Atlantic salmon which had been found by the Department of Commerce to be subsidized by the Government of Norway and sold in the United States at less than fair value.
25. Norway requested the Panel to find that the imposition by the United States of an anti-dumping duty order on imports of fresh and chilled Atlantic salmon from Norway was inconsistent with the obligations of the United States under the Agreement. In particular, Norway requested the Panel to find that:
26. Norway initially requested the Panel to recommend to the Committee on Anti-Dumping Practices that it request the United States to revoke the anti-dumping duty order on imports of fresh and chilled Atlantic salmon from Norway or otherwise bring it promptly into conformity with the obligations of the United States under the Agreement. At a later stage, Norway requested the Panel to recommend to the Committee that it request the United States to revoke the anti-dumping duty order and reimburse
11 56 Fed.Reg., 25 February 1991, pp.7664-7678. ↩
12 USITC Publication 2371 Fresh and Chilled Atlantic Salmon from Norway: Determination of the Commission in Investigation No. 731-TA-454 (Final) Under the Tariff Act of 1930, Together with the Information Obtained in the Investigation, USITC Publication 2371, April 1991 (hereinafter: "USITC Determination"). ↩
[Page 9]
any anti-dumping duties paid. Norway noted that this request was consistent with previous Panel Reports.13
27. The United States requested the Panel to find that the affirmative final determinations of the Department of Commerce and the USITC in the anti-dumping proceeding on imports of fresh and chilled Atlantic salmon from Norway comported with the obligations of the United States under the Anti-Dumping Code. In particular, the United States requested the Panel to find that:
28. The United States also requested the Panel to give a ruling that certain matters raised by Norway were not properly before the Panel (infra, Section IV).
29. However, at the request of the Panel, the United States presented its views on the merits of each of the issues raised by Norway which it considered were not properly before the Panel. The Panel indicated that this request to the United States was without prejudice to the Panel's ultimate decision on the preliminary objections of the United States. On this basis, the United States requested the Panel to find that (i) the initiation of the anti-dumping duty investigation was in accordance with Article 5:1 of the Agreement; (ii) the calculation by the Department of Commerce of the exporters' processing costs was consistent with Articles 6:5 and 6:8 of the Agreement, the inclusion of an eight per cent profit rate in the constructed normal values was consistent with Articles 2:4 and 6:8 of the Agreement, the comparison of an average normal value with individual export prices was consistent with Article 2:6 of the Agreement, and the Norwegian respondents had not been denied national treatment; and (iii) the arguments of Norway based on Article 9:1 of the Agreement were factually incorrect and without a legal basis in the Agreement.
30. The United States requested the Panel to give a preliminary ruling that certain issues raised by Norway were not properly before the Panel. Some of these issues had not been raised by the Norwegian respondents in the administrative proceedings before the investigating authorities in the United States. Other issues were not covered by the Panel's terms of reference and/or had not been raised by Norway during the consultations under Article 15:2 of the Agreement and during the conciliation process which had taken place under Article 15:3 of the Agreement in this dispute.
31. The United States argued that the issues raised by Norway concerning the standing of the petitioner to request the initiation of an investigation on behalf of the domestic Atlantic salmon industry and concerning the comparison of an average normal value to individual export prices had not been raised
13 e.g. "United states - Imposition of Anti-Dumping Duties on Seamless Stainless Steel Hollow Products from Sweden", ADP/47, paragraph 5.24; "United States - Countervailing duties on fresh, chilled and frozen pork from Canada", DS7/R, 38S/30, paragraph 5.2; "Canada - Imposition of countervailing duties on imports of manufacturing beef from the EEC", SCM/85, paragraph 5.6; and "New Zealand - Imports of electrical transformers from Finland", BISD 32S/55, paragraph 4.11. ↩
[Page 10]
by the Norwegian respondents before the investigating authorities in the United States and were therefore not admissible in the proceedings before the Panel.
32. The United States argued that the principle that a party must raise an issue, and present all facts, evidence and arguments on that issue before the investigating authorities and may not present any facts, evidence or arguments in the first instance to a reviewing body was manifest in the Agreement. The Agreement provided domestic investigating authorities with the exclusive authority to gather and consider evidence and make findings of fact and law concerning dumping and injury issues (Articles 3-6). The determinations of the investigating authorities must be made on the basis of the information before the agency (Article 6). The investigating authorities must complete their investigation in one year (Article 5:5). In addition, before the administering authorities took final action, they must give all interested parties, including the foreign respondents, "ample opportunity to present in writing all evidence that they consider useful in respect of the anti-dumping investigation in question" (Article 6:1). Interested parties also had the right to present evidence orally (Article 6:1) and had the right "to see all [non-confidential] information that is relevant to the presentation of their cases" and to present their cases in response (Article 6:2). Finally, Article 6:7 mandated that "all parties shall have a full opportunity for the defence of their interests" through an oral hearing. Throughout the investigative process, therefore, the Agreement required that all parties have the opportunity to state all their arguments in order to influence the investigating authorities. Unless the investigating authorities had all the facts and information (and arguments as to how to interpret those facts and information) they could not take "final action" consistent with the procedural prerequisites of the Agreement. Accordingly, not only was there no provision in the Agreement for presentations ex post facto to a Panel of facts or arguments which had not been raised before the investigating authorities, but the terms of the Agreement, in fact, precluded this. Such untimely presentation of arguments would prevent the investigating authorities from conducting a full investigation, thus denying those authorities the opportunity to consider all the evidence and arguments and render determinations on that basis. In specific, untimely arguments would also deny the other parties their rights under Articles 6:1, 6:2 and 6:7 to present evidence to comment on information in the investigation and to confront parties with opposing views.
33. The United States argued that the principle at issue here, called "exhaustion of administrative remedies" in United States legal parlance, was closely akin to the notion of exhaustion of local remedies, which was well-settled in international law. The exhaustion doctrine had two components, a procedural component and a public policy component. The procedural component was that a party must advance through the appropriate fora in sequence. This was reflected in the requirements of Article 15 of the Agreement that consultations be concluded before a Party could resort to conciliation and that the Committee could not convene a panel until the consultation had ended. Procedurally, the administering authorities must also complete their work before a Party could refer the matter to a panel. Other rationales for the doctrine were that it preserved the integrity of the administrative process and allowed all parties to the administrative proceeding an opportunity to consider and address the facts and arguments raised by other parties. The doctrine prevented a reviewing tribunal from usurping the function of the administrative body which had the expertise to rule in the matter. Another purpose was to avoid duplication of effort and waste of resources by the reviewing tribunal. The public policies behind the doctrine of exhaustion of administrative remedies were virtually identical to the rationales underlying the international law rule of exhaustion of local remedies. Under that rule, if a country offered a remedy under its local laws and procedures, the local remedy should be pursued before the country could be haled before an international tribunal for denying such a remedy.
34. The United States noted in this context that, while in the proceedings before the Panel Norway had claimed that the Agreement required investigating authorities, before initiating an investigation, to take steps to satisfy themselves that a request for the initiation of an investigation was filed on behalf of the domestic industry affected, in the case at hand the Norwegian respondents had never asked the Department of Commerce to take such "steps" either during the period before the initiation of the anti-
[Page 11]
dumping duty investigation or at any time after the initiation of the investigation. In its notice of the initiation of the anti-dumping duty investigation, the Department had invited interested parties to bring to its attention any information related to the petitioner's claim that it had filed the petition "on behalf" of the domestic industry. Yet the Norwegian respondents (all of whom had been represented by the same counsel) had not responded to this invitation. The Department had in recent years rescinded its initiation of investigations after having determined that the petition in question had not been filed on behalf of the relevant domestic industry in the United States.14 However, the Norwegian participants had never once, during nearly a year of investigation and thousands of pages of filings, given any sign, or made any representation, which could have alerted the Department to the concern belatedly expressed by Norway in the proceedings before this Panel. Had any of the Norwegian participants done so, the Department could have addressed the situation.
35. The United States also argued that the issue raised by Norway concerning the comparison of an average normal value to individual export prices had not been raised before the investigating authorities in the United States and was therefore not admissible in the proceedings before the Panel. Norway had argued before the Panel that export prices should have been segregated by fish size and/or quality. This issue, which would involve the collection and evaluation of facts and arguments from all the parties to the investigation, had not been raised by the Norwegian exporters during the investigation. What had been raised by the exporters was the completely unrelated question of whether export prices should be averaged because of price fluctuations common to all salmon from Norway, regardless of size and quality. As the Federal Register Notice of the affirmative final determination of dumping made clear, the basis for the argument of the exporters that average export prices should be used was the existence of price fluctuations due to the alleged perishability of salmon; the exporters had not addressed the question of quality and size differences.
36. Norway contested that the doctrines of exhaustion of local remedies and of exhaustion of administrative remedies applied to dispute settlement proceedings under the Agreement.
37. With respect to the doctrine of exhaustion of local remedies, Norway submitted the following. First, under public international law, the rule of exhaustion of local remedies applied only to cases of diplomatic protection, as distinguished from cases involving "direct injury" to a state. In dispute settlement proceedings under the Agreement, a Party was not bringing a claim on behalf of one of its nationals: the cause of action in such proceedings was the "direct injury" to a Party in the form of nullification or impairment of benefits accruing to that Party or in the form of the impedance of the achievement of any of the objectives of the Agreement. Second, there was no basis in the text of the Agreement for the application of the doctrine of exhaustion of local remedies. Unlike many other international agreements which included an exhaustion of local remedies requirement, the Agreement did not include such a requirement. Had the Parties intended to include such a requirement (which would have drastically changed the procedural steps delineated in the dispute settlement provisions of the Agreement), they would have done so explicitly. Third, there was no GATT practice recognizing the local remedies doctrine. No GATT Panel had even hinted that exhaustion of local remedies was required. In fact, as demonstrated by recent Panel Reports, GATT practice was contrary to such a requirement.15 The Vienna Convention on the Law of Treaties directed in Article 31:3(b) that subsequent practice was to be taken into account when interpreting the provisions of an international agreement. In the case of the General Agreement, such subsequent practice clearly did not require the exhaustion
14 e.g. Gilmore Steel Corp. v. United States, 585 F Supp. 670 (CIT 1984), aff' d sub nom. Oregon Steel Mills v. United States, 862 F 2d 1541 (Fed. Cir. 1988). ↩
15 "United States - Imposition of Duties on Imports of Seamless Stainless Steel Hollow Products from Sweden", ADP/47 and "United States - Definition of Industry Concerning Wine and Grape Products", SCM/71, adopted on 28 April 1982. ↩
[Page 12]
of local remedies. The Vienna Convention did not support the incorporation of unexpressed principles of international law. It did allow parties to rely on supplementary means of when interpreting ambiguous terms of a treaty. However, it was one thing to use customary international law to intepret ambiguous terms of an international agreement; it was quite another to read into the Agreement such a major modification as the local remedies doctrine. If this principle was to be required, the decision had to come through reflective consideration and negotiation by all Parties at the multilateral level.
38. Norway also submitted that the exhaustion of local remedies requirement was a narrow rule in public international law, applicable only to international adjudication, unless otherwise explicitly directed in an international agreement. There was no customary international law rule which required the exhaustion of local remedies in any other kind of international dispute fora. For example, international arbitration agreements were not subject to the requirement of exhaustion of local remedies. Furthermore, international tribunals which had applied the exhaustion of local remedies doctrine had taken a flexible approach in its application and had required exhaustion only after carefully balancing the practical and political pros and cons of doing so. In particular, public international law made the application of the exhaustion of local remedies dependent on criteria of reasonableness and did not require such exhaustion where local remedies were inadequate and ineffective.16 No adequate remedy was available for Norway in the courts of the United States for a breach by the United States of its GATT obligations. US domestic law did "not provide a meaningful legal requirement that GATT law be observed".17 In fact, a US trade statute specifically commanded that no provision of any trade agreement, nor the application of any such provision to any person or circumstance, in conflict with any United States statute, shall be given effect under the laws of the United States.18 In addition, many courts in the United States refused to give full legal effect to the General Agreement.19 Thus, there were no effective local remedies to exhaust in the United States in case of a breach of the General Agreement by the United States.
39. Norway further argued that strong policy considerations dictated that a local remedies doctrine not be applied to dispute settlement proceedings under the Agreement. The imposition of an exhaustion of local remedies requirement would result in years of delay in the dispute settlement process and would therefore be inconsistent with the Agreement's purpose of the effective and timely resolution of disputes. Finally, even if one were to apply the requirement of exhaustion of local remedies to dispute settlement proceedings under the Agreement, account had to be taken of the fact that, as confirmed in a recent judgement of the International Court of Justice, international law permitted the use of a rule of reason in the interpretation of the requirement; under this approach, the exhaustion requirement did not mean that each and every minute aspect of a claim had to be raised in the local fora before the claim could be raised at the international level.20
40. Regarding the principle referred to by the United States of exhaustion of administrative remedies, Norway submitted that this principle was a requirement of US administrative law but not a principle of public international law. Since this principle did not originate in public international law, the reasons for not applying it to dispute settlement proceedings under the Agreement were even stronger than in the case of the exhaustion of local remedies doctrine. In any event, the exceptions established under US jurisprudence to the application of the requirement of exhaustion of administrative remedies weighed
16 Norwegian Loans Case, ICJ Reports (1957), p.9. ↩
17 Hudec, "The Legal Status of GATT in the Domestic Law of the United States", in: Hilf, Jacobs, Petersmann, (eds.) The European Community and GATT (1986) p.193. ↩
18 19 U.S.C. S2504(a) (1983 & Suppl. 1991). ↩
19 e.g. Algoma Steel Corp. v. United States, 865 F.2d 240 (Fed.Cir. 1989) cert. denied, 492 U.S. 919 (1989). ↩
20 Elettronica Sicula S.p.A. Case, 1989 I.C.J. Reports, p.94. ↩
[Page 13]
against the application of this requirement to dispute settlement under the Agreement. Thus, US courts enjoyed a degree of discretion in the application of this requirement and did not apply it when the administrative remedy was inadequate and when resort to agency proceedings would be futile. Given that US trade law was not required to be in conformity with relevant international agreements (supra, paragraph 38) there were no "effective" administrative remedies to exhaust in cases involving an action of the United States inconsistent with its obligations under the General Agreement. Since the United States Department of Commerce and other relevant agencies often did not apply GATT law on any consistent basis, it was also often futile for a contracting party to raise GATT related issues before these agencies.
41. Norway also observed in this context that a major rationale for the application of the requirement of administrative remedies was that it was inefficient and inappropriate to have courts review factual issues which could more effectively be considered by an agency having expertise in that area. In light of this, courts had often excused the exhaustion requirement when reviewing issues of law, as opposed to issues of fact. The issues raised by Norway before this Panel similarly concerned issues of law, not of fact. The questions before the Panel concerned not what the facts were but whether the interpretation and consideration of the facts by the United States were in conformity with the obligations of the United States under the Agreement.
42. Norway did not contest that the issue of the standing of the petitioner in the anti-dumping duty investigation had not been raised before the investigating authorities in the United States by the Norwegian respondents. However, the question of whether the petitioner was acting on behalf of the domestic industry had been raised in a letter to the Department of Commerce from a domestic producer, prior to the initiation of the investigation. The Department had ignored this letter. More importantly, the Panel established by the Committee on Anti-Dumping Practices in the dispute between Sweden and the United States in "United States - Imposition of Anti-Dumping Duties on Imports of Seamless Stainless Steel Hollow Products from Sweden"21 had held that, before initiating an investigation, investigating authorities were required to satisfy themselves that a written request for the initiation of the investigation was filed on behalf of the domestic industry. To satisfy themselves as to industry support, the investigating authorities must take affirmative steps. The Panel had found that it was not sufficient to rely upon statements by petitioners claiming to be acting on behalf of the domestic industry. Thus, the question of the standing of a petitioner did not need to be raised by any party: investigating authorities were under an affirmative obligation to satisfy themselves that a petition was filed on behalf of the domestic industry. Indeed, in the case considered by the Panel in the dispute between Sweden and the United States, there had been no challenge of the petitioner's standing prior to the initiation of the investigation.
43. Norway further argued that it was consistent practice of the United States to assume that a petitioner was acting on behalf of a domestic industry until such time as a substantial proportion of the domestic industry come forth to oppose the petition.22 The United States would not investigate the standing of a petitioner if the challenge came from foreign private respondents or from a foreign government. There had therefore been no reason for the Norwegian respondents to raise this issue during the investigation.
44. Norway argued that the issue of the comparison of an average normal value to individual export prices had, in fact, been raised in the proceedings before the Department of Commerce. In support, Norway referred to comments by the Norwegian respondents reflected in the Federal Register Notice
22 e.g. Certain Electrical Conductor Aluminium Redraw Rod from Venezuela, 53 Fed.Reg. p.24764 (1989). ↩
[Page 14]
of the affirmative final determination of dumping.23 In response to these comments, the Department of Commerce had stated that it had used its "... normal practice of comparing individual US prices to weight-average home market or third country prices".24 This response illustrated the futility of raising in the administrative proceedings issues contesting normal US practice or legislation. Since the United States would continue to apply its "normal practice", it was futile to raise such issues.
45. The United States submitted that it had not argued that the public international law rule of exhaustion of local remedies was applicable to dispute settlement proceedings under the Agreement but that the rationale of this rule was similar to the rationale of the Agreement-based requirements that an issue first be raised in the domestic administrative proceedings. Norway had not addressed the specific language of the Agreement relied upon by the United States to support its view that a matter not raised before the investigating authorities could not in the first instance be raised before a Panel. Rather, it had argued that the GATT system generally did not impose a requirement to go through national authorities before raising an issue in GATT dispute settlement proceedings. However, the Agreement established a rôle for domestic investigating authorities not found under other GATT provisions. Under Norway's argument, the investigating authorities were virtual appendages, which could be ignored at will. This view was inconsistent with the central and exclusive rôle provided under the Agreement for the investigating authorities.
46. The United States considered that, while Norway's discussion of the public international law rule of exhaustion of local remedies was beside the point in that the United States had not argued that this rule applied to dispute settlement under the Agreement, Norway's interpretation of this rule was in any event erroneous. Historically, the rule of exhaustion of local remedies had been used in cases where the national of one country had been injured by another country. In these cases, the national was required to seek redress under the allegedly offending country's system before asking his own government to try to resolve the dispute on a government-to-government level. The doctrine did not apply to disputes solely between countries. This distinction had been clarified in a recent judgement of the International Court of Justice in the Elettronica Sicula S.p.A. case.25 There, the United States had claimed that the doctrine did not apply because the United States was representing itself, not the two American companies involved. The Court had rejected this argument, stating that "the matter which colours and pervades the United States claim as a whole is the alleged damage to Raytheon and Machlett, said to have resulted from the actions of the Respondent".26 The Court had thus ruled that the doctrine of exhaustion of local remedies applied when a nation was primarily representing its nationals, even if some issues of sovereignty were present.
47. In the view of the United States, the interests of the Norwegian exporters "coloured and pervaded" Norway's claim in the proceedings before this Panel. This dispute had arisen only after the United States had imposed anti-dumping duties on Norwegian imports of fresh and chilled Atlantic salmon. The arguments made by Norway were in most instances identical to those which were made or could have been made by the private Norwegian interests during the investigations. In fact, Norway was actually espousing the interests of its nationals in these proceedings. Norway's argument that it was adjudicating its own rights under the Agreement, separate and apart from the interests of its nationals would create an exemption to the local remedies doctrine which would effectively swallow the entire doctrine. By definition, any time one country brought a claim against another, international legal rights, usually treaty rights or the equivalent, were at issue. To argue, as did Norway, that in any such instance the international matter involved an offence by one country against another and was thereby exempt
23 56 Fed.Reg., 25 February 1991, pp.7673-7674. ↩
24 56 Fed.Reg., 25 February 1991, p.7674. ↩
25 1989 I.C.J. Reports, p.15. ↩
[Page 15]
from the exhaustion doctrine ignored the international jurisprudence on this subject. The United States was not claiming that disputes involving fundamentally private interests in which there was an element of government-to-government obligations were not rightfully a subject for international dispute resolution, but only that such disputes would not be exempted from the principle of exhaustion of local remedies.
48. The United States argued that the issues raised by Norway with respect to the alleged denial of national treatment was not within the Panel's terms of reference. In its first submission to the Panel, Norway had for the first time argued that the United States had failed to provide national treatment, as required by Article III of the General Agreement. As the United States understood it, Norway alleged less favourable treatment because its imports were subject to anti-dumping duties, while domestically produced salmon was not subject to anti-dumping duties. Norway had not provided any argumentation based on the Agreement for this novel approach, but had relied solely on Article III of the General Agreement. The issue of a violation by the United States of its obligations under Article III of the General Agreement did not appear in document ADP/65/Add.1. Given that the Panel's terms of reference had been defined in document ADP/69 by reference to documents ADP/65 and Add.1, the Panel's terms of reference excluded this issue. In any event, there was no basis for Norway to raise an issue under the General Agreement as part of panel proceedings under Article 15 of the Agreement. Norway had explicitly stated that its national treatment stemmed from Article III of the General Agreement. The Panel's terms of reference mandated the Panel to "examine [the matter] in light of the relevant provisions of the Agreement on Implementation of Article VI of the General Agreement". Arguments based on the General Agreement were outside this mandate.
49. The United States also argued that the issue raised by Norway regarding the differing treatment of foreign and domestic respondents was outside the Panel's terms of reference. The question of the allegedly disparate treatment of Norwegian and domestic producers was referenced in neither ADP/65 nor in ADP/65/Add.1, which had been specifically referred to in ADP/69 as providing the basis for the Panel's consideration. Accordingly, the Panel's terms of reference excluded this issue. Moreover, Norway's argument concerning the differing treatment of foreign and domestic producers was an issue that arose under Article III of the General Agreement. As such, this issue could not properly be raised before a panel formed under the Agreement.
50. Norway argued that the terms of reference of the Panel, as stated in document ADP/69, provided that the Panel's mandate was to examine "the matter referred to the Committee referred by Norway in documents ADP/65 and Add.1". The "matter" referred to the Committee by Norway was the imposition of anti-dumping duties on imports of fresh Atlantic salmon from Norway.
51. With regard to the issue of the denial of national treatment, Norway noted that it had raised the issue of fair and equitable treatment throughout the consultation, conciliation and panel process. The national treatment requirement was included in the Agreement's requirement of fair and equitable treatment, except to the extent that national treatment was specifically not required by the Agreement (e.g., the imposition of anti-dumping duties after meeting the procedural requirements of the Agreement). Norway pointed out that this matter had been identified in Norway's request for the establishment of a panel (document ADP/65/Add.1, item III).
52. With respect to the comments made by the United States on whether the issue of different treatment of domestic and foreign respondents was within the Panel's terms of reference, Norway referred to its observations on the issue of national treatment.
53. The United States considered that the issue raised by Norway concerning the use by the Department of Commerce of the Norwegian central exporters' organization's processing fees in the calculation
[Page 16]
of the constructed normal values was not within the Panel's terms of reference. This issue was referenced in neither ADP/65 nor ADP/65/Add.1, which were specifically referred to in document ADP/69 as providing the basis for the Panel's consideration. Accordingly, the terms of reference excluded this issue.
54. Norway argued that the question of the use of the central exporters' organization's processing fee had been raised in Norway's request for the establishment of a panel (document ADP/65/Add.1, item II.A).
55. The United States considered that Norway's submission concerning the continued imposition of the anti-dumping duty order was not within the Panel's terms of reference. No reference to this issue appeared in document ADP/65/Add.1.
56. The Panel asked the United States to comment on the reference made in document ADP/65/Add.1, section V.A to the issue of the continued imposition of the anti-dumping duty order. The United States observed that, while Article 9 of the Agreement was nominally referenced in ADP/65/Add.1, at Section V.A, this section only referred to events up to the time of the USITC's final affirmative injury determination. As such, that section did not raise a cognizable issue under Article 9 of the Agreement, which concerned the continuation of anti-dumping duties, rather than their imposition in the first instance.
57. The United States argued that the issues raised by Norway with respect to the denial of national treatment, the allegedly differing treatment of foreign and domestic respondents, the use of minimum amounts for profits and the calculation of processing fees and the continued imposition of the anti-dumping duty order had not been raised by Norway in the consultations and conciliation preceding the establishment of the Panel and were therefore not properly before the Panel.
58. In support of its view that Norway was precluded from raising before the Panel issues not raised during consultations and conciliation, the United States argued that the Agreement embodied a fundamental principle of jurisprudence that certain procedures must be followed before a panel could consider the matter. Article 15 of the Agreement set forth this precept. The first two paragraphs stated the conditions for consultations. If consultations failed, a Party could under Article 15:3 refer the matter to the Committee on Anti-Dumping Practices for conciliation. Article 15:4 provided that the Parties must try to resolve the matter during the thirty-day conciliation period. According to Article 15:5, if, and only if, the Committee could not reach a solution after three months of conciliation, the Committee could convene a panel to resolve the dispute. Therefore, an issue could not be presented in the first instance to a panel. The principle at issue here was closely akin to the notion of exhaustion of local remedies under international law.
59. Norway, referring to its comments on the principles of exhaustion of local remedies and exhaustion of administrative remedies27, rejected the application of these principles to the remedies provided for under the Agreement. In any event, no new issues had been raised in the first instance before the Panel. Moreover, at least when issues had been raised in general in consultations and conciliation, they could be raised more specifically before the Panel without prejudicing the other party to the dispute. In this case, the United States had not been prejudiced because it had had at least one month to respond to any issues Norway might have covered more in detail in its first submission than during the consultations and the conciliation process. The United States had access to all underlying facts in
[Page 17]
the investigation and had had ample opportunities to rebut any arguments Norway might have elaborated upon since consultations and conciliation.
60. With regard to the issue of denial of national treatment and differing treatment of foreign and domestic respondents, the United States pointed out that these issues were not raised during consultations and that the text of Norway's request for conciliation (ADP/61) and the minutes of the meeting of the Committee on Anti-Dumping Practices held in July 1991 for purposes of conciliation under Article 15:3 (ADP/M/33) indicated that these issues had also not been raised during the conciliation process.
61. Norway argued that it had raised the issue of fair and equitable treatment throughout the consultation, conciliation and panel process. The national treatment requirement was included in the requirement of the Agreement of "fair and equitable" treatment, except to the extent that national treatment was specifically not required in the Agreement (e.g. the imposition of anti-dumping duties after meeting the procedural requirements of the Agreement). Norway pointed out that in document ADP/61, page 5, reference had been made to the use by the Department of Commerce of the "best information available" in an arbitrary and unwarranted manner. Moreover, as reflected in document ADP/M/33, paragraphs 28 and 29, at the conciliation meeting Norway had discussed the question of the use of the best information available and the information required from the Norwegian respondents.
62. With respect to the comments made by the United States on whether the issue of different treatment of domestic and foreign respondents was within the Panel's terms of reference, Norway referred to its observations on the issue of national treatment.
63. The United States argued that the issue raised by Norway regarding the use of minimum amounts for profits in the calculation of constructed normal values had not been raised during the bilateral consultations held between Norway and the United States and that the text of Norway's request for conciliation (document ADP/61) and the minutes of the meeting of the Committee on Anti-Dumping Practices held in July 1991 under Article 15:3 (document ADP/M/33) indicated that this issue also had not been raised by Norway during the conciliation process.
64. Norway argued that this issue had been raised during the bilateral consultations held between Norway and the United States. In support, Norway referred to a written question addressed by Norway to the United States during bilateral consultations held on 15 March 1991. In this question, Norway had asked the United States to explain how "the US cost calculation [could] possibly be right in light of the significantly lower EC cost calculation". In its request for conciliation (document ADP/61), Norway had identified this issue when it had alleged that the Department of Commerce had "arbitrarily allocated expenses" (document ADP/61, p.5). At the conciliation meeting held in July 1991, Norway had also referred to this issue (document ADP/M/33, paragraph 29).
65. The United States argued that the issue raised by Norway regarding the calculation of processing fees had not been raised during consultations and that the text of Norway's request for conciliation (document ADP/61) and the minutes of the meeting of the Committee on Anti-Dumping Practices held in July 1991 for purposes of conciliation under Article 15:3 of the Agreement (document ADP/M/33) indicated that this issue also had not been raised by Norway during the conciliation process.
66. Norway argued that the issue of the calculation of the exporters' processing fee had been raised by Norway during bilateral consultations. In support, Norway referred to a written question addressed by Norway to the United States during bilateral consultations held on 15 March 1991. In this question Norway had asked the United States to explain how "the US cost calculation [could] possibly be right in light of the significantly lower EC cost calculation". This issue had also been raised at the conciliation phase, as demonstrated by the reference in Norway's request for conciliation (document ADP/61) to the use of the best information available in an arbitrary and unwarranted manner.
[Page 18]
67. The United States argued that the question of the continued imposition of the anti-dumping duty order had not been raised by Norway during the consultations and the conciliation process preceding the establishment of the Panel.
68. Norway argued that Article VI of the General Agreement constituted an exception to the obligations of Articles I and II of the General Agreement. The interpretative practice of the contracting parties confirmed that exceptions such as Article VI had to be interpreted narrowly and that the contracting party invoking the exception had the burden of proof of demonstrating that it had met all the requirements of the provision in question.28 Article VI of the General Agreement provided that no anti-dumping or countervailing duties could be levied unless certain facts had been established and the contracting party invoking this Article had taken into account all facts necessary to meet the requirements of this Article.29 The contracting party taking action under this Article must establish the existence of these facts when its action was challenged.30 In the matter before this Panel, the United States had not demonstrated that it had met these requirements.
69. The United States considered that the proposition that Article VI of the General Agreement constituted an exception to fundamental rights and obligations under the General Agreement was contradicted by the text structure and the drafting and interpretative histories of the General Agreement. The fact that with respect to no other commercial practice subject to provisions of the General Agreement the drafters had used language as strong as in Article VI, where they had expressly provided that injurious dumping was to be "condemned" was revealing of the key rôle intended by the drafters for the unfair trade remedies within the GATT framework. The structure of the General Agreement also confirmed that Article VI had not been drafted as an exception. Article VI was placed at the beginning of the General Agreement, where the primary subjects of the General Agreement were found. By contrast, where the drafters had intended to craft exceptions, they had placed them at the end of Part II of the General Agreement in Articles XX and XXI. Indeed, the placement of these Articles (grouped together at the end of Part II), their titles (expressly identified as exceptions) and their text (e.g. the requirement that measures taken under these provisions not be a "disguised restriction on international trade) all set them clearly apart from inter alia Article VI. Moreover, application of anti-dumping and/or countervailing duties had also not been encumbered with restrictions and requirements to hold consultations found elsewhere in the General Agreement, e.g. in Article XIX.
70. Norway argued that the proposition advanced by the United States that Article VI was not an exception to fundamental principles of the General Agreement was inconsistent with the plain language of this Article and the overall objectives of the General Agreement. This view was also incompatible with previous panel findings, views of well respected international legal scholars and the drafting history
28 e.g. Report of the Panel in "Canada - Import, Distribution and Sale of Alcoholic Drinks by Canadian provincial Marketing Agencies", BISD 35S/37; Report of the Panel in "Canada - Administration of the Foreign Investment Review Act", BISD 30S/140, and Report of the Panel in "United States - Countervailing Duties on fresh, chilled and frozen pork from Canada", BISD 38S/30. ↩
29 Report of the Panel in "Swedish Anti-Dumping Duties", BISD 3S/81, paragraph 15; Report of the Panel in "United States - Countervailing Duties on fresh, chilled and frozen pork from Canada, BISD 38S/30, paragraph 4.8. ↩
30 Report of the Panel in "Swedish Anti-Dumping Duties", BISD 3S/81, paragraph 15 and Report of the Panel in "New Zealand - Imports of Electrical Transformers from Finland", BISD 32S/55, paragraph 4.4. ↩
[Page 19]
of the General Agreement. In any event, the US argument on Article VI was irrelevant since, whatever the nature of Article VI, in the case before this Panel the United States had not met express requirements of the Agreement.
71. Norway argued that the statement in Article VI of the General Agreement that injurious dumping was to be "condemned" provided no support for the view that Article VI was not an exception to fundamental principles of the General Agreement. When, at the second session of the CONTRACTING PARTIES, the text of Article VI of the General Agreement had been replaced by Article 34 of the Havana Charter, the Working Party had noted that there was no difference in meaning between the original Article VI and Article 34 of the Havana Charter.31 This demonstrated that the inclusion of the word "condemned" was without significance If anything, the drafting history of Article 34 of the Havana Charter indicated that the term "condemned" had been added in order to limit, not expand, the use of anti-dumping measures. In November 1947, at the Havana Conference, Article 34 of the draft Charter had been considered by the sub-committee on general commercial policy provisions. A number of delegations to this committee had wanted to expand Article 34 to include a condemnation of dumping and to cover in addition to "price dumping" all forms of dumping without requiring an injury test. Another group of delegations had believed that the primary objective of the Article should be to restrict the abuse of anti-dumping measures. The result had been the current text of Article VI, which kept the main focus of the Article on limiting the use of anti-dumping duties but which included a statement that dumping was to be "condemned", but only dumping as defined in Article VI, and only if injury was also found. The inclusion of the word "condemned" had been necessary to reach a compromise under which the coverage of the Article was limited to instances of price dumping which caused injury. Norway noted that the United States had been among the delegations which had wanted the focus of the Article to be on restricting the use of anti-dumping duties, not on limiting the use of dumping in general.
72. Norway considered that Article II:2(b) of the General Agreement supported the view that Article VI was an exception. The language and placement of this provision demonstrated that the imposition of anti-dumping and countervailing duties was intended to be an exception to, not a fundamental right of, the General Agreement. In fact, one author had described the reference to anti-dumping and countervailing duties in Article II:2(b) as an "exception".32
73. In response to the argument of the United States with respect to the placement of Article VI within the General Agreement, Norway observed that this argument overlooked the fact there were exceptions scattered all over the text of the General Agreement, e.g. in Articles I:2, II:2(a), (b) and (c), III:3, III:6, IV, XII, XIV and XIX. There was therefore no basis for the view that the placement of Article VI in the General Agreement indicated that the Article was not an exception. With respect to the argument that the absence in Article VI of provisions regarding consultation confirmed that this Article was not an exception, Norway noted that this argument ignored the practice of many signatories to require consultations before imposing duties under Article VI. Moreover, this argument failed to take into account that consultation procedures had been included in the Agreement and in the Agreement on Interpretation and Application of Articles VI, XVI and XXIII of the General Agreement, which had been designed to elaborate upon the requirements of Article VI.
74. The United States further argued that the negotiating history of the General Agreement demonstrated that remedies for dumped and subsidized goods had from the beginning been a fundamental aspect of the General Agreement. As described in a recent GATT publication, the promotion of fair competition (defined as curbing government subsidies, dumping and "other distortions of international
32 Jackson, World Trade and the Law of GATT (1969), p.210. ↩
[Page 20]
competition") had been and remained one of the fundamental objectives of the General Agreement.33 This recent description of the fundamental nature of the rights under Article VI reflected the negotiating history of the General Agreement. The drafters of the General Agreement had recognized in 1947 that distortions to international competition caused by unfair trade practices could be so severe that effective remedies to curb such distortions were essential: indeed, as essential to an overall programme of liberalization of international trade as, for example, the m.f.n. principle and the national treatment principle. The essential balance reflected in the text of the General Agreement was that contracting parties would open their markets - principally through tariff reductions - in exchange for reciprocal access and the right to take action against unfairly traded imports. Without such disciplines, tariff reduction would have been of little or no value. The importance of disciplines governing unfair trade practices was reflected in the formal announcement by the United States Department of State of the accession of the United States to the General Agreement. Describing Part II of the General Agreement, which contained the commercial policy provisions (including Article VI), the Department had explained that:
"Part II deals with barriers to trade other than tariffs .... The provisions of Part II are intended to prevent the value of the tariff concessions from being impaired by the use of other devices, and also to bring about the general relaxation of non-tariff trade barriers, thus assuring a further quid pro quo for the action taken with respect to tariffs."34
75. Regarding the negotiating history of the General Agreement, the United States also observed that injurious dumping had been viewed with such concern during the original GATT negotiations that proposals had been considered to permit imposition of tougher countermeasures than merely offsetting duties.35 However, in the end the Article VI remedy had been limited to such duties. This choice indicated clearly that the drafters of Article VI had been capable of narrowing the anti-dumping/countervailing duty instrument in the General Agreement in whichever way they chose. The negotiating record revealed that the drafters had narrowed the remedy. By contrast, there was no support for the view of Norway that the application of that remedy should be further narrowed by, inter alia, establishing a burden of proof or persuasion on parties invoking their fundamental rights under Article VI. Indeed, the narrow nature of the remedy suggested the contrary: if the imposition of offsetting duties alone was to bear the burden of remedying the harm caused by, and deterring these anti-competitive and unfair trade practices, the application of the remedy should be broadly construed. In particular, it should not be restricted except as expressly required by the terms of Article VI. The drafting history also demonstrated that no special burden of proof had been contemplated with respect to contracting parties imposing duties under Article VI. Early proposals that the importing country be required to prove dumping allegations had been rejected in favour of the weaker and broader language of Article VI as adopted, which simply provided that imposition of anti-dumping duties or countervailing duties should occur only after a determination by a contracting party that dumping and injury existed.36
76. Norway argued that the main objective of the General Agreement was the reduction of tariff rates on an m.f.n. basis. By contrast, nothing in the General Agreement obligated nations or firms to refrain from dumping37 While "promoting fair competition" might be one of the objectives of the General Agreement, the GATT had never defined this to mean "curbing government subsidies, dumping and
33 GATT What It Is, What It Does (1990), p.4. ↩
34 17 Dept. State Bull, 1042, 1045 (1947). ↩
35 Jackson, World Trade and the Law of GATT, (1969), p.421, note 55. ↩
36 E/PC/T/C.II/32 (1946) (Note of the Benelux countries). ↩
37 Report of the Panel in "Canada - Administration of the Foreign Investment Review Act, BISD 30S/140, 164. ↩
[Page 21]
other distortions of international competition", as claimed by the United States. In fact, the arbitrary imposition of anti-dumping duties was regarded as a protectionist device in its own right, retarding the promotion of fair competition.38 The GATT publication referred to by the United States did not define the promotion of fair competition in the manner suggested by the United States. Rather, it pointed out in a neutral fashion that "increasingly, the GATT is concerned with subsidies and dumping," and then pointed out more specifically that "the rules under which governments may respond to dumping in their domestic market by overseas competitors are contained in the GATT and an Anti-Dumping Code." Thus, the General Agreement and the Agreement provided for limitations on the use of anti-dumping measures and did not regulate the practice of dumping as such. This publication further referred to Article I of the General Agreement as "the key article" of the General Agreement and described Article VI as a "technical article designed to prevent or control possible substitutes for tariffs". The publication also referred to Article VI as "lay[ing] down the conditions under which anti-dumping duties may be imposed", which again confirmed that Article VI dealt with limits on the use of anti-dumping measures. Moreover, the United States was incorrect in referring to dumping as "unfair trade" in its discussion of this GATT publication. This publication did not describe dumping as "unfair trade"; nor was dumping described as "unfair trade" in the text of the General Agreement or in the Anti-Dumping Code.
77. In the view of Norway, the State Department publication referred to by the United States did not provide support for the view that Article VI remedies were a fundamental right of contracting parties to the General Agreement. The paragraph quoted by the United States referred to "non-tariff trade barriers" and did not discuss dumping. Interestingly, dumping was not mentioned in the introductory paragraph in which the State Department discussed the scope of the General Agreement, or in the paragraph which summarized the provisions of Part II of the General Agreement. By contrast, rules on tariffs, preferences, quotas, internal controls, customs regulations, state trading and subsidies were all identified in the introduction as key features of the General Agreement. This absence of a reference to the imposition of anti-dumping duties contradicted the view that from the outset the application of such duties had been considered a fundamental right under the General Agreement. To the contrary, the position of the State Department as reflected in this publication supported the view that Article VI had been intended to limit the application of anti-dumping measures, rather than to discipline dumping, as claimed by the United States.
78. The United States, referring to the Panel Reports in "Swedish Anti-Dumping Duties"39, and in "New Zealand - Imports of Electrical Transformers from Finland"40, argued that the interpretative history of the General Agreement strongly supported the view that Article VI should be construed as a remedial provision, rather than as an exception. The former Report was significant in that the Panel had held that a principle as important as the m.f.n. principle was not applicable to duties imposed under Article VI. By contrast, true exceptions, such as Article XX and XXI contained "soft" m.f.n. provisions, generally requiring or urging compliance with m.f.n. principles to the extent not inconsistent with the exception itself. Equally important was the Panel's holding that a party invoking Article VI bore no special burden of proof. Rather, the Panel had simply found that it:
"would be reasonable to expect that [a] contracting party should establish the existence of [dumping] when its action is challenged."41
38 See, e.g. Petersmann, "Need for Reforming Anti-Dumping Rules and Practices", in 45 Aussenwirtschaft 179 (1990). ↩
[Page 22]
To understand what the Panel had meant by "establishing the existence of dumping", it was instructive to note the context of the Panel's comment. The Panel had noted that the Swedish authorities "had not established that the export prices of the Italian exporters were less than the normal value".42 In the words of the Panel: "no definitive evidence had been brought forward to support the conclusion [of dumping]".43 In other words, the Swedish authorities had not even collected the most rudimentary evidence of dumping; indeed, they appeared confused as to whether they believed that dumping had occurred on the basis of a comparison between home market prices and export prices, third country prices and export prices, or constructed values and export prices.44 The facts of this case thus demonstrated that the Panel was to be taken at the plain meaning of its words when it had written that dumping must be "established" before action under Article VI was permitted.
79. With respect to the dispute in "New Zealand - Imports of Electrical Transformers from Finland", the United States observed that, while the Report of the Panel in this dispute was most frequently cited for reiterating the words of the Panel in the Swedish Anti-Dumping Duties case that a party invoking Article VI "must establish the existence" of injurious dumping, in fact the holding of the Panel in this case was far richer and consisted of two essential elements. First, the Panel had discussed the parties' respective allegations with regard to New Zealand's determination of dumping. After describing in detail the arguments of the parties to the dispute, it had reached the following conclusion:
"[The Panel] also noted that Article VI did not contain any specific guidelines for the calculation of cost-of-production and considered that the method used in this particular case appeared to be a reasonable one. In view of this ... the Panel considered that there was no basis on which to disagree with the New Zealand authorities' finding of dumping."45
The Panel's conclusion clearly indicated that, absent an express provision of the General Agreement which a complaining party could demonstrate to have been violated, and in the absence of evidence demonstrating that the factual basis of the determination did not conform to the requirements of the General Agreement, the party taking action under Article VI could be considered to have acted within its rights. In other words, the burden of producing evidence to the effect that the determinations were not made on a justifiable factual or legal basis rested with the complaining party.
80. Turning to the conclusions of the Panel on the injury determination, the United States noted that the Panel had faced an absolutist argument from New Zealand that the General Agreement did not permit any body other than a national investigating authority to make an Article VI determination or to review the basis for such a determination.46 Not surprisingly, the Panel had rejected this attempt to escape GATT review, finding that what New Zealand was asking for would be "complete freedom and unrestricted discretion in deciding anti-dumping cases without any possibility to review the action taken in the GATT".47 Once again, it was in the context of responding to this argument that the Panel had concluded that a contracting party was under an obligation "to establish the existence" of dumping and injury.
81. The United States concluded that the two above-mentioned Panel Reports revealed that a contracting party acting under Article VI must be able to illustrate the factual basis of its determinations. In other words, the authorities must establish, quite literally, the facts on which their decision was founded.
[Page 23]
However, it was up to the party asserting a GATT and/or Code violation to demonstrate the basis - based on the express requirements of the GATT or the Codes - for the finding of a violation.
82. The United States argued that the conclusory statement - in dicta - by the Panel in "United States - Countervailing Duties on fresh, chilled and frozen pork from Canada"48 concerning the scope of Article VI and its status as an "exception" to fundamental rights and obligations under the General Agreement found no support in the text of the General Agreement. The sources relied upon by this Panel when making this statement did not even relate to the interpretation and application of Article VI: the Report of the Panel in "Canada - Administration of the Foreign Investment Review Act"49 concerned an interpretation of Article XX; the Report of the Panel in "Canada - Import, Distribution and Sale of Alcoholic Drinks by Canadian Provincial Marketing Agencies"50 involved Article XXIV:12, and the Report of the Panel in "Canada - Import Restrictions on Ice Cream and Yoghurt"51, involved an interpretation of Article XI:2(c)(i) of the General Agreement. Accordingly, the Panel's statement regarding Article VI as an exception was fundamentally in error and should be rejected by the Panel in this case.
83. Norway argued that previous Panel Reports supported the position that Article VI of the General Agreement was an exception to fundamental rules of the General Agreement. The Panel in "United States - Countervailing duties on fresh, chilled and frozen pork from Canada" had specifically stated that:
"Article VI:3, an exception to the basic principles of the General Agreement, ha[s] to be interpreted narrowly."52
The sources cited by this Panel in its statement on Article VI were relevant in that the Panel Reports referred to by this Panel had involved various exceptions to the basic rules of the General Agreement and had described how such exceptions were to be interpreted. Each of these exceptions required the contracting party invoking the exception to justify the use of the exception with specific evidence. The United States could not ask this Panel to ignore the findings of the Panel in the pork case given that GATT panels were to make their judgements based upon the provisions of the General Agreement and past panel reports. The Report of the Panel in the pork case had confirmed that Article VI was an exception (as done by other Panel Reports). Contrary to what had been argued by the United States, Norway was not asking this Panel to subject Article VI remedies to stricter scrutiny than actions taken under other provisions of the General Agreement. Rather, Norway was asking the Panel that, in accordance with previous panel cases referred to in the pork decision involving other exceptions to the General Agreement, this Panel require the party applying the exception to justify in factual detail the consistency with the Agreement of its determination of the existence of dumping and injury.
84. Norway further observed in this context that the Report of the Panel on "EEC - Regulation on Imports of Parts and Components"53 had also described Article VI as an exception.
85. Norway argued that the Panel in "Swedish Anti-Dumping Duties"54 had found that the m.f.n. requirement did not apply to measures taken under Article VI not, as suggested by the United States,
52 BISD 38S/30, paragraph 4.4. ↩
[Page 24]
because of the fundamental nature of the rights of contracting parties under Article VI, but precisely because of the nature of Article VI as an exception to the m.f.n. requirement.
86. The United States noted that, although Norway had not referred to these sources, recent advocates seeking to circumscribe the scope of action under Article VI might have in mind a statement in the first Report of the Group of Experts that anti-dumping and countervailing duties "were to be regarded as exceptional and temporary measures to deal with specific cases of injurious dumping or subsidization".55 This statement could, however, not provide an argument in support of the view that Article VI was an exception. The word "exception" denoted "the act of excepting: EXCLUSION; a case to which a rule does not apply". By contrast, the word "exceptional" denoted something "RARE" or "deviating from the norm".56 Anti-dumping and countervailing duties might have been intended to be "exceptional" in the sense that most products should not be subject to such measures because unfair trade should be the exception rather than the norm. However, to say that these measures were exceptional was completely different from asserting that the drafters had intended Article VI to be an exception to fundamental rights and obligations of contracting parties under the General Agreement, causing a party taking action under this Article to bear a special burden of proof to justify its action. Moreover, it was notable that the same Group of Experts, in a second Report, had clearly reaffirmed the broad nature of Article VI remedies when it had observed that:
"The fact that the granting of certain subsidies was authorized by the provisions of Article XVI of the General Agreement clearly did not debar importing countries from imposing, under the terms of Article VI, a countervailing duty on the products on which subsidies had been paid."57
If Article VI had been intended to be read narrowly, then surely subsidies specifically authorized elsewhere in the General Agreement would be among the first items read out of the purview of Article VI.
87. Norway argued that the first Report of the Group of Experts on Anti-dumping and Countervailing Duties confirmed that Article VI was an exception when it stated that anti-dumping duties and countervailing duties "were to be regarded as exceptional and temporary measures to deal with specific cases of injurious dumping or subsidization".58 The United States had attempted to advance a semantic argument differentiating the term "exception" from "exceptional". However, the first meaning of the term "exceptional" in Websters Third New International Dictionary (the unabridged version of the dictionary cited by the United States), American Heritage, and Oxford English dictionaries was "forming an exception" or "being an exception". Both words meant the same: a deviation from the central principles of the General Agreement.
88. The United States concluded that an examination of the text as well as the drafting and interpretative histories of the General Agreement led to two basic conclusions concerning the status of Article VI in the framework of rights and obligations of the General Agreement. First, Article VI accorded rights to act against unfair anti-competitive trade practices which were essential to the establishment, essential balance and continued successful functioning of the GATT system. Second, the right under Article VI to impose offsetting duties was remedial in nature. In the case before the Panel, the United States had more than amply illustrated that the facts in the records of the US authorities established the existence of injurious dumping within the meaning of Article VI and the Agreement, consistent with the findings in the Panel Reports on "Swedish Anti-Dumping Duties" and "New Zealand - Imports of Electrical
56 Webster' s Ninth New Collegiate Dictionary at 432 (1990). ↩
57 BISD 8S/194,200 (paragraph 32). ↩
[Page 25]
Transformers from Finland". The basis for Norway's claim in the present case appeared to be the view that it was sufficient for a contracting party challenging an action under Article VI to raise issues - whether or not founded on express requirements under the General Agreement - and then shift the burden onto the contracting party taking action under Article VI to prove the consistency of its action. However, Norway had not referred to specific legal requirements under the Agreement which would have been violated. Rather, Norway's entire argumentation was founded on the premise that, as the Party taking action, the United States bore some additional burden of proof. It was on the basis of this higher obligation of proof that Norway asked the Panel to find fault with the US determinations.
89. The United States considered that there were three basic problems with the approach taken by Norway in these proceedings. First, there was no basis for Norway's view that Article VI was an exception to fundamental rights and obligations under the General Agreement. Second, as the New Zealand Transformers Panel had held, a violation existed only when a determination was shown to be inconsistent with an express requirement. Norway had not shown that in the present case any express requirement of the Agreement had been violated. Finally, Norway's proffered rôle for panels as triers of fact was in fundamental conflict with the express provisions of the Agreement, which explicitly and exclusively empowered "the competent national authorities" to conduct the investigation. By contrast, dispute settlement provisions of the Agreement clearly contemplated that the important rôle reserved for panels was to resolve disagreements over interpretations of provisions of the Agreement.
90. Norway argued that even if one (incorrectly) assumed that Article VI was not an exception to fundamental GATT principles, a contracting party imposing anti-dumping or countervailing duties had to demonstrate that its determinations were consistent with the requirements of the General Agreement and with the requirements of the Agreements implementing and interpreting Article VI. The United States had argued that the contracting party taking action under Article VI need only meet a test of "reasonableness" and that it was up to the party asserting a violation of Article VI to demonstrate the basis for a finding of a violation. Under this proposed rule, the United States did not need to present all facts to the Panel to affirmatively demonstrate the "reasonableness" of its determinations but only needed to describe the methodology used and the conclusions it had reached and could then ask the Panel to assume that the determinations made were consistent with the requirements of the Agreement. Even if this standard of "reasonableness" were the correct standard, the United States had failed to demonstrate that its actions in the investigation of Atlantic salmon from Norway met this standard. Thus, the United States had failed to ask the petitioners even the most basic questions to determine whether they had in fact filed the petition on behalf of the domestic industry affected and the United States had imposed extremely onerous standards of response on Norwegian respondents but not on domestic respondents. While the United States claimed that Norway had not identified express requirements of the Agreement alleged to have been violated by the United States, Norway had in fact demonstrated how the actions of the United States violated specific requirements of the Agreement, including those contained in Articles 5:1 and 3:4.
91. Norway argued that, despite the claim of the United States, previous Panel Reports had not adopted a standard of "reasonableness" when reviewing actions taken under Article VI of the General Agreement. Rather, these Reports confirmed the view that the party taking action under this Article had to demonstrate that its actions were in conformity with the requirements of this Article and that it had established the requisite facts before imposing duties. In the proceedings before this Panel, the United States had neither provided the facts that formed the basis of its anti-dumping determination nor demonstrated that its anti-dumping duty was in conformity with the Agreement. Norway noted the argument of the United States that the Panel Report on "Swedish Anti-Dumping Duties"59 case had concluded that a party taking action under Article VI bore no special burden of proof. However,
[Page 26]
Norway was not asking for a "special" burden of proof. Rather, it was asking that the Panel apply the same rule applied by previous Panels, i.e. that the United States demonstrate to the Panel that its determinations were in conformity with the Agreement. In discussing the Swedish Anti-Dumping Duties case, the United States had asserted that the Panel had simply found that it:
"would be reasonable to expect that [a] contracting party should establish the existence of [dumping] when its actions is challenged."
However, this was not "simply" what the Panel had found. The full statement of the Panel read as follows:
"It is clear from the wording of Article VI that no anti-dumping duties should be levied unless certain facts have been established. As this represented an obligation on the part of the Contracting Party imposing such duties, it would be reasonable to expect that that Contracting Party should establish the existence of these facts when its action is challeged."60
Thus, the Swedish Anti-Dumping Duties Panel required much more than clarity in the importing country's determination that dumping existed and confirmed the affirmative obligation of the contracting party imposing duties to demonstrate the existence of "certain facts".
92. Norway considered that the United States had failed to provide the full quotation from the New Zealand Transformers Case when describing that panel's conclusion regarding the standard of review. The full text of the second sentence in the statement quoted by the United States read as follows:
"In view of this and having noted the arguments put forward by both sides as regards the costing of certain inputs used in the manufacture of the transformers, the Panel considered that there was no basis on which to disagree with the New Zealand authorities' finding of dumping."61
The part of this sentence omitted by the United States was essential as it demonstrated that the Panel had accepted the view of New Zealand, not because Finland had failed to meet some burden of proof, but because the Panel had required New Zealand to demonstrate the specific facts underlying its decision and had evaluated that decision on the basis of those facts. The United States was incorrect in paraphrasing the above standard as meaning that "the burden of producing evidence to the effect that the determinations are not made on a justifiable factual or legal basis rests with the complaining party". The Panel could not have been more direct in confirming that it was the contracting party imposing anti-dumping duties which was obliged to establish to the satisfaction of the Panel the factual basis and GATT-consistency of its determinations of dumping and injury. Norway further argued in this context that, if ever there were any doubts as regards the implications of the Swedish Anti-Dumping Duties and New Zealand Transformers cases with respect to the question of the obligation of a contracting party imposing duties under Article VI, this matter had been settled by the Panel in "United States - Countervailing Duties on fresh, chilled and frozen pork from Canada" when it had concluded that it was "up to the ... party invoking the [Article VI] exception, to demonstrate that it had met the requirements of Article VI:3".62 This obligation of a contracting party to demonstrate that it had met the necessary requirements of Article VI was not conditioned on Article VI being an exception to fundamental GATT principles.
61 BISD 32S/55, paragraph 4.3. ↩
62 BISD 38S/30, paragraph 4.4. ↩
[Page 27]
93. Norway argued that the initiation by the United States of the anti-dumping duty investigation on imports of Atlantic salmon from Norway was inconsistent with Article 5:1 of the Agreement as a consequence of the failure of the United States' authorities to satisfy themselves before initiating the investigation that the request for the initiation of this investigation was filed on behalf of the domestic industry.
94. The United States argued that the petition had provided a satisfactory statement of industry support. In light of the certified statement that the major proportion of the domestic industry supported the petition, and the lack of significant opposition to the petition, the Department of Commerce had considered the petition to have been filed on behalf of the domestic industry and had satisfied itself of industry support prior to the initiation of the investigation.
95. Norway noted that the investigation had been initiated following a petition received by the United States' authorities on 28 February 1990 from the Coalition for Fair Atlantic Salmon Trade (FAST). This Coalition had requested the initiation of an anti-dumping and a countervailing duty investigation "on behalf of the United States' producers of fresh Atlantic salmon".63 The petition had described FAST as "a limited trade association organized for the purpose of pursuing relief from unfairly traded Atlantic salmon from Norway under the US international trade laws". Its address was "c/o Ocean Products, Inc.", a firm which, shortly after the petition was filed, had been taken over by a Canadian firm. The petition listed in support of the petition 21 member companies which "to the best of the petitioner's information ... currently accounts for well over a majority of all production of this product in the United States".64 In consultations held between the United States and Norway after the imposition of the anti-dumping duty order, the United States had indicated that in the case at hand it had followed its standard practice with respect to the question of the standing to file an anti-dumping duty petition: unless a substantial portion of the domestic industry came forth to oppose a petition, the Department of Commerce reasonably assumed that the domestic industry in question, or a major proportion thereof supported the petition. The United States interpreted the term "substantial portion" as more than half of the industry.
96. Norway considered that the assumption that, absent express opposition to a petition by domestic producers accounting for at least 50 per cent of production, the industry, or a major proportion thereof, supported the petition had been rejected in the Report of the Panel on "United States - Imposition of Anti-Dumping Duties on Imports of Seamless Stainless Steel Hollow Products from Sweden".65 In its Report the Panel had stated that:
"... it did not consider that absence of opposition by domestic producers was a factor which, by itself, demonstrated that a written request for the initiation of an investigation was filed on behalf of the domestic industry."66
The Report also stated that a request for the initiation of an anti-dumping duty investigation:
63 Fresh Chilled and Frozen Salmon from Norway: Petition for the Imposition of Anti-Dumping and Countervailing Duties Pursuant to Sections 701, 702, 731 and 732 of the Tariff Act of 1930, as Amended, on behalf of the Coalition for Fair Atlantic Salmon Trade, 28 February 1990 (Public version), p.1. ↩
[Page 28]
"...must have authorization or approval of the industry affected before the initiation of an investigation."67
Furthermore, according to the Report, investigating authorities were required, prior to the opening of an investigation, to take steps which could reasonably be considered to be sufficient to ensure that the initiation of the investigation was consistent with the obligation of the authorities to satisfy themselves that the request was filed on behalf of the domestic industry affected.68
97. The United States considered that Norway's argument placed inappropriate support on the unadopted Panel Report on "United States - Imposition of Anti-Dumping Duties on Imports of Seamless Stainless Steel Hollow Products from Sweden". The Panel had noted in its Report that the Agreement did not provide precise guidance with regard to standing and that the question of how this requirement was to be met depended on the circumstances of each particular case. The Panel's conclusion was that the initiation of an anti-dumping investigation in the circumstances of the case before the Panel was inconsistent with the obligations of the United States under the Agreement. The standards set forth by the Panel had been satisfied by the petition which had led to the initiation of the investigation of imports of salmon from Norway. The Panel had concluded that a written request filed on behalf of the industry affected "implies that such a request must have the authorization or approval of the industry affected before the initiation of an investigation". The petition filed in the salmon case had provided exactly such an authorization when it stated that:
"The members of these two trade associations include substantially all of the US growers of fresh Atlantic salmon."
Thus, the authorization which the Panel had not found in the Swedish steel case had been expressly presented to the investigating authorities in the salmon case. Also, no reason had been presented to the authorities to revisit the issue, despite the explicit request by the Department of Commerce for comments on the standing issue. The Panel Report on the dispute between Sweden and the United States described a factual scenario vastly different than that in the present case and Norway's reliance on this Report as the sole basis for its arguments on the question of standing was therefore misplaced. In any event, the Report had not been adopted. Moreover, even if the Report had been adopted, it could not be given retroactive applicability to the present case.
98. Norway observed that there was no information indicating that the United States' authorities had taken any steps to satisfy themselves prior to the initiation of the investigation (or at any other time) that the petition had been filed on behalf of the domestic industry affected, despite the fact that one domestic producer had notified the Department of Commerce before the initiation of the investigation of its disagreement with the petition. The United States had thus been aware that a significant portion of the industry opposed the petition. Norway referred in this context to a letter received by the Department of Commerce on 19 March 1990 (one day before the initiation of the investigation) from a domestic producer, Global Aqua stating that this producer did not support the petition and did not agree with the allegations contained therein. The only meaningful interpretation of the requirement that a written request for the initiation of an investigation be "by or on behalf of the industry affected" was that such a request required affirmative support by the industry. A lack of express support, for whatever reason, was different from active support. Nowhere did the Agreement imply that lack of opposition by domestic producers to a written request for the initiation of an investigation was a sufficient basis for the initiation of an investigation. Norway noted that the facts of the salmon case made an even more compelling argument that the United States had not met its obligations under the Agreement
[Page 29]
than the facts of the case considered by the Panel in "United States - Imposition of Anti-Dumping Duties on Imports of Seamless Stainless Steel Hollow Products from Sweden". In the latter case, the Panel had found that the United States was under an obligation to satisfy itself that the petition was filed on behalf of the industry, even though the domestic industry had never provided any indication that it was opposed to the petition. In contrast, in the salmon case, at least one domestic producer in the United States had written to the Department of Commerce before the initiation of the investigation to state its disagreement with the petition.69
99. Norway also pointed to other facts which called into question the petitioner's claim to act on behalf of the domestic industry. First, while the petition had listed twenty-one firms as members of FAST, in January 1991 FAST had submitted a brief to the Department of Commerce in which only thirteen firms were listed as members of the association. Second, the petition had asserted that the request for the initiation of an investigation was supported by the Washington State Fish Growers Association (WFGA), whose members resided principally in the State of Washington. Had the United States' authorities investigated this assertion, they would have found that this Association was not supporting the petition, as was evident from a letter dated 16 March 1990 from the President of the WFGA to counsel for the petitioner.70 Third, during the course of its investigation, the USITC had obtained information calling into question the assumption of industry support for the petition. The Annex to the final determination of the USITC indicated that producers representing approximately 50 per cent of the domestic industry (by production) either opposed or did not express support for the petition. Producers accounting for over one-third of production had expressed opposition to the petition. This figure was based on the 1988/89 harvest season and the 1987/88 smolt harvest. The USITC had noted that, based on earlier harvest seasons, the firms expressing opposition to the petition produced more Atlantic salmon than did the firms in support of the petition. Finally, the largest domestic producer, Ocean Products (the assets of which had been purchased by a Canadian firm during the investigation), while claiming to support the petition, had not provided a questionnaire response in the final investigation of the USITC, either as Ocean Producers or as its successor, Connors Brothers, and had thus expressed lack of interest in the outcome of the investigation.71
100. The United States noted that the petitioner, the Ad-Hoc Coalition for Fair Atlantic Salmon Trade had described itself in the petition as an organization consisting of 21 members who represented a major proportion of domestic production of fresh Atlantic salmon. The petition had stated that:
"Most of the coalition members are also members of either the Cobscook Bay Finfish Grower Association whose members reside principally in th[e] State of Maine, or the Washington State Fish Growers Association, whose members reside principally in the State of Washington. The members of these two trade associations include substantially all of the US growers of fresh Atlantic salmon. Both organizations have voted to support the petition."72
It was important to note that the petition did not state that the WFGA was a co-petitioner but rather that this association supported the petition. The petition contained certifications by both a member of the petitioner coalition and petitioner's legal counsel as to the completeness and accuracy of the
69 Letter from Global Aqua to FAST, 14 March 1990. ↩
70 Letter from the Washington Fish Growers Association to Michael Coursey, 16 March 1990. ↩
71 USITC Determination p.A-19, note 49. ↩
[Page 30]
statements presented therein.73a) The petition also listed those firms which had expressed no opinion about the petition, including Global Aqua, a domestic producers of Atlantic salmon owned by a Norwegian farm. No firm had expressed opposition to the petition (although Norway had portrayed Global Aqua's statement of non-support as opposition). Global Aqua had never stated that other producers might or did oppose the petition and had never requested the Department of Commerce to revisit the prima facie showing of industry support contained in the petition.
101. The United States pointed out that, after the petition had been filed, the WSFGA had indicated that it did not support the petition. The Association, however, had not expressed opposition to the petition. Promptly upon receiving this notice, counsel for the petitioner had notified the Department of Commerce of this change and had amended the petition accordingly.74 Norway had failed to even mention this amendment, creating the misimpression that the petitioner had ignored the change in the Washington Grower's sentiments. This was not the case. Norway was therefore wrong in arguing that the Washington Growers had not originally supported the petition and that their position had been misrepresented in the petition. The Washington Growers had supported the filing of the petition and had assumed a neutral stance only after the petition had been filed. In fact, the president of the Washington Growers had been the major proponent of the commencement of an investigation.75 Many of the companies in the Washington Growers Association were owned by Norwegian salmon interests. The organization's actions after the filing of the petition were accounted for by pressure from the Norwegian owners to oppose or maintain a neutral stance in the investigations. The fact that in its brief filed in January 1991 FAST had listed thirteen, rather than twenty-one firms as members, reflected the decision of certain Washington State producers to take a position with respect to the investigation after filing of the petition. Even after some west-coast producers had changed their position to one of neutrality, the petition still had the support of a majority of the domestic industry, as it did throughout the entire investigation. The correctness of Commerce's original finding of the petitioner's standing had therefore not been affected by the post-filing statement of neutrality by the Washington Growers.
102. The United States further noted that in its notice of the initiation of the investigation, the Department of Commerce had specifically asked respondents for additional comments so that, if necessary, it could revisit its initial finding that the petitioner had filed the petition on behalf of the domestic industry. No comments had however been received in response to this initiation and the Department had thus not been presented with any reason to revisit its finding on the question of the standing of the petition.
73 The United States explained in this connection that, pursuant to amendments made in 1988 to the United States statute, the Department of Commerce required that factual information provided by parties be certified as accurate. This provision had been added to the law in order to ensure that proceedings ↩
"... are not initiated or conducted based upon frivolous allegations and arguments which are either not supported by the facts alleged, or decided based on arguments that omit important facts known or reasonably available to the party making the submission of fact." S.Rep. No. 71, 100th Cong., 1st Sess. 114 (1987).
The Department of Commerce was currently working on proposals for the enforcement of this provision. To the extent a party's legal representative was found to have falsely certified information, there would be implications for that representative's standing with the bar.
74 Letter from Michael Coursey to Robert A. Mosbacher, Sr. and Kenneth R. Mason, 16 March 1990. ↩
75 The United States referred in this connection to an article in Seafood Trend, 13 November 1989, p.4. ↩
[Page 31]
103. The United States also submitted that the determination of the USITC demonstrated that the industry had supported the petition.76 The data in the Report of the USITC included domestic producers who were related to exporters of the product under investigation and who therefore could have been excluded from the definition of the domestic industry under Article 4:1 of the Agreement. Had such producers been excluded from the industry, the extent of industry support for the petition would have been even higher.
104. Regarding the issue raised by Norway with respect to the participation of Ocean Products in the USITC's injury investigation, the United States pointed out that this company had responded to the questionnaire in the preliminary investigation of the USITC. However, the company had ceased operating and had been liquidated by September 1990. The USITC questionnaire in the final injury investigation had been sent in October 1990. There simply no longer was a corporate entity to respond. However, an official of the former Ocean Products provided the USITC with the necessary information, as was specifically noted in the USITC Report.77 Connors Aquaculture, which had purchased the assets of Ocean Products, had provided a questionnaire response in the final investigation.
105. Norway noted that the Annex to the determination of the USITC stated in footnote 49 on page A-19 that one firm (unidentified but obviously Ocean Products) "would be unable to provide a questionnaire response in the final investigations". The note went on to state that the "data for Ocean Products presented in this report are based on its preliminary questionnaire and on those additional documents". Thus, the data were not based on a response by Ocean Products to the USITC's questionnaire in the final investigation. Moreover, in footnote 50 the USITC Report stated that "Connors Aquaculture was unable to provide data relating to the operations of Ocean Products" and thus did not answer the final questionnaire. This was the only information available to Norway and it indicated that Ocean Products had not answered the final questionnaire. The United States now claimed that Ocean Products had answered that questionnaire. Since the United States had access to data to which neither the Panel nor Norway was privy, Norway could not determine whether the USITC Report stated the facts incorrectly or whether the United States was now stating the facts incorrectly. Obviously, the two statements were contradictory.
106. Norway also noted in this context that Ocean Products had not been alone in not responding or in not providing a full questionnaire response. The USITC Report indicated that many of the approximately 25 firms farming Atlantic salmon in the United States had not submitted complete responses.78 Thus, in contrast to the treatment of the Norwegian farmers and exporters, the domestic producers in the United States were not required to submit all the information requested by the investigating authorities and no adverse inferences had been made when the requested information was not supplied.
107. In summary, Norway argued that the affirmative final determination of dumping by the Department of Commerce in respect of imports of fresh and chilled Atlantic salmon from Norway was inconsistent with the obligations of the United States under the Agreement as a result of (1) a failure to follow fair and equitable procedures, (2) the calculation of constructed normal values in a manner which had led to an overstatement of the margins of dumping contrary to Articles 2:6 and 8:3 of the Agreement, and (3) a failure to effect a fair comparison of the normal value and export price, contrary to Articles 2:6 and 8:3 of the Agreement.
76 USITC Determination p.A-14. ↩
77 USITC Determination, p.A-19, note 49. ↩
78 USITC Determination, p.A-14, A-24 and A-29. ↩
[Page 32]
108. Norway argued that the affirmative final determination of dumping made with respect to imports of Atlantic salmon from Norway was inconsistent with the obligations of the United States to accord fair and equitable procedures, as reflected in the Preamble of the Agreement and in the provisions of Articles 5 and 6. In particular, Article 6:1 contained a fundamental principle of fairness in that it required that all interested parties in an anti-dumping duty investigation have ample opportunities to provide evidence in writing. In support of this claim Norway listed six grounds, pertaining to: (1) the denial of ample opportunities to the Norwegian respondents to present evidence in writing, contrary to Article 6:1 of the Agreement; (2) the imposition of onerous requirements with respect to questionnaires and verification procedures; (3) differing treatment of domestic and foreign respondents, in violation of Article III of the General Agreement; (4) the treatment of sales below fully allocated costs of production as not being "in the ordinary course of trade"; (5) calculation of costs of production on the basis of the costs of production of the salmon farmers, rather than on the basis of the acquisition price paid by the salmon exporters; and (6) inclusion in the constructed normal value of a "freezing charge".
109. Norway argued that the United States had acted inconsistently with Article 6:1 of the Agreement in that the Norwegian exporters of salmon had been given only fifteen days to respond to Section A of the questionnaire issued by the Department of Commerce on 30 April 199079 instead of the thirty days provided for in the Recommendation adopted by the Committee on Anti-Dumping Practices on the subject of time limits for responses to questionnaires.80 Section A of the questionnaire had contained a request for detailed information on, inter alia, corporate structure, accounting practices, distribution system and sales process. It had also requested respondents to provide a list of all farms from which they had purchased salmon during the period 1 September 1989- 28 February 1990 (the period covered by the Department's investigation). The request for data on suppliers was particularly difficult to meet because the Norwegian salmon exporters did not keep records of farms which had supplied salmon to the exporters during a particular period of time. For example, an individual exporter who purchased salmon from 18 to 100 different farms in a given year did not necessarily have records on the specific farms from which salmon had been purchased in a particular month.
110. Norway considered that where, as in this case, an exporter purchased salmon from a number of farms, fifteen days were not sufficient to determine accurately the exact date of purchase of salmon for sale to the United States. As a result, the lists of farms submitted by the exporters in response to Section A of the questionnaires had not always been accurate and had included farms which had not supplied salmon to the exporters during the six-month period specified by the Department of Commerce. Given the difficulty of gathering this data, the incompleteness or inaccuracy of the lists could not be considered to mean that the exporters were not co-operating. The exporters should have been given the opportunity to correct the lists initially submitted. Although, at the time of the submission of these lists, the Department of Commerce had been informed that the lists were not completely accurate because of the difficulty of obtaining the information in the form requested, the Department had never asked the respondents to correct the lists. Instead, almost four months after the receipt of these lists, the Department had concluded that the lists were flawed because they included farms which had not supplied salmon to the exporters during the period of investigation. This had
[Page 33]
significantly affected the methodology employed by the Department in the construction of samples of farms for purposes of its costs of production investigation.81
111. In support of its contention that, at the time of the submission by the exporters of the responses to Section A of the questionnaires, the Department of Commerce had been informed that the lists of suppliers were not entirely accurate, Norway pointed to the following. The Department had asked that the exporters list all farms from which they had purchased salmon for export to the United States during the period 1 September 1989-28 February 1990. It was clear from the wording of the responses submitted on 16 May 1990 by three exporters that they had not been able to provide full information on this issue. Thus, one exporter, Skaarfish Mowi, had replied as follows:
"Please note, however, that Exhibit H [the list of suppliers] is a complete list of salmon farms from whom Skaarfish purchased. Skaarfish's records do not permit it to break out the farms authorized for US sales, or farms purchased from during the POI, during the short time allowed for the Section A response."82
A second exporter, Chr. Bjelland Seafoods A/S, had stated in its response:
"A list of farmers and packers that Chr. Bjelland dealt with during the period of investigation is contained in Exhibit C."83
Finally, a third exporter, Sea Star International A/S, had replied as follows:
"A list of farmers and packers that Sea Star dealt with during the period of investigation is contained in Exhibit C."84
On the basis of these responses, it was clear as early as mid-May 1990 that the lists of farms supplied by the exporters were not entirely accurate.
112. The United States argued that the exporters had been given sufficient time to file their responses to Section A of the questionnaire issued by the Department of Commerce on 30 April 1990. The errors in the exporters' responses to Section A of the questionnaire could not be attributed to the allegedly insufficient period of fifteen days to respond to this Section. The exporters had requested, and had been granted an extension, and could have sought additional time, had they so desired. They had not done so, thus demonstrating that the response time was not a problem. Moreover, the Department of Commerce had welcomed amendments and corrections to these Section A responses for months afterward and the exporters had taken advantage of this option. The Department of Commerce had allowed the exporters to correct the lists of farms subsequent to their initial submission on 16 May 1990. Thus, in mid-June 1990, three exporters had submitted corrected lists of the farms with which they had dealt during the period of investigation.85 On several other occasions, various exporters had
82 Response of Skaarfish Mowi A/S to Section A of the Questionnaire of the US Department of Commerce, 16 May 1990, p.9. ↩
83 Response of Chr. Bjelland Seafoods A/S to Section A of the Questionnaire of the US Department of Commerce, 16 May 1990, p.9 (emphasis by Norway). ↩
84 Response of Sea Star International A/S to Section A of the Questionnaire of the US Department of Commerce, 16 May 1990, p.8 (emphasis by Norway). ↩
85 Letter from David Palmeter to Robert Mosbacher on behalf of Sea Star International A/S, 12 June 1990; Letter from David Palmeter to Robert Mosbacher on behalf of Chr. Bjelland Seafood A/S, 13 June 1990; Letter from David Palmeter to Robert Mosbacher on behalf of Salmon A/S, 14 June 1990. ↩
[Page 34]
submitted amendments to correct or revise other parts of their Section A responses. Each such submission had been accepted by the Department of Commerce without objection from the petitioner. Thus, the exporters had been provided, and had exercized, a very broad ability to amend their Section A responses throughout the period May-August 1990. The exporters had provided the false information and had failed for months thereafter to correct this information. In a letter dated 30 August 1990, counsel for the Norwegian respondents had reported for the first time to the Department that several of the exporters had given erroneous information in their Section A responses and that several of the farms selected for purposes of the Department's costs of production investigation had not sold any salmon during the period of investigation to the exporters to which they were linked.
113. The United States considered that the Department of Commerce had acted consistently with the Recommendation of the Committee on Anti-Dumping Practices86 that respondents be given thirty days to respond to questionnaires. In fact, the respondents had been given well over thirty days to respond to the questionnaires. In response to Norway's argument that the Department of Commerce should have notified the exporters of the deficiencies in their lists of farms, the United States argued that only the exporters had known the deficiencies; the Department had been unaware of these deficiencies until the exporters' belated admission of error and therefore could not have advised the exporters of the flaw.
114. Norway considered that the contention of the United States that the Department of Commerce had complied with this Recommendation was incorrect; the Norwegian exporters had not been given "well over 30 days" to respond to Section A of the questionnaire issued on 30 April 1990.
115. The United States considered that Norway's argument that the Department of Commerce had provided less time for responses to the Section A questionnaire than was called for by the Recommendation of the Committee on Anti-Dumping Practices overlooked the fact that Section A was part of a questionnaire, not the whole questionnaire. The Department of Commerce had provided the Norwegian exporters with more than three months to respond to the entire questionnaire and had also provided ample opportunity to the exporters to amend and correct their submissions. The Recommendation of the Committee on Anti-Dumping Practices did not require thirty days response periods for individual parts of a questionnaire; even so, when the liberal amendment policy of the Department of Commerce was considered, the exporters effectively had been given well over thirty days to respond to Section A.
116. In response to the argument of the United States that the Recommendation of the Committee on Anti-Dumping Practices did not require thirty day response periods for individual parts of a questionnaire, Norway argued that, under the logic of this argument, the United States could have divided its questionnaire into thirty parts and give respondents one day to respond to each part. If respondents then failed to fully respond in that one-day time period, the United States would be justified in applying a punitive best information standard. This could not be permissible under the Agreement.
117. In response to the argument of the United States that the exporters could have requested an extension of the period within which to respond to the Section A questionnaire, Norway submitted that under US practice such extensions were not automatic and were rarely granted for more than a few days. In addition, this argument of the United States improperly placed the burden on the Norwegian respondents. Article 6:1 of the Agreement, by contrast, placed the burden on the investigating authorities to provide adequate time to respond to questionnaires. With respect to the information submitted by the United States regarding the requests granted for extensions of the time
[Page 35]
period to respond to the questionnaire, Norway argued that it was inconsistent with Article 6:1 that a respondent had to request an extension, rather than being provided with sufficient time in the first place. Article 6:1 provided that foreign suppliers and all other interested parties "shall be given ample opportunity" to provide evidence in writing. To require respondents to request an extension created great uncertainty. It was by no means certain that the Department of Commerce would grant an extension, or grant an extension for the entire period requested, as illustrated by a letter dated 14 September 1990, from the Department to counsel for the Norwegian respondents in which the Department had only partially granted a request for an extension.87
118. The United States argued that Norway's portrayal of the Department of Commerce as unwilling to grant extensions of time was demonstrably false. The Department of Commerce had granted every request for additional time filed by the Norwegian respondents.88 In only one instance had a requested extension not been granted for the full period of time requested.
119. The United States contested that some exporters might have been unaware of the liberal extension and amendment policy of the Department of Commerce in this case and could not therefore be faulted for their failure to provide correct information. First, the Department had granted every single request for extension or amendment submitted by the exporters. Second, all exporters had been represented by the same US law firm which had sought extensions for responses, and which had provided amendments on behalf of the exporters. The record amply demonstrated that the law firm knew of the Department's policies and had informed at least some of the exporters that amendments could be made. It was far-fetched to assume that the law firm had informed only some exporters, but not others, of the ability to amend.
120. In response to Norway's argument that, while Article 6:1 required that investigating authorities provide sufficient time to respond, the Department of Commerce had placed a "burden" on respondents by requiring them to request additional time, the United States argued that the policies followed by the Department of Commerce, including more than sufficient initial time to respond and the liberal extension and amendment policies which had invaryingly been applied, met the requirement of Article 6:1. The only "burden" placed on the Norwegian parties to seek any additional time they needed was to make a request for such additional time. Unless the respondents made such a request, the Department could not have been aware of the need for additional time.
121. Norway contested that, as argued by the United States, on 30 August 1990 the Department of Commerce had for the first time been informed of the inaccuracies in the lists of farms provided in response to the Section A questionnaire and referred in this respect to the responses submitted by three exporters on 16 May 1990.89
122. The United States considered as unpersuasive Norway's argument that in the responses of the exporters referred to by Norway the Department of Commerce had been notified that the responses were incomplete. Both Chr. Bjelland Seafoods A/S and Sea Star International A/S had stated in their Section A responses that "a list of the farmers and packers that [the exporter] dealt with during the period of investigation is contained in Exhibit C". These responses on their face appeared to state that the Department's request had been fully met. There was no statement, explicit or implicit, that time constraints had forced the exporters to respond incorrectly. Moreover, both exporters had subsequently submitted amended lists, an option which Skaarfish could also have followed.
87 Letter from David L. Binder to David Palmeter, 14 September 1990. ↩
88 The United States provided to the Panel copies of the letters in which these requests were granted. ↩
[Page 36]
123. Norway also submitted that the Department of Commerce had itself been responsible for the errors in the information provided by the exporters because it had asked for the wrong information. The United States had explained in the proceedings before the Panel that for the second farm sample the Department of Commerce had determined to select farms from a list of all the farms serving an individual exporter.90 However, in Section A of the questionnaire issued on 30 April 1990 the Department of Commerce had only requested information on farmers supplying salmon to the exporters for export to the United States during the period of investigation. Consequently, even if the lists submitted by the exporters in response to the Section A questionnaire would have contained the information requested by the Department of Commerce, they would not have contained the information required by the Department when it had decided to construct a sample of farms for purposes of calculating the costs of production of the farms. Given that the Department was investigating whether sales of salmon to the EEC were at prices below costs of production, information on farms supplying salmon for export to the United States would have been irrelevant.
124. The United States denied that, by requiring exporters to report the farms from which during the period of investigation they had purchased salmon for export to the United States, the Department of Commerce had asked for the "wrong information". In developing eight separate samples of farms (one for each exporter) for its sales below-cost investigation91, the Department had decided to use the universe of farms for each exporter submitted in response to question 4.A.vii of Section A of the questionnaire issued to the exporters on 30 April 1990. This question had asked the exporters to "provide the name and address of each salmon farm from whom you purchased salmon for export to the United States during the POI." Accordingly, the Section A lists of farms submitted by the exporters had comprised only farms from which the exporters had purchased salmon during the period of investigation for export to the United States. Theoretically at least, a list for one or more exporters might have included fewer than all the farms from which the exporter purchased salmon during the period of investigation, since only about half the farms in Norway had been approved to have their product ultimately shipped to the United States.
125. The United States argued that the Department of Commerce had reasonably limited the lists to farms which had received approval, and whose product had actually been shipped by the relevant exporter to the United States during the period of investigation. Indeed, the Department could not have made another choice. A constructed normal value was based upon the costs of production and included selling, general and administrative expenses, profit and packing costs. The costs of production were used to determine whether sales in the home market or to a third country market were made at below cost prices and thus were inappropriate for use in deriving the normal value. The costs of production measured the costs of producing the product sold in the home or third country market (in this case, countries in the EEC). Constructed value, on the other hand was used as normal value in those cases where the sales volume in the home market or third country market was insufficient or where there were insufficient sales above costs of production and therefore normal value could not be based upon price. Constructed value measured the costs of the product sold in the relevant export market (in this case the United States). The Norwegian respondents had stated that, while about half of the farms in Norway had been approved to have their production ultimately shipped into the United States, only a very small portion of the production of an approved farm was actually shipped to the United States. If only 11 per cent of total salmon production went to the United States and if only 45 per cent of the total farms in Norway were licensed to export to the United States, the actual number of farms which did export to the United States was probably something much less than the 360 farms who were licensed. This meant that the vast bulk of all farm production, including those
[Page 37]
farms whose product had been approved for ultimate shipment to the United States, was actually shipped to non-US markets, mainly in the EEC.
126. The United States pointed out that, as a result, virtually any sample of farms would result in a cost of production based in large part on the costs of producing product bound not for the United States but for the EEC, since even those farms which had been approved to have their product shipped to the United States only sent a small portion of their production to the United States. Since only half the farms had been approved to have their product shipped into the United States, there was a significant chance that any sample drawn from a universe of farms which had both received and had not received this approval would result in a sample that would not include any farm whose product was sold by the exporter in question to the United States. It was important to point out, however, that the costs of producing salmon sold in the EEC were exactly the same as the costs of producing salmon sold in the United States. This had raised the prospect that such a sample would be challenged if the costs of production developed from it were used as normal value (i.e. as the costs of production of product shipped to the United States), since none of the product on which the costs of production were based would actually have been sold in the United States during the period of investigation. The Department of Commerce had quite reasonably concluded that it could avoid this problem by limiting the universe of farms for each exporter to those farms from which the exporter had purchased salmon during the period of investigation for shipment to the United States. Since it had been established on the record that the vast bulk of the salmon produced by farms which had received approval for shipment to the United States was actually not sold to the United States but to the EEC, limiting the universe of farms for each sample in this manner would ensure that the costs of production calculated from the selected farms included the costs of producing salmon both for the EEC market and the United States, and thereby enable it to be used as the constructed value both as benchmark against which to measure prices to the EEC and as the normal value benchmark against which prices to the United States could be measured.
127. In response to the argument of the United States that on several occasions during the period May-August 1990 the Norwegian exporters had been allowed by the Department of Commerce to correct their initial lists of farms with which they had dealt during the period of investigation, Norway pointed out that in a Memorandum of 25 July 1990 the Department of Commerce had decided to look at farmers' costs, rather than the exporters' costs.92 Moreover, the Department of Commerce had informed the exporters that Section A of the questionnaire was intended only to
"'familiarize' the Department with their [i.e. the respondents'] corporate structure and accounting practices; to define the merchandise sold in the United States and identify any identical or similar merchandise sold in other markets; and to establish the size of each company's various markets to determine the best measure of foreign market value."93
Thus, the three exporters who had not been able to provide entirely accurate lists of farms had had no reason to believe that the information requested on the farms with which they had dealt during the period of investigation was essential, or particularly relevant, to the investigation. Norway also considered that the fact that several exporters had attempted to provide corrected information after the submission of their initial responses only underlined that the exporters had tried to fully co-operate in the investigation and to respond to an onerous questionnaire. It did not in any way mitigate the fact that the Department knew that these were potential problems with this portion of the responses. If anything, the corrections should have highlighted the potential problems.
92 Memorandum to Frank J. Sailer from Roland L. MacDonald, 25 July 1990. ↩
93 Department of Commerce questionnaire, Section A, "General Information." ↩
[Page 38]
128. In response to Norway's reference to the Memorandum of 25 July 1990, the United States presented three arguments. First, this Memorandum had appeared well over two months after the Section A responses had been first submitted on 16 May 1990. Both Chr. Bjelland and Sea Star had provided amended lists within this time. Skaarfish surely could also have done so. If Norway's argument was that Skaarfish was in the process of working on a correction and decided not to pursue the matter after 25 July, this meant that Skaarfish had recognized that it had an ongoing obligation to provide correct information. If not, Norway's discussion of the Memorandum was irrelevant. Since both Chr. Bjelland and Sea Star had provided corrected responses in mid-June, there was no basis for Norway's statement that these two exporters had shared Skaarfish's alleged reaction to the Memorandum. Second, Norway's interpretation of the Memorandum was erroneous. The Memorandum had stated that, based on a review of the first sample of farms, the Department had determined to continue with the exporters as the proper respondents. The first sample had not been based on the Section A information requested from the exporters but on a completely different request issued to the fish farmers' sales organization on 12 June 1990. The issue had nothing to do with the question of reviewing farms' data for a costs of production investigation, under which consideration of the Section A responses would be relevant. The Memorandum had specifically noted the possibility that costs of production investigation of the farms might be undertaken. Third, the Memorandum had stated that the Department would not continue to pursue information from the eight fish farms in the first sample. There was no basis for inferring from this statement that the exporters were somehow excused from responding to a completely separate request for information for a completely separate purpose. Rather, the Memorandum had shown an intensified concern by the Department with the question of the farms' costs.
129. Norway considered that, in refusing to give the Norwegian respondents enough time to present evidence in writing and in relying on the best information available, the United States had not only violated Article 6:1 but also Article 8:3 of the Agreement by imposing an anti-dumping duty far in excess of the margin of dumping, if any.
130. Norway submitted that in the investigation of Atlantic salmon from Norway the United States had also failed to use equitable and open procedures and to provide ample opportunities to present evidence as a result of unreasonable and onerous questionnaire and verification procedures. In addition to the demanding nature of Section A of the questionnaire issued to the exporters, completing the Sections B and D of the questionnaires had been similarly onerous. To respond to these questionnaires required that the salmon farmers have access to and use computers on a regular basis. This alone was a difficult undertaking for such small salmon farmers. Moreover, Section D required responses on floppy disks, compatible with Lotus 1-2-3, which the majority of the salmon farmers did not have, nor were at all familiar with. The procedures used at verification had been equally burdensome. For example, the Department of Commerce had required the farmers/exporters to make available equipment such as photocopiers. This requirement had created considerable hardships since the Norwegian farms frequently were one-to-two person operations in isolated locations. The information obtained by the Department of Commerce from the questionnaire responses and the verification had served as the basis for the Department's determination. As a result of the averaging technique used by the Department in its costs of production calculation, all exporters who had correctly and completely responded to the questionnaires had been adversely affected by the use of the best information available in the case of the responses of the farmers which the Department had found to be insufficient. These exporters had had no control over the responses of the farmers and had had no opportunity to rebut this evidence. They therefore had not been allowed an opportunity to present evidence, as required by Article 6:1 of the Agreement.
[Page 39]
131. Norway considered that, in granting more favourable procedural treatment to domestic respondents than to Norwegian respondents, the United States had acted in violation of the requirement of Article III:4 of the General Agreement. A recent Panel had applied the national treatment rule of Article III:4 to the procedures of Section 337 of the United States Tariff Act of 1930, as amended, with respect to "unfair methods of competition and unfair acts in the importation of articles."94 In that case, the denial of national treatment had resulted from the fact that different procedures applied under Section 337 of the Tariff Act to producers of foreign products than the procedures applied to domestic producers under a different statute. By contrast, in the matter before the Panel, the United States had provided more favourable procedural treatment to domestic producers than to foreign producers under the same law.
132. In support of its view that Article III:4 applied to the anti-dumping proceeding before the Panel, Norway argued that the anti-dumping law of the United States affected the price at which imports were sold in the domestic market of the United States and that the retrospective duty assessment procedures used by the United States created uncertainty regarding the ultimate costs of the imported products. Both these aspects of the US anti-dumping law could adversely modify the conditions of competition between domestic and imported products. In the view of Norway, the investigation of salmon by the United States concerned salmon already imported into the United States. The US system of retroactive determination of the ultimate liability for payment of anti-dumping duties ensured that a product would always have been imported before the United States began an investigation or a review under its legislation. The fact that the anti-dumping law was nominally enforced at the border did not prevent Article III:4 of the General Agreement from applying.95 Article VI of the General Agreement and the provisions of the Agreements interpreting Article VI, specifically permitted the application of anti-dumping legislation only to foreign products, thus permitting legislation which would otherwise violate the requirements of Article III:4 of the General Agreement. However, matters not specifically dealt with in Article VI of the General Agreement or in the provisions of the Agreement were subject to the national treatment rule.
133. Norway pointed to the following procedural aspects of the anti-dumping proceedings on imports of Atlantic salmon from Norway which amounted to a violation of Article III:4 of the General Agreement. The United States had imposed more stringent requirements and procedures on the Norwegian respondents than on the domestic respondents with respect to the submission of responses to questionnaires and the procedures used to verify the information provided by the respondents. Thus, the United States had requested the Norwegian respondents to submit responses on computer tape or floppy disk while no such requirement had been imposed on domestic producers. When the Norwegian exporters or farmers had failed to fully respond to the questionnaires, the Department of Commerce had based its determination on the best information available. However, in its injury investigation, the USITC had not made adverse inferences against domestic producers who had failed to fully respond to the questionnaire.96 The United States had violated the national treatment requirement in its verification procedures in that, while an extensive examination had been conducted of the financial records of each Norwegian respondent who had been required to make available such equipment as photocopiers, a similarly burdensome verification procedure had not been applied to the domestic producers.
94 "United States - Section 337 of the Tariff Act of 1930", BISD 36S/345. ↩
96 USITC Determination, p.A-19, note 49. ↩
[Page 40]
134. The United States denied that, in the investigations conducted by the Department of Commerce and the USITC, national treatment had been denied to the Norwegian respondents.97 Regarding the question of the medium in which the domestic producers had been required to provide information to the USITC, the United States pointed out that, although the USITC had required extensive information from the domestic producers, the USITC had not allowed them to file their information on computer tape or floppy disk format, regardless of whether that would have been more convenient for the domestic producers. Rather, members of the domestic industry had been required to submit information on paper, the form most convenient for the USITC. Different types of information were necessary to analyse the existence and extent of dumping versus the information necessary to assess the condition of a domestic industry and the volume and price effects of imports and their impact on the domestic industry. To the extent that one group of farms or persons providing data might be treated differently, it was because that group provided a different type of data warranting such differential treatment.
135. The United States explained in this context that the calculation of a margin of dumping required information in the possession of foreign producers and/or exporters of the product under investigation. Such information consisted e.g. of pricing information and data necessary to properly adjust that information and the costs of producing the product. No other firm or persons besides the producers and/or exporters could provide that information. The analysis of margins of dumping also typically involved a firm-by-firm enquiry. The form in which the information was submitted was the most readily usable form for the Department of Commerce to conduct its analysis of the data. By contrast, an investigation into the condition of a domestic industry producing goods like those being dumped required information of a substantially broader category and from a wider range of entities: data not only on pricing and (where appropriate) production costs and profits but also on such factors as capacity, capacity utilization, shipments, employment, research and development, and capital investment, of "domestic producers as a whole" of the like product, not just the petitioners. Furthermore, consideration of the volume and price effects of the dumped imports, and their impact on the domestic industry required that data be provided by importers concerning such matters as import volume, prices, shipments and inventory. Consideration of the existence of a threat of material injury required data from foreign producers of the product under investigation concerning such factors as foreign production, capacity and shipments to third-country markets.
136. The United States further pointed out that the USITC also sought information from purchasers of both the domestic like product and the dumped imports to determine the reasons why purchases were made and the extent to which imports and the domestic like product competed for the same purchasers on the basis of price. The USITC required the submission of information from all responding entities in the form most readily usable in the compilation and analysis of that information. Thus, while Norway would have the Panel believe that the only entities which had submitted information to the USITC were the domestic producers and the Norwegian exporters, such was not the case. Also, the complaint that "US Producers" were treated differently than foreign exporters under the practise of the United States was invalid because the differences were based on the different kinds of data necessary for the injury and dumping determinations, the differing categories of information provided, and the different forms in which such information must be submitted to be analyzed by the various investigating authorities.
137. On the issue of verification of responses submitted in the investigation of the USITC, the United States noted that the USITC had sought questionnaire responses from a number of sources. These sources had included importers of Atlantic salmon from Norway, purchasers of Atlantic salmon from all sources, US producers, and the Norwegian respondents. The Commission had supplemented
97 See Section IV of this Report for the views of the United States on the admissibility of Norway's claim. ↩
[Page 41]
the questionnaire responses, when necessary with telephone contacts with the submitters of the questionnaire information. The Commission had not considered it to be necessary or appropriate to verify responses on-site for any category of questionnaire respondents.
138. On the issue of the consequences of a failure of respondents to provide full information to the USITC, the United States pointed out that the US industry had been required to respond to questionnaires in both the preliminary and final investigations. In the final injury investigation the questionnaire responses had provided nearly 95 per cent coverage of the US Atlantic salmon industry.98 All of the US producers had been required to produce voluminous information. For example, Ocean Products, the large Marine company which had been forced to liquidate itself during the investigation had filed a response of 468 pages. This near perfect participation by US producers stood in stark contrast to the participation of the importers of Norwegian salmon. Fewer than half of the importers, representing just over half of all Atlantic salmon imports from Norway, had even returned their questionnaires.99 Despite this poor response rate, the USITC had not made adverse inferences against the Norwegian importing interests.
3.1.4 Treatment of third-country sales below costs of production as not being "in the ordinary course of trade"
139. Norway, referring to the finding of the Department of Commerce that for a number of exporters sales to EEC markets were made at prices below costs of production and therefore could not be used as a basis to establish the normal value for purposes of the final determination, argued that the United States had acted inconsistently with Article 2:4 of the Agreement and had denied Norway the equitable and open procedures prescribed by the Agreement by assuming arbitrarily that sales of Norwegian salmon to the EEC were not "in the ordinary course of trade", without examining what was the "ordinary course of trade" in the salmon industry. The EEC constituted the world's largest market for farmed salmon and Norway was the largest supplier to this market. Norwegian exports of fresh salmon to the EEC accounted for more than half of the Norwegian exports. It was therefore difficult to understand how the United States could assert that the exports of salmon from Norway to the EEC were not "in the ordinary course of trade" without an investigation of what was "the ordinary course of trade" in the salmon markets.
140. The United States argued that the Department of Commerce had considered using export prices of Norwegian salmon to the EEC for purposes of establishing the normal value but had found that these prices overwhelmingly were below the costs of production of salmon. For seven exporters, the Department had found that over 90 per cent of the sales to the EEC were below costs of production. The Agreement permitted, and US law required, in such cases that a constructed normal value be used. For the eighth exporter, the Department had found that 70 per cent of sales to the EEC markets were below costs of production; the normal value for this exporter had been based on the remaining "above cost" export sales prices. Article 2:4 of the Agreement did not establish a preference for the use of export prices to a third country over the use of a constructed normal value. The text of this provision also made it clear that the use of a constructed normal value in lieu of export prices to a third country was not contingent on a finding that the sales to the third country were not "in the ordinary course of trade". Because the Agreement established export prices to third countries and constructed value as equally acceptable alternatives, the Department had not been required to consider third country sales before looking to constructed value. Norway's argument that the United States had acted in an arbitrary manner by failing to examine what was "the ordinary course of trade" in the salmon industry therefore failed to allege a violation of the Agreement.
98 USITC Determination, p.A-21. ↩
99 USITC Determination, p.A-20. ↩
[Page 42]
141. Norway pointed out that it was not disputing that Article 2:4 of the Agreement permitted the use of either export prices to a third country or constructed value for purposes of establishing the normal value. However, if an importing country had a stated preference for the use of third country sales in the ordinary course of trade, the Agreement did not allow the country to arbitrarily determine that sales to a third country were not in the ordinary course of trade without examining what was the ordinary course of trade for the industry in question. Given the significance of the volume of exports of salmon from Norway to the EEC, the assumption that certain sales were not in the ordinary course of trade solely on the basis that they were at prices below costs of production was unreasonable and inequitable.
142. Norway also argued that in this context the United States had failed to take account of the fact that salmon was a perishable product. The United States applied in its anti-dumping practice a special rule with respect to the conditions under which sales of perishable products at prices below costs of production were considered not to be in the ordinary course of trade. In the investigation of Atlantic salmon from Norway, the USITC had found salmon to be a perishable product.100 However, the Department of Commerce had declined to consider salmon a perishable product, despite the fact that, at verification, it had found that once an exporter purchased salmon, the salmon had to be sold within fourteen days or it began to lose its commercial value.101 This inconsistency in the treatment of salmon by the USITC and the Department of Commerce had prejudiced Norway twice: first, when the Department of Commerce had rejected the use of export prices to the EEC markets in favour of constructed value, in part because it had found that salmon was not perishable; second, when the USITC had based its affirmative injury determination in part on the fact that salmon was a perishable product. These inconsistent actions had denied a fair and equitable treatment to Norway.
143. The United States considered that Norway was apparently arguing that, because there was a preference under United States' law for the use of export prices to a third country as a basis for the establishment of the normal value, the Agreement required the United States to follow this preference. However, under United States law, when, as in this case, sufficient third country sales were made at prices below cost, the Department of Commerce was required to base its determination of the normal value on the costs of production. Thus, the Department had followed the United States law in this investigation. In any event, if the Department had failed to act consistently with United States law, the proper remedy was in United States courts, not in dispute settlement proceedings under the Agreement. The United States further argued that the Department of Commerce had not overlooked the issue of whether Atlantic salmon was a perishable product but had determined, based on the record evidence, that Atlantic salmon was not perishable. As explained in the public notice of the final determination of dumping, the Department had found that Norwegian Atlantic salmon farmers had the ability to control the time of sale of their output by 'holding over' inventory and, since January 1990, by freezing fresh salmon. The Department had, in addition, found at verification that Atlantic salmon was not perishable in the hands of the exporters because exporters co-ordinated their salmon requirements in weekly telephone converences with their customers, with farmers and with other exporters. By doing so, the exporters could communicate their salmon requirements two weeks into the future to the farmers so that farmers could begin to prepare the salmon harvest two weeks prior to harvest. The United States also pointed out in this regard that the Department of Commerce had considered whether live, unharvested fish were perishable commodities for purposes of purchase by exporters from salmon farms prior to export. The USITC, on the other hand, had considered whether dead fish were perishable after importation into the United States and had concluded that they were. Thus, the two agencies had defined "perishability" differently, for different purposes.
100 USITC Determination, p.A-3. ↩
101 Verification Report for Halvard Leroy A/S, 10 December 1990, at II.(B)(1). ↩
[Page 43]
3.1.5 Calculation of costs of production on the basis of the costs of production of the salmon farmers, rather than on the basis of the acquisition price paid by the salmon exporters
144. Norway considered as inconsistent with the requirement of the Agreement that equitable and open procedures be used in anti-dumping duty investigations the fact that, for purposes of determining whether export sales to EEC markets were made at prices below costs of production, the Department of Commerce had investigated the costs of production of the salmon farmers, rather than the acquisition costs of the salmon exporters. Exporters, who negotiated the sales prices in different export markets, did not know the costs of production of the farms from which they purchased salmon, given that they purchased salmon from a number of farms ranging from 18 to approximately 100. The farmers, on the other hand, did not know the ultimate destination of the salmon they sold, and thus did not set the export price in the United States markets. In using the costs of production of unrelated suppliers, the exporters had been penalized.
145. Norway noted in this context that in determining the cost of production of the farmers, the Department of Commerce had used the acquisition price of smolt, if that price was an arms-length price, instead of the costs of production of smolt. Thus, the Department apparently believed that the acquisition price between unrelated parties was the correct measure of the buyer's costs. However, where the use of acquisition prices might have resulted in a lower margins of dumping, as would have been the case if the Department had used the exporters' acquisition prices, the Department had ignored the acquisition price and had relied on an estimate of the costs of producing salmon. The use of the farmers' costs of producing salmon rather than the acquisition price paid by the exporters had made the investigation more expensive and complex and had given rise to problems of verification which would have been avoided if the arms-length prices charged the exporters by the independent farmers had been used.
146. The Panel asked Norway to indicate how the text of Article 2:4 of the Agreement in its view required that the Department of Commerce should have relied on the exporters' acquisition prices for purposes of determining the costs of producing salmon and to clarify how its argument on this issue related to the views expressed in the investigation by the Norwegian exporters, who, according to the Federal Register Notice of the final determination of dumping, had argued that "acquisition prices are not relevant to the COP analysis".102 In response, Norway argued that, in an international dispute settlement proceeding, the Government of Norway was not obligated to present the same arguments as private Norwegian respondents in the administrative proceedings before the investigating authorities in the United States. Furthermore, the Federal Register notice had misrepresented the argument of the Norwegian respondents. The Norwegian exporters had taken the view that the costs of production should be determined on the basis of either (a) the exporters' acquisition price plus processing costs, general and administrative expenses incurred by the exporters and an amount for profit, or (b) the farmers' costs of producing salmon, plus the exporters' expenses and profits. The petitioner had argued that the Department should use the higher of these two alternatives. It was in this context that the exporters had argued that the use of acquisition prices was inappropriate; the costs of producing salmon had to be determined either on the basis of the exporters' acquisition prices or on the basis of the farmers' costs of producing salmon, but could not be selected by choosing whichever method produced the larger margins. In a brief submitted in January 1991 the exporters had reiterated that acquisition prices should be used. Thus, the exporters had stated that:
102 56 Fed.Reg., 25 February 1991, p.7672. ↩
[Page 44]
"Traditionally, the Department has used an exporter's cost of acquisition from an unrelated supplier as a measure of its production cost. Respondents strongly urged, and continue to urge, the Department to maintain this methodology."103
147. The United States argued that the Department of Commerce had properly based its costs of production calculation on the costs of producing salmon in Norway, not on the exporters' costs of acquiring already-produced salmon. Under the Agreement, this was the only way to determine production costs. Article 2:4 explicitly required that a constructed normal value be based on "the cost of production in the country of origin". There was no basis in the Agreement for Norway's view that the Department of Commerce should have relied on the acquisition costs of Norwegian exporters, rather than on the producers' costs of production, in preparing its costs of production calculation. The exporters produced nothing and, therefore, had no production costs. Indeed, with few exceptions, under Norwegian law the exporters were prohibited from producing salmon. Under the Agreement, the Department had been required to base its determination on the farmers' production costs.
148. The United States considered that the fact that exporters might not have knowledge of the farmers' production costs was irrelevant under the Agreement because it was the production costs which must form the basis for the establishment of the normal value. Similarly, the farmers' alleged lack of knowledge concerning the destination of their product was irrelevant to their production costs.
149. The United States considered that the use of the farmers' acquisition costs of smolt for purposes of calculating the costs of producing salmon was proper. The Department of Commerce had sought to determine the costs of producing salmon, not smolt. Therefore, the costs of producing salmon would reflect the price paid by the salmon farmer for smolt; the production costs of the smolt by an unrelated smolt farmer was not relevant to the salmon farmer's costs of production. The Department's choice in this regard was consistent with the requirement under the Agreement to consider the production costs of the exported merchandise and was consistent as well with the calculation of constructed value.
150. The Panel asked the United States to further explain its view that exporters' acquisition prices were irrelevant to the costs of production of salmon and to indicate whether the Department of Commerce had considered the possible use of exporters' acquisition prices for purposes of its costs of production analysis. The Panel noted in this context a statement in a Department of Commerce Memorandum of 20 August 1990 suggesting that the exporters' acquisition prices might "become the starting point in the cost analysis overall in this investigation".104
151. In response, the United States emphasized the unusual structure of the Norwegian salmon industry in which farmers generally were prohibited from exporting themselves a product they grew and harvested for the export market. The relatively few companies licensed by the farmers' co-operative to purchase from the farmers and export the product were generally prohibited from farming. The question of whose costs - the exporters' or the farmers' - should be considered in determining whether export sales were at below cost prices was, under this industry structure, a novel one, meaning that the Department's sampling and related methodology from other investigations of sales below cost cases had not been particularly helpful or relevant. When the Department had initiated its sales below cost investigation on 20 August 1990, it had initially planned to "test" the exporters' acquisition prices against the farmers' costs of production, and to use the higher of the two as the basis for the costs of production. As the investigation had progressed, and the Department analyzed and verified the cost-related information provided by both the farmers and the exporters, the Department had ultimately concluded, in December 1990, that the exporters' acquisition prices were irrelevant for costs of production purposes,
103 Case Brief of the Norwegian Respondents, 14 January 1991, p.2. ↩
104 Memorandum from David L. Binder to Richard Moreland, 20 August 1990, p.3. ↩
[Page 45]
and that the costs of producing the product should be based on the farmers' costs of production, as explained in a Memorandum dated 18 December 1990 from the Director of the Office of Policy. The December 1990 Memorandum explained that the exporters' acquisition costs would be relevant only if it had been alleged that the exporter was selling its product at below the acquisition price. However, since in this case it had been alleged that the prices to be used to determine normal value (the export prices to a third country) were below the farmers' costs of production, the farmers' costs were relevant because the farmers were the producers of the fish. The acquisition prices were just that, prices, and not costs of production.
152. The United States noted in this context that, in a subsequent case, Fresh Kiwifruit from New Zealand, the Department of Commerce had requested costs of production information from the kiwifruit growers. In the kiwifruit case, as in the salmon case, the growers were the actual producers of the product. The growers sold the kiwifruit to the New Zealand Marketing Board (the one exporter in the country). The costs of production had not been based upon the price paid by the Board to the growers but on the actual costs of production of the growers plus the selling, general and administrative expenses incurred by the Marketing Board.
3.1.6 Inclusion in the constructed normal values of a "freezing charge"
153. Norway argued that a fair and equitable treatment had been denied to Norwegian exporters as a result of the treatment by the Department of Commerce of a "freezing charge". This charge, 5 NOK per kilogramme, was collected from the exporters by the FOS in order to finance a freezing programme introduced on a voluntary basis in January 1990 by the Norwegian industry with a view to stabilize world market prices for fresh salmon. Given that this charge was paid by the exporters to the FOS for purposes of financing the freezing programme, it was completely unrelated to the costs of production of fresh salmon. The Department of Commerce had nevertheless included this charge in its calculation of the costs of production of fresh salmon even though the effect of the freezing programme had been to raise prices of fresh salmon in world markets. The freezing charge was not a cost of producing the exported fresh salmon. By definition the charge was collected to cover the costs of freezing salmon, not of producing fresh salmon. Consequently, the freezing fee was not a proper cost to be included among the "reasonable amount" for costs under Article 2:4 of the Agreement. Moreover, even if it were a cost of production, the effect of the fee was to reduce the supply of fresh salmon and increase its price.
154. The United States noted that Norway appeared to argue that, because the proceeds from a charge on fresh salmon were used to pay for a programme concerning another product, frozen salmon, the charge was somehow not a cost of producing fresh salmon. However, regardless of the purpose for which the charge was used, there was no question that it was a charge levied on all sales of fresh salmon for the explicit purpose of dealing with overproduction of fresh salmon. This fee was an expense incurred in the production of fresh salmon and it had therefore been proper for the Department of Commerce to include it, like other production expenses, in its determination of the cost of production.
155. The Panel asked Norway to explain how in its view the treatment by the Department of Commerce of the freezing charge as an element of the costs of production of fresh salmon was inconsistent with Article 2:4 of the Agreement. In response, Norway pointed out that the freezing programme introduced in January 1990 was voluntarily financed by members of the FOS. In the prevailing market situation, the effect of this programme had been that Norwegian exporters had to reduce their profits by 5 NOK per kilogramme. The fee had not been imposed on the producers of salmon. The charge was therefore an extraordinary cost which affected the profits realized by the exporters but not the costs of production. Inventory costs were deducted from the normal rate of profit. The freezing charge was part of inventory costs. Therefore, the decision of the Department of Commerce to treat this charge as an
[Page 46]
element of the cost of production was inconsistent with Article 2:4. An adjustment to account for the freezing charge should have been made in the calculation of "the reasonable amount for profits".
156. Norway also pointed out that the FOS had an intermediary rôle in the trade between farmers and exporters. The exporters bought directly from the farmers. These sales were recorded by the FOS, which invoiced the exporters and paid the farmers. With respect to the financing of the freezing programme, Norway explained that when the FOS invoiced the exporters, it added the freezing charge of 5 NOK per kilogramme. Thus, the freezing charge was paid to the FOS by the exporters. The individual salmon producer neither received nor had any claims to this money.
157. The Panel asked Norway to clarify whether it considered that inventory costs were not part of a constructed normal value. In response, Norway explained that inventory costs would be included within the constructed value as part of the "reasonable amount for administrative, selling and any other costs" provided for in Article 2:4. However, such costs would reduce the "reasonable amount" for profit. Under the approach taken by the Department of Commerce, the freezing charge had been double-counted in the constructed value calculation. First, the Department had included the charge as an element of the costs of production. Second, the Department had added an arbitrary amount for profit of 8 per cent over the cost of production. Thus, the actual amount of profit had been increased (by 8 per cent of the inventory cost) whereas the amount added for profit should have been decreased.
158. The United States argued that Norway was factually incorrect in claiming that the FOS freezing charge had been paid by exporters rather than by the farmers. At verification, the Department of Commerce had determined that the farmers - the actual producers of salmon - paid the freezing fee. This information had been provided by the exporters themselves and by the FOS. The United States provided to the Panel copies of the relevant sections of the verification reports and a copy of a report on a meeting held with FOS representatives on 9 November 1990. There was no question, therefore, that the freezing charge was a cost of producing salmon levied on the farms and had been properly treated as such by the Department of Commerce.
3.2 Calculation of constructed normal values
159. Norway argued that the affirmative final determination of dumping in the investigation of imports of Atlantic salmon from Norway was inconsistent with Articles 2:4 and 8:3 of the Agreement as a result of the methodology used by the Department of Commerce in the calculation of constructed normal values. In support of this claim, Norway made five submissions, pertaining to: (1) the use of statistically invalid techniques in the construction of a sample for purposes of determining the costs of production of Norwegian salmon farms; (2) the use of a simple, rather than a weighted, average of the costs of production data obtained on the basis of this farm sample; (3) the use of the best information available as a basis for the calculation of the costs of production of one of the Norwegian farms; (4) the inclusion in the constructed values of an unreasonable amount for profit, and (5) the use of the FOS processing fee as the best information available in the calculation of exporters' processing costs.
3.2.1 Sampling techniques
160. Norway argued that the sample of farms used by the Department of Commerce for purposes of determining the costs of production of salmon was not in conformity with generally recognised sampling techniques and had resulted in an overstated costs of production figure, contrary to Article 2:4 of the Agreement. The costs of production figure calculated by the Department of Commerce had been much higher than the figure calculated by the EEC in its anti-dumping investigation and the figure calculated by the Norwegian Directorate for Fisheries in an annual survey of costs of production in
[Page 47]
the Norwegian salmon industry.105 Norway pointed in particular to the fact that, as a result of the limited number of farms selected and the failure of the Department of Commerce to stratify the sample by farm size, no account had been taken of differences in costs of production between the farms who supplied the exporters. The use of a single cost of production figure for any exporter was a biased estimator which led to an overestimate of the cost of production of the salmon sold by the exporter unless smaller farms had lower costs of production than larger farms. In fact, the largest farms in Norway had by far the lowest cost of production. The data obtained by the Department of Commerce for the seven sampled farms indicated that for the largest farm, costs of production were 26.24 NOK per kilogramme, compared to a weighted average of the figures for the seven farms of 34.30 NOK per kilogramme and a simple average of 39.39 per kilogramme. Given this negative correlation between cost of production and farm size, the decision of the Department of Commerce not to stratify the sample by farm size was inconsistent with proper statistical procedures.
161. Norway also argued that the farm sample used in the cost of production investigation was inconsistent with proper statistical techniques in that two exporters had been lost to the study because the farms selected from the lists of farms provided by these exporters had not supplied any salmon to the exporters during the period of investigation. In particular in view of the small size of the sample, the Department should have selected additional farms for these exporters. The Norwegian respondents had suggested to the Department that it use the sample of 42 farms used by the EEC in its anti-dumping duty investigation of imports of Atlantic salmon from Norway.106 Norway also pointed out that farms which were located in a region accounting for less than 25 per cent of the number of farms supplying salmon to an individual exporter had no chance of being selected for the sample used by the Department of Commerce in its costs of production investigation.
162. Norway noted that the United States had acknowledged that the farm sample used for purposes of the cost of production investigation in the salmon case had been flawed. Thus, in a more recent case concerning imports of Kiwifruit From New Zealand, the Department of Commerce had taken a sample stratified by size and geographical location. In explaining its sampling methodology in that case, the Department had referred to the deficiencies in the sampling techniques used in the salmon case:
"...considering our recent experience in the case of Salmon from Norway, it would be prudent to ensure that we have a valid universe from which to select."107
163. Norway, responding to a question by the Panel on which factors it believed should have been taken into account in the selection of farms, explained that, since the purpose of the cost of production investigation was to determine the average cost of production per kilo salmon and not per farm, the United States should have taken into account the size of the Norwegian farms, as measured by their production volume when allocating the farms between regions. The Department of Commerce had the information necessary to stratify the farms by size.108 Even the petitioner had objected to the use of only seven farms and to the failure to account for differences in farm size.109
164. The Panel requested Norway to explain whether its statement that the use of the costs of production of a single farm for any exporter was a biased estimate reflected a criticism of the number of farms in the sample or whether this point pertained to Norway's argument that a weighted average
105 Annual Survey (1989) by the Norwegian Directorate of Fisheries. ↩
106 Letter from David Palmeter to Richard Moreland, 22 August 1990. ↩
107 Memorandum from Richard W. Moreland to Francis J. Sailer, 18 July 1991, p.8. ↩
108 Memorandum from C. Olivieri and T. Oakes to the file, 17 August 1990. ↩
109 Petitioner's case brief of 14 January 1991, cited in respondent's rebuttal brief at 27. ↩
[Page 48]
of the costs of production of the farms in the sample should have been used.110 In response, Norway noted that smaller farms were over-represented in number compared to their contribution to total production volume, thus creating a high probability that they would be over-represented in the sample. This would lead to an overestimate of the costs of production, unless smaller farms had lower costs of production than larger farms. However, this point was also relevant to the question of the use of a weighted average of the costs of production of the sampled farms. By weighting the average to reflect more closely the actual distribution of production supplied to exporters, the Department of Commerce could have adjusted for some of the bias inherent in the use of such a small sample.
165. The United States argued that, in its analysis of the cost of production of the Norwegian salmon farms, the Department of Commerce had applied proper sampling techniques in light of the information before it, as authorized by Article 6:8 of the Agreement. In August 1990, after the Department had decided to examine whether export sales to EEC markets were made at prices below cost of production, the Department had decided to develop a sample of farms for each of the eight exporters under investigation, each sample to be drawn from a universe of farms with which the exporter had actually dealt during the period of investigation. In order to ensure that the information to be provided by the farms could be verified in time, the number of farms selected for each exporter had been limited to, at most two. The farms selected would be drawn from the lists of farms provided by the exporters in their responses to Section A of the questionnaire issued on 30 April 1990. As explained in a Memorandum dated 4 September 1990, (a copy of which was provided to the Panel) the farms on each list had been divided into two regional categories: the northern and the southern regions in Norway. This had been done because evidence on the record showed that production costs were typically much higher in Norway's northern region than in its southern region. The Memorandum also explained that one farm from each region was selected for an exporter only if each region included more than 20 per cent of the farms on the list; otherwise, only one farm was selected for an exporter and that selection was made from the region which included 80 per cent or more of the farms on the list. This procedure was used in order to ensure that the sole farm selected for an exporter did not come from a region where the small minority of farms with which the exporter dealt were located. Under these procedures, three exporters had two farms selected, and five exporters had one farm selected, for a total of eleven farms. On 21 August, questionnaires had been issued to these eleven farms, with an instruction to respond by late September 1990.
166. The United States pointed out that on 30 August 1990 the Department of Commerce had been informed for the first time that the lists of farms provided by several exporters in response to the Section A questionnaires contained erroneous information and that several of the farms selected by the Department for purposes of its cost of production analysis actually had not sold any salmon during the period of investigation to the exporters to which they were linked. Based on this new information, the Department had concluded that the exporters' lists were flawed because they contained the names of farms from which the exporters had not purchased any salmon during the period of investigation. The Department had determined that it could not develop new samples for each exporter because there was no way to determine from the flawed lists which farms actually had sold to the individual exporters during the period of investigation and because there was not sufficient time left in the investigation to develop a new sample, present questionnaires to new farms, and analyse and verify the responses. Instead, the Department had decided to proceed to collect cost of production information from the remaining farms selected for the survey, i.e., the farms which actually had supplied salmon to the exporters during the period of investigation and to develop an average cost of production from the remaining farms.111 Because of the erroneous information provided by the exporters, the Department
111 Memorandum from Richard W. Moreland to Francis J. Sailer, 13 September 1990. ↩
[Page 49]
had been placed in an untenable position and had had no reasonable option other than to rely solely on data from the remaining seven farms.
167. In response to Norway's argument that the Department of Commerce should have selected additional farms after having learnt that several of the farms on the lists provided by the exporters had not sold salmon to these exporters, the United States argued that it had been too late in the investigation to add farms. The list from which the sample had been drawn and from which any additions to the sample would have to have been drawn was flawed. The Department could not have relied on such a flawed document and could not have relied on the exporters to provide correct information in a timely manner; after all, it had taken the exporters fourteen weeks to report the first set of errors in their responses. The exporters' own errors had forced the Department of Commerce to rely on the seven farms and had precluded the Department from augmenting the sample.
168. The United States considered that the reliance by the Department of Commerce on the information before it was authorized under Article 6:8 of the Agreement which provided that:
"In cases in which any interested party refuses access to, or otherwise does not provide, necessary information within a reasonable period of time or significantly impedes the investigation, preliminary and final findings, affirmative or negative, may be made on the basis of the facts available."
In the case under consideration, the Norwegian exporters had not provided necessary information within a reasonable period of time, and the Department had properly relied on the facts available in constructing its sample.
169. The United States further argued that the survey of farms as originally chosen by the Department of Commerce was statistically valid. The farms had been chosen using a random number generator process, which was a widely used sampling technique. The sample as originally structured had included farms of different sizes and located in both the southern and the northern parts of Norway, thus taking account of all the factors which Norway had indicated would account for cost differences between farms.
170. In response to a question by the Panel, the United States emphasized that the original intention of the Department of Commerce had not been to construct a single sample of farms for the eight exporters under investigation but a set of samples, one for each of the eight exporters, so that the cost of production developed for each exporter could be developed from information on farms with which the exporter had actually dealt during the period of investigation. However, after the Department had been informed of the errors in the lists of farms submitted by the exporters, it had abandoned this original plan and had decided to treat the seven remaining farms who had actually sold salmon during the period of investigation as a single sample representative of all salmon farms in Norway. The decision to proceed with an average of the cost of production data of these seven farms was not only consistent with Article 6:8 of the Agreement but was also supported by evidence provided by the Government of Norway indicating that the overwhelming number of farms in Norway ranged from 8-12,000 cubic meters. Except for two of the seven farms, the remaining five farms included in the sample were of sizes ranging from 8-12,000 cubic meters. The remaining farms therefore could reasonably be considered to be representative of the industry.
171. The Panel asked the United States to explain whether any other facts had been considered by the Department of Commerce as a possible basis for the calculation of the costs of production of the Norwegian salmon farms. The United States responded that there were two possible sets of other data which the Department might have used, but had correctly chosen not to use, in its final determination. First, the Norwegian respondents had asked that the Department use cost of production information which had been submitted in the EEC's anti-dumping duty investigation of imports of Atlantic
[Page 50]
salmon from Norway. As explained in a letter dated 11 September 1990 to the Norwegian exporters, the Department had declined to use this information for a number of reasons.112 The exporters had failed to provide the Department with adequate information describing the sampling methodology employed in the EEC investigation, which had prevented the Department from detecting any biases which might have been inherent in the EEC samples. In addition, the Department had reviewed the questionnaire sent by the EEC authorities to certain farms in Norway and had concluded that the responses to this questionnaire would be inadequate for purpose of its investigation. Moreover, the sample used in the EEC investigation had been prepared in part by the Norwegian parties to that investigation, raising doubts about whether the farms chosen were representative of the Norwegian industry. Second, the Department also could have relied on a study conducted in 1988 by the Government of Norway of the cost of production of Norwegian salmon farms. Indeed, the petitioner had strongly urged the Department to use the results of this study as a measure of adverse best information available. The Norwegian exporters had never urged the Department to rely in any manner on this study. Since this study had been for the period of one year prior to the period of investigation, and had been derived through procedures not wholly understood by the Department or in accordance with the Department's cost of production methodology, the Department had decided instead to attempt to develop an actual cost of production from farmers who had actually sold to the exporters during the period of investigation.
172. Norway rejected the argument of the United States that the Department of Commerce had been placed in an untenable position because of the erroneous information provided by the exporters in their responses to section A of the questionnaire. Any errors in the lists of farms provided by these exporters were the result of the fact that the Department had asked for the wrong information.113 Norway also rejected the argument of the United States that the Department of Commerce had only been informed of these errors on 30 August 1990.114
173. In response to the argument of the United States that the sampling methodology originally chosen by the Department of Commerce was statistically valid, Norway argued that, while it was correct that the use of a random number generator was a widely used sampling technique, this alone was not sufficient to make the sample statistically valid. Other aspects of the procedures followed by the Department were inconsistent with recognized statistical techniques. Thus, in designing the original samples, the Department had not consulted its own Office of Statistical Standards and Methods. The Department had also failed to consult with the FOS or other Norwegian salmon industry organizations in order to develop a representative sample. While the United States had argued that the sample originally chosen had included farms of different sizes and in different geographic locations, the sample actually used had never been stratified by size and the stratification by geographical location had been nullified when the Department of Commerce had removed four of the eleven farms.
174. In light of Norway's statement that the costs of production figures resulting from the sample drawn by the Department of Commerce were far above the figures resulting from studies conducted by the Government of Norway, the Panel asked Norway to indicate whether the Norwegian respondents had ever suggested that the Department use the data of the Government of Norway in its costs of production analysis. In response Norway noted that, following the allegation of the petitioner that sales to third countries were at prices below costs of production, the Norwegian respondents had pointed out to the Department of Commerce that the costs of production used by the petitioner was based on data for 1988, contained in a study by the Norwegian Government, and that the costs of production
112 Letter from Francis J. Sailer to David Palmeter, 11 September 1990. ↩
[Page 51]
had actually declined since then.115 The Norwegian respondents thus had considered that the cost figures for 1988 would have to be adjusted to take into account the recent declines of costs.
175. Norway argued that, as demonstrated by an affidavit of a statistical expert, dated 11 December 1991, the farm sample used by the United States for purposes of determining the cost of production of Norwegian salmon had a probability of 93 per cent of overstating the costs of production. While the use of sampling techniques might be appropriate in order to expeditiously carry out an anti-dumping duty investigation, such techniques had to be reasonably designed to achieve their purpose - in this case calculating the costs of production in as accurate a manner as possible. This required the use of statistically valid sampling techniques. In the proceedings before the Panel, the United States had failed to provide any explanation of how the sample used by the Department of Commerce had resulted in an accurate measurement of the farms' costs of production. The small size of the sample in itself had produced a high risk of an incorrect average cost of production figure, given the large variations in cost of production between farms. In other investigations in which sampling techniques had been used by the Department of Commerce, emphasis had been put on the need to have as large a sample as possible with as broad a geographical base as possible. In those cases, the samples used had been two to four times larger than the sample used in the salmon case.116
176. With respect to the affidavit of the statistical expert provided to the Panel by Norway, the United States considered that Norway was attempting to insert new evidence into the record before the Panel. However, under Article 6:8 of the Agreement, the Department of Commerce had been authorized to base its determination on the information before it. It was too late for the Government of Norway to attempt to provide new information which could and should have been presented by the Norwegian respondents to the investigating authorities. The affidavit of the statistical expert had never been provided to, and therefore could not have been used by, the Department of Commerce. This document was therefore not relevant to the Panel's review of what the Department had done on the basis of the information before it.
177. In response to a question by the Panel to comment on Norway's argument (based on the above-mentioned affidavit) that the sample of farms used for the cost of production analysis had a 93 per cent probability of being inaccurate, the United States argued that the analysis leading to the conclusion in this affidavit that the second farm sample had overestimated the cost of production contained a number of flaws which reflected a misunderstanding of the rôle of the sample in the anti-dumping investigation. First, the affidavit suggested that the original method of selecting exporters had been biased. However, the Department of Commerce had not intended to select the eight largest exporters, accounting for at least 60 per cent of all exports to the United States, as a representative sample of all Norwegian exporters. The Department normally attempted to analyze all exports of the product under investigation to the United States. When the total number of exporters subject to investigation was relatively large (in this case more than 70 exporters) the limited resources which the Department could devote to any one case made 100 per cent sales coverage impossible. Thus, the Department must in such a case select from the pool of all exporters those who would be investigated. To maximize the share of total export sales analyzed, the Department investigated the largest exporters. This limited coverage of export sales did not, however, prejudice the interests of non-selected exporters. The Department would include in the investigation all exporters who submitted sales and price data on a voluntary basis. Therefore, any non-selected exporter who believed that he would not be fairly represented by the selected
115 Letter from David Palmeter to Robert Mosbacher, 10 August 1990. ↩
116 Fresh Kiwifruit from New Zealand, 56 Fed.Reg., p. 60,092 (1991); Certain Fresh Winter Vegetables from Mexico, 45 Fed.Reg., p.20,5122 (1980); Certain Fresh Cut Flowers from Colombia, 55 Fed.Reg., p.20,491 (1990), and Fall-Harvested Round White Potatoes from Canada, 48 Fed.Reg., p.51,669 (1983). ↩
[Page 52]
exporters was free to join the investigation and have its own dumping margin calculated. There had been no Norwegian exporters who had requested such treatment in this case. Second, Norway's consultant had criticized the selection by the Department of eleven fishfarms in the sample used for purposes of the cost of production analysis. The Department had determined that the methodology employed would provide a representative sample which could still be analyzed and verified in time. To satisfy Norway's consultant, the Department might have had to sample 20 or 30 fishfarms, which had simply not been possible if the Department were to satisfy the requirements of the Agreement. If investigating authorities were given the time to conduct scientifically exact samples, the requirement that investigations be concluded within one year could not be met. Finally, the consultant had criticized the Department's decision not to weight the farms by size. However, the Department's intention had been to estimate cost of production for each of the eight exporters based on a sample from the pool of farmers supplying each exporter. Information on the record had shown that large, low-cost farms represented only about 4-5 per cent of total production volume in Norway. The Department had had no information on the volumes of salmon shipped from these farms to the eight exporters. Therefore, with respect to the cost of salmon shipped to the eight exporters, there had been no basis to attach greater significance to the cost of large farms relative to the cost of small farms.
178. Norway noted that, during the meeting of the Committee on Anti-Dumping Practices held under the provisions of Article 15:3 of the Agreement, the United States had claimed that its sampling methodology had been "designed to arrive at representative costs for each of the eight exporters based on their individual experiences." (ADP/M/33, paragraph 40). In fact, the Department of Commerce had considered neither the costs of the individual exporters (which would have entailed an examination of their acquisition prices) nor the individual costs of their suppliers. Instead, the Department had calculated one costs of production figure for all salmon farmers and had applied this figure to all exporters, regardless of the individual experiences of the exporters or of their suppliers.
179. The United States argued that Norway missed the point made by the United States that the sample as originally planned by the Department of Commerce would have provided individual costs of production calculations of each of the eight Norwegian exporters but that the reduction in the sample size due to the erroneous questionnaire responses had forced the Department to change this plan and to calculate a single cost of production applicable to all eight exporters. Had the Department followed its original plan, it would have had to rely on the best information available for three of the eight exporters (Skaarfish, Chr. Bjelland, and Saga). The United States noted in this context that the Norwegian respondents in the investigation had never raised any objections to the sample used by the Department, even though they had the opportunity to do so, given that the methodology of the Department had been available for review and comment by all interested parties during the investigation. Rather, the Norwegian exporters had generally objected to any investigation of the farmers and had suggested that, if the Department were to proceed with investigating the farmers' costs it should use the cost information gathered by the EEC in its investigation of Norwegian salmon exports to the EEC. After the Department had announced on 7 September 1990 that it would combine the cost of production of the seven remaining farms, the Norwegian respondents had not objected to this proposal then or for the remainder of the investigation. There was nothing in the record of the investigation to indicate that they were dissatisfied with the use of the seven farms or with the identities of the farms in the survey. Consequently, none of the objections raised in the proceedings before the Panel by the Government of Norway had ever been placed before the Department of Commerce for its consideration. Therefore, the Panel was being asked by Norway to act as an investigating authority and to review factual issues in the first instance. Norway's approach lost sight of the very different rôles accorded to the investigating authorities and a dispute settlement panel under the Agreement.
180. Norway contested that, as argued by the United States, the Norwegian respondents had not objected to the sampling methodology used by the Department of Commerce. Thus, in a letter dated 10 August 1990, counsel for the Norwegian respondents had stated that:
[Page 53]
"While petitioners postulate a single cost of production in Norway, the fact is that there are 700 different costs."117
On 22 August 1990, counsel for the Norwegian respondents had requested the Department of Commerce to use the sample of forty-two farms used by the EEC in its anti-dumping investigation of imports of salmon from Norway.118 Finally, in a letter dated 30 August 1990, counsel for the Norwegian respondents had stated inter alia that:
"Choosing one farmer to represent the costs of a particular exporter makes no theoretical or practical sense."119
181. Norway also argued that the sampling technique used by the Department of Commerce in its cost of production analysis in the salmon investigation differed significantly from sampling techniques used by the Department in similar cases. Thus, in Fall-harvested round white potatoes from Canada120 the Department had examined the costs of both unrelated and related growers and growers/distributors. A random sample of nineteen farms had been taken, stratified by size of the farm, type of farm and by geographical location. The Department had determined that this sample provided a statistically valid 95 per cent certainty of accuracy. In Certain Fresh Cut Flowers From Colombia121 the Department had selected at random fifteen farms after stratifying by size and taking relative market strength into account, while in Certain Fresh Winter Vegetables From Mexico122 the Department of Commerce had determined that a sample of thirty five farms was necessary to ensure accurate results "by all conventional statistical measures". This sample had been enlarged by about 50 per cent after the preliminary determination in order to increase its accuracy. In that case, the Department had expressly recognized the importance of stratifying the sample by farm size. Finally, in the recent case concerning Fresh Kiwifruit from New Zealand123 the Department had taken a sample consisting of nineteen farms, stratified by size and geographical location.
182. In response to Norway's comment that in other cases where sampling techniques had been used by the Department of Commerce, the samples had been larger than in the salmon case, the United States argued that samples had to be tailored to the facts of each case. Simply because more farms had been included in other samples did not mean that these samples were necessarily more representative. For example, in Fresh Kiwifruit From New Zealand, the Department had sampled 20 farms from the more than 2000 kiwifruit growers in New Zealand. The 20-member sample also had taken into account differences in size and geography because it had been reported that these factors had a significant effect on cost. In the salmon case, the Department had sampled eleven farms from a universe of 700 salmon growers. Norway had reported that almost all farms were identical in size and the sample had therefore not included size as a stratification factor. The sample had taken differences in geographic location into account, because the information provided to the Department had indicated that there were differences in costs between farmers in the north and farmers in the south of Norway.
183. In response to a question by the Panel on the factual basis of Norway's argument that costs of production varied by farm size, Norway first referred to the letter, dated 10 August 1990, from counsel for the Norwegian respondents in which it had been stated that, contrary to what had been
117 Letter from David Palmeter to Robert Mosbacher, 10 August 1990, p.12. ↩
118 Letter from David Palmeter to Robert Mosbacher, 22 August 1990. ↩
119 Letter from David Palmeter to Robert Mosbacher, 30 August 1990, p.3. ↩
120 48 Fed.Reg., 51,669 (1983). ↩
121 55 Fed.Reg., 20,491 (1990). ↩
122 45 Fed.Reg., 20,512, (1980). ↩
123 56 Fed.Reg., 229, (1991). ↩
[Page 54]
alleged by the petitioner, there were in fact 700 different costs of production in the Norwegian salmon industry. Second, the sample of forty-two farms used by the EEC in its anti-dumping investigation had shown variations in the costs of production from NOK 12.60 to NOK 53.74. Third, the annual study conducted by the Norwegian Directorate of Fisheries in 1989, based on an examination of information provided by 293 salmon farms, showed that the costs of production of average farms in eight different regions varied from NOK 28.68 to NOK 37.21 per kilogramme. Moreover, this study had found significant differences in costs of production of farms located within the same regions. This study also indicated that in 1989 the costs of production of the average Norwegian farm were NOK 32.32 per kilogramme but that the costs of production of the average farm in the group of fifteen farms with the lowest costs of production wer NOK 19.41 per kilogramme. Finally, the investigation conducted by the Department of Commerce of the costs of production of the seven sampled farms indicated that these costs varied from NOK 26.24 to NOK 48.06 per kilogramme. The Department of Commerce therefore could not have been unaware of the need for a sample including a far larger number of farms.
3.2.2 Use of a simple, rather than a weighted, average of the cost of production data obtained on the basis of the farm sample
184. Norway argued that the failure of the Department of Commerce to weight average the cost of production figures resulting from the sample of seven farms was not in conformity with recognized averaging techniques and had led to an artificially high constructed normal value. This led to an estimated average cost of production per kg. per farm. Instead, a weighted average should have been used in order to estimate a cost of production per kg., to be compared to the import price. The use of a simple instead of weighted average of the cost of production data had more than doubled the margin of dumping from around 9 per cent to 23.8 per cent. The cost of production data of salmon farms differed considerably. These varieties in costs had been disregarded by the Department of Commerce when it applied a simple average. Thus, the largest farm in the sample had a cost of production of 26.24 NOK/kg., while the smaller farms had costs of production of up to 48.06 NOK/kg. The decision to use a simple average of these figures had not merely been a matter of administrative convenience, given that the Department of Commerce had the information to apply either a simple or a weighted average. Nothing in the Agreement permitted the United States to use a meaningless estimate as the basis of the "cost of production" in Article 2:4.
185. Norway pointed out that, on 13 February 1991, the Department of Commerce had reconsidered the question of whether the seven farms should be stratified according to size and be assigned a weight proportional to each farm's share of total Norwegian production. The Department had claimed that there was no basis to weight costs of farms of different sizes because there was no information in the record concerning production. However, the only information needed for this purpose would be basic data on the production volumes of the individual farms included in the sample. Such data on production volumes had been provided by the respondents.
186. Norway also noted that the decision of the Department of Commerce not to weight-average the cost of production data was contrary to its practice in other cases. Thus, in Fall-Harvested Round White Potatoes from Canada124 the Department had applied a weighted average of the costs of production of the growers included in the sample used in that case.
187. In response to a question by the Panel on the factual basis of Norway's statement that the Department of Commerce had possessed the necessary information to apply a weighted average of
124 48 Fed.Reg., 51669 (1983). ↩
[Page 55]
the cost of production data, Norway referred to a Memorandum dated 17 August 1990.125 In response to a question by the Panel on the factors which should have been taken into account in the weighing of the cost of production data, Norway explained that these data should have been weighed by the production volume of the individual farms in the sample. The Panel asked Norway to explain how Norway's argument that the Department had before it sufficient information to enable it to apply a weighted average of the costs of production data could be reconciled with Norway's statement that the Department had failed to request this information in its questionnaire. Norway explained that its statement regarding the Department's failure to request the relevant information addressed an argument made in an internal Memorandum of the Department, dated 13 February 1991, that it was not possible to calculate a weighted average because of the lack of relevant data.126 However, those data, i.e., data regarding the size of the seven farms, would have been easier to obtain than much of the information actually obtained by the Department. If this information was not available in the form desired by the Department, it was because the Department had not requested the information in this form.
188. Norway pointed out that, by using a simple average of the costs of production data for the seven farmers in the sample, the Department of Commerce had determined the costs of production for the investigation period to be NOK 39.39 per kilogramme of gutted salmon. Converted into round weight, this simple average corresponded to NOK 35.45 per kilogramme. That a weighted average of these data would have been more correct was confirmed by the fact that an annual survey conducted by the Norwegian Directorate of Fisheries of the costs of production of salmon showed that the costs of production per kilogramme round weight were NOK 30.47.127 In this survey, 293 out of 700 salmon farms had been examined.
189. The United States argued that the Department of Commerce had properly used a simple average of the farms' costs of production to reflect the fact that 96 per cent of salmon production in Norway took place in small farms, with large farms accounting for only four per cent of salmon production in Norway. One of the seven farms in the sample, the Bremnes farm, was one of the largest in Norway, and thus had represented a far greater share of the total production in the sample than large farms represented of total production in Norway. Weight-averaging would have provided much greater importance to this larger, more efficient farm relative to the sample than large farms occupied relative to the Norwegian salmon industry as a whole. The Department therefore had applied a simple average to avoid the distortion of costs attributable to Bremnes. However, even under the simple average this farm had accounted for 14 per cent of the cost sample, giving larger, lower-cost farms more prominence in the sample than they actually had in the Norwegian industry.
190. The Panel asked the United States to indicate whether the decision not to weight-average the cost of production data had been taken before the Department had obtained these data. The United States replied that this decision had been made after counsel for the Norwegian respondents had reported that certain farms which had claimed to have supplied the exporters during the period of investigation in fact did not.
191. With respect to the reconsideration on 13 February 1991 of the decision not to weight-average the cost of production data, the United States observed that, in response to comments by the parties to the investigation the Department of Commerce had reviewed its decision to use a simple average of the seven farms' costs. Given that the overwhelming number of farms were in the same size range (8-12,000 cubic metres) and that there was no reliable information on the total Norwegian production
125 Memorandum from Caroline Olivieri and Tracey Oakes to the file, 17 August 1990. ↩
126 Memorandum from David Mueller to David L. Binder, 13 February 1991. ↩
127 Annual Survey (1989) by the Norwegian Directorate for Fisheries. ↩
[Page 56]
volumes which would allow stratification by size, the Department had concluded that a simple average of the production costs was preferable.
192. In response to Norway's reference to the use of a weighted average of growers' production costs in Fall-Harvested Round White Potatoes from Canada, the United States argued that in that case the Department of Commerce had not been faced with a situation in which the exporters were not the growers of the production under investigation. No attempt had been made to create a sample for each individual exporter in that case since the exporters themselves were the growers.
193. In response to the argument of the United States that weight-averaging the cost of production data would have given much greater importance to the Bremnes farm relative to the sample than large farms occupied relative to the Norwegian industry as a whole, Norway argued that Bremnes was not one of the ten largest salmon farms in Norway.128 In any case, what mattered was the proportion of production supplied to exporters to the United States, not the proportion of total production in Norway.
194. The United States noted that in its questionnaire response filed on 28 September 1990, Bremnes had reported its production volume at 70,600 cubic metres, an assertion which had not been withdrawn or corrected at any time during the investigation. In its response of 15 May 1990 to the questionnaire in the countervailing duty investigation, the Government of Norway had reported that there were only thirteen farms with a size larger than 12,000 cubic metres. Norway had listed the ten largest farms, whose sizes ranged from 20,000 to 52,000 cubic metres and had stated that these farms accounted for about 4 per cent of Norwegian production.129 The Department of Commerce had quite reasonably concluded that, given Bremnes' reported size, it was among the largest farms in Norway. Thus, the information provided by Bremnes itself refuted Norway's claim that Bremnes was not one of the ten largest farms in Norway. There was, therefore, no question about the factual underpinning for the conclusions drawn by the Department concerning the relative sizes of the small and large sectors of the Norwegian industry and the size of the Bremnes farm. The information relied upon had been provided either by the private Norwegian interests or by the Government of Norway in the countervailing duty investigation. Norway could not now disavow that information. In using a simple average of the cost of production data, the Department had chosen a more representative approach than would have been the case if it had weight-averaged the farms' cost of production, thereby over-representing the few lower-cost farms such as Bremnes, and under-representing the many higher-cost, small-sized farms. There was no requirement in the Agreement which precluded the Department from insuring that the results of its sampling more accurately reflected the composition of the sampled industry.
195. Norway considered that the argument of the United States regarding the size of the Bremnes farm reflected an inconsistency in the position of the United States. Thus, in a Memorandum dated 17 August 1990, the Department of Commerce had stated that "... since most of the fresh farms are relatively similar in size ... the size factor does not become an issue in creating a sampling strata".130 However, on the other hand, the Department had invoked the size of the Bremnes farm as a reason to apply a simple, rather than a weighted-average, of the cost of production data.
196. In response to Norway's argument that there was an inconsistency between the position taken in the Memorandum of 17 August 1990 that farm size was not a relevant factor in the construction of the sample and the later decision not to weight-average the cost of production data of the seven farms,
128 Questionnaire Response of the Government of Norway in the Countervailing Duty Investigation of Fresh and Chilled Atlantic Salmon from Norway, 14 May 1990. ↩
130 Memorandum from Caroline Olivieri and Tracey Oakes to the file, 17 August 1990. ↩
[Page 57]
the United States argued that the information available to the Department of Commerce indicated that the main difference in costs between farms was related to geographic location. Therefore, geographic location had been accounted for in the sampling methodology. The size of the farms had not been considered significant to the creation of the sampling strata, given that in May 1990 the Government of Norway had reported to the Department that 96 per cent of all salmon farms in Norway ranged in size between 8-12,000 cubic metres. When deciding whether to calculate a simple or a weighted average of the sample, it had been necessary to take note of the fact that two of the sampled farms, Bremnes and Austevoll, were significantly larger than the 8-12,000 cubic metre range which represented the overwhelming majority of the industry. Therefore, the unusual size of these two farms had properly been taken into account by the Department in calculating the average cost of production.
197. Norway made the following comments on the argument of the United States that the evidence before the Department of Commerce had indicated that the Bremnes farm was one of the ten largest salmon farms in Norway. First, what the United States had referred to as one farm, Bremnes, actually was two farms: Bremnes, with a reported size of 46,600 cubic metres, and Ola Svendsen, referred to as Jondal, with a reported size of 24,000 cubic metres. Second, the reponse of the Government of Norway regarding the size of the farms in Norway had been based on a standard depth per farm of 5 metres. Thus, the numbers reported for the ten largest farms had been based on that standard. Bremnes had a depth greater than 5 metres and had reported its actual size. If one used the Norwegian Government's standard 5 metre depth, Bremnes' size was only 18,000 cubic metres. Jondal also had a depth greater than 5 metres. Using the Government's standard, its size was only 12,000 cubic metres. Third, the report of the Norwegian Government had been based on individual concessions, and had listed Bremnes and Jondal as separate farms. Thus, neither Bremnes nor Jondal was one of the ten largest farms in Norway and these companies were not included in the list of the ten largest farms of the Norwegian Directorate of Fisheries. The information provided by the Government of Norway to the Department of Commerce on the size of salmon farms had therefore been correct. Norway noted that the above information had been provided to the Department of Commerce during the course of the anti-dumping duty investigation.131
198. The United States argued that the official data of the Government of Norway understated the size of the Bremnes farm and that the Department of Commerce could rely on the actual size reported by Bremnes itself rather than the fictitious size reported by the Government of Norway. In any event, even under the understated figures of the Government of Norway, Bremnes would be among the largest farms in Norway.
3.2.3 Use of the best information available as a basis for the calculation of the costs of production of one of the Norwegian farms
199. As part of its claim that the United States had calculated constructed normal values in a manner inconsistent with Articles 2:4 and 8:3 of the Agreement, Norway also referred to the treatment of the questionnaire responses filed by one of the seven farms in the sample used by the Department of Commerce in its costs of production analysis. The Department had rejected the questionnaire responses of this farm, Nordsvalaks, in their entirety because this farm had failed to report the costs and expenses of a second farm which the Department considered to be related to Nordsvalaks. As evidence of the existence of a relationship the Department had relied on the fact that one of the owners of Nordsvalaks also owned the second farm (Furberg Yttersian A/S ("F&Y")). The questionnaire had been addressed only to Nordsvalaks and it would have been easy for the Department to add the second farm to the response filed by Nordsvalaks. The Department, however, had rejected the cost information provided by Nordsvalaks and had used the highest costs of production figure found for any other farm in the
131 Respondents' Rebuttal Brief, 14 December 1990, pp.16-17. ↩
[Page 58]
sample as the best information available for the calculation of the costs of production of Nordsvalaks. Norway noted that the questionnaire responses filed by Nordsvalaks had been rejected even though the information provided by this farm had been verified by the Department. The resort to the best information available as the basis for the costs of production calculation for this farm had been particularly detrimental because of the fact that the Nordsvalaks farm was the second lowest cost producer among the seven farms in the sample.
200. The United States argued that the Department of Commerce had acted consistently with Article 6:8 of the Agreement in disregarding the information provided by the Nordsvalaks farm. The questionnaire responses filed by this farm had been rejected after it had been determined at verification in Norway that there were such serious omissions in these responses that an entirely new response would have been necessary. At verification, the Department had discovered that Nordsvalaks had misrepresented that it was not related to other producers and that it actually was related to another salmon farm with which it engaged in extensive inter-company transfers. The Department had concluded that the existence of this related farm raised the question of whether all costs and expenses had been properly allocated between Nordsvalaks and the related farm. This complex issue could not be explored without an entirely new response, including the related farm's data. It had been too late in the investigation to pursue this option. Accordingly, the information provided by Nordsvalaks had been disregarded, as was authorized under Article 6:8 of the Agreement in cases in which exporters impeded investigations through non-response or provision of erroneous information. Contrary to what had been alleged by Norway, the Department of Commerce had not verified the questionnaire response of Nordsvalaks and there was, therefore, no basis for Norway's claim that this farm was the second lowest cost producer in the sample.
201. The United States denied that the costs of production of the Nordsvalaks farm had been determined on the basis of the best information available. Nordsvalaks had simply been dismissed as a responding farm and costs of production had been based on data for the remaining farms. No production costs had been calculated for Nordsvalaks using best information available or any other basis.
202. Norway, responding to the argument of the United States that the Nordsvalaks farm had misrepresented the fact that it was not related to other producers, argued that any errors in the responses submitted by Nordsvalaks had been the result of the ambiguous and unclear wording of the questionnaire issued by the Department of Commerce on 21 August 1990. In this questionnaire information had been requested regarding related suppliers:
"Provide a corporate organizational chart specifically identifying the parent company, subsidiaries, hatcheries, farms, processors and all other related entities from which you obtain smolt, feed, R&D, processing services, equipment or any other inputs needed for the cultivation or sale of the products."132
Thus, this section of the questionnaire had requested Nordsvalaks to identify purchases from related suppliers. However, Nordsvalaks did not make any purchases from related suppliers. Given that the questionnaire was in a foreign language and appeared to address only the issue of related suppliers (as did all subsequent questions) Nordsvalaks had understandably been confused as to the precise information requested in this section. Nordsvalaks therefore should not have been penalized by a punitive use of the best information available (the highest cost of production found for any other farm) when its misunderstanding of the precise information requested was a direct result of the manner in
132 Section D questionnaire 21 August 1990, item II.A, (emphasis added by Norway). ↩
[Page 59]
which the Department of Commerce had formulated its question. Norway noted in this respect that a Recommendation adopted by the Committee on Anti-Dumping Practices stated that:
"even though the information provided may not be ideal in all respects, this factor, in itself, should not justify the investigating authorities from disregarding it since the interested party may have acted to the best of its ability."133
203. In response to the argument of the United States that the issues raised by the relationship between Nordsvalaks and F&Y would have required an entirely new set of responses, Norway argued that the Department of Commerce could have easily corrected the responses provided by Nordsvalaks. At verification, Nordsvalaks had demonstrated to the Department that Nordsvalaks and F&Y divided joint costs and revenues 50/50. Officials of the Department had spent almost two days verifying the information provided by Nordsvalaks.134 The Department of Commerce could have used the verified information provided by Nordsvalaks and include the companion company by multiplying the figure by two. However, instead of adopting this simple approach, the Department had rejected the information provided by Nordsvalaks in its entirety and had inflated the costs of production of Nordsvalaks by using as the best information available the highest costs of production found for any of the other farms in the sample.
204. Norway contested that, as alleged by the United States, no costs of production had been calculated for the Nordsvalaks farm. As was apparent from a document provided by the Department of Commerce which contained an explanation of how the average costs of production for the seven Norwegian salmon farms had been calculated, in the case of Nordsvalaks the Department of Commerce had determined the costs of production to be NOK 48.06 per kilo, which correspond to the highest costs of production figure of any of the six other farms in the sample.135 Thus, Nordsvalaks had not simply been dismissed as a responding farm.
205. The United States clarified its statement that Nordsvalaks had been dismissed as a respondent by explaining that the costs of production figure calculated on the basis of the sample of seven farms had consisted of seven separate costs figures which had been averaged to produce the final costs figure. Nordsvalaks had been attributed the highest costs figure calculated for the other farms in the sample. This figure had been added into the average as the best information available for Nordsvalaks costs of production.
206. The United States pointed out that in the questionnaire issued on 21 August 1990 by the Department of Commerce to the Norwegian salmon farms, the following item was included regarding the corporate structure of the farms:
"Provide a corporate organization chart specifically identifying the parent company, subsidiaries, hatcheries, farms, processors and all other related entities from which you obtain smolt, feed, R&D, processing services, equipment or any other inputs needed for the cultivation or sale of the products. The relationship of your company to the other entities should be determined pursuant to Section 773(e)(3) of the Act. Specify the exact nature of the relationship, including percentage of ownership by you, your parent, and/or other subsidiaries of your parent. See attachment A-#5."
133 BISD 31S/283, paragraph II.5. ↩
134 Norway provided to the Panel affidavits of two participants in the verification. ↩
135 Norway provided to the Panel data on the individual cost of production figures determined by the Department of Commerce for of each of the seven farms in the sample. ↩
[Page 60]
Attachment A to the questionnaire had contained a definition of the term "related parties" as provided for in section 773(e) of the United States Tariff Act of 1930, as amended. In its questionnaire response dated 28 September 1990 Nordsvalaks had stated in response to this question that "Nordsvalaks is a privately owned salmon farm with no related companies". Since the Department of Commerce had had no reason to doubt Nordsvalaks on this point, it had not asked for a clarification in its supplemental questionnaire issued on 10 October 1990 to Nordsvalaks. Nor had Nordsvalaks addressed the issue further in its responses to this supplemental questionnaire dated 23 October 1990.
207. The United States pointed out that during verification in Norway the Department of Commerce had found the following with regard to Nordsvalaks corporate structure. Nordsvalaks was 100 per cent owned by an individual who also owned a 50 per cent interest in another entity, Furberg & Yttersian A/S ("F&Y"). The other 50 per cent of F&Y was owned by his wife. Nordsvalaks and F&Y, each of which purported to be separate salmon farms, "were located at the same facility", on Froya Island, Norway, "with all activities commingled, according to the owner and accountant". The salmon for both companies was kept in the same cages and all sales and losses were attributed 50 per cent to each company. The owner of Nordsvalaks managed both farms simultaneously and the farms used the same accountant who made payments and maintained F.O.S. invoices on behalf of both companies and recorded each transaction, making sure to split all costs evenly, 1/2 to each company. Furthermore, purchasing of smolt was normally transacted by either company with invoices (or debit notes) prepared to bill 1/2 of the charges to the other company. Purchases of feed were normally made by F&Y with a bill prepared to charge to Nordsvalaks for half. Payroll was prepared and paid by F&Y with a charge to Nordsvalaks for half. The book-keeper also reviewed all F.O.S. prepared invoices to ensure that quantities delivered from Nordsvalaks to the exporter were attributed equally to the two companies.136 Considered in the context of these findings at verification, Norway's claim that Nordsvalaks in good faith believed that it did not need to provide information on its relationship with F&Y prior to verification was without any basis. The Department had found at verification that Nordsvalaks was in no meaningful sense a salmon farm, any more than F&Y was. Instead, both were entities which jointly owned and operated a single salmon farm. As such, complete information from both companies about their joint operations should have been submitted to the Department of Commerce. The fact that this information had not been provided had constituted ample grounds to reject Nordsvalaks response at verification as deficient and unverifiable.
208. The United States contested that the Department of Commerce had specifically asked for information only about related parties from which Nordsvalaks had purchased inputs. Nordsvalaks had been asked to identify all related entities from which it had obtained material, not, as suggested by Norway, only material purchased from related parties. The passage of the questionnaire dealing with this issue was not susceptible to ambiguity. Despite the fact that there was only one interpretation possible, Nordsvalaks had provided false information. Moreover, the Department of Commerce had found that Nordsvalaks in fact did purchase inputs - smolt and feed - from F&Y. F&Y typically purchased feed for the two companies, purportedly billing Nordsvalaks for half the cost, and both companies apparently took turns at purchasing smolt for their joint farm, with the purchaser purportedly billing the other party for half the cost. Thus, even under Norway's reading of the questionnaire, Nordsvalaks had been required to report in its response its relationship with F&Y.
209. The United States contested that Nordsvalaks failure to report its relationship with F&Y was a mere technicality that could have easily been remedied at verification. In verifying Nordsvalaks' costs for inputs purchased from F&Y, the Department of Commerce would have had to ascertain F&Y's acquisition costs of these inputs to insure that the price it charged Nordsvalaks was arms-length.137
136 Department of Commerce cost verification report for Nordsvalaks, 14 December 1990, pp.2-3. ↩
[Page 61]
Because for time reasons the Department could not receive substantially new questionnaire responses at verification, it had had no choice but to deem Nordsvalaks' response unverifiable and terminate the verification. The United States also argued in this context that the domestic industry participating in the investigation would have had no opportunity to review substantially new information submitted at verification. Norway's suggestion that the Department of Commerce should have somehow obtained and reviewed information on F&Y at verification would have denied the domestic industry its right of review and comment under Articles 6:1 and 6:2 of the Agreement and ignored the express grant of authority in Article 6:8 for the investigating authorities to base their determination on the information available when an interested party had failed to provide necessary information in a timely manner.
3.2.4 Calculation of the amount for profits
210. Norway argued that the United States had acted inconsistently with Article 2:4 of the Agreement when the Department of Commerce had included in the constructed normal values an eight percent profit rate mandated by the anti-dumping legislation of the United States, without determining, as required by Article 2:4, whether this profit rate constituted a "reasonable amount" for profits in the case of the Norwegian salmon industry. The statutory minimum amount for profits applied by the United States was inconsistent with Article 2:4, as had been stated by one internationally recognized American scholar:
"Indeed, this provision [19 U.S.C. 1677b (e) (1) (B) (ii)] of the United States statute is very likely inconsistent with United States international obligations, since those obligations require that a realistic method be used to compute the constructed cost and prices based on constructive cost."138
What was a reasonable amount for profit varied from industry to industry and over time, depending upon fluctuations in price and other factors. The statutory minimum amount for profit applied by the United States disregarded this reality. Norway noted that other Parties to the Agreement properly attempted to determine, on a case by case basis, what was a reasonable amount for profit. Thus, as indicated by recent anti-dumping cases decided by the EEC authorities, the EEC had found profit margins ranging from 5 to 48 per cent to be reasonable for various industries. The failure of the United States to determine the reasonable amount for profit on a case by case basis violated Article 2:4 of the Agreement.
211. The United States argued that in determining the profit rate to be used for purposes of calculating the constructed normal values it had been necessary for the Department of Commerce to use an 8 per cent profit figure because the Norwegian respondents had refused to provide actual profit data.139 Consequently, the inclusion of this profit rate in the constructed normal values was authorized under Articles 2:4 and 6:8 of the Agreement. In Section D of the questionnaire issued to the farmers, the Department had requested profit data as follows. Specifically, Section D.IV.C of this questionnaire had stated:
"For Constructed Value only, calculate the average profit earned on sales to the exporters for the period under review. Please detail the method in which you calculated profit."
In its questionnaire issued to the exporters, the Department also had requested profit data. Section D.IV of this questionnaire had stated:
138 Jackson, The World Trading System, 1989, p.235. ↩
139 See supra, Section IV of this report for the views of the United States on the admissibility of Norway's claim on this matter. ↩
[Page 62]
"Information concerning profit should be provided. The information for the appropriate market (market used for the FMV) should be provided according to the following hierarchy for the period under investigation, based on your experience:
(a) Profit on third country market sales of the such/similar products; or
(b) Profit on third country sales of merchandise of the same
"general class or kind" by your company. This should be an aggregated percentage for all third country markets.
Separate answers had been received from the farmers and exporters. From the farmers the response had been "N/A" - in other words, the farmers had thought the question was inapplicable. From the exporters, a similar, if lengthier answer had been received:
"Because the Department has not yet decided whether a constructed value approach is appropriate, it is premature to request constructed value information. We will provide this information if the Department decides that the constructed value approach is necessary."140
Despite this non-compliance with the questionnaire request, the Department of Commerce had alerted each farmer in a deficiency questionnaire that constructed value information was required and that this information should be submitted. Nevertheless, no such profit information had ever been provided. Given that the Norwegian respondents had refused to provide actual data on profits, Norway could not now complain that it had been unreasonable for the Department of Commerce to use an 8 per cent profit figure. Without a reported profit figure, there had been no way for the Department's officials at verification to verify respondents' actual profits.
3.2.5 Use of the FOS processing fee as the best information available in the calculation of exporters' processing costs
212. Norway considered that the calculation by the United States of the constructed normal values in the investigation of imports of Atlantic salmon from Norway was inconsistent with Article 2:4 of the Agreement as a result of the manner in which the Department of Commerce had determined the amount for processing costs incurred by some of the exporters. Certain exporters did their processing of salmon in-house, rather than through the central organisation, FOS. For those exporters the Department had verified that these in-house processing costs were lower than the average processing fee charged by the FOS. Nevertheless, the Department had used this average FOS processing fee as the best information available for the processing costs incurred by these exporters, despite the fact that it knew that this average FOS processing fee overstated the actual processing costs incurred by the exporters.141
213. The United States argued that Norway's claim that the United States should have relied on the actual processing costs (i.e. the costs for gutting, bleeding and packing) allegedly incurred by exporters Chr. Bjelland and Skaarfish ignored the fact that during the investigation the Department
140 Reply of Hallvard Leroy A/S to Section D of the Department's Questionnaire, 28 September 1990, p.3. ↩
141 56 Fed.Reg., 25 February 1991, p.7675, Comment 7. ↩
[Page 63]
of Commerce had been presented with conflicting information by these exporters.142 Faced with irreconcilably conflicting statements from the exporters, the Department had properly relied on the only verifiable information before it, which was the FOS average processing fee. For example, Chr. Bjelland had reported in its response of 16 May 1990 that it bought fish from farmers at an ex-cage price. On 27 July 1990, however, the firm had reported that all charges applicable to transporting the merchandise to Chr. Bjelland's distribution warehouse, including standard packing (which included processing) were included in the exporter's costs of purchasing the merchandise. Because it had received inconsistent information concerning who actually paid the processing fee, the Department had applied the standard processing fee listed in the FOS schedule (the benchmark for processing costs for most of the farmers and exporters) as the best information available, in accordance with Articles 5:5 and 5:8. Contrary to Norway's assertion, the Department had been unable to verify the processing fees claimed by those exporters, because of the absence of documentary evidence supporting their claims.
3.3 Comparison of the normal value and export price
214. Norway submitted that the affirmative final determination of dumping in the investigation of imports of Atlantic salmon from Norway was inconsistent with Articles 2:6 and 8:3 of the Agreement as a result of a failure of the United States' authorities to effect a fair comparison between the constructed normal values and the export prices. After having calculated one single average cost of production per pound for all qualities and sizes of salmon, the Department of Commerce had compared the constructed values with individual export prices. Salmon was produced and sold in three qualities (production, ordinary, and superior) and three size categories (2-3 kilos, 3-4 kilos, and 4-5 kilos). All three size categories of "superior" salmon were sold for export to the United States. The Department had been aware of these distinctions in quality and weight, which had been described in the petition. Nevertheless, the Department of Commerce had not allocated joint costs to reflect real differences in value for different qualities and weights.
215. Norway considered that the Agreement and Article VI of the General Agreement nowhere authorized a comparison between an average normal value and individual export prices. This was comparing apples to oranges. In the present case, the comparison of an average normal value with individual export prices had inevitably created margins of dumping where there would have been no margins of dumping if the United States had made a fair comparison, i.e., a comparison between a weighted average normal value and a weighted average of the export prices. The method of comparison used by the United States in this case was biased in favour of the domestic industry. Especially in instances where there were price fluctuations, as in this case, this method would inevitably lead to findings of dumping during certain periods.143 In order to effect a fair comparison between the normal value and the export price, identical methods should have been used in the calculation of the normal value and the export price.
142 See supra, Section IV for the views of the United States on the admissibility of Norway's claim on this matter. ↩
143 To illustrate how this method of comparing export prices and normal values inevitably led to findings of dumping, Norway gave the example of a situation in which in the domestic market three sales were made at different points in time at prices of 80, 100 and 120. If there were three export sales at different points in time, also at prices of 80, 100 and 120, the comparison of an average normal value with individual export prices would inevitably result in a finding of dumping in respect of the first of these three export sales. Norway also referred in this context to a draft report prepared by E.U. Petersmann for the International Law Association, which had pointed to the "systematic bias" of comparing a weighted average normal value to indivual export prices. ↩
[Page 64]
216. With regard to the differences in sizes and qualities of salmon, Norway noted that a smaller fish would fetch a lower price per kilo than a larger fish of the same quality because the larger fish had proportionately more meat and less bones. The United States had calculated one single estimate of the normal value, even though salmon of all three weight categories of superior quality had been sold in the United States. The comparison of this single normal value to individual sales of fish to the United States would almost invariably create margins of dumping even where the shipment as a whole received a total price in the United States well above the correct cost of production (and well above the single estimated normal value) because the smaller fish would fetch prices lower than the single estimated normal value, while the larger fish would fetch prices above that single estimate. The feed costs were the same for all salmon, regardless of the quality grade of the finished product. A smaller salmon, however, would fetch a lower price, due inter alia to the lower ratio of meat/bones of smaller fish. Accordingly, the actual costs per kg. of producing small salmon were higher than the costs per kg. of producing large salmon. In fact, the price/cost ratio for salmon was inversely proportional. The production costs decreased with the size of the salmon, while the price increased. Any batch of salmon produced would contain salmon of varying sizes. A single calculated normal value did not take this difference into account.
217. Norway considered that the method of comparison of the normal value and the export price applied by the United States in this case not only violated the requirement of Article 2:6 of the agreement that a fair comparison be made between normal and export price but, in addition, had resulted in the imposition of anti-dumping duties in excess of the actual margin of dumping, in violation of Article 8:3. The United States had previously recognised that the comparison of weighted average normal values to individual export prices could lead to the creation of margins of dumping where there was no dumping. As had been stated in a report by the United States General Accounting Office:
"Customs regulations permit the calculation of margins on a weighted average basis, which amounts to determining a weighted average home market value and comparing it with individual sales to the export market. This method of determining margins between home and export market sales tends to enlarge existing margins or to create margins where none existed."144
The creation of margins of dumping was especially suspect in the present case since small fish sold for less per pound than large fish, as had been admitted by the United States.145 Moreover, for the third country weighted average, the Department of Commerce had included only those third country sales prices found to be above the overestimated normal value. All sales below this value had been disregarded. The consequence of this methodology was that dumping margins could be invented even if the prices were identical in the markets compared.
218. The United States considered that Norway's objection to the comparison of an average estimated normal value to individual export prices was without merit.146 Norway had pointed to no provision in the Agreement which prohibited such a comparison, or which mandated another type of comparison. The comparison of an average estimated normal value to individual export prices was practised by most major users of anti-dumping laws. The Agreement required that the price comparison be "fair" but imposed no specific methodology upon a Party to implement this requirement. The practice of the United States was based on the premise that it was appropriate to compare a fair value benchmark
144 US General Accounting Office, Report to the Congress by the Comptroller General of the United States, U.S. Administration of the Antidumping Act of 1921, 15 March 1979 (emphasis added by Norway). ↩
145 USITC Determination, p.A-48 and p.A-50. ↩
146 See supra, Section IV for the views of the United States on the admissibility of Norway's claim on this matter. ↩
[Page 65]
against individual export transactions. If this were not the case, a producer who engaged in dumping half the time, but exported at a price above the home market price half the time, would not be found to be dumping. Under the practice of the United States, compensatory duties would be paid for each individual sale below the average home market price but would not be paid for sales at or above the fair value. Norway's argument that only an average normal value and an average export price should be compared would make it difficult to remedy instances in which dumping was occurring in only particular product lines, or with respect to particular customers or regional markets. Under Norway's interpretation, such dumping would go undetected, so long as the exporter was making other sales at fair value prices. Irrespective of whether one believed that the methodology suggested by Norway achieved a "fair comparison", the question to be addressed was whether the Agreement mandated any particular methodology to satisfy the "fair comparison" standard. In the view of the United States, this was not the case.
219. In response to a question by the Panel whether the Department of Commerce had given consideration to the possibility to assess separate normal values for different sizes and qualities of salmon, the United States noted that most of the normal value calculations had been based on costs of production calculations. The Department of Commerce had first determined that the Norwegian home market was not viable for each of the eight exporters and had then initially determined that the normal value would be based on sales prices to third countries in the EEC. The Department next had compared these sales prices to EEC countries to the cost of production. The Department had not developed separate costs of production benchmarks for the three grades, and the several different weight bands, of salmon, because the cost of production did not vary by these factors. In other words, it cost as much to produce and ordinary-grade salmon as a premium-grade salmon, and to produce a small salmon as a large salmon. Furthermore, the Norwegian exporters had not asked the Department to somehow allocate costs along weight or grade differences, and certainly had not provided any suggestion on how such an allocation of costs could be made. After finding that substantially all (over 90 per cent) of the third country sales of seven of the exporters were at prices below costs of production, the Department had based the normal value for these seven exporters on the costs of production. Since a sufficient number of third country sales of the eight exporter had been made at above-cost prices, the normal value for that exporter had been based on its sales prices to third countries. However, its sales in the United States and the third country market had been compared only if the sales related to salmon of the same quality, the same weight, and the same condition (gutted or ungutted) and sold in the same month. This last factor had been accounted for in response to the request of the Norwegian respondents that the Department not compare export prices with a weigthed average normal value covering the entire period of investigation.
220. Norway contested that, as had been argued by the United States, costs of production of salmon did not vary by quality and weight. This view virtually guaranteed that small or low quality fish would always be found to be sold below a single average of estimated costs of production. In fact, the United States Court of International Trade had recognised the absurdity of using a single cost of production in this situation and had instructed the Department of Commerce to calculate different costs of production for products made from the same process, thereby contradicting the view taken by the United States in the present case that costs of production for salmon of different qualities and weights did not vary. The methodology required by the Court of International Trade was known as the net realisable value method and was designed to allocate costs to reflect market realities. The United States had failed to use this method in the salmon case. The Norwegian exporters had indicated to the Department of Commerce during the investigation that there were problems with comparing a single constructed normal value to individual export prices because of the differences in market value based on quality and weight differences.147 The United States had acknowledged that there were problems with such a comparison when it had used third country prices as the surrogate for the normal value.
147 Respondents Case Brief, 14 January 1991, pp.10-11. ↩
[Page 66]
In those cases, prices had been compared only of sales of salmon of the same quality, the same weight, the same condition and sold in the same month. This demonstrated that the United States was aware that the normal value would reflect differences based on quality and weight factors; these differences therefore should have been taken into account in the calculation of the normal value based on the costs of production.
221. Norway argued that at the time of its investigation of imports of Atlantic salmon from Norway the United States had been aware that its practice of calculating one single estimated normal value did not effect a fair comparison as required by Article 2:4 of the Agreement. In 1989, the United States Court of International Trade had ruled that this practice of allocating costs on a simple average basis to all grades of a product did not comply with the requirements of United States legislation regarding a fair comparison between the normal value and export price.148 The United States had subsequently revised its practice to allocate costs based on the net realisable value method, under which joint costs were allocated in a manner reflecting actual differences in value, and had specifically discussed the appropriateness of such an approach when dealing with agricultural products with uniform grading standards, such as existed for salmon.149
222. In light of Norway's statement that the practice of the United States regarding the calculation of a single normal value did not effect a fair comparison as required by Article 2:4 of the Agreement, the Panel asked Norway to explain whether its claim with respect to the issue of differences in quality and size of the salmon was based on Article 2:4 or Article 2:6 of the Agreement. In response, Norway stated that the question of quality and weight differences had affected the calculation of the costs of production and therefore was an issue under Article 2:4 of the Agreement. However, the constructed normal values to which the export prices had been compared was based on that calculation of the costs of production and the issues relating to that comparison were based on Article 2:6 of the Agreement. Thus, Norway's claim regarding the differences in quality and size related both to Article 2:4 and Article 2:6.
223. The Panel asked Norway to indicate which specific provisions of the Agreement required an allocation of joint costs to reflect real differences in value for different qualities and weight of salmon, how in the case under consideration this allocation could and should have been made, and whether the Norwegian respondents had at any time during the investigation requested the Department of Commerce to make such an allocation. The Panel also asked Norway to explain whether by "differences in value" it meant differences in prices or differences in costs of production.
224. In response, Norway noted that there was not one accounting method which would determine the costs of production in all contexts. Under Article 2:4, the costs of production were a surrogate for the normal value, which typically was based on sales prices in the home market. When products produced by the same process had noticeably different market values due to physical differences, such as quality differences, cost accounting provided ways to allocate the joint costs to reflect those different market values. The United States valued co-products using the net realisable value method. However, this method had not been used in the investigation of imports of Atlantic salmon from Norway. The use of this method would have better reflected the true costs of production of each quality and weight category in accordance with the requirement of Article 2:4 that the authorities calculate the "cost of production". Costs of production calculated on the basis of this method would also have better approximated the normal value of the salmon. Norway observed that the Norwegian exporters had
148 Ipsco Inc. v. United States, 714 F. Supp. 1211 (CIT 1989). ↩
149 Oil Country Tubular Goods from Canada, 56 Fed.Reg., 21,659-21,660 (1991) Oil Country Polyethylene Terephthalate Film, Sheet and Strip from South Korea, 56 Fed.Reg., 16,305, 16,311-312 and 16,315-316, (1991). ↩
[Page 67]
raised with the Department of Commerce the problem of calculating just one cost of production for all categories of salmon. The exporters had suggested that the Department at least attempt to avoid this problem by comparing the single cost to a single average United States price, based on all sales of all qualities and weight categories of salmon. Moreover, the United States Court of International Trade had already instructed the Department of Commerce to change its practice of calculating one single cost of production for co-products (to be compared to the different export prices of the different co-products) and to use a method such as the net realisable value method. Therefore, the Department of Commerce had been aware of the better approach to calculating the true cost of production in such situations and, indeed, since the salmon case had used the net realisable value method.
225. In response to a question by the Panel as to whether Norway's argument regarding the allocation of costs pertained to Article 2:4 or to Article 2:6 of the Agreement, Norway observed that the issue of differing qualities and sizes of salmon also related to the requirement under Article 2:6 of the Agreement of a fair comparison between the normal value and the export price. Where prices varied per kilo based on quality and weight factors, a comparison of individual export prices to a single cost of production was inconsistent with this requirement. During the investigation, the Norwegian respondents had suggested to the Department of Commerce that, if a single cost of production were used, the Department should compare that single cost of production to a single export price which would include all categories of qualities and weights of salmon.
226. The Panel asked Norway to indicate how in its view the decision of the United States Court of International Trade in Ipsco, Inc. v. United States150 was relevant to the interpretation of Article 2:4 and/or Article 2:6 of the Agreement. In response, Norway noted that this decision demonstrated that even United States courts did not find that the use of a single cost of production for co-products resulted in an accurate calculation of the cost of production, as required by Article 2:4, or in a fair comparison between the normal value and the export price, as required by Article 2:6 of the Agreement.
227. Norway also considered as inconsistent with the requirement in Article 2:6 of a fair comparison between normal value and export price the fact that the United States had failed to take into account the perishable nature of salmon, despite the fact that the USITC in in its affirmative final determination had found that salmon was a perishable product.151 The perishability of the salmon meant that it had to be sold within a very short period of time after the exporter ordered the salmon from the farm and that the exporter had no choice but to accept the price offered for the salmon on the market. The nature of the salmon as a perishable product was an additional reason why a comparison of a weighted average cost or price with individual export prices was not fair.
228. The United States considered that, by implying that the Department of Commerce had somehow overlooked the question of whether salmon was to be treated as a perishable product, Norway misrepresented what had actually occurred in the investigation. Based on the evidence on record, the Department had determined that Atlantic salmon was not a perishable product. Specifically, after the perishability issue had been discussed at the administrative hearing, the Department had requested that the parties to the investigation further brief this issue in a post-hearing submission. Counsel for the Norwegian respondents had on that occasion argued that after one or two years in the sea, a maturing salmon began to "turn grey", which caused a substantial loss in value. The petitioner had argued that this initial maturing period was only temporary, and that Atlantic salmon soon regained its flesh colour. The petitioner also had presented evidence that Norwegian farmers routinely "held over" their salmon
150 714 F. Supp. 1211 (Court of International Trade 1989). ↩
151 USITC Determination, p.A-3. ↩
[Page 68]
stocks beyond the initial maturation process. In its final determination, the Department of Commerce had made the following statement on the perishability issue:
"We agree with petitioners that fresh salmon is not a perishable commodity for purposes of the cost analysis. Norwegian Atlantic salmon farmers have the ability to control the time of sale of their output by 'holding over' inventory and, since January 1990, by freezing fresh salmon. Regarding respondent's assertion that salmon is perishable in the hands of the exporters, the Department found at verification that the opposite is true. Exporters coordinate their salmon requirements in weekly telephone conferences with farmers, and with other exporters. By doing so, exporters can communicate their salmon requirements two weeks into the future to the farmers so that farmers can begin to "starve" (prepare for harvest) the salmon two weeks prior to harvest. Accordingly, there appears to be no perishability problem at the exporter level."152
229. Norway made the following comments in response to the description provided by the United States of the manner in which the Department of Commerce had arrived at its conclusion ton whether or not Atlantic salmon was to be treated as a perishable product. First, the United States ignored the statement in the reports on the verifications conducted by the Department in Norway that salmon began to loose its value within four days of being slaughtered.153 Second, nothing in the Agreement permitted a signatory to reach opposite conclusions on the same issue in order to justify the imposition of duties, as the United States had done in the present case with respect to the issue of perishability. The USITC had stated that Atlantic salmon was highly perishable, while the Department of Commerce had stated that Atlantic salmon was not perishable.
230. In response to a question by the Panel as to how, in the view of Norway, the Department of Commerce should have taken account of the allegedly perishable nature of Atlantic salmon in making comparisons between the normal value and the export price, Norway noted that the Norwegian exporters had pointed out that they sold on a weekly basis.154 Therefore, the Department of Commerce could have made comparisons of weekly average prices on both sides of the equation.
4. Determination of the existence of injury (Article 3)
231. In summary, Norway argued that the affirmative final determination of injury made by the USITC in its investigation of imports of fresh and chilled Atlantic salmon from Norway was inconsistent with the requirements of Articles 3:1, 3:2 and 3:3 regarding the examination of the volume of the allegedly dumped imports, the effect of these imports on domestic prices in the United States of the like product, and the consequent impact of the imports on domestic producers in the United States of the like product. This final determination was also inconsistent with the requirements of Article 3:4 of the Agreement as a result of the failure of the USITC to determine that the allegedly dumped imports were, through the effects of dumping, causing present material injury to the domestic industry and to ensure that injuries caused by other factors were not attributed to these imports.
232. In summary, the United States argued that the consideration by the USITC of the volume of the imports subject to investigation, the price effects of these imports, and of the impact of the imports on the domestic industry were in conformity with the requirements of Articles 3:1, 3:2 and 3:3 of the Agreement and that the conclusions drawn by the USITC with respect to these factors were fully supported by the evidence before the USITC. The United States argued that Norway was asking the Panel to reweigh the facts before the investigating authorities. However, the United States pointed
152 56 Fed.Reg. 25 February 1991, p.7673. ↩
153 Verification Report for Hallvard Leroy A/S, 10 December 1990, at II (b) (1). ↩
154 Respondents Case Brief, 14 January 1991, p.18. ↩
[Page 69]
out, the Agreement did not envision this rôle for dispute resolution panels. Rather, under the Agreement, factual issues were entrusted exclusively to the investigating authorities. Therefore, the United States argued, the Panel should decline Norway's invitation to reweigh the facts, and instead consider whether the USITC had considered the Agreement factors and possessed positive evidence concerning its conclusions. This final determination was also consistent with Article 3:4 in that the USITC had determined that the subject imports were, through the effects of dumping, causing present material injury to the domestic industry, as required by Article 3:4. The USITC had linked the effects of the imports from Norway to the materially injured condition of the domestic industry, and thus had not improperly attributed to the imports any injury from other factors.
4.1 Request by Norway for certain data
233. Norway asked the Panel to request the United States to provide all information relied upon by the USITC in its determination which, because of its confidential nature, had not been disclosed in the published text of the determination or the Annex thereto. Norway specifically requested that the United States make available to the Panel monthly data for the period 1987-1991 regarding production and domestic consumption of Atlantic salmon in the United States, prices in the US market for various sizes of Atlantic salmon, and market penetration of imports of Atlantic salmon from Norway, Canada and Chile. In addition, Norway requested that the United States provide the Panel with data on imports from Norway, by volume and value, during the first months of 1991.
234. The United States responded that it had provided Norway with monthly data on 1989 and 1990 imports from Norway and 1990 imports from all other countries, and that the USITC had not gathered consumption or market penetration figures on a monthly basis. The same was true for figures on US production. With regard to prices, the Annex to the USITC's determination, at pages A-52-54, contained charts displaying publicly available prices for several weights of Atlantic salmon on a weekly basis for the years 1987 through 1989. The USITC had also collected pricing data in questionnaires, as described at page A-51 and pages A-59-61 of the Annex. However, the actual pricing figures in the questionnaires were not releasable because they were business confidential information. As noted at pages A-59-60 of the Annex, price trends and price comparisons had been similar for the questionnaire and publicly available data. With regard to monthly data on imports from Norway, the record of the USITC's investigation did not contain monthly figures for 1991.
4.2 Volume of the allegedly dumped imports (Articles 3:1 and 3:2)
235. Norway submitted that the affirmative final determination of the USITC in its investigation of fresh and chilled Atlantic salmon from Norway was inconsistent with the requirement of Article 3:1 that there be an objective examination of the volume of allegedly dumped imports, and with the requirement of Article 3:2 that the authorities consider whether there has been a significant increase in the volume of imports, either in absolute terms, or relative to production or consumption in the importing country.
236. In its determination, the USITC had concluded that the volume of imports from Norway over the period of investigation and the increases in the volume of the imports from 1987 to 1989 were significant. The USITC had also referred to the "effects of the large increase in salmon imports from Norway ... during the period of investigation through 1989".155 Norway contested these assertions as partly incorrect and partly misleading. The evolution of the volume of imports into the United States of fresh and chilled Atlantic salmon from Norway had to be analyzed in the context of the recent development of the domestic market for this product in the United States. Norway had developed the
155 USITC Determination, p.18 and p.21. ↩
[Page 70]
United States market for fresh salmon and had been before 1984 for all practical purposes the only supplier to the US market. Norway provided to the Panel monthly statistical data covering the period 1986-1991 on indicators of the development of the salmon market in the United States. Domestic consumption of salmon in the United States had fluctuated somewhat but had shown a considerable growth in the long term which appeared to be continuing. This growth of consumption had gained momentum in mid-1988. During the six months prior to the filing of the petition in this investigation (end of February 1990) imports of Norwegian salmon into the United States had totalled 5,984 tons, compared to 6,132 tons during the period September 1988-February 1989. Moreover, whereas market penetration of the Norwegian imports had decreased steadily during this period, imports from all other countries had nearly tripled their market share both by value and by volume.156 By any measure (i.e., either in absolute terms or relative to consumption) imports from Norway had fallen from 1988 to 1990, the period covered by the investigation of the USITC. According to Article 3:2, it was the increase in the volume of imports which must be significant.
237. In response to a question of the Panel as to whether Norway contested that the data on import volume reported in Tables 17 and 18 of Annex A to the determination of the USITC were factually correct, Norway observed that these Tables contained only annual data. Data on monthly import volumes for 1989 and 1990 showed that imports from Norway had begun to decline significantly before the filing of the petition.157 These Tables therefore did not provide a complete picture of the information available to the USITC. Absent a chance to review all the information before the USITC, Norway - and the Panel - could not evaluate whether the statements of the facts in the Report of the USITC were correct.
238. Norway further argued in this context that from the information available it appeared that in its analysis the USITC had failed to take account of the decrease in imports over the third part of the investigation period. When this decline, which could not be explained by the initiation of the anti-dumping and countervailing duty investigations, was seen in conjunction with the decline of the Norwegian market share throughout the period of investigation, the case became even stronger that the USITC had not carried out an objective examination of the evolution of the volume of imports from Norway. The evidence before the USITC showed an increase in the absolute volume of imports from Norway only during the first two years of the investigation period. In the last part of this period, prior to the initiation of the anti-dumping and countervailing duty investigations, imports had declined. To determine whether an increase in the volume of imports was "significant" within the meaning of Article 3:2 of the Agreement, the increase had to be seen in context. In the case of the investigation of imports of Atlantic salmon from Norway the context was that Norway's market share had been declining over the investigation period and that market shares of third countries and of domestic producers in the United States were increasing.
239. Norway did not contest that, as observed by the USITC on pages 16-17 of its determination, the absolute volume of imports of Atlantic salmon from Norway had increased from 1987 to 1989. However, the significance of this information was limited. First, the increase had not been of a continuing nature as monthly data on import volume from Norway showed that the volume of imports from Norway had declined in the last four months of 1989.158 Second, the investigation period which was the basis of the USITC's determination included the year 1990, in which the absolute volume of imports from Norway had declined significantly before the initiation of the investigation. The information in the Report on the increase from 1987 to 1989 in the absolute volume of imports therefore did not give an appropriate picture of the period investigated.
156 USITC Determination, p.A-45. ↩
157 See Annex 3 to this Report. ↩
158 See Annex 3 to this Report. ↩
[Page 71]
240. Norway explained that it was not arguing that, as a matter of law, Article 3:2 of the Agreement permitted a finding of a "significant increase" of the volume of imports under investigation only when the volume of imports at the end of a period of investigation was higher than the volume of imports at the beginning of that period. Article 3:2 referred to the significance of the increase of the volume of imports, either in absolute terms or relative to domestic consumption or production. Where there was no increase in the absolute volume of imports, the investigating authorities were required to examine two questions. First, what accounted for any decline of absolute import volumes toward the end of the investigation period or for the absence of an increase in the absolute volume of imports, and second, whether imports had increased in relative terms. In the present case, the evidence did not support the conclusion that the decline in the absolute volume of imports in the last part of the investigation period was due to the imposition of the provisional measures. In the course of the investigation by the USITC it had been repeatedly pointed out by the Norwegian respondents that this decline in absolute import volume was caused by a combination of several factors: lower domestic prices in the United States, the appreciation of the Norwegian kroner relative to the US dollar, and rising prices in alternative export markets. Regarding the second point, Norway reiterated that over the investigation period the market share held by Norwegian imports in the United States had declined. In this situation, where import volume was not higher at the end of the investigation period than at the beginning of that period, where the facts demonstrated that the decline in absolute import volume was not the result of the initiation of the investigation or of the imposition of provisional measures, and where the import volume declined in relative terms throughout the period of investigation, Article 3:2 of the Agreement did not, in the view of Norway, permit a finding of a "significant increase" in the volume of imports.
241. Responding to a question of the Panel, Norway explained as follows its views on the legal relevance of imports from third countries to an examination of whether, in the analysis of the evolution of the volume of imports of Atlantic salmon from Norway, the USITC had acted consistently with Articles 3:1 and 3:2 of the Agreement. First, Article 3:1 required the investigating authorities to examine the volume of the dumped imports. Article 3:2 explained that the examination is to enable the authorities to determine whether there has been a "significant increase" in absolute terms or relative to production or consumption in the importing country. Third country imports directly affected the level of consumption in the importing country and were thus relevant to an objective examination of the volume of the dumped imports. Second, Article 3:1 also required the investigating authorities to examine the effect of the dumped imports on prices in the domestic market. Article 3:2 stated that this included a consideration of whether there had been significant price undercutting or price depression. In the case under consideration, the question was whether the allegedly dumped Norwegian imports had led to price depression. Third country imports at prices lower than those from Norway had an impact on domestic prices and must be considered in determining whether the price depression is the effect of the dumped imports or the effect of imports of lower priced salmon from other sources. Finally, Article 3:1 required the investigating authorities to consider the consequent impact of these imports on domestic producers of the like product. Article 3:3 provided guidance on how the investigating authorities were to determine the consequent impact and required them to consider all relevant economic factors, including market share and factors affecting domestic prices. Third country imports affected both the market share of the domestic producers and the domestic prices and should thus appropriately be considered under these Articles.
242. Responding to a question of the Panel, Norway explained as follows its views on how the information which it had provided on the expansion of the domestic salmon market in the United States was legally relevant to an examination of whether the USITC had examined the volume of imports from Norway in a manner consistent with the requirement of Articles 3:1 and 3:2 of the Agreement. Articles 3:1 and 3:2 together provided for a requirement of an objective examination of the volume of dumped imports relative to production or consumption in the importing country. Domestic production and market share must be considered as part of such an examination. Moreover, Article 3:1 also required an objective examination of the impact of the dumped imports on prices in the domestic market.
[Page 72]
The increasing supply of domestic salmon could also have an effect on the prices in the domestic market and must be considered to determine the impact of the dumped imports. Finally, Article 3:1 required the authorities to make an objective examination of the consequent impact of the dumped imports on domestic producers of such products. An objective examination must include a consideration of whether the domestic producers were able to expand production and gain market share or whether domestic production or market share declined. Such factors were specifically mentioned in Article 3:3 of the Agreement. Article 3:3 stated that the examination of the impact on the domestic industry shall include an evaluation of all relevant factors including those "having a bearing on the state of the industry such as actual and potential decline in output, sales, market share...". If the decline in such factors was relevant, so was the increase.
243. The United States noted that the USITC had determined that there had been a flood of exports from Norway to the United States in 1988 and 1989:
"Imports of Atlantic salmon from Norway surged from 1987 to 1989. Imports rose from 7.6 million kilogrammes in 1987, to 8.9 million kilogrammes in 1988, and then jumped further in 1989 to 11.4 million kilogrammes, for an overall increase of fully 50 per cent."159
Putting the magnitude of the increase in perspective, the USITC had noted that:
"...the amount of the increase in imports of Atlantic salmon from Norway alone was greater than the total amount of US-produced salmon shipped in harvest seasons 1988-89 or 1989-90."160
Over calendar year 1990, imports of Atlantic salmon from Norway had declined to 7.7 million kilogrammes. The USITC had considered this decline but, based on record evidence, had concluded that it was largely the result of the filing of the petition in February 1990 and the subsequent imposition of provisional countervailing duties in July 1990 and anti-dumping duties on 3 October 1990.161 The USITC had explicitly considered the Norwegian respondent's alternative explanations that the 1990 decline resulted from the appreciation of the Norwegian Krona against the dollar or the institution of a "freezing programme" by the Norwegian industry. It had found that such factors did not wholly explain the decline of imports in 1990, noting, for example, that the freezing programme resulted only in a slight decline in supplies of fresh Norwegian Atlantic salmon from 1989 to 1990. Because the decline in import volume in 1990 occurred concurrently with, and in apparent reaction to, the institution of the anti-dumping and countervailing duty investigations and imposition of provisional measures and was not the result of normal market forces, the USITC had given less weight to this decline. Moreover, even in 1990 Norway had remained by far the largest single supplier of Atlantic salmon to the United States, with Norwegian imports accounting for 42.2 per cent of the United States' market. In light of the evidence presented, the USITC had concluded that, although the relative market share of Norwegian Atlantic salmon had decreased since the investigation had begun:
"... the volumes of imports from Norway over the period of investigation, and the increases in those volumes from 1987 to 1989, are significant. The subject imports are particularly significant when viewed together with information concerning the nature of the US industry, the industry's condition over the period and information on prices for the like product."162
159 USITC Determination, pp.16-17. ↩
160 USITC Determination, p.19, footnote 83. ↩
161 USITC Determination, pp.17-18. ↩
162 USITC Determination, p.18. ↩
[Page 73]
244. The United States argued that in its analysis of the volume of imports of Atlantic salmon the USITC had done precisely what was required by Article 3:2 of the Agreement by determining that there had been a significant increase in dumped Norwegian imports, which had surged fully 50 per cent in the period 1987/1989 and had remained above their 1987 level. The United States considered that the increased imports from third countries did not in any way affect the consistency with the Agreement determination of the USITC regarding the volume of imports from Norway. Countries other than Norway had exported relatively little salmon to the United States in 1987, the first year of the period covered by the investigation of the USITC. Obviously, any increase in their exports to the United States in 1988 or 1989 would necessarily represent a relatively larger percentage growth than the growth in the already huge Norwegian imports. The facts of the case remained that Norway was the dominant factor in the United States' market throughout the investigation, both in sheer volume of imports and in import market share.
245. The United States considered that Norway's argument regarding the decline of Norwegian imports in 1990 as compared to the level of imports in 1988 was without merit in fact or in law. Despite this "decline", Norwegian salmon had been imported in the first half of 1990 at an annual rate higher than the 1988 import rate and had only declined after the imposition of provisional measures in July 1990. Accordingly, the USITC had determined that the overall 1990 decline in Norwegian imports was attributable, at least in part, to the initiation of the investigation and the imposition of provisional measures and thus warranted less weight than the significant volume increases of Norwegian imports between 1987 and 1989. Even in 1990 Norway had remained the largest single supplier of the US fresh Atlantic salmon market. The United States also observed in this context that Norway's argument overlooked the purpose of provisional duties under Article 10:1 of the Agreement: "... to prevent injury being caused during the period of investigation". It was axiomatic that the provisional duties would ameliorate injury by reducing the volume and/or raising prices of the imports under investigation. Article 10:1 would be meaningless if the investigating authorities could not take into account the injury-preventive nature of provisional duties in evaluating import volume and other evidence.
246. In response to the view of Norway that in the case under consideration the USITC had considered the significance of the volume of imports of Atlantic salmon from Norway, rather than the significance of any increase in that volume, the United States submitted that the USITC had plainly considered whether there had been a significant increase in the volume of dumped imports from Norway, as required by the Agreement. Under United States law the USITC was required to consider whether the volume of imports, or any increase in that volume, either in absolute terms or relative to production or consumption in the United States was significant. In this case, the USITC had found both that the volume of imports was significant, and that there was a significant increase in the volume of imports over the period of investigation. The USITC had linked this increase in import volume with price depression, and with the negative effects on the United States' domestic industry. For example, the USITC had found that "the sheer volume of the increase in Norwegian Atlantic salmon imports in 1989" had led to significant price depression and that "the effects of the large increase in Atlantic salmon imports from Norway during the period of investigation through 1989 are being felt presently by the young US Atlantic salmon industry in such forms as financial losses, a scaled-back size, and difficulty in obtaining capital".163 Thus, the USITC had satisfied the requirement of the Agreement to consider whether there has been a significant increase in the volume of dumped imports.
247. In response to a question by the Panel as to how the USITC had arrived at its conclusion that a number of factors mentioned by the respondents in the investigation to explain the decline of the volume of imports from Norway in 1990 were less important in causing this decline than the initiation of the anti-dumping and countervailing duty investigations and the imposition of provisional measures,
163 USITC Determination, p.21. ↩
[Page 74]
the United States noted the following. First, the USITC had referenced its long experience in the dampening effects on import levels which can be caused by an investigation, by preliminary determinations, or by the imposition of provisional measures. Second, the USITC had examined the specific circumstances surrounding the decline of the volume of imports of Atlantic salmon from Norway in 1990. It had linked the timing of the investigation to the development of import volumes, describing "the precipitous nature of the drop of the subject imports by the end of 1990, from record levels in 1989"164 The Commission had cited further evidence that the investigation had played a rôle in the decline in the volume of imports, observing that "the drop in subject imports has been most pronounced since July 1990, subsequent to Commerce's preliminary CVD determinations".165 Third, although there was no provision in the Agreement addressing the issue, the determination of the USITC had explicitly noted the two alternative explanations suggested by the Norwegian respondents for the 1990 decline in import volume from Norway: the institution of a freezing programme by the Norwegian industry, and the appreciation of the Norwegian kroner against the US dollar. The Commission determined that these factors might have played a part in, but did not entirely cause the decline. With regard to the freezing programme, the Commission had observed that this programme "is believed to have resulted only in a slight decrease in supplies of fresh Norwegian Atlantic salmon from 1989 to 1990".166 Thus, this programme could not explain the "precipitous" decline in Norwegian exports to the United States found by the Commission. As for the exchange rates, the Commission's staff report revealed that the kroner-dollar exchange rate had fluctuated strongly over the period of investigation, and yet, until 1990, there had been a steady annual increase in imports from Norway. What had been different in 1990 was the investigation itself.
248. Norway considered that the argument of the United States that the decline of the volume of imports of Atlantic salmon from Norway in 1990 was concurrent with either the initiation of the anti-dumping and countervailing duty investigations or the imposition of provisional measures in these investigations was contradicted by information on monthly import volumes for 1989 and 1990. These data demonstrated that in January and February 1990, i.e., before the filing of the petition and months before the imposition of provisional measures, imports of Atlantic salmon from Norway were 23 per cent lower than in January and February 1989. Norway also reiterated in this context that in the period September 1989-February 1990 imports of Atlantic salmon from Norway had been lower than in the period September 1988-February 1989. This was due, inter alia, to a considerable fall in the US dollar exchange rate to the Norwegian kroner. Moreover, in the Report on its investigation, the USITC had acknowledged that both the institution of a freezing programme and the appreciation of the kroner had helped to cause the decline in the volume of imports from Norway in the latter part of 1990. Nevertheless, before this Panel the United States was ignoring the evidence that imports had declined in absolute terms as well as relative to consumption during the period of investigation and was claiming that this decline was irrelevant because it had occurred after the initiation of the investigation and the imposition of provisional measures. This claim could not be supported.
249. On Norway's argument that imports from Norway had begun to decline prior to the filing of the petition on 28 February 1990, the United States observed that, as revealed by the data regarding monthly import figures167, Norway was correct that imports in January and February 1990 had been lower than imports in the immediately preceding months in late 1989. However, this short-lived decline had not marked the beginning of a longer-term pattern of decline. Rather, import volumes in the several months just after January-February 1990 had steadied or had even increased slightly. The decline in January and February 1990 had thus been a temporary phenomenon. This transitory decline
164 USITC Determination, p.18. ↩
165 USITC Determination, pp.17-18. ↩
166 USITC Determination, p.18, footnote 78. ↩
167 See Annex 3 to this Report. ↩
[Page 75]
could have occurred for a number of reasons. It was possible, for example, that the announcement of the freezing programme by the Norwegian industry in early 1990 had caused a temporary slowing-down in the volume of exports to the United States. However, the effect of this programme could not have lasted: as noted by the USITC in its Report, the freezing programme had ultimately resulted in only a slight decrease in available stocks of fresh salmon from Norway, and thus could not have accounted for the drastic decline in imports of Norwegian salmon experienced by the end of 1990.
250. The United States considered that the record did not bear out Norway's assertion that imports of Atlantic salmon from Norway in the last four months of 1989 had also been at reduced levels. As shown by the figures on monthly import volumes, imports from Norway in September through December 1989 had been at levels as high as they had ever been; in three of those four months imports from Norway had exceeded one million kilograms. In sum, the only pre-filing decline in import volume had been the short-lived drop in January-February 1990. As the Commission had noted, the effects of the investigation were most clearly seen during the second half of 1990, subsequent to the affirmative preliminary countervailing duty determination of the Department of Commerce, which had resulted in the imposition of provisional measures. Not only was the January-February period separated in time from the events in the second half of 1990, but the amount of the temporary decline in those two months paled in comparison to the magnitude of the decline which began in the second half of 1990. By December 1990 imports from Norway had been below 200,000 kilogrammes. These facts of record plainly refuted Norway's claim that unspecified events prior to the investigation caused the decline of imports during 1990.
4.3 Price effects of the imports under investigation (Articles 3:1 and 3:2)
251. Norway submitted that in determining that imports of Atlantic salmon from Norway had significantly depressed prices for the like domestic product, the USITC had acted inconsistently with the requirement in Article 3:1 of an objective examination of the effect on prices of imports under investigation, and with the requirement in Article 3:2 that the investigating authorities consider inter alia whether the effect of the dumped imports is to depress prices to a significant degree.
252. In its determination, the USITC had made the following statement regarding the price depressing effect it had found to have been caused by the imports subject to investigation:
"In sum, given the sheer volume of the increase in Norwegian Atlantic salmon imports in 1989, falling prices for those imports, closely tracking US and Norwegian Atlantic salmon price trends, and information suggesting significant substitutability between Norwegian and US Atlantic salmon, we find that imports of Norwegian Atlantic salmon have significantly depressed prices for the like product.168
253. Norway argued that the above-mentioned conclusion of the USITC regarding the existence of price depression caused by the subject imports was without any basis. In support of this view, Norway pointed to the following. As demonstrated by data in the Annex to the determination of the USITC and by data gathered by Norway, during the period of investigation Norwegian salmon had been priced at a level higher than salmon of domestic producers in the United States. In mid-1990, prices of Norwegian salmon had begun to rise. Prices of domestic salmon in the United States had not followed this rise but had actually fallen. If the USITC had been correct in its finding that prices of domestic salmon "closely tracked" those of imported Norwegian salmon, prices of domestic salmon should have risen when the prices of imported Norwegian salmon had begun to rise. As was demonstrated by data
168 USITC Determination, p.20. ↩
[Page 76]
on page A-56 of the Annex to the USITC determination, after mid-1990 the gap between prices of domestic salmon and imported Norwegian salmon had widened. Moreover, if the USITC had been correct in its conclusion that Norwegian imports had depressed domestic prices, one would logically expect that domestic prices for salmon would increase after the Norwegian imports had disappeared from the United States market in 1991. This, however, had not happened. Since February 1991, Norwegian imports of salmon had been brought to a halt as a result of the final anti-dumping duty determination of the Department of Commerce. During the first half of 1991, prices in the United States' market of salmon from all remaining suppliers had fallen as compared to the first half of 1990. This confirmed that the USITC had been incorrect in concluding that imports of Norwegian salmon had caused depression of domestic prices in the United States.
254. The United States pointed out that the USITC had found, and Norway had not contested, that prices for all Atlantic salmon in the United States market - regardless of origin - had dropped dramatically during the period of investigation: by "a third or even more between mid- to late-1988 and the end of 1989".169 The USITC had further found that United States domestic prices closely followed Norwegian prices because "US and Norwegian Atlantic salmon exhibit a high degree of substitutability, as Atlantic salmon is a near-commodity type product".170 This high degree of substitutability had further strengthened the link between the imports of Atlantic salmon from Norway and the adverse effects on domestic prices in the United States. The USITC had also determined that "the subject imports' presence in the market place, even at premium prices, acted to keep domestic producers from pricing to recover costs and meet cash flow needs".171 In short, the evidence of record overwhelmingly showed that, as found by the USITC, the decline in domestic prices in the United States was caused in large part by the large and growing glut of Norwegian imports. The USITC had observed that the collapse in prices for US Atlantic salmon closely tracked the downward spiral in prices for Norwegian salmon sold in the United States market.
255. Responding to Norway's argument that its imports had not caused the decline of domestic salmon prices in the United States because the Norwegian product was, in general, priced above the US produced Atlantic salmon, the United States considered that this argument suffered from a number of key weaknesses. First, the evidence of record before the USITC demonstrated that the sheer volume of Norwegian imports had forced prices down, a fact that Norway had not contested. Second, the evidence of record showed that Atlantic salmon was a highly substitutable product regardless of its source. Consequently, domestic producers had been forced to lower their prices in response to Norwegian price declines or face losing sales. Finally, there had been numerous instances in which Norwegian imports were priced below the prices for domestic salmon, notwithstanding the price premium that Norwegian salmon had typically commanded over the domestic product. It was for this reason, among others, that the Agreement, contrary to Norway's argument, did not require price undercutting as the basis for a finding of price depression or suppression. Rather, the Agreement provided that the administering authorities must consider whether there is significant price undercutting or significant price suppression or depression. The USITC had found the latter to exist and had come to this conclusion on the basis of the evidence of record.
256. With respect to Norway's argument that there was no relationship between Norwegian and domestic prices because domestic prices had not continued to rise after mid-1990, the United States observed that from mid-1990 onward, there had been a decline in the volume of imports of salmon from Norway, so that Norwegian salmon imports no longer provided the downward pressure that had caused all Atlantic salmon prices in the United States to decline. In any event, the divergence of US
169 USITC Determination, p.18. ↩
170 USITC Determination, p. 19. ↩
171 USITC Determination, p. 20. ↩
[Page 77]
and Norwegian prices in a period of declining market share of Norwegian imports was irrelevant to whether Norwegian production and prices had forced down US domestic prices during the earlier period. The United States noted in this context that Norway had presented extra-record information to the Panel to support its argument concerning current price levels. Such data had been compiled outside the period of investigation and were irrelevant to the proceedings before the Panel.
257. Norway contested the statement of the United States that the evidence before the USITC overwhelmingly showed that the decline in US domestic prices of Atlantic salmon had been caused in large part by the large and growing "glut" of Norwegian imports of Atlantic salmon. There had not been such a glut. Throughout the period of investigation, the US domestic market had grown faster than the volume of imports from Norway: from 1988 to 1989 apparent domestic consumption in the United States had grown by 55 per cent, while imports from Norway had increased by only 28 per cent. Norway's declining market share throughout the period of investigation thus showed that there had not been a "glut" of Norwegian imports.
258. In response to the argument of the United States that US domestic producers had been forced to lower their prices in response to price declines of the Norwegian imports or face losing sales, Norway observed that the US domestic industry had from 1987 to 1989 tripled its share of a domestic market characterized by strong growth in domestic demand, as demonstrated by the data on page A-45 of the Annex to the USITC determination. Norway also argued that the fact that domestic prices in the United States had not increased after Norwegian salmon had been excluded from the domestic market in the United States indicated that the imports from Norway were not the cause of price depression in the United States market.
259. In response to a question of the Panel as to whether Norway considered that the data on pages A-52-54 of the Annex to the USITC determination on price depression were factually incorrect or whether it considered that these data, while factually correct, did not provide evidence in support of the USITC's conclusion on price depression, Norway observed that it could not contest the correctness of data which it did not have. Pages A-52-54 of the Annex summarized some underlying data available only to the USITC. The data, even if correct, did not however provide any evidence as they did not show whether lower prices were a result of competition for increased market shares, and in that case who was leading the downward price trend, or whether the declining prices resulted from supplies beyond demand developments. However, an examination of those pages indicated that the USITC had compared prices of Norwegian salmon to prices of United States and Canadian salmon to determine price trends and price depression. Articles 3:1 and 3:2 of the Agreement required the investigating authorities to make an objective examination, based on positive evidence, of the price effects of the dumped imports in the domestic market of the importing country. The USITC had apparently not relied on positive evidence. A combination of United States and Canadian prices did not provide the requisite link between dumped imports and price depression in the domestic market of the United States. Norway referenced the Panel conclusion on Grain Corn172 which had found that Canada had not met the requirements of Article 6:2 of the Agreement on Interpretation and Application of Articles VI, XVI and XXIII of the General Agreement because it had relied on US grain corn prices instead of on Canadian prices. The Panel in that dispute had rejected the argument that US prices were sufficient even though the Canadian authorities had found that Canadian prices had tracked the US prices. Because the tables on pages A-52-54 of the Annex were not prices of United States producers, these tables did not provide evidence that imports of dumped salmon from Norway had depressed prices of domestic salmon in the United States. Thus, regardless of whether these tables accurately reflected the published weekly prices (an issue Norway could not address since it did not have the underlying data) they did not support a finding that dumped imports of Norwegian salmon had depressed United States domestic
172 SCM/140, adopted on 26 March 1992. ↩
[Page 78]
prices. Norway noted that the use of the Urner Barry price figures which combined US and Canadian prices demonstrated, at most, that Canadian prices were likely to have a profound effect on the United States prices but did not demonstrate the effect of prices of imports from Norway.
260. The United States made the following comments in response to Norway's argument that in its analysis of price depression the USITC had relied on a comparison of United States/Canadian prices with Norwegian import prices. In an effort to gather as complete pricing data as possible, the USITC had sought data on US prices from two sources. The first source was the responses to questionnaires which the Commission had sent to producers and purchasers. These data were explicitly limited to prices for US produced salmon, and did not include any Canadian prices. Thus, through the questionnaires, the USITC had specifically relied on data limited to US prices. The second set of data was published data of the Urner Barry company, an established industry authority. These data were combined United States and Canadian prices. However, the inclusion of Canadian prices in the Urner Barry figures had had no material effect on the USITC's analysis. First, the Commission had been aware that the data included Canadian prices, and had specifically addressed the issue, noting that "prices for Atlantic salmon from the two countries are believed to be comparable".173 Second, the Annex indicated that the questionnaire prices (which were limited to US prices) revealed the same trends over time, and the same pattern of overselling and underselling, as the Urner Barry data. Thus, this Annex noted that "Monthly net f.o.b. price data collected through questionnaires for US- and Norwegian-produced Atlantic salmon generally showed the same decline in price as the published price data" and "Similar to published price data and to reports from industry representatives, Norwegian importers' prices were generally higher than US producers' prices".174
261. The United States also noted in this context that, although Norway now took issue with the use by the USITC of the Urner Barry figures, the Norwegian respondents in the investigation had explicitly urged the Commission to use those figures while the matter was before the Commission. In arguing that the Commission should employ the Urner Barry data, the Norwegian respondents had described Urner Barry as "the recognized price authority in the industry".175
262. Norway contested that, as stated by the USITC on page 20 of its determination, "... until late 1990 prices for Norwegian and United States Atlantic salmon followed a very similar pattern".176 Norway noted again that it had no access to the information underlying the data on which the USITC based its conclusions. All comparisons between Norwegian price trends and domestic price trends in the United States appeared to be based on United States and Canadian price information. If the USITC had based itself on this information, its determination was not based on positive evidence. At most, this information showed that Canadian prices were likely to have a profound impact on domestic prices in the United States. The Annex to the USITC's determination stated that "United States/Canadian and Norwegian price trends for Atlantic salmon were similar from mid-1988 through mid-1989 (figures 5-7). In 1990, the two trends began to diverge."177 This statement implied that after mid-1989 the price trends in two countries had not followed a "very similar pattern". Moreover, figures 5-7 supported the interpretation that the divergence had begun in mid-1989, not in late 1990, although it had become more pronounced in late 1990. Finally, figures 8-10 in the Annex demonstrated that United States/Canadian prices had tracked Chilean prices much more closely than they had tracked Norwegian prices after mid-1989.
173 USITC Determination, p.19. ↩
174 USITC Determination, p.A-59 and p.A-66. ↩
175 Prehearing Brief of Norwegian Respondents, 20 February 1991 at p.35, n.57. ↩
176 USITC Determination, p.20. ↩
177 USITC Determination, p.A-55. ↩
[Page 79]
263. Regarding Norway's argument on the timing of the divergency of the price movements of Norwegian imported salmon and domestic salmon, the United States noted that in the Annex to the determination of the USITC it had been observed that "US/Canadian and Norwegian price trends for Atlantic salmon were similar from mid-1988 through mid-1989. In 1990, the two trends began to diverge...".178 Contrary to what Norway attempted to read into these sentences, they did not state that price trends began to diverge at any time in 1989; they stated that prices had begun to diverge in 1990. Indeed, the text of the opinion of the Commission described Norwegian and US prices as following similar trends into 1990. This was confirmed by the price charts found at pages A-56-57 of the Annex. Even a cursory examination of those charts revealed that prices for Norwegian and US Atlantic salmon had exhibited similar trends through 1989 and the early part of 1990, and had only diverged to some degree starting in the second half of 1990, during the Commission's investigation.
264. In response to a question of the Panel, Norway explained that it was not arguing that, as a matter of law, the fact that imported products were priced above domestic products precluded a finding of price depression under Article 3:2 of the Agreement. However, Article 3:2 required that it be shown that price depression was the effect of the imports under investigation. When imported products were priced above domestic products it was obviously more difficult to demonstrate that those higher priced imports had caused price depression. Norway considered that in the present case the USITC had not demonstrated that price depression had been the effect of the Norwegian imports subject to investigation.
265. Regarding the manner in which the USITC had taken account of the substitutability between Norwegian imported salmon and domestically produced salmon, Norway considered that, if all imports of Atlantic salmon were highly substitutable and imports from third countries were both lower priced and increasing their market share, the logical conclusion was that it was the lower priced product which was depressing domestic prices in the United States, not the higher priced product. If the products were highly substitutable, buyers would buy the lower priced item, not the higher-priced one. Thus, the lower priced product would be dragging down the higher prices, not vice versa. United States prices had been constrained by the increasing volume of lower priced imports, not by the higher priced imports. Norway also argued in this context that the United States had not presented any valid explanation of why domestic prices in the United States had followed the development of prices of imports from Norway instead of Norwegian suppliers having to reduce their prices due to constant price undercutting by competitors from third countries. The United States had also not provided any data demonstrating that price developments of Norwegian salmon had a time lead on price developments for salmon produced in the United States. An analysis on the basis of such data would be the best way to determine who was "depressing prices".
266. The United States argued that it was a fundamental principle that price depression could occur even when the imported product was priced above the domestic product. If two products were substitutable for each other at a given price differential, the narrowing or increasing of the differential would have an effect on the demand and/or price for each product. In this case, as the price for Norwegian salmon declined, US producers had been forced to lower their prices to maintain the differential; if they had not lowered their prices, they would have lost yet more sales to the Norwegian imports. Thus, the Commission's citation to the fact that Atlantic salmon - including Norwegian and US salmon - was a "near-commodity type product" lent support to the Commission's finding of price depression by Norwegian salmon.
267. The United States further recalled in this context that the finding by the USITC of price depression had been based on several factors, including the significant increase in the volume of imports of Atlantic salmon from Norway through 1989, the substitutability between US and Norwegian salmon,
178 USITC Determination, p.A-55. ↩
[Page 80]
and the similar price trends exhibited by US and Norwegian salmon. The USITC's determination made clear that the price depression finding was not dependant on any source being a "price leader" through undercutting the prices of other sources. Rather, the Commission's finding of price depression was grounded in increased supply of salmon to the US market, an increase to which Norwegian salmon had been the major contributor. It should come as no surprise that when supply of a commodity increased substantially, there might be adverse effects on prices. Not only did the analysis of the USITC comport fully with basic economic principles, but the Agreement expressly anticipated this kind of analysis. The Agreement mandated an examination of whether imports undercut domestic prices, "or" whether imports "otherwise" depressed or suppressed prices. Thus, price undercutting and price depression/suppression were treated in the Agreement as separate elements of an examination of price effects. A finding of price depression was not dependent on a finding of price undercutting. The present case was a good example of a situation in which imports under investigation "otherwise" depressed prices for the like product, through the imports' substantial contribution to increased market supply of a commodity type product. In sum, (1) substitutability between Norwegian and US Atlantic salmon had provided support for the Commission's finding of price depression; and (2) any notion that investigating authorities must look to see which supplier was undercutting to determine which was causing price depression was not supported by economic logic or by the text of the Agreement.
4.4 Impact of the imports under investigation on domestic producers of the like product (Articles 3:1 and 3:3)
268. Norway submitted that the analysis of the USITC of the impact of the imports under investigation on domestic producers of the like product was inconsistent with the requirements of Articles 3:1 and 3:3 of the Agreement. Article 3:1 required an objective examination of the consequent impact of the dumped imports on domestic producers, while Article 3:3 required that such an examination include an evaluation of all relevant economic factors and indices having a bearing on the state of the industry. The list of those factors contained in Article 3:3 was not exhaustive, nor could one or several of those factors necessarily give decisive guidance. Norway considered that the conclusion of the USITC regarding the negative impact of the imports on the domestic producers was unfounded. The record showed that the allegedly injured industry had experienced a most impressive growth since its start in 1984, as shown by data on the annual increases in the volume of production by the domestic producers.179 During the period 1987-89, the capacity of United States' firms to produce juvenile Atlantic salmon had risen substantially. US shipments had increased more than fourfold in this period. Production of "adult" salmon had expanded by more than 200 per cent from harvest season 1987-88 to 1989-90. Data available also showed that the number of production and related workers had increased steadily, as had the hours worked, total compensation, and hourly compensation.180 In the view of the foregoing, Norway submitted that the USITC had not carried out an objective examination of the impact on the domestic producers of the imports under investigation.
269. In response to a question of the Panel as to whether Norway considered that the information relied upon by the USITC in its analysis of the impact of the imports on domestic producers was factually incorrect, that the conclusions drawn by the USITC regarding this impact were not supported by the facts in the record of the investigation, or that the analysis by the USITC had not involved the correct application of a legal requirement imposed by the Agreement, Norway observed that it could not determine whether the information relied upon by the USITC was factually incorrect because it did not know what information the USITC had relied upon. From the information available to Norway, it did not appear that the conclusions drawn by the USITC were supported by the facts in the record of the investigation. The United States had had ample opportunity to provide the facts relied upon
179 USITC Determination, p.A-22. ↩
180 USITC Determination, p.A-29. ↩
[Page 81]
by the USITC in order to dispel Norway's belief. Finally, Norway considered that the USITC had not correctly applied a legal requirement imposed by the Agreement in that it had not made a determination based on an objective examination of positive evidence.
270. In response to a question of the Panel as to whether Norway considered that the factors which it had mentioned (supra, paragraph 268) had not been considered by the USITC or whether it was of the view that the USITC had not given adequate weight to these factors, Norway stated that Article 3:3 provided a list of factors to be examined in an analysis of the impact of imports on the domestic producers of the like product and noted that "no one or several of the factors necessarily give decisive guidance". The USITC, however, had based its conclusion regarding the impact of the imports on domestic producers on just a few financial indicators, rather than on a thorough review of all factors. Thus, the USITC had allowed a few factors to give decisive guidance.
271. On the statement of the USITC that "the financial performance of the domestic industry stands in stark contrast to the production and trade figures", Norway observed that certain facts before the USITC discounted the financial indicators as evidence of harm from dumped imports. The pre-hearing brief on behalf of the Norwegian respondents had described many other factors which affected the financial performance of the domestic producers.181 Thus, while the financial indicators might have been poor, their value as indicators of the consequent impact of subsidized imports was limited in this case.
272. The United States argued that, as required under Article 3:3, the USITC had considered the injurious impact which the volume and price effects of Norway's imports had on the domestic industry. The USITC had found that the price depressive effect of the large and increasing volume of Norwegian imports was directly reflected in the injured financial condition of United States producers:
"Lower prices for the like product have meant lower sales revenues in 1989, which contributed to substantial gross and operating losses for the domestic industry. Depressed prices have also exacerbated cash-flow pressures that are inherent in the Atlantic salmon industry."182
The USITC had described the financial condition of the domestic industry as follows:
"The financial state of the US Atlantic salmon industry declined precipitously in 1989. Net sales decreased from 1988 to 1989 while cost of goods sold rose and general, selling, and administrative costs increased. Operating losses in 1989 were enormous. US producers experienced a severe negative cash flow in 1989. The number of firms reporting operating losses increased from 1988 to 1989. For the period January-September 1990, net sales were well above the level recorded in the same period in 1989; nevertheless, the industry recorded a significant operating loss and negative cash flow. As a result of financial setbacks, the largest US producer, Ocean Products, Inc., ceased operations."183
The USITC had also noted that the domestic industry's operating losses in 1989 totalled $4.3 million, or more than half of the industry's net sales for that year.184 As a specific example of negative cash flow effects caused by depressed prices, the USITC had mentioned the experience of the largest US Atlantic salmon producer, Ocean Products, which had been forced into bankruptcy as a result of the impact of ever-decreasing prices, due to the downward spiral of Norwegian prices.
181 Pre-hearing Brief on behalf of the Norwegian Respondents, 20 February 1991, pp.27-47. ↩
182 USITC Determination, p.20. ↩
183 USITC Determination, p.14. ↩
184 USITC Determination, p.A-30, table 7. ↩
[Page 82]
273. The United States noted that the USITC had also described other negative effects of the depressed prices on the industry:
"It is likely that the leveling off of production of juvenile salmon in 1990 was a response to the depressed prices prevailing in 1989. Moreover, there is record information to suggest that banks became more unwilling to provide financing to US producers at least in part because of the low prices prevailing in the market or because of Norwegian oversupply, and that this reluctance continues."185
All of the above-mentioned effects were specifically-enumerated factors under Article 3:3. The USITC had explained that the negative price effects due to the large volume of Norwegian imports were not past effects, but were present effects that were being experienced by US producers through 1990:
"In view of the particular nature of Atlantic salmon production in the United States, the effects of the large increase in Atlantic salmon imports from Norway during the period of investigation through 1989 are being felt presently by the young US industry in such forms as financial losses, a scaled-back size, and difficulty in obtaining capital."186
In sum, the USITC had demonstrated in step-by-step fashion how the subject imports had caused material injury, first describing volume of imports from Norway, relating that volume to negative price effects in the US market, and relating those price effects to the injured condition of US producers. It had found that price depression attributable to the Norwegian imports had resulted in lower sales revenues, which in turn had caused massive financial losses, substantially decreased cash flow, and significantly diminished production of juvenile salmon.
274. The United States considered that Norway ignored the negative financial data which underlay the determination of the USITC, arguing instead that production, shipments, and certain employment data showed increases. Thus, Norway argued that the domestic industry could not have been injured. Norway's argument was without merit for three reasons. First, Norway had focused on isolated factors and bits of information, including new information which had not been on the record before the USITC. The USITC, by contrast, had considered all of the factors specified in the Agreement and all of the evidence of record in reaching its determination. Factors ignored by Norway included profits, cash flow, growth, ability to raise capital, and factors affecting domestic prices. Second, the USITC had explained why the factors that Norway had presented were consistent with a finding of material injury by pointing out that an increase in capacity, production, and employment indicators was only to be expected in a new industry, especially one where there was a delay of several years between the decision to expand production and the actual harvesting of the mature product.187 Third, Norway's argument disregarded the express admonition in Article 3:3 of the Agreement that "this list is not exhaustive, nor can one or several of these factors necessarily give decisive guidance". The USITC, by contrast, had considered all the factors specified in the Agreement and all the evidence in reaching its determination. In sum, the seemingly positive indicators cited by Norway were belied by the industry's dire financial condition, which stemmed directly from the collapse in salmon prices caused by the oversupply of Norwegian imports. The USITC had considered the factors mandated by the Agreement and had determined that the domestic industry was materially injured by reason of the dumped Norwegian imports. Its conclusions concerning the industry's condition were supported by positive evidence and were, for the most part, not even contested by Norway.
185 USITC Determination, p.21. ↩
186 USITC Determination, p.21. ↩
187 USITC Determination, p.14. ↩
[Page 83]
4.5 Causal relationship between the allegedly dumped imports and material injury to the domestic industry (Article 3:4)
275. Norway submitted that the affirmative final determination of the USITC in its investigation of imports of fresh and chilled Atlantic salmon from Norway was inconsistent with the requirements of Article 3:4 of the Agreement for the following reasons: first, the USITC had failed to isolate the effect of the allegedly dumped imports from Norway from the effects of other factors injuring the domestic industry. Second, the USITC had failed to demonstrate that the allegedly dumped imports from Norway had caused injury to the US domestic industry "through the effects of dumping". Third, the USITC had not shown that the imports from Norway had been causing material injury to the US domestic industry at the time the USITC made its determination.
4.5.1 Other factors affecting the domestic industry
276. Norway argued that an interpretation of Article 3:4 in accordance with the ordinary meaning of its terms indicated that the effects of the dumped imports, by themselves, must be sufficient to have caused material injury. The Vienna Convention on the Law of Treaties required in Article 31:1 that a treaty be interpreted in accordance with the ordinary meaning of its terms in context and in light of the object and purpose of the treaty. When Article 3:4 was read as a whole, the ordinary meaning of the phrase "through the effects of the dumping, causing injury" was that the effects of the dumped imports themselves must be causing injury. This was confirmed by the next sentence in Article 3:4 which provided that any injury caused by other factors could not be attributed to the dumped imports. Thus, according to the authoritative rules of treaty interpretation, an anti-dumping measure could not be imposed under the Agreement unless, after all injury caused by other factors was removed from consideration, material injury was caused by the effects of the dumped imports. Thus, those effects must be sufficient to cause injury in and of themselves. This interpretation of the language in Article 3:4 was consistent with the object and purpose of the Agreement which sought to prevent unjustifiable impediments to the flow of international trade, as stated in the Preamble. Therefore, anti-dumping duties were an exception to basic principles of the General Agreement and as such must be interpreted narrowly. Consequently, a strong demonstration was required that the injury to be prevented was caused by the effects of the dumped imports and thus, that the remedy would in fact offset this material injury. If the injury were to be caused by other factors, the anti-dumping duty would not offset that injury and would impede trade for no lawful purpose. Norway referenced that the standard applied by the United States did not conform to the requirement of Article 3:4. The USITC had stated that its standard of causation was to determine whether "imports are a cause of material injury". In the salmon case, the USITC had expressly relied on several US court cases which had articulated this standard. Norway mentioned in this context LMI - La Metalli Industriale, S.p.A. v. United States, 712 F. Supp. 959, 971 (CIT 1989) in which it had been stated that "it is sufficient that the imports contribute even minimally to material injury" and Maine Potato Council v. United States, 613 F. Supp. 1237), 1244 (CIT 1985) in which it had been stated that the USITC had to make an affirmative finding of injury if it found that imports were more than a "de minimis" cause of injury.
277. In response to a question of the Panel, Norway explained that it was not arguing that the causation standard of Article 3:4 of the Agreement was met only when the dumped imports were the sole cause of material injury to a domestic industry. That there could be several causes of material injury was recognized in the text of Article 3:4 and in footnote 5, which referred to other factors which might be causing injury to an industry. However, Article 3:4 stated that "injuries caused by other factors must not be attributed to the dumped imports". Read together with the requirement to demonstrate that the dumped imports, through the effects of the dumping, must be causing material injury, this meant that the dumped imports alone must be sufficient to cause material injury. This interpretation was confirmed by experts in the area. Thus, Beseler and Williams had analyzed the causation standard
[Page 84]
contained in the revised Agreement on Implementation of article VI of the General agreement (1979) as follows:
"The new Code provides more realistic criteria in that the initial requirements that the dumped imports should be 'demonstrably the principal cause' of the injury suffered by the domestic industry, outweighing all other factors combined, is now replaced by a requirement to segregate the injury caused by dumping from the injuries caused by other factors and then to make an assessment of injury caused by dumping alone."188
Thus, even though other factors may have caused more injury, the causation requirement of the Agreement was met as long as an adequate determination was made that the effects of the dumping alone - without injury caused by other factors - were sufficient to cause material injury.
278. Norway argued that in the present case the USITC had not singled out the effect of the allegedly dumped imports under investigation from the effects of other factors which had affected the domestic industry in the United States, thus potentially attributing injury caused by other factors to the dumped imports. During the consultations preceding the establishment of the Panel, Norway had asked the United States several questions aimed at determining how the USITC had distinguished between injury caused by the effects of the dumped imports and injury caused by other factors. In fact, the United States had refused to answer these questions on the ground that the questions concerned issues which might be raised in a panel.
279. Norway considered that, if the United States fresh Atlantic salmon industry had been materially injured, one or a combination of several factors not related to the subject imports accounted for the alleged material injury to that industry. Among such factors were the strong increase in imports from third countries, and growing supplies of close substitute products such as wild Pacific salmon. In support of its view that the information on Pacific salmon harvests was relevant to an examination of possible alternative causes of injury to the United States Atlantic salmon industry, Norway, responding to a question by the Panel, observed that nothing in Article 3:4 or in footnote 5 required that other factors which could cause injury to the industry be limited to sales of like products. Article 3:4 merely stated that "other factors" might be injuring the domestic industry. Footnote 5 provided an illustrative, not an exhaustive list of items which might constitute other factors in a given case. The term "like product" was found nowhere in Article 3:4 and in footnote 5. Moreover, while the USITC had found that Atlantic and Pacific salmon were not like products, it had found that there was some competition between Atlantic and Pacific salmon. Thus, the impact of Pacific salmon on the domestic Atlantic salmon industry was relevant as a possible alternative cause of injury. The effects of internal problems in the United States industry itself also did not appear to have been properly considered in light of the requirements of Article 3:4. These included problems due to mismanagement and the fact that the United States industry did not market its product on a year-round basis (as did the Norwegian industry). This of course affected continuity in contacts with purchasers. Such factors had been recognized during the proceedings before the USITC but had been disregarded when the USITC had drawn its conclusions. Thus, the USITC had concluded that:
"Although some of these factors may have adversely affected the US industry, we determine that an industry in the United States is materially injured by reason of subsidized and LTFV imports of fresh and chilled Atlantic salmon from Norway."189
188 Beseler and Williams, Anti-Dumping and Anti-Subsidy Law: The European Communities (1986), p.15. ↩
189 USITC Determination, p.22. ↩
[Page 85]
This conclusion was inconsistent with Article 3:4 under which Parties were obliged to exclude any injuries caused by factors other than the dumped imports under investigation. This necessitated a thorough examination of all possible causes of alleged injury.
280. In support of its view that Article 3:4 of the Agreement required that investigating authorities conduct a thorough examination of all possible causes of the alleged injury, Norway, responding to a question of the Panel, explained that, in order to ensure that the investigating authorities did not attribute injury caused by other factors to the effects of dumped imports, the investigating authorities must be able to segregate the effects of other factors from the effects of the dumped imports:
"Following the negotiations, the need to demonstrate that the dumped imports were the principal cause of the injury suffered was abandoned, as was the requirement to weigh the effect of the dumping against the effect of all other factors adversely affecting the industry. Instead, a new approach was adopted which consisted of isolating the injuries caused by each of the factors, including the dumping, and to treat each as a separate injury. It had then to be shown that the effect of the dumped imports was such as to cause injury within the meaning of the Code."190
In order to isolate the injuries caused by each factor, the investigating authorities must examine each such factor. Article 3:4 required that it "be demonstrated" that the effects of the dumped imports were causing material injury. This placed an affirmative obligation on the investigating authority to so demonstrate. A part of that demonstration included demonstrating that the investigating authority had not improperly attributed the injury caused by other factors to injury caused by the effects of the dumping. Nothing in the language of the Agreement created an obligation for the Party opposing the duties to demonstrate the negative, i.e., that the effects of the dumped imports were not causing material injury. In the present case, the United States had failed to provide any information on how the USITC had ensured that it did not attribute the injury caused by other factors to the effects of the dumped imports and had failed to demonstrate that the dumped imports, through the effects of the dumping, were causing material injury.
281. In response to a question of the Panel as to whether Norway considered that the possible alternative causes it had identified had not been considered by the USITC, or whether it considered that these possible alternative causes had not been given sufficient weight by the USITC, Norway stated that, while the USITC was not obliged to weigh the different factors of injury, it was required to avoid attributing to the dumped imports injury caused by other factors. While the USITC might perhaps have considered some of these other factors, it had made no effort to avoid attributing injury caused by those other factors to the effects of the dumped imports.
282. The United States argued that the determination of the USITC amply demonstrated that Norway's surging exports of Atlantic salmon to the United States had caused material injury to the domestic industry. In the face of this evidence, Norway pointed to other factors which, it believed, might have caused material injury to the industry. The USITC, however, had determined that material injury was caused by the Norwegian imports; it had expressly considered and rejected the alternative causes proffered by Norway. The determination of the USITC therefore met the requirements of Article 3:4 of the Agreement. Contrary to what was argued by Norway, Article 3:4 did not require a Party to "exclude any injuries caused by factors other than dumped imports". Rather, the Agreement admonished investigating authorities to consider whether other factors might be injuring the domestic industry. Thus, the investigating authorities must find a causal link between the imports and the injury to the
190 Beseler and Williams, Anti-Dumping and Anti-Subsidy Law: The European Communities, (1986), p.167). ↩
[Page 86]
domestic industry, a requirement reflected in both the Agreement and the United States legislation and which had been applied by the USITC in the case at hand.
283. The United States argued that in its analysis the USITC had applied the appropriate Agreement standard in finding a causal link between the dumped imports and material injury to the domestic industry. The Agreement provided that the standard was whether imports were "causing" injury. This was exactly what the USITC had found in the present case: it had found that injury to the domestic industry had been caused "by reason" of the dumped imports, or, stated in another way, that imports were a cause of injury. Norway's argument that the Agreement required the authorities to determine whether dumped imports were, by themselves, the cause of material injury found no support in the language of Article 3:4. A standard along the lines of the standard advocated by Norway had been contained in the 1967 Anti-Dumping Code, which in Article 3 provided that dumped imports must be the "principal" cause of injury. If "principal" cause was no longer the standard, it followed that imports need not be "the" cause of injury by themselves, which was an even higher standard. The test in Article 3:4 was whether dumped imports "were causing material injury within the meaning of this Agreement". The meaning of this language had to be understood in the context of the change which had occurred in the causal link standard in moving from the 1967 Code to the present Agreement. A number of commentators had concluded that the explicit removal of the "principal cause" standard in the present Agreement was a lessening of the causation standard to a standard requiring that the imports be a "contributing cause of injury".
284. In response to the points made by Norway regarding other factors which might have injured the domestic industry, the United States submitted that the USITC in its investigation had found that, although these other factors might have had an effect on the domestic industry, injury was caused by the dumped Norwegian imports. With respect to Norway's argument on imports from third countries as a possible alternative cause of injury, the United States considered that this argument ignored the dominant position held by Norway in the United States market despite the volume increases of imports from third countries. Imports in 1989 from the next largest importer, Canada, had been only one quarter those of Norway, the increase in Norway's import volume had dwarfed the increase in the volume of any other country's imports and the increase in imports from Norway was larger than the total import volume of Canada, the next largest importer. The USITC had properly focused on the overwhelming and increasing volume of Norwegian imports, rather than on the rate of increase of the volume of imports from the other, far smaller exporters of Atlantic salmon. In sum, Norway's argument was based on an invalid assumption that a smaller importer could have an injurious effect while its imports, which had represented 65 per cent of the market in 1989, did not have such an effect.
285. In the view of the United States, Norway's argument that the Pacific salmon catch had injured the domestic Atlantic salmon industry in the United States ignored the fact that Atlantic and Pacific salmon were commercially competitive only to a limited extent, as the USITC had found in defining the like product. Norway had not contested the USITC's finding that Atlantic and Pacific salmon were not like products. Norway had cited the Pacific salmon harvest totals for 1987-1989 but had failed to note that virtually all of this Pacific salmon was either frozen or canned and had thus been marketed to completely different purchasers than fresh Atlantic salmon. Norway also had failed to note that most of the remaining fresh Pacific salmon was exported from the United States and that nearly all of the 1989 increase in the Pacific salmon catch was chum or pink salmon, which were low quality fish sold in different markets than Atlantic salmon. Norway had not contested these facts; it had merely failed to note them.
286. With respect to Norway's argument that the domestic industry had been adversely affected by mismanagement, the United States considered that this argument overlooked the fact that low prices were the root cause of the industry's injured financial condition. Norway had also pointed to the US industry's marketing of Atlantic salmon on a less than year round basis. As the USITC had found,
[Page 87]
the domestic industry had been forced to sell its mature salmon right after harvest in order to maintain cash flow in the face of low prices. The inability to sell for a longer portion of the year was, therefore, a symptom of the injurious price effect of Norwegian imports rather than an alternate cause of the injury.
287. In response to a question of the Panel, the United States explained as follows how the USITC had arrived at the conclusion that, while other factors might have adversely affected the US domestic industry, the industry was materially injured by reason of imports from Norway. The USITC had conducted a thorough analysis of evidence concerning the volume of imports from Norway, their effects on prices in the United States, and their effects on US domestic producers, as provided in the Agreement. Article 3:1, 3:2 and 3:3 specifically envisioned that the focus of an investigation be on those factors. The determination of the USITC also contained findings relating to other suggested factors affecting the industry. As to non-subject imports, the USITC had found that the price depression which had injured the US industry "was due in large part to oversupply in the US market" and that it was "imports from Norway [that] accounted for a large portion of the increased imports in 1989".191 This was fully supported by the facts before the Commission. With regard to Pacific salmon, the USITC had described in detail the many differences between Atlantic salmon and Pacific salmon which restricted their substitutability - and thus their degree of competition with each other. These differences included the form in which the salmon was marketed, distribution channels, prices, and geographical and seasonal differences. Third, as to possible production difficulties or the seasonal marketing of US Atlantic salmon, the USITC had explicitly taken into account these factors which related to the industry's young age, in its determination. For example, the USITC had concluded that the industry's financial performance was "worse than would be anticipated even taking into account start-up conditions".192 In sum, the USITC's determination provided a detailed explanation of how Norwegian imports were causing material injury. This explanation had its focus on the volume of imports from Norway, their price effects, and their effects on US producers, as required by the Agreement. The determination also contained an explicit recognition of respondent's arguments concerning other factors affecting the industry, and contained findings supporting the USITC's conclusion that these other factors did not detract from the fact that imports from Norway had caused injury.
288. Norway considered that the view of the United States that Article 3:4 of the Agreement "admonishes investigating authorities to consider whether other factors may be injuring the industry" rather than requiring the investigating authorities to exclude any injuries caused by other factors rested on a clear misreading of the ordinary meaning of this provision. Article 3:4 provided in relevant part that "There may be other factors which at the same time are injuring the industry, and the injuries caused by other factors must not be attributed to the dumped imports". This sentence nowhere stated that the investigating authorities were only obliged to consider whether other factors were causing injury. Assuming arguendo that this language was ambiguous, an examination of the drafting history demonstrated that the drafters of this provision did not intend this sentence to require only a consideration of whether other factors were causing injury to a domestic industry.
289. In support of its contention on this latter point, Norway pointed out that in the Draft Subsidies Code, dated 10 July 1978 (document MTN/NTM/W/168), the provision now appearing in Article 6:4 of the Agreement on Interpretation and Application of Articles VI, XVI and XXIII of the General Agreement read as follows:
191 USITC Determination, p.19. ↩
192 USITC Determination, p.15. ↩
[Page 88]
"The subsidized products must be [an important contributing factor in causing or threatening] [a principal cause of] [the cause of] injury. All other relevant factors adversely affecting the industry shall be considered in reaching a determination."
This language indeed "admonished" the investigating authorities to consider other factors. However, this was not the final language. Had the signatories intended the interpretation proposed by the United States, they would not have changed the language to state that "injuries caused by other factors must not be attributed to the subsidized imports." The United States had presented no evidence that the causation analysis of the USITC was consistent with the requirements of Article 3:4 of the Agreement. Thus, the United States had failed to demonstrate that it had conducted an injury investigation in accordance with the requirements of Article 3:4.
290. The United States submitted that Norway's arguments regarding the requirements of Article 3:4 with respect to other factors which might be causing injury to a domestic industry were without merit in view of the text of that provision. Norway had argued that the Agreement required the investigating authorities to conduct a "thorough examination of all possible causes of alleged injury" and that, "in order to isolate the injuries caused by each factor, the investigating authorities must examine each factor". Norway had not cited any specific provision in the Agreement requiring its preferred analysis. What the Agreement stated was that investigating authorities must not attribute the effects of other factors to the effects of the subject dumped imports. It did not require any particular analysis of other factors and the language of the Agreement did not support Norway's interpretation that a "thorough examination" of each possible other factor must be undertaken.
291. The United States considered that apparently Norway's argument was that the sentence in Article 3:4 concerning other factors implied that a specific examination of all other factors was required. However, no such inference could be drawn from this language. As shown by the detailed text of Articles 3:2 and 3:3, required analyses were specifically set forth in the Agreement. The fact that no particular analysis had been set forth regarding the other factors was telling. It was not surprising that the Agreement was structured in this way. It was natural that the mandated focus of the analysis was on the effects of dumped imports, rather than on some other factors; this was what anti-dumping duty investigations were all about - the subject imports. Norway would apparently turn the issue on its head and require that the investigating authorities examine, and eliminate, all other possible factors affecting the domestic industry and then decide whether what was left was sufficient for an affirmative determination. In this respect, the standard proposed by Norway was similar to the standard found in the 1967 Anti-Dumping Code. Article 3(c) of that Code provided that "in order to establish whether dumped imports have caused injury, all other factors which, individually or in combination, may be adversely affecting the industry shall be examined". Similarly, Article 3(a) of the 1967 Anti-Dumping Code provided that "the authorities shall weigh, on one hand, the effect of the dumping and, on the other hand, all other factors taken together which may be adversely affecting the industry". This language had been dropped from the present Agreement. The 1967 Code had been denounced in Article 16 of the present Agreement.
292. The United States further submitted in this context that, to the extent there was a standard in the Agreement for not misattributing effects of other factors, it was fulfilled through an examination of the effects of the subject imports, as provided in Articles 3:1, 3:2 and 3:3 which addressed the causal link to imports. This view was amply illustrated by the recent decision of the Panel established by the Committee on Subsidies and Countervailing Measures in the dispute between the United States and Canada regarding countervailing duties imposed by Canada on grain corn from the United States.193 This Panel had found that the Canadian authorities had given primary emphasis to the effects of the
193 SCM/140, adopted on 26 March 1992. ↩
[Page 89]
United States subsidy programme on the world price for corn and had given no consideration to the effects of imports. The Panel had found a failure by Canada to meet the requirements of Articles 6:2 and 6:3 of the Agreement on Interpretation and Application of Articles VI, XVI and XXIII of the General Agreement. The Panel had also found that, because Canada had explicitly based its finding on the effects of something other than the subject imports - the world price for corn - Canada had violated the requirement of Article 6:4 not to attribute the effects of other factors to the subject imports. That decision presented a classic case in which the requirement of Article 6:4 was violated: a signatory's failure to offer any case that it was subsidized imports which were causing injury. The present case, however, was in marked contrast to the facts underlying the Panel's decision in the Grain Corn case. In the salmon case, the USITC had undertaken a detailed analysis of the effects of the subject imports - their volume, effects on prices, and consequent effects on domestic producers as required by Article 3. The findings of the USITC regarding these effects were amply supported by the evidence before the USITC.
293. The United States further pointed out that under United States legislation the effects of other factors could not support an affirmative finding of injury. In this case, the USITC had explicitly considered the other factors suggested by the Norwegian respondents, including various US industry production difficulties, non-subject imports, the inability of United States domestic producers to market their product year-round, and the effects of Pacific salmon. The USITC had ultimately determined that the subject imports from Norway had caused material injury to the domestic industry in the United States and that, while other factors might have had some adverse impact on the industry, they did not detract from the fact that Norwegian imports were injurious.
294. Norway also objected in this context to the USITC having made one collective injury determination for both the anti-dumping and the countervailing duty case. This also violated the requirement under Article 3:4 of the Agreement to exclude injuries caused by factors other than the dumped imports under investigation. The Agreement contained no basis for this type of cross-cumulation. In fact, there had been no investigation and determination concerning the alleged material injury caused by the effects of the dumped imports without regard to injury caused by the subsidized imports. Consequently, the USITC had failed to demonstrate that material injury to the domestic industry had been caused through the effects of the allegedly dumped imports.
295. The United States considered as misplaced Norway's objection to the issuance by the USITC of one injury determination for both the anti-dumping and the countervailing duty investigations. In accordance with Article 3:4, the USITC had considered whether "the effects of dumping" as defined by the Agreement, i.e., the volume and price effects of the imports on the domestic industry, as set forth in Articles 3:2 and 3:3 were "causing injury within the meaning of this Agreement". Article 6:4 of the Agreement on Interpretation and Application of Articles VI, XVI and XXIII of the General Agreement required the USITC to consider these identical factors in determining whether the effects of subsidies were causing injury. Both Agreements required the investigating authorities to evaluate the impact which the imports were having on the domestic industry, and provided substantively identical criteria for making that evaluation. The dumped and subsidized imports from Norway were one and the same. The period of investigation was identical for both the anti-dumping and countervailing duty investigations. Consequently, the import volume and price effect and impact on the domestic industry had been identical for both investigations. In view of this complete congruity between the subject imports in both investigations, Norway's argument was without any basis.
296. Responding to a question by the Panel, the United States further submitted in this context that the issuance by the USITC of one injury determination for the purpose of both the anti-dumping and the countervailing duty investigation was not incompatible with the reference in footnote 5 ad Article 3:4 to "the volume and prices of imports not sold at dumping prices" as a possible "other factor" causing injury to a domestic industry. In the present case the USITC had not, as a result of its "combined"
[Page 90]
analysis, attributed the effects of imports "not sold at dumping prices" to the dumped imports. This was because the exact same set of imports from Norway had been found to be both subsidized and dumped. Of course, even in a case in which the subsidized and dumped imports were not identical, the effects of subsidized, but not dumped imports could render the domestic industry more vulnerable to injury from the dumped imports. However, the present case did not involve differing dumped and subsidized imports.
4.5.2 Material injury caused by the dumped imports, through the effects of dumping
297. Norway further submitted that the standard applied by the USITC in the case under consideration did not conform to the requirements of Article 3:4 in that the USITC had failed to examine the effects of dumping in determining whether a domestic industry was materially injured and had only made a finding that a domestic industry was materially injured (or threatened with material injury, or the establishment of a domestic industry had been materially retarded) "by reason of imports of that merchandise". Since a domestic industry would always be more able to charge higher prices if supply was restricted (e.g. by eliminating imports), imports could always be found to be causing some injury to the domestic industry, even if minimal. Thus, the interpretation of the United States would allow the imposition of anti-dumping duties any time the domestic industry was materially injured by any cause, as long as there were imports. This would make a mockery of the causation standard in Article 3:4 and defeat the purpose of the Agreement.
298. In support of its claim that the USITC did not consider the effects of dumping in determining whether dumped imports were causing material injury to a domestic industry, Norway also pointed out that the Courts in the United States had upheld the approach of the USITC, while acknowledging that the GATT would appear to require the investigating authorities to consider the effects of the dumping. Specifically, the United States Court of International Trade, in discussing what the "effects" language in Article 3 of the Agreement required and how this language was implemented in the legislation of the United States had held:
"Whatever the ideal embodied in GATT, Congress has not simply directed ITC to determine directly if dumping itself is causing injury."194
The interpretation of Article 3:4 advocated by the United States in the proceedings before this Panel would have the Panel ignore the "through the effects" clause of Article 3:4 in its entirety. Such an interpretation was inconsistent with the ordinary meaning of the words and with the drafting history of the paragraph. The Vienna Convention on the Law of Treaties required that "a treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose". Thus, international interpretive practice was to give meaning to all phrases in a text.
299. In response to a question of the Panel as to whether Norway considered that the term "through the effects of the dumping" in Article 3:4 of the Agreement required the investigating authorities to consider factors other than those identified in Articles 3:2 and 3:3, Norway submitted that the investigating authorities must certainly consider the factors listed in Articles 3:2 and 3:3 but that a consideration of only those factors was not sufficient to meet the requirements of Article 3:4. For example, it would be odd not to consider the level of dumping found to exist. This view had been recognized by United States scholars. Thus, one author had written with respect to Article 6 of the Agreement on Interpretation and Application of Articles VI, XVI and XXIII of the General Agreement that:
194 CIT Algoma Steel Corp. v. United States, 688 F. Supp. 639, 645 (CIT 1988). ↩
[Page 91]
"The GATT Subsidies Code explicitly states, 'It must be demonstrated that the subsidized imports are, through the effects of the subsidy, causing injury within the meaning of this Agreement'. This would seem to establish an international obligation to pursue a causal connection that would relate to the actual subsidization - i.e., the margin. A similar clause exists in the Anti-Dumping Code."195
Moreover, this interpretation was consistent with the object and purpose of the Agreement. The Agreement sought to prevent unjustified impediments to the flow of international trade. Consequently, the Agreement required a strong showing that the injury to be prevented was caused by the effects of dumping and thus, that the remedy (the anti-dumping measures) would, in fact offset this material injury. If the injury were to be caused by other factors, the anti-dumping measures would not offset the injury and would impede trade to no lawful purpose.
300. The United States considered that Norway erred in arguing that United States law did not require a consideration of the effects of dumping. In fact, both the Agreement and the United States' law required the USITC to consider identical factors in examining the effect of the dumping on the domestic industry. Specifically, both required an evaluation of the volume and price effects of the imports on the domestic industry. Contrary to Norway's assertion, the USITC did not issue an affirmative injury determination whenever the domestic industry was injured and imports were present in the market, as numerous USITC investigations made clear. United States law required precisely what the Agreement required: that dumped imports cause material injury through volume and price effects, as specified in Article 3:2, and that material injury attributed to other causes cannot be the basis of an affirmative finding. Norway's argument was readily refuted by the number of negative determinations issued by the USITC in the circumstances described by Norway.
301. The United States argued that as indicated by footnote 4 ad Article 3:4, "the effects of dumping" referred to in Article 3:4 of the Agreement were defined in Articles 3:2 and 3:3 as the volume and price effects of the dumped imports, and the consequent impact of these imports on the domestic industry. The meaning of this language was clearly defined in the Agreement and there was no basis to attribute some other meaning to this language. Norway had not been able to define exactly what, in its view, the additional analysis was which was required by this language in Article 3:4 and its imprecise method of construing the Agreement stood in contrast to the plain meaning construction put forth by the United States. Contrary to Norway's assertion, the United States was not asking the Panel to disregard the "through the effects of the dumping" language in Article 3:4. Rather, the United States asked the Panel to give that language the precise meaning set forth in the Agreement: the "effects of dumping" were measured through the volume and price effects of the imports and their impact on the domestic industry. While the Agreement specifically defined the meaning of the term "through the effects of dumping" and contained two paragraphs concerning the analysis of imports, it provided no guidance concerning the interpretation of this term beyond analysis of the imports. If the Agreement had required an additional mode of analysis beyond that set forth in Articles 3:2 and 3:3, one would expect at least a further definition of the "effects of dumping" and some guidance on the proper analysis to assess these effects. There was none, however, providing yet another strong indication that the Agreement imposed no requirement other than an examination of import volume and price effects and the impact of the imports on domestic producers.
302. Norway noted that the interpretation of the "through the effects of ..." language in Article 3:4 advocated by the United States had been refuted by Professor Jackson as follows:
"A counter argument has been raised in connection with footnotes to these clauses. These footnotes refer to paragraphs 2 and 3 in a way that have led some to argue that the notion of an obligation
195 Jackson, The World Trading System (1989), p.242. ↩
[Page 92]
to use margin analysis has softened. However, such a conclusion appears to be somewhat improbable."196
303. The United States considered that the statements from Professor Jackson cited by Norway concerning the meaning of the term "through the effects of ..." did not analyze the text of footnote 4 but set forth a policy which Professor Jackson would like to see adopted. These proposals might be of interest to the negotiators of a new Agreement but were certainly not reflected in the text of the current Agreement.
304. Norway further argued in this context that the interpretation by the United States of the term "through the effects of ..." in Article 3:4 was inconsistent with the drafting history of that provision. Since it appeared that the United States found the wording of Article 3:4 ambiguous, it was appropriate to have recourse to the drafting history of this provision. This drafting history supported an interpretation which accorded meaning to the term "through the effects of ...". The Draft Subsidies Code dated 19 December 1978 had contained the following formulation of the provision now appearing in Article 6:4 of the Agreement on Interpretation and Application of Articles VI, XVI and XXIII of the General Agreement:
"It must be demonstrated that the subsidized imports are causing injury to the domestic industry. There may be other factors which at the same time are injuring the industry and the injuries caused by other factors must not be attributed to the subsidized imports."
This draft noted that this formulation had been developed by some but not all of the participating delegations. The mark-up of this draft at the Helsinki meeting of 12-13 February 1979 had resulted in what was virtually the final language:
"It must be demonstrated that, through the effects of the subsidy, the subsidized imports are causing injury within the meaning of this Arrangement. There may be other factors which at the same time are injuring the domestic industry, and the injuries caused by other factors must not be attributed to the subsidized imports."
Thus, the drafters had deliberately inserted the "through the effects" clause in the text of this provision. They must have intended the clause to have meaning beyond mere consideration of the imports; if not, there would have been no reason to insert this language. The interpretation advocated by the United States would read Article 3:4 to have the meaning found in draft of 19 December 1978, rather than in the final text. This could not be a proper interpretation of the Agreement requirements.
305. The United States denied that, as suggested by Norway, it considered the text of Article 3:4 to be ambiguous. On the contrary, the United States believed that this text was susceptible to only one interpretation. In any event, the documents referred to by Norway pertaining to the drafting history of Article 6:4 of the Agreement on Interpretation and Application of Articles VI, XVI and XXIII of the General Agreement did not support the conclusion drawn by Norway. Rather, they demonstrated the opposite: that the drafters considered the earlier draft standard, that "imports are causing" injury, to be substantially identical. Norway had neglected to mention the relevant footnote in document MTN/NTM/W/210, which stated that "this formulation has been developed by some but not all of the participating delegations" and that "other delegations have suggested alternate texts for consideration". The standard ultimately adopted was simply another way of stating the earlier standard.
196 Jackson, The World Trading System, (1989), p.242. ↩
[Page 93]
306. In response to a question of the Panel, Norway stated that footnote 4 ad Article 3:4 did not detract from the need to consider the effects of dumping. If Article 3:4 only required an analysis of the effects of the imports as stated in Articles 3:2 and 3:3, there would be no distinction between the determination of the existence of injury and the determination of the cause of the injury. In that case, the "through the effects of dumping" language in Article 3:4 would not have been necessary. Thus, Article 3:4 had to be interpreted to require more than a consideration of the effects of the imports as stated in Articles 3:2 and 3:3.
4.5.3 Whether the imports under investigation were causing present material injury to the domestic Atlantic salmon industry in the United States
307. Norway considered that the affirmative final determination of the USITC in its investigation of imports from Norway of fresh and chilled Atlantic salmon was inconsistent with Article 3:4 of the Agreement in that imports of Atlantic salmon from Norway had not been a cause of present material injury to the domestic industry in the United States at the time this determination was made. Article 3:4 required that it be demonstrated that the dumped imports under investigations "are ... causing" material injury. It followed from the present tense of the wording of Article 3:4 that material injury must be found to exist at the time the decision was taken to impose anti-dumping duties. The purpose of the imposition of such duties was not to punish past behaviour but to prevent future harm to the domestic industry resulting from imports which were currently causing material injury.
308. In the view of Norway, the majority of the USITC had ignored this requirement to focus on present injury caused by imports under investigation when it had given less weight to the decline in the volume of imports from Norway in 1990 than to the earlier increase in that volume. However, the acting Chairman had explicitly stated that the crucial question before the Commission was whether "material injury is being caused as of the day of our determination, not the date of the petition". She had taken this view based inter alia on relevant GATT provisions, such as Article 3:4 of the Agreement and in light of the necessity to interpret domestic legislation in conformity with international obligations of the United States.
309. Norway considered that, even if one were to assume that the domestic industry had been injured at the time of the filing of the petition in February 1990, such a conclusion was definitely not justified at the time of the final determination of the USITC in spring 1991. Norway reiterated in this respect that the market share held by Norwegian imports in the United States had been declining during the period covered by the USITC's investigation, mainly to the benefit of imports from third countries. This decline in market share had been caused inter alia by the combined effect of the large depreciation of the US dollar and declining prices in the US market. There was no evidence to suggest any kind of strategic behaviour of the exporters, as had been suggested by the voting majority of the USITC.
310. In response to Norway's argument that an affirmative final determination of injury was not justified because imports of Atlantic salmon from Norway had no longer been injuring the US domestic industry at the time of the USITC's determination, the United States made the following points. Norway reached this conclusion based on the decline in import volume and increase in prices in 1990, following the initiation of the investigation and the imposition of provisional measures. The decline in import volume was simply the expected result of the pendency of the investigation and, especially, the imposition of provisional measures, rather than of market forces. Moreover, Norway's argument ignored that the USITC had determined that the domestic industry was materially injured by Norwegian imports at the time of its determination. In particular, the USITC had pointed to the continuing injurious effects of the Norwegian imports, in the form of financial losses, reduced size, and difficulty in obtaining financing. The United States also observed that the grave financial losses suffered by the domestic industry - on the order of 50 per cent of net sales in 1989 - could not be expected to disappear some months later in early 1991. The negative effects of the industry's reduced production of young salmon
[Page 94]
which began in 1990 as a result of the price decline through 1989 was especially pernicious. Because production of Atlantic salmon for sale, i.e., the industry's capacity to produce marketable salmon, was the result of prior years' production of younger salmon, this reduction continued to injure the domestic industry throughout the period of investigation and beyond. Another ongoing negative effect cited by the USITC was the continuing reluctance of banks to lend to domestic producers.
311. In the view of the United States, the Agreement allowed Parties to take account of these continuing, present injurious effects on the industry's capacity and ability to raise capital attributable to recent imports. In an analogous context, the Agreement expressly contemplated examination of future effects of imports. Thus, Article 3 permitted the imposition of anti-dumping duties in cases in which imports had not yet caused injury but threatened to do so. If the future effects of present imports could thus be considered, it followed that the present, ongoing impact of imports which had entered in the recent past could also be taken into account.
312. The United States further considered that Norway's argument represented a flawed interpretation of the Agreement. Norway's theory would allow exporters to ensure a negative determination by reducing their exports and raising their prices. An unscrupulous exporter could guarantee the outcome of any investigation and simply resume its injurious dumped exports once a negative determination had been entered. It would make no difference that their exports had caused injury at the time the case was filed. The Agreement did not provide for such a loophole. Article 3:2 directed investigating authorities to consider whether there had been a significant increase in dumped imports and whether there had been a significant price undercutting by the dumped imports. This provision on its face permitted a retrospective analysis. Moreover, the intended consequence of provisional remedies under Article 10 was to remedy injury during an investigation, through a reduction of import volume or an increase in import prices. Norway's theory would undercut the purpose of provisional measures, for if injury were avoided within the meaning of Article 10:1, it would in all cases mandate a negative determination under Article 3:4. The Agreement did not envision such an absurd result.
313. Norway considered that the United States had mischaracterized Norway's position in arguing that Norway had concluded that there was no present injury caused by Norwegian imports based on the decline in import volume and increases in prices in 1990, following the initiation of the investigation and the imposition of provisional measures. Norway's position that there was no basis for a determination of present material injury caused by Norwegian imports at the time of the determination of the USITC was based on (1) the fact that the volume of imports from Norway had declined prior to the initiation of the investigation; (2) the decline in the market share held by the Norwegian imports throughout the period covered by the USITC's investigation; (3) the fact that Norwegian salmon commanded a price premium over United States salmon; (4) the fact that US domestic producers had tripled their market share in the same period; (5) the fact that the decline of the Norwegian import volume after the imposition of the provisional measures was essentially due to other factors such as changes in exchange rates, and (6) the failure of the United States to take action to prevent injury caused by other factors from being attributed to the imports from Norway.
5. Continued imposition of the anti-dumping duty order (Article 9:1)
314. Norway argued that the continued imposition of anti-dumping duties by the United States on imports of fresh and chilled Atlantic salmon from Norway was inconsistent with the requirements of Article 9:1 of the Agreement, which provided that an anti-dumping duty shall remain in force only as long as, and to the extent necessary, to counteract dumping which is causing material injury. At the time of the affirmative final determination of the USITC in the salmon case, no material injury was caused to the domestic industry in the United States by reason of imports from Norway. In addition, imports of Atlantic salmon from Norway were certainly no longer causing any present injury to the domestic industry in the United States. Consequently, the United States was under an obligation to terminate the imposition of anti-dumping duties on imports of salmon from Norway.
[Page 95]
315. The United States submitted that Norway's argument that the imports from Norway were not causing injury at the time of the USITC's determination was factually incorrect.197 Furthermore, as to events occurring subsequent to the completion of the investigation, there were no such facts on that issue on the record of the USITC, simply because the USITC's investigation ended within the deadline set by statute for a final determination concerning the existence of material injury. Norway could seek a review investigation by the USITC, which, if warranted, would concern later developments. In any event, a lack of further injury following imposition of an anti-dumping duty order would not be surprising since the Agreement presumed that an order might remove the injury to the domestic industry caused by the subject imports. Apparently, Norway was arguing that once an order was imposed, it must be removed immediately. This was absurd on its face.
VI. ARGUMENTS PRESENTED BY THE EEC AS AN INTERESTED PARTY
316. The EEC presented to the Panel its views on the following aspects of the dispute before the Panel: (1) the scope of the Panel's review; (2) the standing of the petitioner to request the initiation of an anti-dumping investigation on behalf of the domestic industry in the United States; and (3) the determination of dumping made by the Department of Commerce.
317. With respect to the scope of the Panel's review, the EEC noted that it appeared that some of the arguments presented to the Panel by Norway had not been raised during the course of the investigation carried out by the United States' authorities. Therefore, the United States, and equally importantly the petitioners, had not been given the opportunity to consider and address these issues. If Parties were given the right to raise before a panel matters not raised during the course of the investigation, the investigating authorities and the domestic petitioners would be placed in an impossible situation. With respect to such matters a panel proceeding would resemble a proceeding before a court of appeal where the plaintiff, i.e., the petitioner, was absent and could not present his views. The Agreement provided clear guidance on the procedures to be followed in anti-dumping investigations. Article 6 provided for the right of all parties, including the government of the exporting country concerned, to present evidence and arguments to the investigating authorities. In addition, Article 15 of the Agreement clearly set out the basis on which a panel could decided on a dispute. This Article specifically provided that the panel's examination was to take place on the basis of "a written statement of the party making the request" and on the basis of the "facts made available in conformity with appropriate domestic procedures to the authorities of the importing country". The latter provision in particular clearly indicated that only points raised during the course of the investigation could be brought before a panel. Given the status of the Government of Norway as an interested party in the investigation conducted by the United States authorities, Norway could have raised before the investigating authorities in the United States the issues now raised before the Panel.
318. With regard to the question of the standing of the petitioner to request the initiation of an investigation on behalf of the domestic industry, the EEC noted that the facts presented by the parties to the dispute seemed to be conflicting. However, it appeared that the petition had stated that most of the petitioners were members of two trade associations which had voted to support the petition and that the members of these two associations represented substantially all of the producers of fresh Atlantic salmon in the United States. In these circumstances, it would be reasonable to assume that the petition had been lodged by or on behalf of the affected industry, as defined in the Agreement. That the petition was supported by producers accounting for a major proportion of the domestic industry was confirmed by information provided to the USITC during the course of the investigation. It was not appropriate to demand a higher proof requirement with respect to the issue of the standing of a petitioner than with respect to the information to be provided by a petitioner on the existence of dumping, injury and
[Page 96]
causation. The Agreement required that sufficient evidence of the existence of these elements be provided in support of a request for the initiation of an investigation but did not require that a pre-initiation verification be carried out of the information provided on these elements. The position taken by Norway in the proceedings before this Panel would require investigating authorities to carry out an investigation to determine whether a petitioner had the requisite standing. The Agreement did not contain such a requirement.
319. With regard to the determination of dumping made by the Department of Commerce in its investigation of imports of Atlantic salmon from Norway, the EEC presented its views on the determination of normal values on the basis of constructed values, rather than on export prices to third countries, the use of the farmers' costs of production, rather than the exporters' acquisition prices as the basis for the calculation of the costs of production of salmon, the use of sampling and averaging techniques, the inclusion in the constructed normal values of minimum amounts for profits, and the comparison of average normal values to individual export prices.
320. The EEC noted that the Agreement did not provide for a hierarchy between the use of export prices to third countries and constructed normal values as a basis for the determination of the normal value in cases where the normal value could not be determined on the basis of sales prices in the domestic market of the exporting country. In the case under consideration, it appeared that the Department of Commerce had made every effort to rely on export prices to third country markets as a basis for the calculation of the normal value but that the sales to these third country markets predominantly had taken place at prices less than costs of production. This was not surprising insofar as sales to EEC countries were concerned, given that these sales had been subject to an anti-dumping investigation conducted by the EEC in which a finding of dumping and injury had been made. The EEC noted that, as was the case under the practice of the United States, the EEC did not use sales made at prices below costs of production as a basis for the determination of the normal value, if such sales took place in substantial quantities.
321. With respect to the calculation by the Department of Commerce of the costs of production of salmon based on the costs incurred by the farmers, rather than on the basis of the exporters' acquisition prices, the EEC argued that costs of production could be obtained only from the firms which actually produced the product, in this case the salmon farmers. As the exporters neither produced nor processed the product, there had been no other alternative to the United States but to base the constructed normal values on the costs of the salmon farmers. This methodology was fully in accordance with the guidance provided in the Agreement. Furthermore, it would seem inappropriate to use as a surrogate for home market prices the acquisition costs of products destined for export. In the view of the EEC, the Norwegian salmon farmers were aware that their sales to exporters would subsequently be exported, even though they might not have known the precise destination of the exports. Therefore, given that the purpose of the calculation of constructed normal values was to establish a surrogate for domestic prices in Norway, the prices charged between the farmers and exporters were not valid.
322. With respect to the sampling techniques used by the Department of Commerce in its investigation of the farmers' costs of production, the EEC observed that any investigating authority contemplating the use of a sample was faced by a dilemma, given that the sample had to be representative while at the same time the investigation normally had to be completed within one year. The important question with regard to the change to the original sample was whether the Norwegian exporters had provided reliable information in sufficient time to allow for a timely completion of the investigation. If this was not the case, the United States was entitled to rely on Articles 6:8 and 6:9 of the Agreement. In this case, there was no evidence that the sample used by the Department of Commerce had resulted in overestimated normal values. To the contrary, if 96 per cent of salmon farms in Norway were small, the simple average of the costs of production of the seven farms included in the sample underestimated the costs of production of the Norwegian farmers.
[Page 97]
323. On the issues raised by Norway regarding the use of statutory minima for profits in the calculation of constructed normal values, the EEC argued that it had to be borne in mind that the Department of Commerce did not have access to data on domestic sales in Norway of fresh Atlantic salmon and that therefore a reasonable estimate had to be made of the amount for profit. In the absence of any other information, it did not seem unreasonable to use a figure which after all represented only about 5 per cent on turnover. In addition, the EEC considered these figures as not inappropriate for the Atlantic salmon industry.
324. Finally, the EEC considered that the comparison of average normal values to individual export prices was the only methodology which prevented exporters from practising targeted dumping, i.e., offsetting dumped sales in one region by higher priced sales in other regions. In fact, the wording of Article 2:1 of the Agreement supported this type of comparison in that it stated that dumping occurred when the export price was less than the normal value of the like product. The term "normal value" as used in the Agreement referred to one value in the country of origin. It did not even require that a separate normal value be calculated for each investigated firm. This supported the view that the normal value was a benchmark against which export transactions had to be measures in order to verify whether they were dumped.
325. In response to the views presented by the EEC on the use of the farmers' costs of production as a basis for the calculation of the costs of production of salmon, Norway argued that the statements of the EEC on this issue were at variance with the manner in which the EEC had calculated the normal values in its own anti-dumping investigation of imports of Atlantic salmon from Norway. The EEC had used the acquisition prices between farmers and exporters, not the costs of production of the farmers. In September 1990, the EEC had explained that it would not use the costs of production of the farmers because the farmers did not know the final destination of the exported salmon.
326. Norway also considered that the statement by the EEC that there was no evidence that the sampling methodology used by the Department of Commerce had led to overestimated constructed values was in contradiction with the information gathered by the EEC in its own anti-dumping investigation of Atlantic salmon from Norway on the costs of production of salmon in 1989. In its calculation of the costs of production of the farmers, the Department of Commerce had relied on a sample of seven farms. By contrast, the EEC had for this purpose used a sample of 41 farms. Estimates showed that the costs of production calculated by the United States were approximately 50 per cent higher than the costs of production calculated by the EEC for the same year. Thus, based on the analysis by the EEC in its own anti-dumping investigation of imports of Atlantic salmon from Norway, the costs of production calculated by the Department of Commerce had been overstated.
327. In response to the points made by Norway in paragraph 325, the EEC referred to the explanation given in Commission Decision 91/142/EEC of 15 March 1991 of the method used by the EEC to calculate normal values. Paragraph 10 of that Decision stated that the usual method of calculating normal value on the basis of transactions by the producers was not appropriate in this case. Since the producers sold their entire output to exporters for resale on both domestic and export markets, the prices used were those of the exporters. And as practically all domestic sales by these firms were made at a loss, within the meaning of Article 2(4) of Council Regulation (EEC) No. 2423/88, normal value was reconstructed on the basis of the production costs of the Norwegian salmon farmers, to which were added their profits, the overheads of the exporters investigated and a reasonable profit margin on the resale activities.
[Page 98]
VII. FINDINGS
1. INTRODUCTION
328. The Panel noted that the issues before it arise essentially from the following facts: On 12 April 1991, the United States imposed an anti-dumping duty order on imports of fresh and chilled Atlantic salmon from Norway following an affirmative final determination of dumping by the United States Department of Commerce and an affirmative final determination of injury by the United States International Trade Commission (USITC) with respect to these imports. The investigation leading to these determinations was initiated by the Department of Commerce on 20 March 1990 in response to a petition for the initiation of an investigation submitted by the Coalition for Fair Atlantic Salmon Trade, comprised of domestic producers of fresh and chilled Atlantic salmon.
329. Norway requested the Panel to find that the imposition by the United States of the antidumping duty order was inconsistent with the obligations of the United States under the Agreement on Implementation of Article VI of the General Agreement on Tariffs and Trade (the "Agreement"). In particular, Norway requested the Panel to find that:
Norway requested that the Panel recommend that the Committee request the United States to revoke the anti-dumping duty order and reimburse any anti-dumping duties paid.
330. The United States requested that the Panel:
2. PRELIMINARY OBJECTIONS
331. The United States had raised two general types of objections to admissibility of certain claims of Norway.198 Firstly, the United States had argued that certain claims of Norway were not admissible because Norway had not raised them during the consultations and the conciliation phase which had preceded the establishment of the Panel, and that certain claims of Norway were not within the terms of reference of the Panel. Secondly, the United States had argued with regard to certain other claims by Norway that the failure of Norway or private Norwegian respondents to raise these claims before the investigating authorities and "exhaust administrative remedies" precluded Norway from raising them before the Panel.
198 Supra, paragraphs 30-67. ↩
[Page 99]
332. The Panel considered the first group of these objections in the light of the provisions of Article 15:2 through 15:7 of the Agreement concerning consultation, conciliation and panel proceedings. The Panel noted that in each paragraph the drafters of the text had chosen to refer to the subject matter of the dispute in identical terms as "the matter". Consultations would be requested under Article 15:2 "with a view to reaching a mutually satisfactory resolution of the matter"; if a Party considered that such consultations failed to achieve a mutually agreed solution it could refer "the matter" to the Committee for conciliation; in conciliation, the Committee would meet "to review the matter"; and if no mutually agreed solution emerged, a panel had to be established "to examine the matter" if any party to the dispute so requested. This choice of words reflected, in the view of the Panel, the decision to establish a three-step process of settlement of a dispute between Parties concerning a single "matter" and the individual claims of which a matter is composed, in which panel examination of a matter would be preceded by consultations concerning that same matter and conciliation concerning that same matter.
333. The Panel further observed that at the consultation phase, the parties to a dispute were required to consult and thereby provide at least an opportunity for reaching a mutually satisfactory resolution of the matter in dispute. At the conciliation phase, during the Committee's review of the matter, the parties to the dispute were required to go further and "make their best efforts to reach a mutually satisfactory solution throughout the period of conciliation." The Panel therefore considered that the Agreement provided that before a party to a dispute could request a panel concerning a matter, the parties to the dispute had to have been given an opportunity to reach a mutually satisfactory resolution of the matter. This condition would not be meaningful unless the matter had been raised in consultations and conciliation.
334. The Panel noted that Paragraph 4 of the Understanding regarding Notification, Consultation, Dispute Settlement and Surveillance, which applies mutatis mutandis to disputes under the Agreement by virtue of Article 15:7 of the Agreement, provided that "Any requests for consultations should include the reasons therefor". The Panel however considered that whereas the greatest degree of precision could be expected in the definition of specific claims in a panel request, the complaining Party could not be expected to define its specific claims with the same degree of precision at the time of its request for consultations.
335. With reference to conciliation, the Panel further noted the provisions of Footnote 15 to Article 15:3, that "the Committee may draw Parties' attention to those cases in which, in its view, there are no reasonable bases supporting the allegations made". The Panel also noted that Article 15:5 referred to a "detailed examination by the Committee under paragraph 3". The Panel considered that these provisions implied that the conciliation process envisaged was one which would examine legal claims and their bases and in which each member of the Committee would be able to express its views on these legal issues. Such a process would not be possible unless the request for conciliation identified the matter and the claims composing it. Furthermore, the requirement to make best efforts "throughout" the conciliation period to reach a mutually satisfactory solution to the matter could not be fulfilled unless the matter had been identified at the start of the conciliation period. The Panel therefore concluded that a matter, including each claim composing that matter, could not be examined by a panel under the Agreement unless that same matter and claim had been referred to the Committee for conciliation in accordance with Article 15:3.
336. The Panel then examined the relation between the scope of the matter before it and the terms of reference. The 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 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 observed that terms of reference often were standard terms of reference,
[Page 100]
as in the present dispute, in which the definition of the matter had been supplied by a written statement prepared entirely by the complaining Party. In the light of these considerations, the Panel concluded that a matter, including each claim composing that matter, could not be examined by a panel under the Agreement unless that same matter was within the scope of, and had been identified in, the written statement or statements referred to or contained in its terms of reference. The Panel further observed that Article 15:5 provided that the Committee "shall ... establish" a panel based on such a written statement, and considered that it could therefore not be assumed that the Committee by establishing this Panel with standard terms of reference had decided that the Panel should examine any claim in the written statement, regardless of whether that claim had been the subject of consultations between the parties and conciliation in the Committee.
337. In the view of the Panel the foregoing conclusions were particularly appropriate in view of the nature of disputes concerning antidumping actions, relative to the powers accorded to panels by the Agreement. The requirement to engage in consultations and conciliation served an essential purpose in clarifying the facts and arguments in dispute, and framing the dispute concerning the matter in terms which a panel would be best equipped to resolve.
338. In light of the foregoing considerations, the Panel was of the view that, for a claim to be properly before the 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.
(1) Preliminary objections of the United States regarding matters allegedly not within the Panel's terms of reference or not raised during consultations and conciliation
339. The Panel then proceeded to examine on the basis of the preceding conclusions the first group of preliminary objections raised by the United States as to the admissibility of certain claims of Norway. The United States had argued that (i) Norway's claim regarding the use by the Department of Commerce of the FOS processing fees as "the best information available" for the calculation of processing costs of certain exporters was not within the Panel's terms of reference, and had made the same objection with regard to (ii) Norway's claim under Article 9:1 of the Agreement regarding the continued imposition of the anti-dumping order, and (iii) Norway's claims regarding the denial of national treatment under Article III of the General Agreement and the differing treatment of foreign and domestic respondents.
340. The Panel recalled that its terms of reference were "to examine, in the light of the relevant provisions of the Agreement on Implementation of Article VI of the General Agreement, the matter referred to the Committee by Norway in ADP/65 and Add.1 and to make such findings as will assist the Committee in giving recommendations or in giving rulings." The Panel noted that on page 2 of document ADP/65/Add.1 Norway did claim that the United States had overstated the constructed value "by adding arbitrarily amounts for profit and cost" and that on page 3 of this document Norway did refer to the continued imposition of the anti-dumping order and specifically to article 9:1. The Panel therefore considered that Norway's claim regarding the United States' use of FOS processing fees, and Norway's claim under Article 9:1 of the Agreement regarding the continued imposition of the anti-dumping order were within its terms of reference.
341. As for Norway's claims regarding denial of national treatment and differing treatment of domestic industry and foreign respondents in anti-dumping investigations, the Panel considered that these in effect amounted to the same claim. The Panel noted that Norway did not refer to this claim, however characterized, in ADP/65 or Add.1. The Panel noted Norway's argument that this claim was included in the reference in ADP/65/Add.1 to denial by the United States of "equitable and open procedures". The Panel however observed that, while Section III of ADP/65/Add.1 identified four specific aspects of Norway's claim regarding the lack of "equitable and open procedures", these did not include either denial of national treatment or differing treatment of domestic and foreign respondents. The Panel
[Page 101]
therefore considered that the claim in question was not identified in ADP/65 nor in Add.1, and thus reasonable notice had not been provided to the defending party nor to third parties that the claim would be raised in this dispute.
342. The Panel noted Norway's argument that the "matter" before the Panel consisted of the imposition of anti-dumping duties by the United States on imports of Atlantic salmon and that Norway's claim regarding denial of national treatment and differing treatment of domestic and foreign respondents was therefore included in this matter. The Panel however considered that the "matter" referred to the Committee by Norway in its request for the establishment of a panel (ADP/65 and Add.1) was not the imposition of anti-dumping duties by the United States on imports of fresh and chilled Atlantic salmon from Norway; rather, this "matter" consisted of the specific claims stated by Norway in these documents with respect to the imposition of these duties by the United States. The Panel considered that the logical implication of the definition advanced by Norway of the "matter" before the Panel was that whenever a panel was established in a dispute concerning the imposition of anti-dumping duties, such a panel could examine any aspect of the procedures followed and determinations made by the investigating authorities of the party which had imposed the anti-dumping duties, regardless of whether that aspect had been referred to in the complaining party's request for the establishment of a panel. There would then be practically no limit to the claims which could be raised before a panel without any advance notice to the defending party or to third parties. The Panel recalled in this connection its observations in paragraph 336 regarding the functions of panels' terms of reference.
343. The Panel noted that, insofar as Norway's claim concerning the alleged denial of national treatment was based on Article III of the General Agreement, this claim would in any event not have been covered by the Panel's terms of reference as Article III of the General Agreement was not one of the "relevant provisions of the Agreement on Implementation of Article VI" referred to in the Panel's terms of reference. The Panel concluded that it could not examine the merits of Norway's claim regarding the alleged denial of national treatment and the differing treatment of foreign and domestic respondents because this claim was not within its terms of reference.
344. The Panel then considered the preliminary objections of the United States to the admissibility of four claims raised by Norway on the basis that these issues had not been raised in consultations or conciliation. The claims in question related to: (i) the United States' use of statutorily-mandated profit percentages in computing constructed value; (ii) use of FOS processing fees for all exporters in computing constructed value; (iii) continued application of the anti-dumping order inconsistent with Article 9:1; and (iv) the alleged denial of national treatment and different treatment of foreign and domestic respondents.
345. Examining Norway's request for conciliation in document ADP/61, the Panel noted that, while this document did identify relatively specific claims pertaining to the treatment of freezing charges or the failure to take account of the perishable nature of the product, document ADP/61 contained no reference to the claim regarding the United States' use of statutorily-mandated profit percentages in computing constructed value, nor did it refer to use of FOS processing fees for all exporters. The Panel noted Norway's arguments in support of its view that these two issues had been identified by Norway during the conciliation stage of this dispute settlement proceeding.199 The Panel considered that the reference made by Norway on page 5 of document ADP/61 to the allegedly arbitrary allocation of expenses in the calculation of constructed normal values could not reasonably be interpreted to cover the claim raised by Norway regarding the use of statutory minimum amounts for profits. The Panel further considered that the reference made on page 5 of document ADP/61 to the use of best information available in an arbitrary and unwarranted manner could not reasonably be interpreted to include Norway's
199 Supra, paragraphs 64 and 66. ↩
[Page 102]
claim regarding the use by the Department of Commerce of the FOS processing fees. The Panel noted in this context that while Norway's claim regarding the processing fees concerned the treatment of expenses incurred by exporters, in document ADP/61 the issue raised by Norway concerning the use of the best information available pertained to the treatment of information on costs of farmers. Accordingly, the Panel concluded that it could not examine the substantive merits of the claim of Norway concerning the United States' use of statutorily-mandated profit percentages in computing constructed value, nor the claim of Norway concerning the use of FOS processing fees for all exporters, because these claims had not been raised in document ADP/61, Norway's request for conciliation under Article 15:3 of the Agreement.
346. The Panel noted that a claim regarding the continued application of the anti-dumping order inconsistent with Article 9:1 was stated on page 9 of document ADP/61. The Panel therefore concluded that Norway had identified this claim in its request for conciliation, and rejected the United States' objections to the admissibility of this claim based on alleged failure to raise it in conciliation. In the light of its conclusion in paragraph 343 above, the Panel did not consider it necessary to examine the argument of the United States that the issues raised by Norway regarding the denial of national treatment and the differing treatment of foreign and domestic respondents had not been raised during conciliation.
(2) Preliminary objections of the United States regarding matters not raised before the investigating authorities
347. The Panel also noted that the United States had argued, with regard to Norway's claims concerning the initiation of the antidumping duty investigation and the comparison of average normal value with individual export prices, that the failure of Norway or private Norwegian respondents to raise these issues before the investigating authorities precluded Norway from raising them before the Panel. It was not contested by the United States that these issues had been raised in Norway's request for the establishment of a panel, and had also been raised in consultations and conciliation. In the view of the United States, the principle of preclusion of arguments not raised to the administering authorities was manifest in the following provisions of the Agreement:
348. The United States had argued that the rationale behind this concept of "exhaustion of administrative remedies" was akin to the rationale behind the public international law doctrine of exhaustion of local remedies. However, when Norway argued against application of the legal doctrine of exhaustion of local remedies in this dispute, the United States had clarified that it had not sought application of this doctrine. Consequently, the issue of application of the doctrine of exhaustion of local remedies to dispute settlement under the Agreement was not before the Panel.
349. The Panel noted its finding above that the dispute settlement provisions of the Agreement in Article 15 provided for a three-step process of settlement of a dispute between Parties concerning a single "matter" and the individual claims which make up that matter, in which panel examination of a matter would be preceded by conciliation and consultations concerning that same matter. Article 15:2 provided that "if any Party considers that any benefit accruing to it, directly or indirectly, under this
[Page 103]
Agreement is being nullified or impaired, or that the achievement of any objective of the Agreement is being impeded, by another Party or Parties, it may, with a view to reaching a mutually satisfactory resolution of the matter, request in writing consultations with the Party or Parties in question." The Panel did not find in this provision any basis for it to refuse to consider a claim by a Party in dispute settlement under the Agreement merely because the subject matter of the claim had not been raised before the investigating authorities under domestic law. The Panel considered that, had the drafters of the Agreement intended a limitation on the scope of dispute settlement of the nature advocated by the United States, they would have included a clear statement to that effect in the Agreement; no such statement existed in Article 15 or elsewhere in the Agreement, nor could one be implied from the provisions of this Article.
350. The United States had cited certain Agreement provisions and argued that if a panel were to address claims of the type it had objected to on this basis, respondents and governments would be encouraged not to raise such arguments to the investigating authorities and the ability of governments to comply with these provisions would be undercut. In this respect the Panel noted that its conclusion pertained only to the question of admissibility, and did not imply that in reviewing the merits of a claim a panel should not take account of whether or not the issues to which the claim relates were raised before the investigating authorities in the domestic antidumping duty proceedings. The Panel considered therefore that a review of such claims would not in any way interfere with the ability of Parties to exercise their rights or comply with their obligations under those provisions.
351. The Panel concluded that an examination of the merits of the claims of Norway concerning the initiation of the antidumping investigation and the comparison of average normal values with individual export prices was not precluded by the alleged failure of the Norwegian Government or the Norwegian respondents to raise these issues before the investigating authorities.
3. MERITS
A. INITIATION OF THE ANTI-DUMPING INVESTIGATION
352. The Panel then turned to the merits of the issue raised by Norway with regard to the initiation of the antidumping investigation under Article 5:1 of the Agreement. Norway had argued that the initiation by the United States of the antidumping investigation was inconsistent with Article 5:1 because the United States authorities had failed to satisfy themselves before the initiation that the request for the initiation had been filed on behalf of the domestic industry.
353. In particular, Norway had argued that the findings of the panel on "United States - Imposition of Antidumping Duties on Imports of Seamless Stainless Steel Hollow Products from Sweden" (hereinafter "Swedish Steel Pipe")200 with respect to initiation of antidumping investigations were relevant to the present dispute. Norway stated that the practice applied by the Department of Commerce in this case, that unless a substantial portion of the industry comes forth to oppose a petition, the Department reasonably assumes that the industry, or 'a major proportion' thereof, supports the petition, was inconsistent with United States obligations under Article 5:1 in the light of the findings of the "Swedish Steel Pipe" panel. Norway argued that these findings meant that the Department of Commerce was required to conduct an investigation to satisfy itself that the petition was made on behalf of the industry and that there was no information on the record indicating that the United States authorities had taken any steps to satisfy themselves prior to the initiation of the investigation (or at any other time) that the petition had been filed on behalf of the industry affected. Norway also pointed to certain facts
[Page 104]
which in Norway's view called into question the petitioner's claim to act on behalf of the domestic industry.
354. The United States argued that the petition had provided a satisfactory statement of industry support. In light of the certified statement that the major proportion of the domestic industry supported the petition, and the lack of significant opposition to the petition, the Department of Commerce had, prior to initiation, considered itself to be satisfied that the petition was filed on behalf of the domestic industry. Furthermore, facts obtained by the Department of Commerce and the USITC during the investigation had supported the decision to initiate. The United States argued against reliance on the findings of the "Swedish Steel Pipe" panel because the report of this panel had not been adopted by the Committee. The United States also argued that even if this Panel should take those findings into consideration, the standards set forth in those findings had nevertheless been satisfied in the present case. In the view of the United States this case presented a factual scenario quite different from that in the "Swedish Steel Pipe" dispute.
355. The Panel noted the following facts with regard to the initiation of this investigation:
[Page 105]
and developing the Atlantic salmon farming industry and the market for its products in the United States."
356. The Panel noted that Article 5:1 provides in relevant part as follows:
"An investigation to determine the existence, degree and effect of any alleged dumping shall normally be initiated upon a written request by or on behalf of the industry9 affected. The request shall include sufficient evidence of the existence of (a) dumping; (b) injury within the meaning of Article VI of the General Agreement as interpreted by this Code and (c) a causal link between the subsidized imports and the alleged injury ..."
Since the written request in this case had been filed not "by" but "on behalf of" a domestic industry in the United States, Norway's claim concerned the requirement that a written request for the initiation of an investigation be "on behalf of" the industry affected.
357. The Panel noted that Footnote 9 identified the term "industry" as meaning "As defined in Article 4." Article 4, on "Definition of Industry," defined the term "domestic industry" as "referring to the domestic producers as a whole of the like products or to those of them whose collective output of the products constitutes a major proportion of the total domestic production of those products,...". A "written request ... on behalf of the industry affected" therefore meant a request on behalf of the domestic producers as a whole of the like products or those of them whose collective output of the products constitutes a major proportion of the total domestic production of those products.
358. The Panel then turned to the question of the duty incumbent on investigating authorities to ensure that their actions with regard to the treatment of written requests for the initiation of anti-dumping investigations were consistent with their obligations under Article 5:1. The Panel considered that, in light of the requirement in Article 5:1 that a written request be by or on behalf of the industry affected and contain certain evidence, the investigating authorities could not, consistently with Article 5:1, initiate investigations automatically in response to any written request received. The requirements of Article 5:1 clearly implied a duty for the authorities to evaluate each such written request to ascertain whether it contained the required information, and to screen out those requests that failed to provide it. The investigating authorities therefore had to evaluate whether a written request for the initiation of an investigation was made "on behalf of" the industry affected.
359. In this respect, the Panel observed that the parties had not advanced conflicting interpretations of the meaning of the term "on behalf of" in the first sentence of Article 5:1. Referring to the "Swedish Steel Pipe panel report, Norway had submitted that the requirement that a written request be "on behalf of" the industry implied that the request must have the authorization or approval of that industry before the initiation of the investigation. While the United States had argued against reliance on this unadopted panel report, it had submitted that in the case before the Panel the standard set by that report - authorization or approval by the industry - was nevertheless met. The United States had specifically
[Page 106]
argued that the written request "had provided a satisfactory statement of industry support".201 It therefore appeared to the Panel that the issue in dispute concerning the initiation of this investigation by the United States did not pertain to the interpretation of the term "on behalf of" in the first sentence of Article 5:1 but to the question of how the United States had evaluated that the written request for the initiation of this investigation had been made with the authorization or approval of the industry in question.
360. The Panel noted that the Agreement did not provide precise guidance as to the procedural steps to be taken for such an evaluation, and considered that the question of how this requirement is to be met depends on the circumstances of each particular case. In the Panel's view, this question, or in this case the steps the United States was required to take as a prerequisite to initiating an investigation, had to be evaluated on the basis of the information before the investigating authorities at the time of the initiation decision. The Panel examined whether in the case before it the United States had taken such steps as could reasonably be considered sufficient to ensure that the written request for initiation of an investigation had been made with the authorization or approval of the industry affected.
361. The Panel examined this matter on the basis of the facts in paragraph 355 and the analysis in paragraphs 356-360 above. The written request for the initiation of an anti-dumping investigation had been made with a legal certification as to its accuracy and completeness. It had been submitted by twenty-one firms representing well over the majority of all domestic production of Atlantic salmon. As of the date of the initiation decision, none of these firms had made known a change in its position; in the Panel's view, changes in position either way by firms in the domestic industry were irrelevant to its examination of the initiation decision under Article 5:1 if such changes took place after that decision had been made.
362. The Panel considered that under these circumstances, the Department of Commerce could reasonably have relied on the statements in the certified petition that these firms accounted for well over a majority of production of Atlantic salmon and that these firms supported and had authorized the petition. Under these circumstances, the Department could assume that these firms would continue to support the petition unless they had notified the Department of a change in position. Although one firm not in the petitioner group had made a statement which could be interpreted as nonsupport or opposition, as of the date of the initiation decision the twenty-one members of FAST still approved the petition and still represented well over the majority of all domestic production of Atlantic salmon. Under these circumstances the Department of Commerce could, in the Panel's view, reasonably treat this request as being "on behalf of the industry affected.".
363. The Panel therefore concluded that the initiation of the antidumping investigation was not inconsistent with the obligations of the United States under Article 5:1 of the Agreement.
364. The Panel recalled that both parties to the dispute had presented arguments regarding the relevance to this case of the report of the panel in the "Swedish Steel Pipe" dispute202 interpreting Article 5:1 of the Agreement. In the Panel's view, the "Swedish Steel Pipe" panel had not ruled out that a written request on its face could provide sufficient indication that it is "by or on behalf of" the relevant domestic industry; rather, that panel had found that in that dispute, the information presented by the United States did not permit the conclusion that such was the case. The Panel considered that in this respect the factual situation presented to it differed significantly from the factual situation presented to the "Swedish Steel Pipe" panel.
[Page 107]
B. DETERMINATION OF THE EXISTENCE OF DUMPING
365. The Panel then proceeded to examine whether the imposition by the United States of the anti-dumping duty order on imports of fresh and chilled Atlantic salmon from Norway was inconsistent with the obligations of the United States under the Agreement by reason of the affirmative final determination of dumping issued by the Department of Commerce on 25 February 1991.203
366. With regard to this affirmative final determination of dumping, Norway had presented a number of claims under three main headings: firstly, the United States had failed to follow fair and equitable procedures in making this determination. Secondly, the United States had calculated constructed normal values in a manner which overstated the margins of dumping, in violation of Articles 2:4 and 8:3 of the Agreement. Thirdly, the United States had failed to effect a fair comparison of normal values and export prices, contrary to Articles 2:6 and 8:3 of the Agreement.
367. The Panel recalled that it had concluded that certain claims raised by Norway under the first two headings were not properly before it.204
368. The Panel noted that the first group of claims presented by Norway with respect to the affirmative final determination of dumping involved an alleged failure of the United States to follow fair and equitable procedures in making this determination. It appeared to the Panel that Norway derived this requirement of fair and equitable procedures from the statement in the preamble of the Agreement that "it is desirable to provide for equitable and open procedures as the basis for a full examination of dumping cases".205 Given this reliance by Norway on a statement in the preamble of the Agreement, the Panel was faced with the question of the rôle to be accorded to the preamble for the purpose of the Panel's examination of whether the affirmative final determination of dumping was inconsistent with the obligations of the United States under the Agreement.
369. The Panel considered that the statement in the preamble relied upon by Norway could guide the Panel's interpretation of specific operative provisions of the Agreement and noted in this respect that Article 31:2 of the Vienna Convention on the Law of Treaties expressly referred to the preamble of a treaty as part of "the context for the purpose of the interpretation of a treaty". However, this statement in the preamble did not by itself constitute a legal obligation of Parties to the Agreement.206 The question of whether the United States had acted inconsistently with its obligations under the Agreement in respect of the issues raised by Norway under the first group of claims therefore had to be examined by the Panel in the first place on the basis of specific operative provisions of the Agreement which reflected the intentions stated in the preamble.
370. A number of claims presented by Norway regarding the determination of dumping related to the procedural aspects of the investigation conducted by the Department of Commerce while other claims of Norway related to the substantive methodology used by the Department in making its determination. The Panel decided to first address Norway's claims with regard to the procedural aspects of this investigation.
203 56 Fed. Reg., 25 February 1991, pp.7661-7678. ↩
204 Supra, paragrahps 343 and 345. ↩
206 "Preambular provisions, cast in general wording are generally not intended to constitute substantive stipulations. Since they are mere statements, preambles do not create any legal commitment above and beyond the actual text of the treaty." Treviranus, in Encyclopaedia of Public International Law, Vol. 7, p.394 (1984). ↩
[Page 108]
(1) Procedural issues raised by Norway with respect to the investigation conducted by the Department of Commerce
371. The Panel noted that Norway had raised two specific procedural aspects of the investigation conducted by the Department of Commerce in support of its general claim that the United States had failed to follow fair and equitable procedures: firstly, the period of time given by the Department to exporters to submit responses to a part of the questionnaire and, secondly, the onerous nature of the questionnaire and verification procedures used by the Department.207 The Panel first considered the issue raised by Norway concerning the period of time given by the Department of Commerce to respond to a part of the questionnaire.
372. Norway had argued that the United States had acted inconsistently with Article 6:1 of the Agreement because the Norwegian exporters under investigation had been given fifteen days to respond to Section A of the questionnaire issued by the Department of Commerce on 30 April 1990, instead of a period of thirty days, as provided for in a Recommendation of the Committee on Anti-Dumping Practices regarding time limits for responses to questionnaires.208 Norway had in particular contended that, as a result of the insufficient time given to respond to this Section A of the questionnaire, the Norwegian exporters had been unable to identify in an accurate manner the farms from which they had purchased salmon for export to the United States during the period of investigation. Although, at the time of the receipt of the responses to Section A of the questionnaire, the Department of Commerce had been informed by the respondents that the lists of these farms were not entirely correct, it had not taken any steps to seek corrections to these lists. However, at a later stage of the investigations the Department had concluded that the lists of farms provided by the exporters were flawed because these lists included farms which had not supplied salmon to the exporters during the period of investigation.
373. The United States had argued that the exporters had been given sufficient time to respond to Section A of the questionnaire and that the errors in the information provided by these exporters could therefore not be attributed to the allegedly insufficient period of fifteen days given to the exporters to respond to this Section. Some exporters had requested, and had been granted, an extension of this response period and the Department of Commerce had allowed exporters to correct the lists of supplying farms subsequent to their initial responses. Given that Section A was only part of the questionnaire and that the exporters had been able to amend their initial responses, the Department had acted consistently with the period of thirty days mentioned in the Recommendation of the Committee on Anti-Dumping Practices on the question of time limits for responses to questionnaires. The United States had also contested that the exporters had informed the Department of Commerce at the time of their initial submissions that the lists of farms contained in these submissions were not accurate and had stated that it was only on 30 August 1990 that the Department had been notified that these lists included some farms from which the exporters had not purchased salmon for export to the United States during the period of investigation.
374. The Panel noted that Norway's claim regarding the allegedly insufficient period of time allowed by the Department of Commerce to exporters to respond to Section A of the questionnaire was based on Article 6:1 of the Agreement. This provision stated that:
"The foreign suppliers and all other interested parties shall be given ample opportunity to present in writing all evidence that they consider useful in respect of the anti-dumping investigation in question. They shall also have the right, on justification, to present evidence orally."
207 Supra, Section V, 3.1.1 and 3.1.2 respectively. ↩
[Page 109]
The question before the Panel was whether, in granting a period of fifteen days for the responses to Section A of the questionnaire, the Department of Commerce had denied the Norwegian exporters "ample opportunity to present in writing all evidence" considered useful in respect of this investigation.
375. While Article 6:1 did not specify minimum periods of time which had to be allowed by investigating authorities to exporters to respond to questionnaires, Norway had referred to a Recommendation adopted by the Committee on Anti-Dumping Practices which provided for a period of thirty days to be given to parties to respond to questionnaires.209 The Panel considered that Norway's reference to this Recommendation could raise a question as to the legal status to be accorded to this Recommendation in the interpretation of Article 6:1 of the Agreement. However, in the light of its analysis below, the Panel did not find it necessary to pronounce itself on this question.
376. The Panel found the following facts relevant to its consideration of Norway's arguments under Article 6:1 of the Agreement. On 16 May 1990, eight Norwegian exporters under investigation had submitted their responses to the Section A questionnaire issued by the Department of Commerce on 30 April 1990. While these responses were initially due on 15 May 1990, the Department of Commerce had granted a request received on 11 May from counsel for the Norwegian respondents for a one day extension until 16 May.210 In June 1990, three Norwegian exporters informed the Department that in their responses filed on 16 May 1990 they had not properly responded to the Department's request for a list of farms with which the exporters had dealt during the period of investigation for export to the United States and had provided corrected lists of these farms.211
377. The Panel noted that the corrections provided by these three exporters in June 1990 to their initial responses to Section A of the questionnaire had not been rejected by the Department of Commerce as untimely. Thus as a matter of fact the Department had provided these exporters with more than thirty days to provide information in response to Section A of its questionnaire. Nothing in the information before the Panel indicated that, if other exporters (who were represented by the same legal counsel as the three exporters who had submitted the corrected lists) had at the same time submitted similar corrections to the initial lists of farmers provided in May 1990, the Department would have rejected such corrections. Consequently, even if Article 6:1 was interpreted in the light of the period of thirty days mentioned in the Recommendation of the Committee on Anti-Dumping Practices referred to by Norway and this period was interpreted to apply to parts of a questionnaire, the Panel could not find on the basis of the facts before it that the Department of Commerce had acted inconsistently with this provision.
378. The Panel noted Norway's argument that it was inconsistent with Article 6:1 if respondents to a questionnaire had to request for additional time to submit their responses; in the view of Norway, this improperly placed a burden on the respondents. Under Article 6:1 the burden was on the investigating authorities to provide sufficient time for respondents to submit their responses to questionnaires. The Panel considered, that under Article 6:1 investigating authorities were required to give "ample opportunity" to interested parties to present evidence in writing or, upon justification, orally. The Panel found that it could not reasonably be argued that it was inconsistent with this requirement if investigating authorities set an initial time period for responses to questionnaires and
210 Letter from David L. Binder to David Palmeter, 14 May 1990. ↩
211 Letter from David Palmeter to Robert Mosbacher on behalf of Sea Star International A/S, 12 January 1990; Letter from David Palmeter to Robert Mosbacher on behalf of Chr. Bjelland Seafood A/S, 13 June 1990, and Letter from David Palmeter to Robert Mosbacher on behalf of Salmonor A/S, 14 June 1990. ↩
[Page 110]
then left it to respondents to request an extension of this period, if considered necessary by the respondents.
379. In any event, the Panel found that the relevance of this argument to the facts of the case before it was limited. As noted above, at least three exporters had provided the Department of Commerce with corrections to their initial questionnaire responses. These corrections had been submitted well after the expiration of the initial period for the filing of the questionnaire responses. There was no information before the Panel that these exporters had been obliged to somehow make a special request to the Department to be allowed to submit these corrections. Rather, the exporters had simply submitted these corrections, and these corrections had been accepted by the Department of Commerce. As noted above, all exporters had been represented by the same legal counsel and there was nothing in the information before the Panel to indicate that corrections made by other exporters would not have been accepted. The Panel failed to see how under these circumstances the Department of Commerce had somehow put an unreasonable burden on the exporters.
380. In the light of the foregoing considerations, the Panel concluded that the United States had not acted inconsistently with its obligations under Article 6:1 of the Agreement with respect to the time period granted to the Norwegian exporters to respond to Section A of the questionnaire of the Department of Commerce.
381. The Panel then proceeded to examine Norway's argument relating to the allegedly onerous questionnaire and verification procedures Norway had, inter alia, pointed to the fact that in order to respond to the questionnaires Norwegian respondents needed to have access to computers and that during verification, the Norwegian respondents had been required to make available photocopiers. Norway had also argued that as a result of the calculation of constructed normal values of the exporters on the basis of costs of production of farmers, in those instances in which the Department of Commerce had found the responses provided by the farmers to be insufficient and had relied on "the facts available" for purposes of calculating these costs of production, the exporters had not been allowed to present their views on this information, contrary to Article 6:1 of the Agreement.
382. With respect to this claim of Norway regarding the allegedly onerous nature of the questionnaires and of the verification procedures used by the Department of Commerce, the Panel noted that there was no provision in the Agreement which specifically addressed the question of the type of technical aspects of questionnaire and verification procedures raised by Norway. For example, there was no provision in the Agreement regulating the medium in which responses to questionnaires were to be submitted by respondents. The Panel considered, however, that Article 6:8 of the Agreement could be relevant in this context. If investigating authorities made their findings in a particular case "on the basis of the facts available" within the meaning of Article 6:8, a review by a panel of whether the authorities had acted within their rights under this provision could take into account as a relevant factor the nature of the information requirements imposed by the investigating authorities on respondents.212 However, the issue raised by Norway with regard to the questionnaire and verification procedures was not presented in these terms to the Panel. Rather, it appeared to the Panel that Norway considered that the allegedly onerous questionnaire and verification procedures were per se inconsistent with the general requirement of equitable and open procedures in anti-dumping investigations. The Panel recalled in this respect its observations in paragraph 369 regarding the legal meaning to be accorded to this statement in the preamble and concluded that, as presented by Norway, the claim regarding the
212 The Panel considered that its view on Article 6:8 as the proper legal basis for a review of the type of issues raised by Norway was supported by the fact that the Committee on Anti-Dumping Practices had addressed the question of the technical aspects of questionnaire procedures in a Recommendation on the use of "the facts available" under Article 6:8 of the Agreement. See BISD 31S/283. ↩
[Page 111]
information requirements imposed by the Department of Commerce did not constitute a ground to find that the United States had acted inconsistently with its obligations under the Agreement.
383. Regarding Norway's argument that the Norwegian exporters had been denied an opportunity to present evidence, as required under Article 6:1, in those instances in which the Department of Commerce had relied upon "the facts available" for purposes of calculating costs of production of Norwegian salmon farmers, because the exporters had not had the opportunity to rebut the information used by the Department of Commerce, the Panel considered that, insofar as this argument pertained to an alleged lack of access of the exporters to relevant information, it raised an issue under Articles 6:2 and 6:3, rather than under Article 6:1 of the Agreement. While Article 6:1 provided for a right of interested parties to provide evidence (in writing, and upon justification orally), Articles 6:2 and 6:3 specifically dealt with the question of access to information for interested parties. An alleged failure of investigating authorities to provide access to relevant information therefore had to be examined under Articles 6:2 and 6:3. Even assuming that Articles 6:2 and 6:3 were not exclusive in this regard and that a failure to provide access to relevant information could also raise an issue under Article 6:1, in such a case Article 6:1 was nevertheless subject to Articles 6:2 and 6:3. Obviously, if information was not accessible to interested parties for reasons consistent with Articles 6:2 and 6:3, such a limitation of access to information could not be interpreted to amount to a violation of Article 6:1. Therefore, a limitation of access to information could lead to an infringement of Article 6:1 only if this limitation was inconsistent with the provisions of Articles 6:2 and 6:3. However, Norway had not indicated in the proceedings before the Panel what specific information regarding the calculation to the Department of Commerce of the costs of production of Norwegian salmon farmers had not been accessible to the Norwegian exporters in a manner inconsistent with Articles 6:2 and 6:3 of the Agreement.
384. Accordingly, the Panel concluded that the United States had not acted inconsistently with Article 6:1 of the Agreement with respect to the issue raised by Norway concerning the opportunities for exporters to present evidence concerning the calculation of costs of production of the Norwegian salmon farmers.
(2) Issues raised by Norway regarding the substantive methodology used by the Department of Commerce in its determination of dumping
385. The Panel then turned to Norway's claims regarding the substantive aspects of the methodology followed by the Department of Commerce in its final determination of dumping of imports of Atlantic salmon from Norway. Norway had contested the consistency with the Agreement of this determination on grounds pertaining to the following issues:
(i) use of constructed values, rather than export prices to third countries for purposes of determing normal values;
(ii) calculation of costs of production on the basis of the costs of production of the salmon farmers, rather than on the basis of the acquisition prices paid by the exporters;
(iii) sampling techniques used by the Department of Commerce in the selection of the Norwegian salmon farmers for purposes of its costs of production investigation;
(iv) use of a simple, rather than a weighted average of the costs of production data obtained on the basis of the sample;
(v) use of "the facts available" as a basis for the calculation of the costs of production of one of the Norwegian salmon farms;
[Page 112]
(vi) inclusion in the constructed normal values of a "freezing charge"; and
(vii) comparison of normal values and export prices.
(2)(i) Export prices to third countries versus constructed normal values
386. The Panel first examined the merits of Norway's claim that, by determining the normal value of the imports of Atlantic salmon under investigation on the basis of constructed values rather than on the basis of prices at which Atlantic salmon was sold for export from Norway to third countries, the United States had acted inconsistently with its obligations under the Agreement.
387. The Panel noted that in its affirmative preliminary determination of dumping in this investigation the Department of Commerce had found for seven out of the eight investigated Norwegian exporters that the volume of sales of Atlantic salmon in the domestic market in Norway was too small for these sales to constitute a viable basis for the calculation of the normal value and had determined the normal value for these exporters on the prices at which they sold Atlantic salmon to EEC markets. However, in its affirmative final determination of dumping, the Department of Commerce had calculated constructed normal values for these exporters, on the ground that their export sales of Atlantic salmon to the EEC markets had been found to be at prices less than the costs of production.213
388. Norway had not contested the determination by the Department that the domestic market in Norway was too small for home market sales to constitute a viable basis for the determination of normal values but had requested the Panel to find that the decision by the Department of Commerce in its final determination to rely on constructed normal values rather than on export prices of Atlantic salmon sold to EEC markets was inconsistent with Article 2:4 of the Agreement and with the general requirement that "equitable and open" procedures be used in anti-dumping duty investigations.
389. In support of this claim, Norway had argued that the Department had acted in an arbitrary manner by assuming that export sales of Atlantic salmon from Norway to the EEC markets were not in the ordinary course of trade. Given that the EEC was the world's largest market for Atlantic salmon, that Norway was the largest supplier to this market and that Norway's exports to the EEC accounted for more than half of all Norway's exports of Atlantic salmon, there was no basis for the finding that these export sales were not in the ordinary course of trade. In the view of Norway where, as in the case of the United States, a Party had a preference in its anti-dumping duty regulations for the use of export prices to third countries over constructed normal values (if the normal value could not be established on the basis of domestic sales prices) the Agreement did not permit that Party to arbitrarily find in a given case that export sales to third countries were not in the ordinary course of trade without examining what was the "ordinary course of trade" for the industry in question. Finally, Norway had also argued that the Department of Commerce had failed to follow in this case its established administrative practice for determining whether sales of perishable products at prices less than costs of production were to be considered to be in the ordinary course of trade. Norway had pointed out in this connection that the Department's finding that Atlantic salmon was not a perishable product was inconsistent with facts established by the Department during verification and was also in contradiction with the affirmative final determination of injury made in this investigation by the USITC, in which Atlantic salmon had been found to be a perishable product.
390. The United States had argued that under Article 2:4 of the Agreement export prices to third countries and constructed values were equally acceptable alternatives. Therefore, the Department of Commerce had not been required to consider the possible use of export prices to third countries before
213 56 Fed. Reg., 25 February 1991, pp.7663-7664. ↩
[Page 113]
resorting to constructed normal values. Article 2:4 did not make the choice of constructed normal values instead of export prices to third countries contingent upon a finding that export prices to third countries were not in the ordinary course of trade. Consequently, Norway's argument that the United States had not considered what was the ordinary course of trade in the Atlantic salmon industry failed to allege a violation of the Agreement. With respect to Norway's reference to the preference in the anti-dumping regulations of the United States for the use of export prices to third countries over constructed values, the United States had argued that under its domestic law export prices to third countries found to be below costs of production could not be used as a basis for determining normal values. In any event, any inconsistency with domestic laws and regulations was a matter to be addressed in judicial proceedings under United States domestic law, not in dispute settlement proceedings under the Agreement. Finally, the United States had argued that the Department of Commerce had not failed to consider the question of whether Atlantic salmon was a perishable product but had properly determined, based on the evidence of record, that live, unharvested Atlantic salmon was not a perishable product for purposes of the Department's costs of production analysis. This determination was not inconsistent with the finding by the USITC that dead Atlantic salmon was a perishable product.
391. In examining Norway's claim, the Panel noted that the pertinent provisions regarding the determination of normal values were Articles 2:1 and 2:4 of the Agreement. Article 2:1 provided that:
"For the purpose of this Code a product is to be considered as being dumped, i.e. introduced into the commerce of another country at less than its normal value, if the export price of the product exported from one country to another is less than the comparable price, in the ordinary course of trade, for the like product when destined for consumption in the exporting country."
Article 2:4 provided that:
"When there are no sales of the like product in the ordinary course of trade in the domestic market of the exporting country or when, because of the particular market situation, such sales do not permit a proper comparison, the margin of dumping shall be determined by comparison with a comparable price of the like product when exported to any third country which may be the highest such export price but should be a representative price, or with the cost of production in the country of origin plus a reasonable amount for administrative, selling and any other costs and for profits. As a general rule, the addition for profit shall not exceed the profit normally realised on sales of products of the same general category in the domestic market of the country of origin."
It followed from these provisions, read together, that the normal value was in the first place to be established on the basis of "the comparable price, in the ordinary course of trade, for the like product when destined for consumption in the exporting country". Only in the circumstances defined at the beginning of Article 2:4 was it permissible to resort to the use of the alternative methods for determining normal value which were specified in that provision. In those circumstances, Article 2:4 provided for the use of "a comparable price of the like product when exported to any third country..." or "the cost of production in the country of origin plus a reasonable amount for administrative, selling and any other costs and for profits." Article 2:4 thus meant that there was no order of preference between the use of export prices to a third country and a "constructed" normal value in cases where the normal value could not be established on the basis of domestic sales prices in the exporting country. This provision did not require that investigating authorities resort to the use of a constructed normal value only after having given consideration to the possible use of export prices to a third country; nor did it condition the use of constructed values upon a finding that export sales to third countries were not in the ordinary course of trade.
[Page 114]
392. The Panel noted in this connection that, as far as the absence of an order of preference between the two alternative methods for establishing the normal value was concerned, Article 2:4 of the Agreement was identical to Article VI:1(b) of the General Agreement. A Report of the Group of Experts on Anti-Dumping and Countervailing Duties, adopted on 13 May 1959, stated that:
"The Group was of the opinion that paragraph 1(b)(i) and paragraph 1(b)(ii) laid down alternative and equal criteria to be used at the discretion of the importing country but only after it had failed to establish a normal market value under paragraph 1(a) of Article VI."214
393. The Panel thus found that under Article 2:4 the United States was not under an obligation to first consider the use of export prices to third countries as a basis for the establishment of normal values before resorting to the use of constructed normal values. In the case under consideration, the Department of Commerce had decided not to establish normal values for a number of exporters on the basis of export prices to third countries after it had found that these export prices were below the costs of production. The Panel considered that nothing in the text of Article 2:4 indicated that the Department could not have relied on this factor in declining to use these export prices for the purpose of the determination of the normal values and that the text of this provision contained no criteria on the basis of which the Panel could review the methodology underlying the Department's finding that these export prices were below costs of production.
394. The Panel noted the argument presented by Norway pertaining to the preference in the anti-dumping duty regulations of the United States to use export prices to third countries rather than constructed normal values in cases where normal values could not be established by reference to domestic sales prices in the exporting country. The Panel considered that the text of Article 2:4 did not support the view that a preference under domestic regulations of a Party for one of the alternative methods for determining normal values somehow limited the discretion of Parties under this provision with respect to the choice in a particular case between the two alternative methods of establishing normal values. Consequently, even if in the case under consideration the United States had deviated from a methodology preferred under its domestic regulations (an issue on which the Panel did not consider it appropriate or necessary to pronounce itself) this would not implicate the obligations of the United States under Article 2:4 of the Agreement.
395. For similar reasons the Panel saw no merit in Norway's argument that in finding that export prices of Atlantic salmon sold for export from Norway to EEC markets were below costs of production, the Department of Commerce had failed to take account of the fact that Atlantic salmon was a perishable product. As explained above, the text of Article 2:4 did not provide specific criteria to review the methodology used by the Department of Commerce in determining that export prices of Atlantic salmon sold to EEC markets were below costs of production. An alleged failure of the Department of Commerce to follow its established practice regarding the treatment of perishable products for purposes of determining whether sales were at prices below costs of production was a matter under United States domestic law but did not implicate the obligations of the United States under Article 2:4 of the Agreement.
396. The Panel further noted in this connection that the Department of Commerce had considered whether live, unharvested Atlantic salmon was a perishable product whereas the USITC had considered whether dead Atlantic salmon was a perishable product. There was therefore no inconsistency between the conclusions reached by these different agencies as to the perishable nature of Atlantic salmon. The question of whether such inconsistency would have been legally relevant under the Agreement therefore did not present itself to the Panel.
214 BISD 8S/145, 148 (emphasis added). ↩
[Page 115]
397. Norway had also argued that the decision of the Department of Commerce not to use export prices to third countries as the basis for the establishment of the normal value of the imports of Atlantic salmon from Norway was inconsistent with the requirement that "equitable and open procedures" be used in anti-dumping procedures.
398. In this respect the Panel recalled its conclusion that a statement in the preamble of the Agreement could not constitute an independent legal basis upon which the Panel could review the consistency of the actions taken by the United States with the obligations of the United States under the Agreement.215 Norway had not presented arguments as to how the decision by the Department of Commerce not to use export prices to third countries for the purpose of the determination of normal values was in conflict with specific operative provisions of the Agreement other than Article 2:4. Therefore, as presented by Norway, this argument relating to this alleged failure of the United States to follow equitable and open procedures did not provide the Panel with a legal basis to review the consistency with the Agreement of the actions taken by the United States in deciding to use constructed normal values instead of export prices to third countries. For instance, Norway had not argued that in making this decision the Department of Commerce had violated provisions of the Agreement regarding procedural rights of parties to an investigation such as those laid down in Article 6. The Panel noted in this connection that the public notice of the affirmative final determination of dumping made in this case reflected detailed arguments of interested parties, including the Norwegian exporters, with respect to various aspects of the methodology used by the Department of Commerce in determining that export prices of Atlantic salmon to EEC markets were below costs of production; this public notice also indicated that these comments had been considered by the Department.216
399. In the light of the foregoing considerations, the Panel concluded that by using constructed normal values rather than export prices of Atlantic salmon sold to third countries for the purpose of determining normal values for seven of the exporters under investigation, the United States had not acted inconsistently with its obligations under Article 2:4 the Agreement.
(2)(ii) Calculation of costs of production on the basis of the costs of production of the salmon farmers rather than on the basis of the acquisition prices paid by the exporters of salmon
400. The Panel then proceeded to examine Norway's argument that in calculating (for purposes of determining constructed normal values) the costs of production of Atlantic salmon as the costs of production of the Norwegian salmon farmers, rather than as the acquisition prices paid by the Norwegian exporters, the United States had acted inconsistently with its obligations under the Agreement.
401. The Panel noted that in its affirmative final determination of dumping in the investigation of imports of Atlantic salmon from Norway, the Department of Commerce had calculated constructed normal values for exporters under investigation as the sum of (1) the simple average of the costs of production of a number of Norwegian salmon farmers and (2) the exporter's selling, general and administrative expenses, profit and packing.217
402. Norway had argued that the inclusion in these constructed normal values of the costs of production of the Norwegian salmon farmers instead of the acquisition prices paid by the Norwegian exporters of salmon was inconsistent with the requirement that equitable and open procedures be followed in anti-dumping investigations. In support of this argument, Norway had pointed out that the exporters set the prices of their export sales to various markets and had no knowledge of the costs of production
216 56 Fed. Reg., 25 February 1991, pp.7672-7673. ↩
217 56 Fed. Reg., 25 February 1991, p.7663. ↩
[Page 116]
of each of the many individual farmers from which they purchased salmon and that the farmers had no knowledge of the ultimate destination of the salmon sold to these exporters. Norway had also observed that in determining the costs of production of the salmon farmers, the Department of Commerce had relied on the acquisition prices paid by these farmers in their purchases of smolt (where these prices were arms-length prices) and had argued that the Department's refusal to rely on the acquisition prices paid by the exporters for the salmon purchased from the salmon farmers was inconsistent with the use of acquisition prices of smolt for the purpose of the calculation of the farmers' costs of production.
403. The United States had pointed out that under Article 2:4 of the Agreement constructed normal values had to be based on "the cost of production in the country of origin" and that, therefore, there was no basis in the Agreement for Norway's view that the Department of Commerce should have relied on the acquisition costs incurred by the Norwegian salmon exporters rather than on the costs of production of the Norwegian salmon farmers. Given that exporters did not produce Atlantic salmon, the only manner in which the Department could calculate the costs of production consistently with Article 2:4 was by using the costs of production incurred by the actual producers, i.e. the Norwegian salmon farmers. The United States had also argued that it was irrelevant in this context whether the exporters had knowledge of the costs of production of individual salmon farmers and whether the farmers had knowledge of the destination of the salmon sold to exporters.
404. With regard to the legal basis in the Agreement invoked by Norway in support of its claim with respect to this aspect of the methodology applied by the Department of Commerce, the Panel recalled its earlier observations on the reference made by Norway to the term "equitable and open procedures" in the preamble of the Agreement.218
405. In the view of the Panel, relevant to its consideration of this claim by Norway was Article 2:4, which referred as follows to the elements composing a constructed normal value:
"...the cost of production in the country of origin plus a reasonable amount for administrative, selling and any other costs and for profits." (emphasis added)
The Panel considered that, read in its context, the term "cost of production" referred to the cost of production "of the like product", i.e. in this case Atlantic salmon. The question before the Panel was whether in the circumstances of the case before it the term "cost of production in the country of origin" in Article 2:4 necessarily had to be interpreted as meaning the acquisition prices paid by the exporters, rather than the costs of production incurred by the salmon farmers.
406. It had not been disputed by the parties that Norwegian exporters of salmon did not produce salmon and that Norwegian salmon farmers did not export salmon. Norway had not argued before the Panel that the Norwegian salmon exporters, rather than the salmon farmers, were the actual producers of Atlantic salmon subject to the investigation by the Department of Commerce. The purchase of Atlantic salmon by the exporters from the salmon farmers could therefore not be considered to amount to the purchase of an input for use by the actual producers of Atlantic salmon. While the acquisition price paid by the exporters to purchase Atlantic salmon from the salmon farmers obviously represented a cost to the exporters, it would be inconsistent with the plain meaning of the term "cost of production in the country of origin" to interpret this term as requiring investigating authorities to determine the cost of production on the basis of this cost to the exporters of acquiring the product, rather than on the basis of the cost of production incurred by the actual producers of the product. The Panel noted in this connection that, while it had specifically requested Norway to present arguments as to why in
[Page 117]
the case before the Panel, the text of Article 2:4 mandated the use of acquisition prices paid by exporters, Norway had not presented such arguments.219
407. With respect to Norway's argument concerning the lack of knowledge of exporters of the costs of production of individual salmon farmers and the lack of knowledge of the farmers of the ultimate destination of their sales of Atlantic salmon, the Panel found that their was no information before it indicating that in the circumstances of this case these factors were relevant to the calculation of "cost of production in the country of origin" under Article 2:4. For instance, there was no evidence that costs of production of salmon in Norway varied by destination of the sales.
408. In the light of the foregoing considerations, the Panel concluded that by including in the constructed values the costs of production incurred by the Norwegian farmers of Atlantic salmon, rather than the costs of acquisition incurred by the Norwegian exporters of Atlantic salmon, the United States had not acted inconsistently with its obligations under Article 2:4 of the Agreement.
(2)(iii) Sampling techniques used by the Department of Commerce in the selection of the Norwegian salmon farmers for purposes of its cost of production investigation
409. The Panel then turned to the Norway's claim that the calculation by the Department of Commerce of the cost of production of the Norwegian salmon farmers was inconsistent with the obligations of the United States under Articles 2:4 and 8:3 of the Agreement as a result of the sampling methodology used by the Department of Commerce. The Panel recalled that in its calculation of constructed normal values for investigated exporters, the Department of Commerce had determined the costs of production of Norwegian salmon on the basis of a simple average of the costs of production figures obtained through a sample of seven investigated salmon farmers in Norway.220
410. Norway had argued that this sample of seven salmon farmers was statistically invalid and had resulted in an overstated cost of production figure, in violation of Articles 2:4 and 8:3 of the Agreement, in particular because of the limited number of the farmers included in the sample and the failure of the Department of Commerce to stratify the sample by size of farm. In this latter respect, Norway had argued that the largest farms in Norway had by far the lowest cost of production per kg. As evidence of the fact that the cost of production calculated by the Department of Commerce was excessive, Norway had pointed to the fact that this cost of production figure was much higher than the cost of production figure calculated by the EEC in its anti-dumping investigation of imports of Atlantic salmon from Norway and than cost of production data reported in annual surveys by the Norwegian Directorate for Fisheries.
411. The United States had argued that the Department of Commerce had initially constructed individual samples for each exporter under investigation of Norwegian farms which had supplied salmon to that exporter during the period of investigation. These individual samples together comprised a total of eleven farms. While these samples had been stratified by geographic location, the evidence before the Department indicated that, since a large majority of Norwegian farms were within a similar size range, there was no basis to stratify these samples by farm size. However, the Department of Commerce had been forced to abandon its plan to use individual samples after learning in August 1990 that the lists of farms (provided by the exporters) from which these farms had been drawn contained farms which had not actually supplied salmon to exporters during the period of investigation. Given that at that time it was too late to develop new samples and that, in any event, there was a high probability that any new samples would also include farms which had not supplied salmon to the exporters
220 56 Fed. Reg., 25 February 1991, p.7663. ↩
[Page 118]
during the period of investigation, the Department of Commerce had decided to treat the seven remaining farms as a single sample and to develop an average cost of production figure for these seven farms. The erroneous information provided by the exporters on farms from which they had purchased salmon during the period of investigation had left the Department of Commerce with no choice but to proceed on the basis of the information before it, as authorized under Article 6:8 of the Agreement. Furthermore, in the light of information before the Department regarding the size of most salmon farms in Norway, this single sample of seven farms could reasonably be considered to be representative of the Norwegian industry. The Norwegian parties to this investigation had never objected to the Department's decision to proceed with a single sample of seven farms.
412. The Panel noted that Norway's claim regarding the inconsistency of the farm sample with Articles 2:4 and 8:3 pertained not to the use of samples per se but to the consistency with the Agreement of the specific sampling methodology used by the United States under the circumstances of the case before the Panel.
413. The Panel considered that the fact that the Agreement contained no specific provisions explicitly addressing the use of sampling techniques in anti-dumping investigations did not mean that there was no basis in the Agreement upon which the Panel could review those aspects of the sampling methodology employed by the United States in this case which had been raised by Norway. The point of departure of the Panel's analysis was the text of Article 2:4 which defined the elements of a constructed normal value as:
"... the cost of production in the country of origin plus a reasonable amount for administrative, selling and any other costs for profits." (emphasis added)
The question before the Panel was whether, as a result of alleged defects of the sampling methodology used by the Department of Commerce, "the cost of production in the country of origin" had been calculated inconsistently with Article 2:4.
414. In the view of the Panel, a review of this sampling methodology had to examine whether this methodology could reasonably be considered to be sufficient to serve its stated purpose, i.e. to determine "the cost of production in the country of origin" under Article 2:4 of the Agreement. Given that, by definition, the purpose of a sample was to obtain information on the characteristics of the population from which the sample was drawn, the Panel considered that in the present case it was appropriate for it to review whether the Department of Commerce had reasonably considered facts before it relevant to the representativeness of the sample(s) in relation to the total population of Norwegian salmon farms. In resorting to a sampling procedure, for purposes of calculating the costs of production of Atlantic salmon in Norway, the Department of Commerce had to be satisfied on the basis of the information before it that the results yielded by this sampling procedure would not be significantly different than the results obtained through an investigation of the total population of Norwegian salmon farms.
415. Bearing in mind the foregoing considerations, the Panel found the following facts relevant to its examination of the issues raised by Norway with respect to the sampling technique used by the Department of Commerce.
416. When, in August 1990, the Department of Commerce had decided to investigate the costs of production of the Norwegian salmon farmers, it had first attempted to develop for each of the exporters under investigation a sample of farms which supplied Atlantic salmon to that individual exporter. A Department of Commerce Memorandum dated 4 September 1990 (which was part of the public record of this investigation and was provided to the Panel) explained the methodology followed by the Department in the construction of these individual samples. After noting that a random number generator was used to select the individual fishfarms, this memorandum stated that:
[Page 119]
"To ensure complete verification of all sampled farms in a timely manner, we limited the sample for each exporter to a maximum of two farms. For each exporter, we compiled a list of all the farms serving that exporter. Information on the record indicated that production costs in the north exceed such costs in the south. Accordingly, we stratified the group of farms of each exporter."
"Using the lists of farms, we determined the percentage of farms in each region. We then allocated the sample for each exporter on the basis of these percentages. For example, if 60% of the farms were in the north and 40% were in the south, we multiplied 60% by 2 to get 1.2, which when rounded equals 1. We multiplied 40% by 2 to get .8, which when rounded equals 1. We then selected 1 farm for each region. We limited the total sample size to the extent possible by selecting two farms for an exporter only when the allocation scheme indicated that both regions should be represented in the sample. For example, where the percentage of farms in one region was less than 25%, e.g. 20%, .20x2 = .4 (which equals 0 after rounding), we selected only one farm from the other region."
"On the basis of this methodology, we selected a total of 11 farms. A list of the farms selected for each exporter (with north/south farm percentages) is attached."221
The eleven salmon farmers referred to in this Memorandum had been provided with a cost of production questionnaire on 21 August 1990.
417. On 30 August 1990, counsel for the Norwegian respondents in this investigation sent a letter to the Department of Commerce in which it was noted inter alia that information received from seven out of the eight exporters indicated that the farmers selected for five of these exporters had supplied little, and in some cases, no salmon to the exporter in question during the period of investigation. This letter contained the following observations regarding the sampling methodology used by the Department of Commerce. First, it was argued that:
"... the 'sample' has no validity whatsoever. Choosing one farmer to represent the costs of a particular exporter makes no theoretical or practical sense. Even in theory, one farmer cannot constitute a "sample". In practice, as illustrated above, one farmer is likely to represent little or no purchase in the POI."
Second, the letter observed that:
"We understand that in selecting the farmers, the Department used the list of farmers supplied by each of the companies in the May 16 responses. In preparing those responses, which asked exporters to identify farmers from whom fish was purchased and sold in the U.S. in the POI, the companies made every effort to comply with the Department's request; however, as we have stated repeatedly in this case, the tracing of specific-export sales to specific farmers is very difficult. The lists therefore inevitably contained the names of farmers who supplied fish near the time of the POI but not necessarily in the period itself. The selection of farmers that have little or no sales to the eight exporters points up the invalidity of the "sample" approach chosen. A "sample" can have no statistical validity if one does not know what percentage of the possible universe those selected represent. Since the Department did not request further information regarding the actual purchases from fish farmers or any other additional information on the representativeness of the fish farmers, its "sample" methodology was statistically and legally invalid from the outset and remains so."222
221 Memorandum from Louis Apple to the file, 4 September 1990. ↩
222 Letter from David Palmeter to Robert Mosbacher, 30 August 1990, p.2, footnote 2. ↩
[Page 120]
The Panel further noted that in an earlier letter, dated 22 August 1990, counsel for the Norwegian respondents had provided the Department of Commerce with information on a sample of salmon farmers used by the EEC for the purpose of its anti-dumping investigation of imports of Atlantic salmon from Norway and had requested the Department to use this sample. In a letter dated 11 September 1990, the Department of Commerce had informed counsel for the Norwegian respondents of the reasons why it had decided not to use this sample.
418. The Department of Commerce had abandoned its attempt to rely on specific samples of fish farmers for each individual exporter in the light of information provided by counsel for the Norwegian respondents on 30 August 1990 that some of the farmers included in the initial samples had not supplied any salmon during the period of investigation to the exporter to whom they had been linked. This issue was discussed in a Department of Commerce memorandum dated 13 September 1990. This memorandum noted the time limits for the preliminary and final determinations to be made in this investigation and formulated the following recommendation:
"The Department should proceed to collect cost of production information from the remaining fish farms selected for the survey.
The Department can average the costs of the remaining firms in the survey.
"We do not recommend constructing a new sample because the lists of farms are flawed. There would be a substantial likelihood of our selecting additional farms that also did not sell during the period of investigation. Further, there would be insufficient time to present questionnaires, analyse and verify their responses."223
As a result, the Department had treated the seven remaining farms as a single sample and based the constructed normal values on an average of the costs of production of these seven farms.
419. The Panel then turned to Norway's criticism of the limited number of farms in the sample used by the Department of Commerce in its cost of production investigation.
420. From the information available, it appeared to the Panel that in determining the number of farms to be included in the original samples the Department of Commerce had been guided by considerations relating to the time available for the completion of its investigation within the statutory time-limits.224 While such considerations pertaining to the need for a timely completion of anti-dumping investigations were relevant and legitimate225 the Panel found it significant that there was no information before it indicating if and how, in addition to considerations regarding the time available for the completion of its investigation, in determining the number of farms to be included in the samples for the exporters, the Department of Commerce had also taken into account how many farms per exporter needed to be selected with a view to ensuring that these samples could reasonably be considered to be representative of the populations of farms in question.
421. In this connection, the Panel attached importance to the fact that from the outset the Norwegian respondents had raised a concern regarding the number of farms selected per exporter. As noted above, the respondents had pointed out that even theoretically an examination of one farm per exporter could
223 Memorandum from Richard Moreland to Francis J. Sailer, 13 September 1990. ↩
225 In this connection the Panel noted that Article 5:1 of the Agreement provided that "Investigations shall, except in special circumstances, be concluded within one year after their initiation". ↩
[Page 121]
not constitute a sample. The respondents also had urged the Department to use a sample of forty-one salmon farms developed by the EEC for purposes of its anti-dumping investigation. In addition, the Norwegian respondents had argued before the Department that there were wide variations of costs of production between individual salmon farmers in Norway and had referred in this context to information gathered by the Government of Norway in annual surveys of the profitability of the Norwegian salmon industry. The Panel considered that the Department of Commerce had thus been presented with a potentially significant issue as to the number of farms to be included in its samples for the purpose of ensuring that these samples would be representative. On the basis of the information before it, the Panel could not conclude that this issue had been properly considered by the Department.
422. The United States had argued that, when the Department in early September 1990 had decided to limit itself to an examination of the costs of production of the seven farms which had actually supplied salmon to the exporters during the period of investigation, the Norwegian respondents had not objected to this decision. The arguments presented by Norway with regard to the size of this sample therefore related to an issue not raised during the investigation. However, as far as the number of farms was concerned, the Panel found this argument unpersuasive. As noted above, the Norwegian respondents had challenged the validity of the original samples on the basis of inter alia the number of farms selected per exporter and had suggested that the Department use a sample of forty-one farms developed in the anti-dumping investigation conducted by the EEC. It could not reasonably be concluded that this concern with regard to the initial samples (in which a total of eleven farmers had been selected) was no longer relevant after the Department had decided to limit itself to an examination of the costs of production of seven out of these eleven farms. The Panel therefore considered that Norway's argument regarding the number of the farms included in the sample could not be considered to pertain to an issue which had not been raised before the Department of Commerce during the course of its investigation.
423. The Panel noted the argument of the United States that its reliance on a single sample of seven farms, after it had been forced to abandon its original plan for individual samples for each exporter, was a valid exercise of its rights under Article 6:8 of the Agreement.
424. The Panel considered in this respect that Article 6:8 could not be invoked if from the outset the information sought by investigating authorities was not of a type that would make it possible to make a determination consistent with the substantive requirements of the Agreement. The Panel therefore considered that the argument of the United States relating to Article 6:8 might have been relevant if there were information before the Panel indicating that the Department of Commerce, in constructing the original samples, had reasonably considered how many farms per exporter needed to be selected in order to obtain representative results. If, in that situation, the Department of Commerce had encountered difficulties as a result of non co-operation or erroneous information provided by interested parties, Article 6:8 might have been relevant to the Panel's examination of the consistency with Article 2:4 of the sampling procedure used by the Department of Commerce. However, this was not the factual situation before the Panel.
425. The Panel therefore found that under these circumstances the argument of the United States regarding Article 6:8 of the Agreement was not relevant to the Panel's examination of the consistency with Article 2:4 of the sampling methodology of the Department of Commerce.
426. In light of its observations set forth in paragraphs 413 and 414, the Panel concluded that the United States had acted inconsistently with its obligations under Article 2:4 of the Agreement with respect to the calculation of the cost of production in the country of origin, by reason of the apparent failure of the Department of Commerce to consider the question of the number of the farms to be included in the samples from the perspective of how the Department was to ensure that these samples would be representative.
[Page 122]
427. Having concluded that this aspect of the sampling technique used by the Department of Commerce was inconsistent with the obligations of the United States under Article 2:4 of the Agreement, the Panel considered whether, as argued by Norway, this also meant that the Department of Commerce had acted inconsistently with the obligations of the United States under Article 8:3 of the Agreement.
428. The Panel noted that Article 8:3 provided that:
"The amount of the anti-dumping duty must not exceed the margin of dumping as established under Article 2. Therefore, if subsequent to the application of the anti-dumping duty it is found that the duty so collected exceeds the actual dumping margin, the amount in excess of the margin shall be reimbursed as quickly as possible." (emphasis added)
The Panel observed that, while Article 2 governed the determination the existence and extent of dumping, Article 8:3 governed the amount of the anti-dumping duty imposed after the existence and extent of dumping had been determined under Article 2. As noted above, the Panel had found that the United States had determined the "cost of production in the country of origin" in a manner inconsistent with Article 2:4 of the Agreement. Norway's argument under Article 8:3 raised the question of whether, if a Party acted inconsistently with Article 2:4 (or with other provisions relevant to the determination of the existence of dumping), it was ipso facto acting inconsistently with the first sentence of Article 8:3.
429. The Panel considered that an inconsistency with Article 2 of a determination of dumping could possibly give rise to an issue under Article 8:3 if the phrase "as established under Article 2" in the first sentence of Article 8:3 were interpreted to mean "established in conformity with Article 2". However, the Panel did not consider it necessary to decide this issue of interpretation. While the Panel had found in the case before it that the United States had determined normal values inconsistently with Article 2:4, the Panel had no basis to pronounce itself on what margins of dumping would have resulted if the United States had not determined those normal values inconsistently with Article 2:4. Therefore, assuming that a determination of dumping could be challenged both under Articles 2 and 8:3 of the Agreement, in the case before it the Panel could not determine whether, as a result of the determination of normal values inconsistently with Article 2:4, the United States had imposed anti-dumping duties in excess of the margin of dumping in conformity with Article 2.
430. The Panel therefore concluded that, while the apparent failure of the Department of Commerce to consider how many farms needed to be included in the samples for purposes of ensuring their representativeness was inconsistent with Article 2:4, it could not find that this aspect of the methodology of the Department of Commerce was also inconsistent with Article 8:3 of the Agreement.
431. The Panel then proceeded to examine Norway's argument that the Department of Commerce had failed to stratify its sample(s) by size of farm. In this examination, the Panel was guided by the considerations set forth in paragraphs 413 and 414.
432. In reviewing the merits of this argument, the Panel examined the documents provided by the parties to the dispute which contained an explanation by the Department of Commerce of its sampling methodology and comments on this methodology made by the Norwegian respondents. The Panel concluded from this examination that the Department of Commerce had considered the question of whether the individual samples it intended to use for the exporters should be stratified by farm size and had decided that since evidence on the record indicated that most of the farms in Norway were relatively similar in size, averaging from 8,000 to 12,000 cubic meters, farm size was not a factor in creating a sample strata. This decision was well documented in the public record of this
[Page 123]
investigation.226 On the other hand, the Department had found that there was a need to develop sample strata in order to account for differences in location of farms in Norway.227
433. While the Norwegian respondents had objected to several aspects of the sampling methodology developed by the Department of Commerce, the documents before the Panel did not indicate that they had ever stated a specific concern regarding the decision by the Department not to stratify the samples by farm size. There was no information before the Panel indicating that the Norwegian respondents had argued before the Department that costs of production varied systematically by farm size and that they had contested the factual correctness of the Department's statement that most of the salmon farms in Norway were relatively similar in size. Therefore, the Panel was unable to find that, by considering that the size of the Norwegian salmon farms was not a relevant factor in the development of sample strata, the Department of Commerce had acted unreasonably in the light of the facts before it. The Department had not failed to consider this question; rather, it had determined that in the light of the facts before it, there was no need to stratify the samples by farm size. In addition, there was no evidence that at any point in the investigation the factual basis of this determination been challenged by the Norwegian respondents.
434. The Panel concluded that, in deciding not to stratify the samples of farms by farm size, the United States had not acted inconsistently with its obligations under Article 2:4 of the Agreement. The question of a possible inconsistency with Article 8:3 therefore did not arise.
(2)(iv) Use of a simple, rather than a weighted average of the cost of production data obtained on the basis of the farm sample
435. As part of its claim that the United States had determined normal values inconsistently with Article 2:4 and 8:3 of the Agreement, Norway had also raised the question of the calculation by the Department of Commerce of a simple, rather than a weighted, average of the costs of the production of the seven Norwegian salmon farms which had been included in the Department's costs of production analysis.
436. The Panel recalled that when, in early September 1990, the Department of Commerce had abandoned its attempt to develop cost of production information on the basis of individual samples of farms for each exporter under investigation it had indicated that it would "average the costs of the remaining firms in the survey."228 In the public notice of the affirmative final determination of dumping, the Department of Commerce had stated the following in response to arguments of the petitioner and the Norwegian respondents as to whether a simple weighted average of the costs of production of the seven salmon farms should be used:
"We agree with the petitioner that weight averaging would skew the results. Bremnes, one of the seven sampled farms, is one of the largest farms in Norway. Based on public information on the record of this case (response of the Government of Norway to the countervailing duty questionnaire (C-4.03-802)), the largest farms in Norway produce a very small proportion of total salmon production. However, Bremnes' production constitutes a large proportion of the combined production of the seven farms. Therefore, weighted averaging would result in a COP which disproportionately reflects the costs of the largest farms in Norway. In view of this, a simple average of costs is more representative of industry-wide costs than a weighted average".229
226 Memoramdum from Carolina Olivieri and Tracey Oakes to the file, 17 August 1990. ↩
227 Memorandum from Louis Apple to the file, 4 September 1990. ↩
228 Memorandum from Richard Moreland to Francis J. Sailer, 13 September 1990. ↩
229 56 Fed. Reg., 25 February 1991, p.7672. ↩
[Page 124]
437. The Panel noted Norway's argument that the Department of Commerce should have weighted the costs of production of the seven farms in the sample by the relative production volumes of the individual farms in order to account for significant cost differences per kg. between large and small farms. Norway had argued in this connection that the Department of Commerce had all the necessary data to compute such a weighted average. Norway had also contested the factual correctness of the Department's statement that the Bremnes farm was among the largest farms in Norway.
438. The United States had argued that the Department of Commerce had properly decided to use a simple average of the costs of production of the seven Norwegian salmon farms in light of information before the Department (provided by the Government of Norway) indicating that 96 per cent of salmon production in Norway took place in small farms. However, one of the seven farms in the sample, the Bremnes farm, was one of the largest in Norway and had accounted for a greater share of the combined production volume of the seven farms in the sample than the share of total production in Norway generally accounted for by large farms. Consequently, the use of an average costs of production figure weighted by the relative production volumes of each of the farms in the sample would have given much greater importance to this large farm than large farms generally occupied in the Norwegian industry as a whole.
439. The Panel considered that the logic of Norway's argument that greater weight should have been assigned to the costs of large farms in the sample than to the costs of small farms required that there be evidence of record showing that the Department of Commerce had before it information concerning the relative importance of various categories of sizes of farms in the Norwegian salmon industry and concerning differences between large and small farms in costs of production per kg. of salmon. This information would also have to provide a basis to conclude that the small farms were over-represented in the sample, compared to their relative importance in the Norwegian salmon industry overall. Absent such information, there would be no basis to argue that by failing to assign greater weight to the costs of large farms the Department of Commerce had calculated costs of production in a manner that disproportionately reflected the higher costs of production per kg of small farms.
440. In this connection, the Panel noted that the Department of Commerce had concluded, based on information provided by the Governement of Norway, that most salmon farms in Norway were in the same size range and that large farms accounted only for 4 per cent of salmon production in Norway.230 There was no information before the Panel showing that the Norwegian respondents had contested the correctness of the factual basis of this conclusion drawn by the Department. The Panel reviewed the data provided to the Department of Commerce by the Government of Norway and found that the Department's conclusion regarding the relative importance of small and large salmon farms in the Norwegian industry was supported by this information.
441. The Panel noted that the parties had disagreed as to the factual basis of the Department's conclusion that one of the Norwegian salmon farms in the sample, the Bremnes farm, was among the largest farms in Norway. The Panel reviewed the data provided by the parties on this issue and concluded that, even if one adjusted the figures regarding the size of the Bremnes farm in the manner suggested by Norway, the result of this adjustment would not detract from the validity of the Department's conclusion that this farm was larger than the 8-12,000 cubic metre range which represented the majority of salmon farms in the Norwegain industry.
442. In light of the foregoing considerations and the considerations set forth in paragraphs 413 and 414, the Panel found that the Department of Commerce had not acted unreasonably in the light of the
230 Questionnaire Response of the Government of Norway in the Countervailing duty investigation of Fresh and Chilled Atlantic salmon from Norway, 14 May 1990, pp.9-10. ↩
[Page 125]
information before it when it decided that there was no basis to assign a greater weight to the cost of production figures of the large farms in the sample than to the costs of production figures of the small farms. The Panel therefore concluded that in determining that a simple average of the costs of production figures of the farms in the sample would be more representative of industry-wide costs than a weighted average, the United States had not acted inconsistently with Article 2:4 of the Agreement. The question of a possible inconsistency with Article 8:3 therefore did not arise.
(2)(v) Use of "the facts available" as a basis for the calculation of the costs of production of one of the Norwegian farms
443. The Panel then proceeded to examine Norway's claim that the United States had acted inconsistently with its obligations under Article 2:4 and 8:3 of the Agreement when the Department of Commerce had rejected the questionnaire responses of one farm, Nordsvalaks, and had attributed to this farm the highest costs of production figure calculated for any of the other six salmon farms in the sample as "the best information available".
444. The Panel noted that the stated basis of the decision by the Department of Commerce not to accept the information provided by Nordsvalaks concerned this farms' alleged failure to report in its questionnaire information on transactions with a related party.231
445. Norway had argued that Nordsvalaks failure to report transactions with this related party had been caused by the unclear and ambiguous wording of the respective item of the questionnaire. Furthermore, in view of the fact that Nordsvalaks and this related party shared joint costs and revenues on a 50/50 basis and that officials of the Department of Commerce had verified the questionnaire responses provided by Nordsvalaks, the Department could have easily corrected the data for Nordsvalaks to take into account the related party transactions. The imputation to the Nordsvalaks farm of the highest costs of production of any of the remaining six farms in the sample had been particularly detrimental given that this farm was the second lowest cost producer among the farms in the sample.
446. The United States had argued that despite the clear and unambiguous wording of the Department of Commerce's questionnaire, the Nordsvalaks farm had failed to report that it was related to another salmon farm. The existence of this relationship had raised questions regarding the proper allocation of costs and expenses between Nordsvalaks and the related party, questions which would have necessitated an entirely new response both from Nordsvalaks and from the related party. Under these circumstances, it was within the rights of the United States under Article 6:8 for the Department to disregard the information provided by Nordsvalaks and to base its calculation of costs of production for this farm "on the facts available". Finally, the United States had denied that the Department had in fact verified the data provided by Nordsvalaks; there was therefore no basis for Norway's statement that this farm was the second lowest cost producer in the sample.
447. The arguments made by the parties on this issue presented the Panel with the question of the relationship between the substantive provisions of the Agreement invoked by Norway and the provisions in Article 6:8 invoked by the United States. The Panel considered in this respect that the right granted by Article 6:8 to investigating authorities, under the circumstances defined in that provision, to make findings "on the basis of the facts available" had to be interpreted in conjunction with the relevant substantive provisions of the Agreement. In the case under consideration, the substantive provision in question was Article 2:4 of the Agreement, and in particular the reference to the "cost of production in the country of origin" as one of the components of a constructed normal value. Therefore, "the facts available" used by the United States under Article 6:8 had to be relevant to the determination
231 56 Fed. Reg., 25 February 1991, p.7668 ↩
[Page 126]
of the "cost of production in the country of origin" in a manner consistent with Article 2:4. The Panel recalled its observations in paragraphs 413 and 414 regarding the conditions under which sampling techniques could be used for purposes of determining "the cost of production in the country of origin" under Article 2:4. The Panel was of the view that these observations were relevant to its examination of whether in the case before it the United States had properly invoked Article 6:8 with respect to the determination of the costs of production of the Nordsvalaks farm.
448. The Panel therefore considered that, even assuming that the United States could reasonably have found that Nordsvalaks had not provided necessary information within a reasonable period of time and that it was therefore necessary to make its findings regarding the costs of production of Nordsvalaks "on the basis of the facts available", an analysis of whether the United States had acted within its rights under Article 6:8 also required an examination of the data used for Nordsvalaks costs of production in the light of the stated purpose of the sample of seven farms.
449. The Panel observed that, taken literally, it could not be argued that, when the Department of Commerce had imputed to Nordsvalaks the highest (verified) cost of production figure found for any of the remaining six farms, it had not relied on "a fact available". Nevertheless, in this case bearing in mind the consideration set forth in paragraph 414, the Panel found that a reasonable exercise of the discretion enjoyed by the United States under Article 6:8 with regard to the choice of "the facts available" would have required that the Department take into account the purpose of its calculation of costs of production of the seven salmon farms.232 The actual verified costs of production per kg. for the six remaining farms in the sample had differed significantly and the imputation of the highest of these figures to Nordsvalaks had had a significant impact on the average cost of production figure. Given that the sample was used by the Department of Commerce to compute a single average "cost of production in the country of origin" figure to be included in the calculation of the constructed normal values of most of the exporters under investigation, the Department should have considered how its choice of "the facts available" for determining the costs of production of Nordsvalaks would affect the representativeness of the results of the sample. There was no information before the Panel indicating how the Department had considered this aspect in its decision with regard to the choice of "the facts available" for Nordsvalaks.
450. The Panel concluded that the United States had not acted within its rights under Article 6:8 by imputing to Nordsvalaks the highest costs of production figure found for any other farm in the sample without considering how this would affect the representativeness of the results of the sample, and had thereby acted inconsistently with its obligations under Article 2:4 of the Agreement.
451. With respect to Norway's argument that this aspect of the final determination of dumping was also inconsistent with the obligations of the United States under Article 8:3 (in particular the first sentence thereof), the Panel recalled its analysis in paragraph 429 above.
232 Supra, paragraphs 413 and 414. ↩
[Page 127]
(2)(vi) Inclusion in the constructed normal values of a "freezing charge"
452. The Panel then examined Norway's claim concerning the treatment by the Department of Commerce of a NOK 5/kg. freezing charge for the purposes of its calculation of constructed normal values.
453. The Panel noted that the public notice of the affirmative final determination of dumping indicated that "In all cases, for salmon sold on or after January 1, 1990, a five NOK/kg. cost was added to the CV [constructed value] before profit".233 This cost included in the constructed value corresponded to a fee charged in connection with the financing of a programme under which, beginning in January 1990, a part of the Atlantic salmon harvest in Norway was frozen. In the proceedings before the Department of Commerce, the petitioner and the Norwegian respondents had disagreed as to whether this freezing charge should be included in the constructed normal values as an element of the farmers' costs of production of fresh Atlantic salmon.234 The Department of Commerce had formulated its position on this issue as follows:
"This fee is a five NOK/kg. charge assessed on all sales of fresh salmon. Therefore, the amount of the fee incurred by each salmon farmer is completely a function of the amount of fresh salmon it sells. The fact that FOS uses this money to finance a freezing plan is not the deciding factor. The Department considers this fee to be a general expenses and included it as a cost of producing the fresh salmon."235
454. As the legal basis of its claim on this issue Norway had indicated that the treatment by the Department of Commerce of this freezing charge was inconsistent with the requirement of a fair and equitable treatment of the Norwegian exporters but had also referred to the provisions of Article 2:4 of the Agreement. In light of the Panel's observations in paragraph 369, the Panel decided to examine this issue on the basis of Article 2:4 which defined the components of a constructed normal value as follows:
"... the cost of production in the country of origin plus a reasonable amount for administrative, selling and any other costs and for profits."
The legal question before the Panel was whether the treatment by the Department of Commerce of the freezing charge as a general expense to be included in the cost of producing fresh Atlantic salmon by the Norwegian farmers was inconsistent with this part of Article 2:4.
455. In support of its claim on this issue, Norway had argued that the NOK 5/kg. charge was not paid by farmers but by exporters. Therefore, this fee did not represent a cost incurred by producers of Atlantic salmon. In addition, Norway had referred to the objective of the freezing programme; since the fee was charged to finance the freezing of fresh Atlantic salmon, the fee should be treated as part of the costs of freezing salmon and not as part of the costs of producing fresh Atlantic salmon.
456. The United States had argued that the evidence before the Department of Commerce indicated that the freezing charge was paid by the Norwegian salmon farmers rather than by the exporters. Furthermore, the fact that the fee was charged to finance a programme concerning frozen salmon was irrelevant for the purpose of determining whether the fee was a cost of producing fresh Atlantic salmon.
233 56 Fed. Reg., 25 February 1991, p.7664. ↩
234 56 Fed. Reg., 25 February 1991, p.7665. ↩
[Page 128]
457. The Panel noted that a key factual element of Norway's argument was that the freezing charge was paid not by the producers of Atlantic salmon (i.e. the salmon farmers) but by the exporters.
458. The Panel reviewed the documentation before it and considered that, based on information provided by the Norwegian respondents, the Department of Commerce could reasonably have found that that the freezing charge was paid by the salmon farmers, rather than by the exporters. Verification reports for six of the investigated farms included statements by officials of these farms that all freezing charges were for the account of the farmers.236 The Panel noted in this respect that even if the freezing charge had not been paid by farmers but by the exporters it would be far from clear that under Article 2:4 this charge could not have been included as one of the other components of the constructed normal values.
459. The Panel further considered that the fact that the freezing charge was imposed for the purpose of financing the freezing of fresh salmon was not relevant in determining whether or not this charge could have been included as an element of the costs of production of fresh salmon. Relevant was that this charge was levied on all sales of fresh salmon by the farmers to the exporters and that the total amount of charges paid by the farmers thus depended upon the amount of salmon sold to the exporters. As such, this charge could not be considered to be unrelated to the costs of production of fresh salmon, as had been argued by Norway.
460. In light of the foregoing considerations, the Panel concluded that, by including a freezing charge of NOK 5/Kg. in the computation of the costs of production of Atlantic salmon, the United States had not acted inconsistently with its obligations under Article 2:4 of the Agreement.
(2)(vii) Comparison of normal values and export prices
461. The Panel then examined Norway's claim that in comparing average (constructed) normal values to individual prices of Atlantic salmon sold for export to the United States in different weight categories, the United States had acted inconsistently with its obligations under the Agreement.
462. The Panel considered it necessary to distinguish two aspects of this claim presented by Norway. Firstly, whether the United States had acted inconsistently with its obligations under the Agreement by failing to take account of differences in weight categories in the comparison between normal values and export prices. Secondly, whether the comparison of average normal values with individual export prices per se was inconsistent with the obligations of the United States under the Agreement.
(2)(vii)(a) Alleged failure of the United States to take account of differences in weight categories
463. The Panel first examined whether, as contended by Norway, in comparing normal values and export prices the United States had failed to take into account difference in weight categories between Atlantic salmon produced and sold in Norway and Atlantic salmon sold for export to the United States.
464. In this connection, the Panel noted that Norway had explained that its claim with regard to this question of differences in weight was based both on Article 2:4 and Article 2:6 of the Agreement.237 It appeared to the Panel that Norway's argument was that the Department of Commerce should have
236 Verification report for Fremstad Group A/S, p. 11; Verification report for Skaarfish Mowi A/S, P. 15; Verification report for Christian Bjelland Seafood A/S, p. 11; Verification report for Salmonor A/S, p. 9; Verification report for Sea Star International A/S, p. 12; Verification report for Hallvard Leroy A/S, p. 12; ↩
[Page 129]
taken account of these differences either by calculating separate constructed values for each weight category or, if a single constructed value was used, by comparing this single constructed value to an average export price across different weight categories.
465. The Panel considered that the question before it was whether in the present case the Department of Commerce had improperly failed to account for differences in weight categories as a factor affecting the comparability of export prices and (constructed) normal values.
466. The Panel found the following facts relevant to its examination of this issue. Firstly, it had not been disputed by the parties that large salmon was sold at a higher price per kg. than small salmon. Secondly, the public notice of the affirmative final determination contained a statement on how the Department had made product comparisons in those cases in which normal values were based on export prices to third countries:
"For the purpose of this investigation, we have determined that all Atlantic salmon comprises a single category of such or similar merchandise. Product comparisons were made on the basis of grade of salmon (superior, ordinary) and weight bands. We compared U.S. sales of gutted Atlantic salmon to sales of gutted Atlantic salmon sold in third countries because only gutted merchandise is sold in the United States. In addition, U.S. sales were compared only to sales of identical weights and grades of merchandise sold in the third country markets"238
Thirdly, in those cases in which normal values were constructed, the basis of these constructed values was a single cost of production per kg which did not distinguish between different weight categories of salmon. From the information before the Panel it appeared that other elements of these constructed normal values (e.g. the amount for profits) had not been differentiated to reflect price differences per kg between different weight categories of Atlantic salmon.
467. Fourthly, it did not appear from the documents before the Panel that the Norwegian parties participating in this investigation had raised before the Department of Commerce the question of how account should be taken of differences in weight for the purpose of comparing normal values and export prices. While it was clear from the public notice of the final determination of dumping that the respondents had raised a concern regarding the use of individual export prices239, there was no indication that the respondents in this context had referred to differences in weight categories. The Panel further noted Norway's argument that in the respondents' case brief dated 14 January 1991 the Norwegian exporters "had indicated ... that there were problems with comparing a single constructed normal value to individual export prices because of the differences in market value based on quality and weight difference." However, the Panel found that the document referred to by Norway240 did not address this issue.
468. The Panel was of the view that Article 2 of the Agreement provided for two ways in which differences affecting the comparability of export prices and normal values could be taken into consideration. Firstly, the like products for which comparisons were made could be chosen in such a way as to eliminate the effect of such differences on the comparability of export prices and normal values. Secondly, Article 2:6 specifically provided that "due allowance" was to be made, "in each case on its merits for the differences in conditions and terms of sale, for the differences in taxation, and for the other differences affecting price comparability."
238 56 Fed. Reg., 25 February 1991, p.7661. ↩
239 56 Fed. Reg., 25 February 1991, pp.7673-7674. ↩
240 Case brief of the Norwegian respondents, 14 January 1991. ↩
[Page 130]
469. As noted above, where normal values had been based on export prices to third countries, the Department of Commerce had made price comparisons for salmon of identical weight categories. The Panel found that this indicated that the Department was aware that differences in weight categories could affect the comparability between these export prices to third countries and the export prices to the United States.
470. The Panel further observed that it had not been contested by the United States that in those cases in which normal values had been constructed, these constructed normal values had been compared to export prices of salmon sold in different weight categories. While the United States had explained that because of the absence of differences in costs of production between salmon of different weights no separate constructed values for individual weight categories had been calculated, the United States had not put forward any arguments to explain why export prices of individual weight categories had been used in the comparison with the single constructed values. The public notice of the affirmative final determination of dumping was also silent on this point.
471. In the view of the Panel, the comparison of a constructed value across different weight categories with export prices which varied by weight category would be reasonable only if the Department of Commerce considered that differences in weight did not affect the comparability of the constructed value and the export prices. However, the Panel found it significant in this respect that in comparing third country export sales to export prices to the United States, the Department of Commerce had considered it necessary to limits its price comparisons to identical weight categories. The Panel could find nothing in the information of record before it to explain why, if the differences in weight categories were a relevant factor in those instances in which normal values were based on prices, such differences were not considered relevant by the Department of Commerce where normal values were constructed. While it might be factually correct, as pointed out by the United States, that the costs of production per kg. did not vary by weight of salmon, prices of Atlantic salmon per kg. did vary by weight category. Given that under the Agreement a constructed value was a proxy for a price-based normal value, the Panel found that the fact that costs of production per kg. did not vary by weight could not, without further explanation, constitute a basis to conclude that differences in weight did not need to be taken into account when normal values were constructed. In this connection the Panel observed that the provisions in Article 2:6 regarding the comparison of normal values and export prices applied both to cases in which price based normal values were used and to cases in which constructed normal values were used.
472. In the light of the foregoing considerations, the Panel found that the Department of Commerce had not properly considered the rôle of differences in weight as a factor which possibly affected the comparability between the constructed normal values and export prices and for which due allowance might have to be made under Article 2:6 of the Agreement. The Panel concluded that this aspect of the final determination of dumping was inconsistent with the obligations of the United States under Article 2:6 of the Agreement.
473. With regard to Norway's argument that this aspect of the final determination of dumping was also inconsistent with the obligations under Article 8:3 (in particular the first sentence thereof) the Panel recalled its analysis in paragraph 429 above.
(2)(viii)(b) Comparison of average normal values to individual export prices
474. The Panel then proceeded to examine the second aspect of Norway's claim with regard to the manner in which the United States had compared normal values and export prices. Norway had argued that a comparison between average normal values and individual export prices inherently was inconsistent with the requirement of Article 2:6 of the Agreement that a fair comparison be made between normal values and export prices.
[Page 131]
475. In support of its claim with respect to this issue, Norway had argued that neither the Agreement nor Article VI of the General Agreement authorized a comparison between an average normal value and individual export prices. In the case before the Panel this method of comparing normal values and export prices had inevitably created margins of dumping where no margins would have been found if the United States had compared average normal values to average export prices. Norway had also argued that the fact that Atlantic salmon was a perishable product was an additional reason why a comparison between an average normal value and individual export prices was unfair.
476. The United States had argued that there was no provision in the Agreement which prohibited a comparison between an average normal value and individual export prices; while a "fair comparison" was required under Article 2:6 of the Agreement, no particular methodology was mandated to satisfy this standard. The United States had also argued that many Parties applying anti-dumping measures used this method of comparing average normal values to individual export prices and that the methodology advocated by Norway would make it difficult to remedy instances in which dumping was occurring only in particular product lines, or time periods, or with respect to particular customers or regional markets. Finally, with regard to Norway's argument that Atlantic salmon was a perishable product, the United States had argued that the Department of Commerce had properly determined, based on the evidence of record, that Atlantic salmon was not perishable.
477. The Panel observed that the public notice of the affirmative final determination of dumping contained a discussion of the question of how the Department of Commerce should make a comparison between the normal values and export prices. In particular, the Norwegian respondents had advanced a number of arguments in support of their view that "the Department's usual practice of comparing U.S. prices to a weighted average FMV covering the entire period of investigation would result in an inherently unfair comparison of "apples to oranges" or "fish to fowl" and that average normal values should preferably be used on a daily or weekly basis, or in the alternative, on a monthly basis. Furthermore the respondents had argued that export prices should be calculated on "an average basis comparable to that utilized for FMV".241 The response by the Department of Commerce to this argument noted that in view of the significant price fluctuations over the period of investigation, it was appropriate to use weighted-average foreign market values on a monthly basis. At the same time, this response explained the view of the Department that there was no reason not to follow the normal practice of calculating individual export prices. It was noted in this connection that, while the Norwegian exporters had referred to previous cases in which the Department of Commerce had used average export prices to account for the perishable nature of the products in question, this rationale for calculating export prices on an average basis did not apply to the present case. The Department had observed in this respect that:
"... vegetables and flowers were highly perishable products, dominated by sales at auction, and having significant price fluctuations each day. Salmon shares none of these characteristics and, therefore, averaging to eliminate the distortions is unnecessary."242
478. The Panel further observed that elsewhere in this public notice the Department of Commerce had explained its view that, for purposes of its analysis of costs of production, fresh Atlantic salmon was not a perishable product.243
241 56 Fed. Reg., 25 February 1991, pp.7673-7674. ↩
242 56 Fed. Reg., 25 February 1991, p.7674. ↩
243 56 Fed. Reg., 25 February 1991, p.7673. ↩
[Page 132]
479. In reviewing the merits of Norway's claim, the Panel noted that this claim was based on Article 2:6 of the Agreement, which provided in relevant part that:
"In order to effect a fair comparison between the export price and the domestic price in the exporting country (or the country of origin), or if applicable, the price established pursuant to the provisions of Article VI:1(b) of the General Agreement, the two prices shall 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."
Norway interpreted this first sentence as containing a general requirement that the comparison between the normal values and export prices must be fair and that this implied that identical methods be used in the calculation of the normal value and the export price. The United States had not denied the existence of this requirement of a fair comparison but had argued that the Agreement did not set forth a specific methodology to satisfy this requirement.
480. The Panel considered that the text of Article 2:6 did not explicitly address the specific question of the use of averaging techniques in the comparison of normal values and export prices. The requirements of this provision with regard to the treatment of factors affecting the comparability of normal values and export prices pertained to an aspect of the comparison between normal values and export prices which was distinct from the issue raised by Norway. Logically, the question of whether normal values and export prices should be compared on an average-to-average basis could arise only after the comparability of the individual normal values and export prices, in terms of the factors referred to in Article 2:6, had been ensured.
481. While the Panel thus found that Article 2:6 did not explicitly deal with the question of whether normal values and export prices should be compared on an average-to-average basis, it considered that it was possible to interpret the first sentence to reflect a requirement of a "fair comparison" which applied generally to any aspect of the comparison of normal values and export prices. The Panel considered, however, that this interpretation of Article 2:6 would not permit a conclusion that a method whereby average normal values were compared to individual export prices was per se inconsistent with Article 2:6. Rather, the "fairness" of such a method would have to be evaluated in the light of the circumstances of each case.
482. The Panel noted in this respect that an essential element in Norway's claim was the view that a comparison of average normal values with individual export prices inevitably created margins of dumping where no margins of dumping would be found if normal values and export prices were compared on an average-to-average basis. The Panel observed that in these general terms this view was not correct in that the alleged bias resulting from a method under which average normal values were compared to individual export prices depended upon the pattern of prices in the domestic market and in the export market. In particular, for this alleged bias to occur, there would have to be a number of individual export prices above the average normal value. If export prices were uniformly below the average normal value, this bias could not occur.
483. In this connection, the Panel noted that Norway had presented an example of how the comparison of an average normal value to individual export prices inevitably led to the creation of margins of dumping. This example pertained to a situation in which prices changed over time in both the domestic market and to the exporting country but in which at each point in time export prices were identical to domestic prices. However, this example was hypothetical and there was no evidence before the Panel that the factual pattern of prices in the case before it was similar to the pattern described in Norway's example or, more generally, that in the circumstances of this case the pattern of prices was such that, as contended by Norway, the comparison of average normal values to individual export prices
[Page 133]
had created margins of dumping where no such margins would have been found if an average-to-average comparison had been made.
484. The Panel therefore considered that, assuming that the concept of a "fair comparison" in the first sentence in Article 2:6 provided a basis upon which it could review the comparison made by the Department of Commerce of average normal values to individual export prices, the information before it did not permit it to find that under the circumstances of this case this method had been inconsistent with this concept of a "fair comparison".
485. Finally, with regard to Norway's argument concerning the nature of Atlantic salmon as a perishable product, the Panel recalled its observations in paragraphs 395 and 396.
486. In the light of the foregoing considerations, the Panel concluded that in comparing average normal values to individual export prices, the United States had not acted inconsistently with its obligations under Article 2:6 of the Agreement. The question of a possible inconsistency with Article 8:3 therefore did not arise.
C. DETERMINATION OF THE EXISTENCE OF MATERIAL INJURY
487. The Panel then proceeded to examine whether the imposition by the United States of the anti-dumping duty order on imports of fresh and chilled Atlantic salmon from Norway was inconsistent with the obligations of the United States under the Agreement by reason of the affirmative final determination of material injury of the USITC.244
488. Norway had argued that this determination was inconsistent with the requirements of Article 3 of the Agreement on two main grounds. Firstly, the findings of the USITC regarding the volume of imports under investigation, the price effects of these imports and the consequent impact of these imports on the domestic Atlantic salmon industry in the United States were inconsistent with Articles 3:1, 3:2 and 3:3. Secondly, the finding of the USITC of a causal relationship between the allegedly dumped imports from Norway and material injury to the domestic Atlantic salmon industry in the United States was inconsistent with Article 3:4.
489. The United States had submitted that the findings of the USITC regarding the volume of the imports subject to investigation, the price effect of these imports, and the consequent impact of the imports on the domestic industry in the United States were consistent with the requirements of Articles 3:1, 3:2 and 3:3 of the Agreement and that the USITC's finding of a causal relationship between the subject imports from Norway and material injury to the domestic industry in the United States was consistent with Article 3:4 of the Agreement.
(1) Volume of imports subject to investigation, price effects of the imports and consequent impact of these imports on the domestic industry in the United States
490. The Panel first examined the claims presented by Norway regarding the alleged inconsistency with the requirements of Articles 3:1, 3:2 and 3:3 of the USITC's findings regarding the volume of imports subject to investigation, the price effects of the imports, and the consequent impact of these imports on the domestic industry in the United States.
244 Fresh and Chilled Atlantic Salmon from Norway: Determination of the Commission in Investigation No. 731-TA-454 (Final) Under the Tariff Act of 1930, Together with the Information Obtained in the Investigation. USITC Publication 2371, April 1991 (hereinafter: USITC Determination) ↩
[Page 134]
491. In view of the factual nature of some of the disputed issues raised under these provisions the Panel found it appropriate to articulate certain general considerations by which it was guided in its review of the issues raised by Norway.
492. Firstly, the Panel noted the requirement of Article 3:1 of an "objective examination" of the volume of imports, their effect on prices in the domestic market for like products, and the consequent impact of these imports on domestic producers of like products. In the view of the Panel, 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.
493. Secondly, the Panel noted that Articles 3:2 and 3:3 of the Agreement specified how the factors mentioned in Article 3:1 were to be examined by investigating authorities. Article 3:2 required that the authorities "consider" whether there had been a significant price undercutting, price depression or price suppression by the imports in question. Article 3:3 required the investigating authorities to include in their examination of the impact of the imports on the domestic industry "an evaluation of all relevant economic factors and indices having a bearing on the state of the industry" and contained an illustrative list of those "factors and indices". The Panel noted that Article 3:4, which required a demonstration of a causal relationship between the allegedly dumped imports and material injury to a domestic industry, explicitly referred to the factors set forth in Articles 3:2 and 3:3. Therefore an essential element of a review of whether a determination of material injury was in conformity with Article 3 was an examination of whether the factors set forth in Articles 3:2 and 3:3 had been properly considered by the investigating authorities. However, it followed from the last sentence in Article 3:2 and from the last sentence in Article 3:3 that Article 3 did not prejudge the weight to be given in a particular case to any of the factors listed in these provisions.
494. Thirdly, the Panel observed that Article 3:1 required that determinations of material injury be based on "positive evidence". A review of whether in a given case this requirement was met involved an examination of the stated factual basis of the findings made by the investigating authorities in order to determine whether the authorities had correctly identified the appropriate facts, and whether the stated factual basis reasonably supported the findings of the authorities. In this context, the Panel considered that the mere fact that in a given case reasonable, unprejudiced minds could differ as to the weight to be accorded to certain facts was not a sufficient ground to find that a determination of material injury based on such facts was not based on positive evidence within the meaning of Article 3:1. The question of whether a determination of injury was based on positive evidence therefore was distinct from the question of the weight to be accorded to the facts before the investigating authorities. The Panel, however, recalled in this connection its observations in paragraph 492 on the requirement of an "objective examination" as the basis of injury determinations under Article 3.
(1)(i) Volume of the imports under investigation
495. The Panel then examined the issues raised by Norway with respect to the findings made in the affirmative final determination by the USITC on the volume of imports of Atlantic salmon from Norway.
496. Norway had argued that these findings were inconsistent with the requirement of Article 3:1 of an "objective examination" of the volume of imports and that these findings were inconsistent with the requirement of Article 3:2 that investigating authorities consider whether there has been a "significant increase" in the volume of dumped imports. The Panel considered that some of the arguments presented by Norway in support of these two claims also pertained to the question of whether the USITC's findings were based on positive evidence.
[Page 135]
497. The United States had argued that the USITC had properly considered whether there had been a significant increase of the volume of imports of Atlantic salmon from Norway, as required by Article 3:2, and that the USITC's conclusion that these imports had increased significantly was supported by the evidence of record.
498. The Panel first examined whether, as required by Article 3:2, the USITC had considered whether there had been a significant increase in the volume of dumped imports, either in relative or in absolute terms. The Panel noted in this connection Norway's argument that the USITC had considered the significance of the level of the volume of imports from Norway throughout the period of investigation (1987-1990) rather than the significance of any increase in that volume.
499. The Panel observed that in its determination the USITC had made the following statements on the evolution of the volume of imports of Atlantic salmon from Norway during the period of investigation:
"Imports of Atlantic salmon from Norway surged from 1987 to 1989. Imports rose from 7.6 million kilograms in 1987 to 8.9 million kilograms in 1988, and then jumped further in 1989 to 11.4 million kilograms for an overall increase of fully 50 per cent. In value terms, imports also increased strongly, but at a slower rate, from $74.4 million in 1987 to $93.7 million in 1989. Despite increases in absolute terms, in terms of market penetration Norwegian imports fell steadily by quantity from more than 75 per cent in 1987 to 60.2 per cent in 1989. A similar decline was posted in market penetration by value terms, from more than 75 per cent in 1987 to 62.5 per cent in 1989. In 1990, subject imports fell strongly to 7.7 million kilograms, valued at $66.4 million. Subject imports by volume and value accounted for 36.7 per cent and 40.8 per cent, respectively, of apparent US consumption in 1990."245
After explaining why it had accorded less weight to the decline in imports in 1990246, the USITC had concluded its discussion of the volume of imports of Atlantic salmon from Norway as follows:
"We find that the volumes of imports from Norway over the period of investigation, and the increases in those volumes from 1987 to 1989, are significant. The subject imports are particular significant when viewed together with information concerning the nature of the US industry, the industry's condition over the period and information on prices for the like product."247
500. On the basis of these statements, the Panel found that the USITC had specifically considered changes in import volume both in absolute terms and in relative terms and had indicated that it considered the increase in the absolute volume of imports from 1987 to 1989 to be significant. While the USITC had also considered the significance of "the volumes of imports from Norway over the period of investigation", the text of the USITC's determination made it clear that the USITC had not considered the significance of the volumes of imports in lieu of a consideration of the significance of the increase in these volumes.
501. The Panel therefore found that the USITC had not failed to consider whether there had been a significant increase in the volume of the subject imports, as required by Article 3:2.
245 USITC Determination, pp. 16-17, footnotes omitted. ↩
247 USITC Determination, p.18. ↩
[Page 136]
502. With respect to the requirement of Article 3:1 that there be positive evidence as a basis for an affirmative determination of injury, the Panel observed that in its statements on the evolution of the (absolute and relative) volume of imports from Norway over the period of investigation, the USITC had relied on data in Tables 17 and 18 in the Annex to its determination.248 Table 17 contained data on the absolute volume of imports (by quantity and by value) of imports of Atlantic salmon from Norway and other supplying countries for the period 1987-1990, while Table 18 contained data on the relative volume of imports (by quantity and by value) of Atlantic salmon from Norway during this period. The Panel found that the statements made on the volume of imports from Norway in the text of the USITC's determination were supported by the data in these tables and noted in this respect that it had not been argued by Norway that these data were not factually correct.
503. The Panel therefore considered that the statements by the USITC on the evolution of the volume of imports from Norway were based on positive evidence.
504. The Panel noted that Norway's principal claim regarding the USITC's findings on the evolution of the volume of imports was that, when analysed in the context of other facts before the USITC, the increase from 1987 to 1989 in the absolute volume of imports of Atlantic salmon from Norway was not significant within the meaning of Article 3:2.
505. In this connection, Norway had argued that, for purposes of determining the significance of the increase in the absolute volume of imports from 1987 to 1989, the USITC should have taken into account the fact that the market share in the United States of Norwegian imports had declined over the investigation period, while the market share of third countries and of US domestic producers had increased. Furthermore, the absolute volume of imports from Norway had started to decline in late 1989, well before the initiation of this anti-dumping duty investigation and application of any provisional measures. In Norway's view, Article 3:2 of the Agreement did not permit a finding of a significant increase in the volume of imports where (1) the absolute volume of imports at the end of the investigation period was not higher than at the beginning of that period and the facts demonstrated that the decline in absolute import volume was not the result of the initiation of the investigation and application of provisional measures, and (2) the relative volume of imports declined throughout the period of investigation.
506. In examining the legal and factual aspects of Norway's argument that, under the circumstances of this case, Article 3:2 did not permit a finding of a significant increase of import volume, the Panel first observed that Article 3:2 of the Agreement did not contain a requirement that imports from third countries not subject to investigation be considered as part of an examination of the significance of an increase in the volume of imports from a country whose imports were the subject of an anti-dumping duty investigation. A consideration of the volume imports from such third countries might be relevant for the purpose of determining the existence of a causal relationship between the allegedly dumped imports under investigation and material injury to a domestic industry. In that context, such imports might be relevant as one of the "other factors" referred to in Article 3:4. Footnote 5 expressly identified as one of these possible "other factors" "the volume and prices of imports not sold at dumping prices". However, nothing in the text of Article 3:2 indicated that imports from third countries had to be examined as part of the analysis under Article 3:2 of whether the volume of imports under investigation had increased significantly. Likewise, the consideration of the market share of domestic producers was expressly mentioned in Article 3:3 as part of the analysis of the impact of the imports on the domestic industry concerned, but was not a mandatory factor under Article 3:2.
248 See Annexes 1 and 2 to this Report. ↩
[Page 137]
507. The Panel then considered Norway's argument that the significance of the increase in the absolute volume of imports of Atlantic salmon from Norway from 1987 to 1989 was limited, inter alia, because of the subsequent decline in the absolute volume of these imports starting in late 1989. The USITC had made the following comments on this decline:
"We have given less weight to the recent decline in imports in 1990 because it appears to be largely the result of the filing of the petition and/or the imposition of provisional anti-dumping and countervailing duties. The petition was filed in this investigation in February 1990, the Commission issues its preliminary determinations in April 1990: Commerce made its preliminary CVD determination in June 1990, imposing a 2.45 per cent ad valorem provisional duty; and Commerce rendered its affirmative preliminary anti-dumping duty determination in October 1990, imposing interim duties on most firms ranging from 1.6 to 4.9 per cent. The drop in subject imports has been most pronounced since July 1990, subsequent to Commerce's preliminary CVD determinations. In view of the precipitous nature of the drop in subject imports by the end of 1990, from record levels in 1989, it is likely that the Commission and/or Commerce proceedings played a rôle in the import decline.
Respondents claim that the decline in Norwegian imports in 1990 was the result of the appreciation of the Norwegian kroner against the US dollar, and the institution of a freezing programme in Norway to reduce the amount of fresh Norwegian salmon available for export. Although it is possible that these factors may have played some rôle, they cannot entirely account for the drastic decline that occurred in the second half of 1990."249
Thus, the USITC had explained that it had accorded less weight to the more recent decline in the absolute volume of imports of Atlantic salmon from Norway because of the fact that this decline appeared to be largely the result of the filing of the petition and/or the imposition of provisional anti-dumping and countervailing duties.
508. The Panel noted that Norway had contested that, as stated by the USITC, the decline in the volume of imports from Norway was largely the result of the initiation of the investigation and/or the imposition of provisional measures. Norway had argued that this decline had begun well before the initiation of this investigation in March 1990. In support, Norway had presented monthly data on the absolute volume of imports from Norway in 1989-1990. These data, which were included in the record of the USITC's investigation, are reproduced in Annex 3 to this Report. The Panel reviewed these data and found that decline in imports levels in January and February 1990 had been preceded by a period of four months in which imports had increased. In December 1989 imports had been at a higher level than in January 1989. Furthermore, after the filing of the petition in February 1990, the monthly import levels had increased during March and April 1990. Finally, imports had begun to decline in May 1990, with the largest decline taking place in the period July-December 1990. In light of these data, the Panel considered that there was no clearly discernible level of a declining absolute volume of imports in the period prior to the initiation of the countervailing duty investigation and that imports started to decline considerably only in July 1990. The Panel therefore found that the USITC had not made an error of fact in its statements on the evolution of the absolute volume of imports in 1990.
509. In light of its findings in paragraphs 506-508, the Panel considered that there was neither a legal nor a factual basis for the view that, in the circumstances of this case, Article 3:2 did not permit a finding of a significant increase in the volume of imports. In the view of the Panel, where, as in this case, the facts before the investigating authorities indicated an increase of imports during part of
249 USITC Determination, pp.17-18, footnotes omitted. ↩
[Page 138]
the investigation period, followed by a decrease, it was not properly within a panel's task to make a judgement on the relative weight to be accorded to these facts. Rather, in such a situation a panel had to review whether the investigating authorities had carried out an "objective examination", by considering all information and by explaining why the data on the decrease in the volume of imports did not detract from a finding of a significant increase in the volume of imports. In the case before it the USITC had not failed to carry out such an objective examination: the USITC had considered the decline in the volume of imports from Norway in the latter part of the investigation period and had reasonably explained why it had accorded less weight to this decline. In determining that this decline deserved less weight, the USITC had not committed errors of fact.
510. In light of the foregoing considerations, the Panel concluded that the analysis and findings of the USITC with regard to the volume of imports of Atlantic salmon from Norway were not inconsistent with the obligations of the United States under Articles 3:1 and 3:2 of the Agreement.
(1)(ii) Price Effects of the Imports under Investigation
511. The Panel then proceeded to an examination of Norway's claim with respect to the finding of the USITC that imports of Atlantic salmon from Norway had significantly depressed prices of the like domestic product.
512. Norway had argued that this finding was inconsistent with Article 3:1, which required an objective examination of the effect of the allegedly dumped imports on prices for domestic like products and positive evidence as the basis of an affirmative determination, and with Article 3:2, which required that investigating authorities consider, inter alia, whether the effect of the allegedly dumped imports was to depress prices of domestic like products to a significant degree.
513. The United States had argued that, consistently with Article 3:2, the USITC had considered whether the subject imports from Norway had significantly depressed domestic prices of Atlantic salmon in the United States and that its findings on this issue were supported by the evidence of record.
514. The Panel noted that the text of the determination by the USITC contained the following observations on the question of the price effects of the imports from Norway:
"Public and questionnaire information reveal that prices for U.S. Atlantic salmon fell up to a third or even more between mid- to late- 1988 and the end of 1989. Prices rebounded during 1990, then fell back somewhat at the end of 1990, but generally remained at levels below those recorded in September 1988. Prices for the like product closely tracked prices for Norwegian Atlantic salmon over much of the period. Beginning in the middle of 1988, prices for Norwegian Atlantic salmon started to drop and continued to fall even after U.S. Atlantic salmon had left the market in the spring of 1989. Prices for Norwegian Atlantic salmon reached their lowest point at the end of 1989, then climbed somewhat in 1990. Although other factors may have contributed, the decline in U.S. prices for Atlantic salmon in 1988 and 1989 was due in large part to oversupply in the U.S. market. Imports from Norway accounted for a large portion of the increased imports in 1989. This suggests that Norwegian Atlantic salmon played a role in the price decline. It is true that Norwegian Atlantic salmon generally oversold the like product during much of the period of investigation. This fact does not mean, however, that Norwegian Atlantic salmon did not contribute to the price decline for U.S. Atlantic salmon. Indeed, U.S. and Norwegian Atlantic salmon exhibit a high degree of substitutability, as Atlantic salmon is a near-commodity type product. Moreover, until late 1990 prices for Norwegian and U.S. Atlantic salmon followed a very similar pattern. In sum, given the sheer volume of the increase in Norwegian Atlantic salmon imports in 1989, falling prices for those imports, closely tracking U.S. and Norwegian Atlantic salmon price trends, and information suggesting significant substitutability between
[Page 139]
Norwegian and U.S. Atlantic salmon, we find that imports of Norwegian Atlantic salmon have significantly depressed prices for the like product. The subject imports' presence in the market place, even at premium prices, acted to keep domestic producers from pricing to recover costs and meet cash flow needs as described below."250
Thus, on its face, the text of the USITC determination demonstrated that the USITC had not failed to consider the price effects of the imports of Atlantic salmon from Norway in terms of one of the factors explicitly identified in the second sentence of Article 3:2 of the Agreement (i.e. "whether the effect of such imports is otherwise to depress prices to a significant degree").
515. The Panel then examined whether the finding by the USITC of significant price depression caused by imports of Atlantic salmon from Norway was based on positive evidence, as required by Article 3:1.
516. In this connection, the Panel first considered the stated factual basis of the finding of the USITC that domestic prices for Atlantic salmon in the United States had fallen up to a third or even more between mid- to late 1988 and the end of 1989. As indicated in the text of the USITC's determination, in making this statement the USITC had relied upon public information on prices in the US market and on price data gathered on the basis of responses to questionnaires. The public data on prices, presented in a graphical form in figures 2, 3 and 4 in the Annex to the USITC determination, consisted of weekly price data for three different weight categories of Atlantic salmon during the period January 1987-December 1990. While these figures appeared to support the finding by the USITC regarding the extent of the decline of domestic prices in 1988 and 1989, the Panel noted Norway's argument that the data presented in these figures could not be properly relied upon in an analysis of the effects of imports on domestic prices because these data pertained not to US domestic prices but to combined US/Canadian prices. The Panel observed that this information had not been the only source relied upon by the USITC; the USITC had also relied upon price data obtained through responses to questionnaires. Unlike the published price information, the responses to these questionnaires had provided data specifically on US domestic prices. The Panel reviewed the data derived from these questionnaire responses and found that it was factually correct that, as stated in the Annex to the USITC determination,
"Monthly net f.o.b. price data collected through questionnaires for U.S.- and Norwegian-produced Atlantic salmon generally showed the same decline in price as the published price data. Prices generally declined between 20 and 34 per cent during September 1988 - November/December 1989 for most salmon sizes in each channel of distribution, then increased between 5 and 33 per cent during 1990 (table 19). In nearly all weight categories and distribution channels, prices were lower in October 1990 than in September 1988."251
The Panel therefore did not consider it necessary to pronounce itself on the question of whether the use by the USITC of price data which had included combined US/Canadian prices was proper. The price data derived from the responses to the questionnaires provided a sufficient factual basis for the statement made by the USITC regarding the evolution of domestic prices of Atlantic salmon in 1988 and 1989.
517. The Panel then examined the factual basis of the finding of the USITC that "prices for the like product closely tracked prices for Norwegian Atlantic salmon over much of the period" and that "...
250 USITC Determination, pp.18-20. ↩
251 USITC Determination, p.A-59. ↩
[Page 140]
until late 1990 prices for Norwegian and US Atlantic salmon followed a very similar pattern.".252 The Panel noted that the Annex to the determination by the USITC contained the following statement on the pattern of prices of domestic and imported Atlantic salmon:
"US/Canadian and Norwegian price trends for Atlantic salmon were similar from mid-1988 through mid-1989 (figures 5-7). In 1990, the two trends began to diverge, and US/Canadian prices seem to have followed Chilean Atlantic salmon prices more closely (figures 8-10)."253
The Panel considered that the data presented in figures 5-7 of this Annex supported this statement. In particular, these data indicated that the two price trends had begun to diverge only in 1990, with Norwegian prices increasing and domestic prices decreasing. The Panel therefore considered that the findings of the USITC on the similarity of the price trends of domestic and Norwegian Atlantic salmon "over much of the [investigation] period" were based on positive evidence.
518. With respect to the link between imports from Norway and the development of domestic prices, the Panel observed that the USITC had referred to several factors in explaining its finding that the imports of Atlantic salmon from Norway had played a rôle in the decline of domestic prices. Firstly, the USITC had pointed out that the decline in US prices for Atlantic salmon in 1988 and 1989 was due in large part to the oversupply in the US market, and that imports from Norway had accounted for a large portion of the increased imports in 1989. Secondly, the USITC had noted that, while Norwegian Atlantic salmon was generally sold at prices higher than domestic Atlantic salmon, imports of Atlantic salmon from Norway had nevertheless had a depressing effect on domestic prices because of the high degree of substitutability of domestic and Norwegian Atlantic salmon, which the USITC characterised as a "near commodity type product".
519. The Panel found that the USITC's statement regarding the proportion of the increased volume of imports of Atlantic salmon in 1989 accounted for by imports from Norway was supported by the data before the USITC. In this connection the Panel referred to the data presented in Table 17 in the Annex to the USITC determination.254 The Panel also noted that in 1989 imports of Atlantic salmon from Norway had accounted for 62.5 per cent of the US domestic market by value and for 60.2 per cent of the US domestic market by quantity. Furthermore, Norway had not contested the factual correctness of the USITC's statement that domestic and Norwegian Atlantic salmon were highly substitutable.
520. The Panel then turned to the arguments presented by Norway to contest the legal and factual sufficiency of the USITC's finding that imports of Atlantic salmon from Norway had contributed to price depression in the US market.
521. Norway had argued that the evidence before the USITC indicated that during the period of investigation prices of Atlantic salmon from Norway had generally been higher than prices of domestic Atlantic salmon in the United States. When, in mid-1990, prices of Atlantic salmon from Norway had begun to rise, domestic prices had not followed this rise but had actually fallen. Norway had also pointed to the fact that domestic prices in the United States had not risen in the first half of 1991, after the imports from Norway had virtually disappeared from the US market. In the view of Norway, these facts demonstrated that the USITC had been incorrect in concluding that prices of domestic Atlantic salmon "closely tracked" prices of Norwegian Atlantic salmon. In addition, Norway had argued that,
252 USITC Determination, p.19 and p.20. ↩
253 USITC Determination, p.A-55. ↩
254 See Annex 1 to this Report. ↩
[Page 141]
if Atlantic salmon was a highly substitutable product and imports from third countries were both lower priced and increasing their market share, the logical conclusion was that it was the lower priced product that depressed domestic prices in the United States, not the higher priced Norwegian product. If the products were highly substitutable, buyers would buy the lower priced product rather than the higher priced product.
522. Norway had also argued that the USITC had failed to explain why domestic prices in the United States had followed prices of imports from Norway, instead of Norwegian suppliers having to reduce their prices in response to price undercutting by suppliers from third countries. Furthermore, the USITC had not provided any data demonstrating that prices of Norwegian Atlantic salmon had a "time lead" on price developments for domestic Atlantic salmon in the United States.
523. The Panel considered that the fact that domestic prices were lower than prices of imported products did not per se preclude a finding under Article 3:2 that the imports had a significant depressing effect on domestic prices. The USITC had not ignored the fact that prices of Atlantic salmon imported from Norway were generally higher than prices of domestic Atlantic salmon but had found that, because of the high degree of substitutability of domestic and imported Atlantic salmon, this did not mean that the imports had not depressed domestic prices. The Panel further considered that the fact that domestic prices in the United States had fallen after mid-1990 while prices of imports from Norway had risen, did not invalidate the finding of the USITC that domestic prices had closely tracked Norwegian prices "over much of the [investigation] period". This divergent price movement had occurred during a relatively short period in the period of investigation (1987-1990). As to the information provided by Norway concerning price developments in the US market since the beginning of 1991, the Panel considered that, since this information pertained to a period following the period of investigation examined by the USITC, this information by definition could not be taken into account by the Panel for purposes of determining whether the data before the USITC constituted positive evidence in support of the USITC's finding that imports from Norway had contributed to price depression in the US market.
524. The Panel noted Norway's argument that the fact that Atlantic salmon was a highly substitutable product implied that imports from third countries, rather than the higher priced imports from Norway, had depressed domestic prices in the United States. However, the Panel considered that when products sold at different prices were substitutable this did not necessarily imply that consumers would buy the lower priced product. Rather, substitutability meant that an expansion of supply of either product would affect prices of the products for which this product could be substituted. In this respect the Panel noted the increase in the absolute volume of imports of Atlantic salmon from Norway in the United States from 1987 to 1989, as recorded in Table 17 in the Annex to the Determination by the USITC. The Panel further observed that, while it was factually correct that imports from third countries had increased over the investigation period, in each of the calendar years covered by this period Norway had been the biggest supplier to the US market. During 1987-1989, Norway's market share had been larger than the combined market share of all third countries supplying Atlantic salmon to the US market.
525. The Panel considered that Article 3:2 did not require, as a condition of a finding of significant price depression by imports under investigation, that the authorities determine that the suppliers in question were price leader in the market. Even if prices of Atlantic salmon from Norway were influenced by prices of competitors from third countries, this did not imply that the USITC could not reasonably have found (on the basis of the evidence before it regarding the increase in the volume of imports from Norway from 1987 to 1989, the similarity in price trends of these imports and domestic Atlantic salmon and the substitutability of imports from Norway and domestic Atlantic salmon) that imports from Norway had contributed to significant price depression in the domestic market in the United States. Therefore, Norway's argument regarding the possible effect of imports from third countries on prices of imports of Atlantic salmon from Norway did not detract from the fact that the USITC's finding of significant price depression was based on positive evidence.
[Page 142]
526. Given that, as stated above, the Panel did not consider that Article 3:2 required a finding of price leadership as a condition of a finding of price depression by imports, the Panel also saw no merit in Norway's argument that the USITC had not demonstrated that prices of imports from Norway had a "time lead" on prices for domestic Atlantic salmon in the United States. A finding of price depression under Article 3:2 was not conditional upon a finding that price declines of domestic products were preceded in time by price declines of imported products. The Panel also noted in this connection that Article 3:2 treated price undercutting and price depression as separate possible effects of imports on domestic prices, without giving any greater weight to either of the two. The fact that the USITC's determination did not indicate whether the declines of domestic prices had been preceded by price undercutting by the imports from Norway therefore did not mean that the USITC's finding of significant price depression by the imports from Norway was not based on positive evidence.
527. In light of the foregoing considerations, the Panel concluded that the finding of the USITC that imports of Atlantic salmon from Norway had a significant price depressing effect in the US market was not inconsistent with the obligations of the United States under Articles 3:1 and 3:2 of the Agreement.
(1)(iii) Impact of the imports of Atlantic salmon from Norway on the domestic industry
528. The Panel then examined Norway's claim that the examination by the USITC of the impact on the domestic industry of the allegedly dumped imports from Norway was inconsistent with the obligations of the United States under Articles 3:1 and 3:3 of the Agreement.
529. Norway had argued that the USITC's finding of a negative impact of these imports on the domestic industry had not resulted from an "objective examination" (Article 3:1) of "all relevant facts having a bearing on the state of the industry" (Article 3:3). In support of its view that the findings made by the USITC with respect to the negative impact of the imports from Norway on the domestic industry in the United States were unfounded, Norway had referred to several facts before the USITC which in the view of Norway indicated that this industry had expanded significantly since it had first begun production in 1984. Thus, Norway had pointed to data concerning annual increases in the volume of domestic production capacity to produce juvenile Atlantic salmon, shipments, and employment in the Atlantic salmon industry in the United States.
530. The United States had argued that the USITC's finding concerning the impact of the imports from Norway on the domestic industry had resulted from a consideration of all the factors specified in Article 3:3 and was supported by the evidence of record.
531. The Panel noted that in its determination the USITC had discussed several indicators pertaining to the "condition of the industry" and had concluded from this discussion that the US domestic industry was experiencing material injury.255 The USITC then had separately examined the question of whether material injury was caused "by reason of" the imports from Norway.256 As the Panel understood Norway's arguments, Norway's objections raised under Articles 3:1 and 3:3 pertained to the first part of the USITC's analysis, i.e. the analysis of the "condition of the industry".
532. The Panel examined whether the USITC's finding that the domestic industry was experiencing material injury had involved "an evaluation of all relevant economic factors and indices having a bearing on the state of the industry", as provided for in Article 3:3.
255 USITC Determination, pp.11-15. ↩
256 USITC Determination, pp.15-22. ↩
[Page 143]
533. In this connection, the Panel noted that the USITC had first discussed a number of non-financial indicators (consumption, capacity and production, shipments and employment) and had then examined a number of financial indicators. The discussion of these specific indicators of the condition of the industry was preceded by a general comment on what the USITC considered to be "distinctive features" of the domestic industry:
"First, although we have found the industry to be 'established' for purposes of the statute, the industry is nevertheless young and emerging. Second, the Atlantic salmon industry is governed by a three-year production cycle. Some industries are such that firms can respond quickly to changing supply, demand, or other market conditions by adjusting output, employment or prices. Unlike these industries, the supply of U.S. Atlantic salmon, and the corresponding level of labor and other resources necessary to produce that supply, are largely fixed by production decisions made in previous years. Domestic producers' output of adult salmon is essentially a function of the amount of 'juvenile' Atlantic salmon produced in prior years."257
534. With regard to the non-financial indicators, the USITC had made the following observations. Firstly, the US market for fresh and chilled Atlantic salmon had grown strongly over the period of investigation, as indicated by data on annual apparent consumption, by quantity and by value. Secondly, production and production capacity of juvenile Atlantic salmon (eyed eggs, fry and smolt) had risen substantially from 1987 to 1989; however, this production and production capacity had leveled off in the full year 1990. Production of adult Atlantic salmon had expanded by more than 200 per cent from harvest season 1987-1988 to 1989-1990. Thirdly, annual shipments in terms of quantity of juvenile Atlantic salmon had grown from 1987 to 1989, followed by a leveling off in 1990. In terms of value, annual smolt shipments had increased several-fold from 1987 to 1989 and had further increased in 1990. Shipments by quantity of gutted Atlantic salmon had tripled from 1987-1988 to 1989-1990; in value terms these shipments had also reflected growth during the period of investigation. Finally, the number of production and related workers had more than doubled in the period 1987 to 1989 and comparable increases had occurred in the hours worked and total compensation. Employment figures for January-September 1990 had been higher than those for the same period in 1989.
535. With regard to the financial indicators, the USITC had stated that:
"The financial performance of the domestic industry stands in stark contrast to the production and trade figures. From 1987 to 1988, the industry's financial condition improved markedly. Net sales jumped more than four times. After posting a large operating loss in 1987, the domestic industry recorded an overall operating profit in 1988. However, the financial state of the U.S. Atlantic salmon industry declined precipitously in 1989. Net sales decreased from 1988 to 1989 while cost of goods sold and general, selling and administrative costs increased. Operating losses in 1989 were enormous. U.S. producers experienced a severe negative cash flow in 1989. The number of firms reporting operating losses increased from 1988 to 1989. For the period January-September 1990, net sales were well above the level recorded in 1989; nevertheless, the industry recorded a significant operating loss and negative cash flow. As a result of financial setbacks, the largest U.S. producer, Ocean Products, Inc., ceased operations. In August 1990, Ocean Products sold its assets to a Canadian firm, Connors Brothers Ltd., at terms that for purposes of confidentiality we can only describe as favorable. Connors Acquaculture, Inc., began operations in September 1990 using the assets purchased from Ocean Products."258
257 USITC Determination, pp. 11-12 (footnote omitted). ↩
258 USITC Determination, p.14, footnotes omitted. ↩
[Page 144]
536. After discussing these various indicators of the condition of the domestic industry, the USITC evaluated the data before it for purposes of determining whether the domestic industry in the United States was experiencing material injury. With respect to the non-financial indicators, the USITC observed that because the US Atlantic salmon industry was young, it was not unexpected to find expansion in such factors as capacity, production, shipments, and employment, as was seen between 1987 and 1989. It was also noted that steady or increasing employment was expected also because of the three-year production cycle in the industry. The USITC then noted that the increase in capacity and production of juvenile salmon had largely levelled off since 1989, despite increasing domestic demand in 1990 and observed that, given the nature of the production cycle, a flattering in growth of production of young salmon indicated that production of adult salmon would flatten as well. From these observations, the USITC concluded that:
"... the US industry is not presently on the road to further expansion to achieve economies of scale in production which might enable it to lower unit costs and re-establish operating profits."259
With respect to the financial indicators, the USITC considered that, while the financial performance of a new industry might be affected by start-up costs, given that the industry had been profitable in 1988, its more recent financial performance was worse than would be anticipated even taking into account start-up conditions. In addition, the USITC pointed to the fact that in 1990 the industry continued to post a failing financial performance despite having been in operation for several years. The USITC had summarized its conclusions as follows:
"In sum, we find that the U.S. Atlantic salmon industry is experiencing material injury, based on its extremely negative financial performance including the failure of its largest producer in 1990. We also note the leveling of growth in production of juvenile salmon, which suggests a stagnation in the growth of the industry despite growing U.S. demand."260
537. The Panel considered, in light of its review of the analysis undertaken by the USITC that the USITC had not failed to carry out "an evaluation of all relevant economic factors and indices having a bearing on the state of the industry" as provided for in Article 3:3. The factors considered by the USITC (consumption, production, production capacity, shipments, employment sales, profits and operating losses, cash flow) were specifically mentioned in the (illustrative) list of "relevant economic factors and indices" in Article 3:3.
538. The Panel further observed that the statements made by the USITC on the negative financial performance of the industry were supported by the data before the USITC. Table 7 on p.A-30 of the Annex to the USITC Determination contained data showing decreasing net sales, increasing costs of goods sold and general, selling and administrative expenses, and increasing operating losses (which in 1989 amounted to 52.3 per cent of net sales) and negative cash flows. Therefore, these statements could not be considered not to be based on positive evidence.
539. Having found that the statements made by the USITC on the financial performance of the industry were supported by the facts on record, the Panel considered that the arguments presented by Norway on the USITC's conclusions regarding the negative impact of the imports on the industry pertained to the weighing of the evidence before the USITC. However, it followed from the last sentence of Article 3:3 that the positive developments reflected in the indicators referred to by Norway could not per se have precluded the USITC from finding that the domestic Atlantic salmon industry was experiencing material injury. The Panel noted that these indicators had been discussed explicitly in
259 USITC Determination, p.15. ↩
[Page 145]
the USITC's determination. In the view of the Panel, the USITC had provided a reasonable explanation of why, in light of the negative financial performance of the industry, the industry was experiencing material injury, notwithstanding the growth of certain non-financial indicators.261 The Panel therefore could not find that the USITC had not carried out an objective examination of the evidence before it.
540. For the same reasons, the Panel also did not consider that, as contended by Norway, the USITC had improperly "allowed a few factors to give decisive guidance". Rather, the USITC had explicitly discussed all the evidence before it regarding the condition of the domestic industry and had reasonably explained its conclusion regarding the relative weight to be accorded to the facts before it concerning financial and non-financial indicators.
541. In light of the foregoing considerations, the Panel concluded that the findings of the USITC regarding the condition of the domestic Atlantic salmon industry were not inconsistent with the obligations of the United States under Articles 3:1 and 3:3 of the Agreement.
(2) Causal relationship between the allegedly dumped imports from Norway and material injury to the domestic industry in the United States
542. The Panel then proceeded to examine Norway's claim that the affirmative final determination of material injury made by the USITC in its investigation of imports of fresh and chilled Atlantic salmon from Norway was inconsistent with the obligations of the United States under Article 3:4 of the Agreement.
543. Norway had based this claim on three main grounds. Firstly, in making this determination the USITC had failed to ensure that injuries caused by factors other than the imports from Norway were not attributed to these imports. Secondly, the USITC had failed to demonstrate that material injury was caused to the domestic industry in the United States by the imports of Norway "through the effects of dumping". Thirdly, the USITC had not demonstrated that the imports from Norway under investigation were causing present material injury at the time the affirmative determination was made by the USITC.
(2)(i) Factors other than the imports under investigation
544. The Panel first examined Norway's claim that the USITC's treatment of factors other than the allegedly dumped imports from Norway as possible causes of injury was inconsistent with Article 3:4 of the Agreement.
545. Norway had argued that any material injury to the domestic Atlantic salmon industry in the United States was caused by factors other than imports from Norway. In this connection, Norway had mentioned the significant increase of the volume of imports of Atlantic salmon from third countries, increased supplies of substitute products, and internal problems in the United States domestic industry such as the inability of domestic producers to market Atlantic salmon on a year-round basis. These factors had been raised in the proceedings before the USITC but had been disregarded by the USITC in its determination. In the view of Norway, the treatment of these factors by the USITC was inconsistent with Article 3:4, which required that in order to demonstrate that dumped imports were causing material injury to a domestic industry, investigating authorities carry out a "thorough examination" (rather than a mere consideration) of all possible causes of material injury to the domestic
[Page 146]
industry and "isolate" and "exclude" the effects of such other possible causes of injury from the effects of the imports under investigation. By not conducting such an examination, the USITC had failed to ensure that it was not attributing to imports from Norway injury caused by other factors, and had failed to demonstrate that material injury was caused by the allegedly dumped imports from Norway.262
546. The United States had argued that the USITC had properly determined, based on volume and price effects of the imports from Norway, that these imports were causing material injury to the domestic industry in the United States. The USITC had explicitly considered the alternative factors mentioned by the Norwegian respondents and determined that, while these factors might have had an adverse impact on the industry, material injury was caused by the imports from Norway. In the view of the United States, Article 3:4 of the Agreement did not require that imports under investigation be "the" or the sole cause of material injury. Nor did this provision require investigating authorities to carry out a thorough examination of all possible causes of injury in order to exclude injury caused by factors other than imports under investigation.
547. The Panel noted that in its affirmative final determination the USITC had made the following statement with respect to other possible causes of material injury referred to by the Norwegian respondents:
"Respondents claim that any injury being experienced by U.S. producers is a result of factors other than the subject Norwegian imports. Among the alternative causes they suggest are: (1) various U.S. industry production difficulties, (2) non-subject imports, (3) the inability of U.S. producers to market their production year-round, and (4) the effects of Pacific salmon. Although some of these factors may have adversely affected the U.S. industry, we determine that an industry in the United States is materially injured by reason of subsidized and LTFV imports of fresh and chilled Atlantic salmon from Norway."263
In the light of this statement, the Panel found that, as a matter of fact, the USITC had not "disregarded" possible other causes of injury. The USITC had expressly recognized that some of these factors might have "adversely affected" the domestic industry but that this did not detract from the fact that material injury was (also) caused by the imports from Norway subject to investigation. The Panel also noted in this connection that the factors mentioned in the above quoted statement by the USITC were identical to the factors referred to by Norway in the proceedings before the Panel. There was no evidence before the Panel indicating that during the investigation the Norwegian respondents had identified other possible causes of injury which had not been considered by the USITC.
548. Given that, as noted above, the USITC had not ignored the impact of factors other than the imports under investigation, the Panel considered that the basic question before it was whether the manner in which the USITC had treated these other factors was inconsistent with the obligations of the United States under Article 3:4 of the Agreement.
549. The Panel noted that Article 3:4 provided the following:
"It must be demonstrated that the dumped imports are, though the effects4 of the dumping, causing injury within the meaning of this Agreement. There may be other factors5 which at the same
262 Norway had in this context also contested the consistency with Article 3:4 of the fact that the USITC had made one injury determination for the purpose of both its anti-dumping and countervailing duty investigation. See infra, paragraphs 572-574. ↩
263 USITC Determination, pp.21-22. ↩
[Page 147]
time are injuring the domestic industry, and the injuries caused by other factors must not be attributed to the dumped imports."
Footnote 4 provided: "As set forth in paragraphs 2 and 3 of this Article." Footnote 5 provided that:
"Such factors can include, inter alia, the volume and prices of imports not sold at dumping prices, contraction in demand or changes in the pattern of consumption, trade restrictive practices of and competition between the foreign and domestic producers, developments in technology and the export performance and productivity of the domestic industry."
The Panel was presented with divergent interpretations by the parties to the dispute of the nature of the obligations of Parties under Article 3:4 with respect to the treatment of factors other than the imports under investigation which might cause injury to a domestic industry. The basic question of interpretation before the Panel was whether, in order to demonstrate that the allegedly dumped imports caused material injury to a domestic industry, the investigating authorities were required to carry out a thorough examination of all possible causes of injury and "isolate" or "exclude" injury cased by such other factors from the effects of the imports subject to investigation. In this connection, the Panel noted that Norway had not argued that Article 3:4 required that imports under investigation be the sole cause of material injury to a domestic industry. Rather, the issue before the Panel concerned the weight accorded under Article 3:4 to an analysis of the effects of factors other than the imports under investigation for purposes of determining whether the imports under investigation were causing material injury to a domestic industry.
550. The Panel found that two aspects of the text of Article 3:4 were particularly relevant to its analysis of this question. Firstly, footnote 4 to the first sentence of Article 3:4 linked the requirement to demonstrate that the dumped imports are, through the effects of dumping, causing material injury to a domestic industry to a specific analysis of the volume and price effects of the imports and the consequent impact of the imports on the domestic industry, as set forth in Articles 3:2 and 3:3. These latter provisions contained mandatory factors to be considered in each case by investigating authorities. Secondly, the specific and mandatory nature of the analysis required under the first sentence of Article 3:4 (through the reference in footnote 4 to Articles 3:2 and 3:3) contrasted with the second sentence of Article 3:4 which provided that "There may be other factors ( ) which at the same time are injuring the domestic industry, and the injuries caused by other factors must not be attributed to the dumped imports." Furthermore, footnote 5 stated that "Such factors can include, inter alia, ...." Thus, the second sentence of Article 3:4 did not impose an express requirement that investigating authorities examine in each case on their own initiative the effects of all possible factors other than the imports under investigation. Rather, this sentence recognized the possibility that other factors were injuring the domestic industry and required that in that contingency "the injuries caused by other factors must not be attributed to the dumped imports". Furthermore, rather than specifying a priori, which other factors were relevant in this context, footnote 5 provided a non-exhaustive, illustrative list of such factors.
551. In view of this difference between the specific and mandatory nature of the analysis required under the first sentence of Article 3:4 and the manner in which the second sentence of Article 3:4 treated factors other than the imports under investigation, the Panel considered that for purposes of the causation standard in Article 3:4 the rôle of an analysis of possible factors other than the imports under investigation was qualitatively different from the rôle of the analysis of the imports under investigation. To the extent that the second sentence of Article 3:4 could be interpreted to require a consideration of factors other than the imports under investigation, such a requirement was an implicit one, following from the statement that "injuries caused by other factors must not be attributed to the dumped imports." The type of analysis which might be necessary under this sentence was not specified. By contrast,
[Page 148]
Article 3:4 was explicit and specific with regard to the required analysis of the effects of the imports under investigation.
552. The Panel therefore found that the text of Article 3:4 did not support the view that this provision required a thorough examination of all possible causes of injury, which was to be somehow just as important as the analysis under Articles 3:2 and 3:3 of the effects of the imports. The primary focus of Article 3:4 was on the examination of whether allegedly dumped imports caused the effects described in Articles 3:2 and 3:3. The second sentence of Article 3:4 did not contain an express general requirement to consider all possible factors other than the imports under investigation which might be causing injury to the domestic industry. While the need for such a consideration might be implied from the requirement that injuries caused by other factors not be attributed to the imports under investigation, it followed from the wording of the beginning of the second sentence in Article 3:4 that the relevance of a consideration of other factors was to be determined on a case-by-case basis. Furthermore, the focus of the second sentence in Article 3:4 was on the requirement that injuries caused by other factors not be attributed to the imports under investigation, not on a precise identification of the extent of injury caused by these possible other factors.
553. The Panel was of the view that its interpretation of Article 3:4 was not contradicted by the reference made by Norway to the drafting history of this provision. Norway had referred to the following draft of the provision now appearing in Article 6:4 of the Agreement on Interpretation and Application of Articles VI, XVI and XXIII of the General Agreement, contained in one of the draft Arrangements discussed during the Tokyo Round negotiations:
"The subsidized products must be [an important contributing factor in causing or threatening] [the cause of] injury. All other relevant factors adversely affecting the industry shall be considered in reaching a determination."264
The Panel considered that, as far as the rôle of factors other than imports under investigation was concerned, the second sentence of the present Article 3:4 was less categorical than the second sentence of the above quoted draft.
554. The Panel then examined the USITC's finding of a causal relationship between the imports from Norway and material injury to a domestic industry in the light of its analysis above of the requirements of Article 3:4.
555. As noted above265, the Panel considered that the primary focus of the requirement in Article 3:4 of a demonstration of a causal relationship between imports under investigation and material injury to a domestic industry was on the analysis of the factors set forth in Articles 3:2 and 3:3, i.e. the volume and price effects of the imports, and their consequent impact on the domestic industry. In this connection, the Panel recalled its conclusions regarding the findings made by the USITC with respect to these factors. Under Article 3:4 the USITC was required not to attribute injuries caused by other factors to the imports from Norway. In the view of the Panel this did not mean that, in addition to examining the effects of the imports under Articles 3:1, 3:2 and 3:3, the USITC should somehow have identified the extent of injury caused by these other factors in order to isolate the injury caused by these factors from the injury caused by the imports from Norway. Rather, it meant that the USITC was required to conduct an examination sufficient to ensure that in its analysis of the factors set forth in Articles 3:2 and 3:3 it did not find that material injury was caused by imports from Norway when material injury to the domestic industry allegedly caused by imports from Norway was in fact caused
264 MTN.NTM/W/168, 10 July 1978. ↩
[Page 149]
by factors other than these imports. The Panel therefore proceeded to consider whether in its investigation the USITC had conducted such an examination.
556. The Panel noted in this respect that Norway had argued that any material injury to the domestic Atlantic salmon industry in the United States was caused by factors other than imports from Norway, including (i) the significant increase in the volume of imports of Atlantic salmon from third countries; (ii) the effects of the increased supplies of substitute products, and (iii) the effects of internal problems in the domestic industry in the United States.
557. With regard to the first factor mentioned by Norway, the Panel noted that the USITC had before it data on the evolution of the volume of imports from all supplying countries.266 The USITC had stated in its determination, with reference to these data, that:
"Although other factors may have contributed, the decline in U.S. prices for Atlantic salmon in 1988 and 1989 was due in large part to oversupply in the U.S. market. Imports from Norway accounted for a large portion of the increased imports in 1989. This suggests that Norwegian Atlantic salmon played a rôle in the price deline."267
This statement indicated in the view of the Panel that the USITC had specifically found that imports from Norway, by reason of their proportion of the increased imports in 1989, had contributed to price declines in the United States market. The Panel considered that the USITC's finding regarding the proportion of increased imports in 1989 accounted for by imports from Norway was supported by the data before the USITC.268 When the amount of the increase in absolute import volume from Norway from 1987 to 1989 was compared to the amount of the increase in absolute import volume from other supplying countries, it could not, in the view of the Panel, reasonably be found that the USITC had attributed to the Norwegian imports effects entirely caused by imports from other supplying countries.
558. With regard to the second factor mentioned by Norway (the effects of Pacific salmon harvests) the Panel noted that the USITC had in its investigation gathered data on "related species".269 The information before the USITC indicated, inter alia, that the vast majority of Pacific salmon was sold in frozen or canned form270, and that the majority of the U.S. Pacific salmon catch was sold in export markets.271 The USITC had discussed these and other factors and concluded that the similarities between Pacific and Atlantic salmon were limited.272 While this discussion had taken place in the context of the USITC's examination of how to define the "like product", the Panel considered that the specific factors discussed by the USITC suggested that the increased availability of Pacific salmon could have had only a limited effect on domestic prices in the United States of fresh Atlantic salmon.
559. Finally, with regard to Norway's reference to internal industry problems as an alternative cause of injury to the domestic industry, the Panel noted that the USITC had stated that:
"... the financial performance of a newer industry may not be of a similar level or nature as a more mature industry due to start-up costs or other factors. However, given that the industry
266 See Annex 1 to this Report. ↩
267 USITC Determination, p.19, footnotes omitted. ↩
268 See Table 17 in Annex 1 to this Report. ↩
269 See in particular Appendix D at pp.B-45-61 of the USITC Determination. ↩
270 USITC Determination, pp.B-46-47. ↩
271 USITC Determination, p.B-48. ↩
272 USITC Determination, pp.6-7. ↩
[Page 150]
was profitable in 1988, its more recent financial performance is worse than would be anticipated even taking into account start-up conditions."273
560. The Panel considered on the basis of this examination of the data contained or referred to in the USITC Determination with regard to these alternative causes of material injury mentioned by Norway, that the USITC had not failed to conduct an examination of these factors sufficient to ensure that it did not find that material injury was caused by imports from Norway when material injury to the domestic industry allegedly caused by imports from Norway was in fact caused by factors other than these imports.
561. The Panel concluded, in the light of the foregoing considerations, that the analysis by the USITC of factors other than the imports from Norway under investigation was not inconsistent with the obligations of the United States under Article 3:4 of the Agreement.
(2)(ii) Material injury caused to the domestic industry "through the effects of dumping"
562. The Panel then turned to Norway's claim that the USITC's affirmative final determination of injury in this case was inconsistent with the obligations of the United States under Article 3:4 because the USITC had not determined whether material injury was caused by the imports from Norway "through the effects of dumping".
563. The arguments presented to the Panel by the parties offered different interpretations of the meaning of the first sentence of Article 3:4 of the Agreement.
564. Norway's argument was essentially that, in order to give effect to the phrase "through the effects of dumping" in the first sentence of Article 3:4, this sentence had to be interpreted to require that the injury analysis extend to factors other than those described in Articles 3:2 and 3:3. As an example of an additional element the consideration of which was required to give effect to the phrase "through the effects of dumping", Norway had mentioned the margin of dumping found in a given case. Norway had referred to the drafting history of Article 6:4 of the Agreement on Interpretation and Application of Articles VI, XVI and XXIII of the General Agreement in support of its view on the interpretation of this phrase. The United States had argued that footnote 4 ad Article 3:4 defined "the effects of dumping" in the first sentence of Article 3:4 as the effects of the imports under investigation, as described in Articles 3:2 and 3:3 of the Agreement. Under this interpretation, in order to give effect to the phrase "through the effects of dumping" it was not necessary to analyse any factors other than the effects of the imports as set forth in Articles 3:2 and 3:3. The United States argued that the drafting history of Article 6:4 of the Agreement on Interpretation and Application of Articles VI, XVI and XXIII of the General Agreement did not support the interpretation advocated by Norway.
565. The Panel considered that the key legal question in this respect concerned the relationship between the term "through the effects of dumping" and the effects of dumped imports described in Articles 3:2 and 3:3. Under the interpretation presented by Norway, the Agreement required an analysis in each case of whether and how the effects of the imports under Articles 3:2 and 3:3 were the "effects of dumping"; under the interpretation advanced by the United States, the effects of the imports under Articles 3:2 and 3:3 by definition were the "effects of dumping".
566. The Panel noted that, if the text of footnote 4 was included in the first sentence of Article 3:4, this sentence could be rewritten as follows:
273 USITC Determination, p.15, footnote omitted. ↩
[Page 151]
"It must be demonstrated that the dumped imports are, through the effects as set forth in paragraphs 2 and 3 of this Article of dumping, causing injury within the meaning of this Agreement."
567. What needed to be demonstrated according to this sentence was that "the dumped imports are causing injury within the meaning of this Agreement". This demonstration required an analysis of the "effects as set forth in paragraphs 2 and 3 of this Article of dumping". In other words, dumped imports cause injury through the effects described in Articles 3:2 and 3:3. However, this sentence did not state that it must be demonstrated that "the effects as set forth in paragraphs 2 and 3 of this Article" are "the effects of dumping". Rather, it defined "the effects of dumping" as the effects described in Articles 3:2 and 3:3, i.e. the volume and price effects of the dumped imports and consequent impact of these imports on the domestic industry.
568. The Panel noted Norway's argument that, if Article 3:4 required only an analysis of the effects of imports under Articles 3:2 and 3:3, there would be no distinction between the determination of the existence of material injury and the determination of the cause of injury. The principle of effective treaty interpretation ruled out such an interpretation, under which the phrase "through the effects of dumping" would be superfluous.
569. The Panel considered that the principle of effective treaty interpretation required that effect be given to the entire term "through the effects as set forth in paragraphs 2 and 3 of this Article of dumping." Moreover, Article 3 did not treat the factors set forth in Articles 3:2 and 3:3 only as as indicia of the existence of material injury but also as indicia of a causal relationship between the dumped imports and material injury to a domestic industry. The text of the first sentence of Article 3:4 made it clear that "the dumped imports" were at the centre of the causation analysis required under this provision. Therefore, Article 3 did not treat "the effects of the dumping" as the cause of material injury and the effects of the imports under Articles 3:2 and 3:3 as mere indicators of the existence of material injury.
570. The Panel did not consider that the reference made by Norway to the drafting history of Article 6:4 of the Agreement on Interpretation and Application of Articles VI, XVI, XXIII of the General Agreement warranted a different interpretation of the first sentence of Article 3:4. Norway had referred to a draft dated 13 February 1979 which read as follows:
"It must be demonstrated that, through the effects of the subsidy, the subsidized imports are causing injury within the meaning of this Arrangement. There may be other factors which at the same time are injuring the domestic industry, and the injuries caused by other factors must not be attributed to the subsidized imports."274
However, this draft was followed by a draft dated 21 February 1979 in which what was now footnote 19 ad Article 6:4 was added after the word "effects".275 Thus, what needed to be interpreted was not only the fact that the drafters of the Agreement introduced the term "through the effects of the subsidy" but also the fact that they almost immediately qualified this term by inserting a footnote referring to Articles 6:2 and 6:3. Taken together, the "through the effects of the subsidy" language and the footnote established a link between Article 6:4 and Articles 6:2 and 6:3, a link which had been absent from previous drafts. As such, the term "through the effects of the subsidy", together with the footnote,
274 Subsidies/Countervailing Measures, Working Paper prepared by some delegations, 13 February 1979, p.15. ↩
275 MTN.NTM/W/220, 21 February 1979, p.15. ↩
[Page 152]
provided greater precision as to the manner in which the causal relationship between the subsidized imports and material injury to a domestic industry was to be established.
571. The Panel concluded that by treating the "effects of dumping" in the first sentence of Article 3:4 to mean the effects of dumped imports, set forth in Articles 3:2 and 3:3, the USITC had not acted inconsistently with the obligations of the United States under Article 3:4.
572. The Panel then analysed Norway's claim that the USITC had acted inconsistently with Article 3:4 by making one combined injury determination for purposes of both the anti-dumping and the countervailing duty investigation.
573. The Panel recalled its conclusion that the primary focus of the causation analysis required by Article 3:4 was on the effects of the dumped imports, as set forth in Articles 3:2 and 3:3.276 The Panel noted tht Articles 6:2 and 6:3 of the Agreement on Interpretation and Application of Articles VI, XVI and XXIII of the General Agreement described in an identical manner the volume and price effects, and the consequent impact of imports on the domestic industry, to be considered in a countervailing duty investigation. Given that in the anti-dumping and countervailing duty investigations by the USITC of imports of Atlantic salmon from Norway the same imports had been investigated and that the investigation periods had been identical, it appeared to the Panel that there would have been no basis for the USITC to distinguish between the effects of the dumped imports (in terms of Articles 3:2 and 3:3 of the Agreement) and the effects of the subsidized imports under investigation (in terms of Articles 6:2 and 6:3 of the Agreement on Interpretation and Application of Articles VI, XVI and XXIII of the General Agreement).
574. The Panel therefore concluded that, by making one determination of injury for the purposes of both the anti-dumping and the countervailing duty investigation, the USITC had not acted inconsistently with the obligations of the United States under Article 3:4 of the Agreement.
(2)(iii) Whether the imports under investigation were causing present material injury to the domestic Atlantic salmon industry in the United States
575. The Panel then proceeded to consider Norway's argument that the affirmative final determination of injury by the USITC was inconsistent with Article 3:4 because the USITC had failed to determine that at the time of this determination imports of Atlantic salmon from Norway were causing present material injury to the domestic industry in the United States.
576. In support of its claim, Norway had pointed out that Article 3:4 required that it be demonstrated that imports "are ... causing" material injury. It followed from the present tense of the first sentence of Article 3:4 that material injury had to be determined to be caused by the imports at the time of the determination. Norway had argued in this context that the purpose of the imposition of anti-dumping duties was to prevent future harm to a domestic industry resulting from imports which were presntly causing material injury. In the case under consideration, even if imports from Norway were causing injury to the domestic industry at the time of the filing of the petition (March 1990), these imports were no longer causing such injury at the time of the final determination by the USITC (April 1991).
577. Norway had based its argument on the absence of present material injury at the time of the final determination by the USITC on six specific elements: first, the fact that the volume of imports from Norway had declined prior to the initiation of the anti-dumping duty investigation. Second, the decline over the period of investigation of the market share of Norwegian imports. Third, the fact
[Page 153]
that Norwegian salmon commanded a price premium over domestically produced salmon in the United States. Fourth, the fact that domestic producers in the United States had tripled their market share over the investigation period. Fifth, the fact that imports from Norway had declined after the imposition of provisional measures due to factors such as exchange rate changes and finally, the failure of the United States to take action to prevent injury caused by other factors from being attributed to the imports from Norway.
578. The United States had argued that the USITC had in fact determined that the domestic industry was experiencing material injury at the time of its final determination and had referred in this respect to the findings made by the USITC regarding the continuing injurious effects of the Norwegian imports, inter alia, in the form of financial losses. In addition, the United States had argued that the decline in 1990 of the volume of imports from Norway and the increase in prices of the Norwegian imports were the expected result of the investigation and of the imposition of provisional measures. The United States had also pointed out that Article 3:2 explicitly contemplated a retrospective analysis. Finally, if Article 3:4 were interpreted to require a negative final determination whenever imports declined and prices rose following the imposition of provisional measures, the purpose of provisional measures under Article 10 would be undermined.
579. The Panel found that, while Norway had made a separate claim under Article 3:4 as to an alleged failure of the USITC to determine whether imports from Norway were causing present material injury at the time of the determination made by the USITC, in fact each of the specific arguments raised by Norway in support of this claim had already been addressed by the Panel as part of its examination of Norway's claims on other aspects of the injury determination made by the USITC. Thus, Norway's arguments regarding the evolution of the volume of imports had been examined by the Panel under Articles 3:1 and 3:2 of the Agreement. Norway's argument on the premium commanded by imports from Norway had been addressed in the Panel's examination of the USITC's analysis of the price effects of the imports. Norway's argument regarding the increased market share of domestic producers had been addressed by the Panel under Article 3:3. Finally, Norway's argument concerning the alleged failure of the USITC to prevent injury caused by other factors from being attributed to the imports from Norway had already been examined by the Panel under Article 3:4.
580. The Panel further considered that the requirement in the first sentence of Article 3:4 that it must be demonstrated that imports "are ... causing material injury" had to be interpreted consistently with other provisions of the Agreement. An interpretation of this sentence under which investigating authorities would somehow be obliged to continue to collect data up to the time of the final determination would undermine other provisions of the Agreement, in particular those relating to rights of interested parties concerning access to information used by the investigating authorities (e.g. Article 6:2). An adequate protection of procedural rights of interested parties therefore required that determinations of (present) material injury be based on a defined record of facts before the investigating authorities. In this respect, the Panel noted that the factors referred to by Norway in support of its claim pertained to factual developments over the period of investigation which had been considered by the USITC, on the basis of the record before it.
581. In light of the foregoing considerations, the Panel concluded that the United States had not acted inconsistently with its obligations under Article 3:4 with respect to the issue raised by Norway concerning the existence of present material injury caused by the imports from Norway.
582. In light of its conclusions in paragraphs 510, 527, 541, 561, 571, 574 and 581, the Panel concluded that the imposition by the United States of the anti-dumping duty order on imports of fresh and chilled Atlantic salmon from Norway was not inconsistent with the obligations of the United States under the Agreement by reason of the affirmative final determination of injury by the USITC.
[Page 154]
D. CONTINUED IMPOSITION OF THE ANTI-DUMPING DUTY ORDER
583. The Panel then considered the claim presented by Norway under Article 9:1 of the Agreement regarding the continued imposition by the United States of the anti-dumping duty order on imports of fresh and chilled Atlantic salmon from Norway.
584. Norway had argued that the continued imposition of the anti-dumping duty order was inconsistent with the provision in Article 9:1 that "an anti-dumping duty shall remain in force only as long as, and to the extent necessary to counteract dumping which is causing injury." In the view of Norway, it followed from this provision that the United States was under an obligation to terminate the anti-dumping duties on imports of Atlantic salmon from Norway because, firstly, at the time of the affirmative final determination of injury by the USITC no material injury was caused to the domestic industry in the United States by imports from Norway and, secondly, imports of Atlantic salmon from Norway were no longer causing any present material injury to this industry.
585. The United States had argued that Norway was factually incorrect in contending that at the time of the final determination by the USITC no material injury to the domestic industry in the United States had been caused by the imports from Norway. Regarding events occurring subsequent to the imposition of the anti-dumping duty order, the United States had argued that Norway could seek a review by the investigating authorities of the United States of the need for the continuation of this anti-dumping duty order. Finally, the United States had argued that a lack of injury following the imposition of anti-dumping duties was not surprising since the Agreement presumed that these duties might remove the injury caused to the domestic industry by these imports.
586. The Panel considered that the first argument presented by Norway to support its claim under Article 9:1 - the alleged absence of material injury caused by the Norwegian imports at the time of the final determination of injury by the USITC - raised an issue which logically pertained not to the consistency with the Agreement of the continuation of the anti-dumping duty order but to the consistency with the Agreement of the final determination of injury which, together with the affirmative final determination of dumping, had led to the imposition of this order on 12 April 1991. In this respect the Panel observed that this argument of Norway under Article 9:1 was identical to the argument presented by Norway in support of its view that the USITC's determination was inconsistent with Article 3:4 of the Agreement because the USITC had failed to determine that imports from Norway were causing present material injury to a domestic industry at the time of this determination. The Panel recalled its conclusion on this issue in paragraph 581.
587. With respect to the second argument raised by Norway under Article 9:1 - the fact that imports from Norway were no longer causing injury to the domestic industry in the United States - the Panel recalled that the central issue in the dispute before it was the consistency with the Agreement of the steps taken by the United States in imposing this order on 12 April 1991 and that all claims presented by Norway (such as those based on Articles 2 and 3) pertained to the consistency with the Agreement of the initial imposition of this order. Norway's second argument under Article 9:1 differed fundamentally from these other claims in that it referred to developments since the imposition of the anti-dumping duty order and related to the obligations of the United States concerning the continuation of this order, rather than to the obligations of the United States with respect to the imposition of this order.
588. The Panel noted that Norway had stated that imports of Atlantic salmon from Norway were no longer causing any present injury to the domestic industry. The Panel considered that this argument, even if it were factually correct, could not by itself be a sufficient basis to conclude that the United States had acted inconsistently with Articles 9:1 by continuing to impose anti-dumping duties on imports of Atlantic salmon from Norway. If the mere fact that, following the imposition of anti-dumping duties,
[Page 155]
the imports in question were no longer causing injury were sufficient to require a Party to terminate the imposition of these duties, the logical result would be that any anti-dumping duty which was effective in removing injury to a domestic industry had to be withdrawn immediately. The Panel considered that this interpretation of Article 9 would make ineffective the other provisions of the Agreement. An interpretation of Article 9 consistent with other provisions of the Agreement required that in considering whether a Party was acting inconsistently with Article 9:1, account be taken of the effect of the imposition of the anti-dumping duties.
589. In light of the foregoing considerations, the Panel concluded that the continued imposition of anti-dumping duties on imports of fresh and chilled Atlantic salmon from Norway was not inconsistent with the obligations of the United States under Article 9:1 of the Agreement.
VIII. CONCLUSIONS
590. the Panel recalled its conclusions with respect to the preliminary objections of the United States, that:
591. The Panel further recalled its conclusion in paragraph 363 above that the initiation of the antidumping investigation was not inconsistent with the obligations of the United States under Article 5:1 of the Agreement.
592. The Panel further recalled its conclusions with respect to the claims of Norway regarding the final determination of dumping by the Department of Commerce, that:
[Page 156]
593. The Panel further recalled its conclusion in paragraph 582 above that the imposition by the United States of an antidumping order on imports of fresh and chilled salmon from Norway was not inconsistent with the obligations of the United States under the Agreement by reason of the affirmative final determination of material injury by the USITC.
594. The Panel further recalled its conclusion in paragraph 589 above that the continued imposition of anti-dumping duties on imports of fresh and chilled Atlantic salmon from Norway was not inconsistent with the obligations of the United States under Article 9:1 of the Agreement.
595. The Panel finally recalled its conclusions that:
[Page 157]
had acted inconsistently with its obligations under Article 2:6 of the Agreement (paragraph 472).
The Panel therefore concluded that to this extent the imposition by the United States of an antidumping order on imports of fresh and chilled salmon from Norway was inconsistent with the obligations of the United States under the Agreement.
596. In considering the recommendation to be addressed to the Committee on Anti-Dumping Practices, the Panel took into account that the grounds upon which it had found that the United States had imposed anti-dumping duties inconsistently with its obligations under the Agreement pertained specifically to certain aspects of the methodology for calculating margins of dumping. It could not be presumed that a methodology of calculating dumping margins consistent with the Panel's findings on these aspects would necessarily result in a determination that no dumping existed rather than in a determination that duties were to be imposed at a different rate. The Panel therefore found that in this situation it could not recommend that the Committee request the United States to revoke the anti-dumping duty order and reimburse any duties paid or deposited under this order, as requested by Norway.
597. The Panel therefore recommends that the Committee request that the United States bring its measures with respect to the imposition on 12 April 1991 of an anti-dumping duty order on imports of fresh and chilled Atlantic salmon from Norway into conformity with its obligations under the Agreement and that, to this end, the United States reconsider the affirmative final determination of dumping, consistent with the Panel's findings under Articles 2:4 and 2:6, and take such measures with respect to this anti-dumping duty order, as imposed on 12 April 1991, as may be warranted in the light of that reconsideration.
[Page 158]
ANNEXES
1. FRESH ATLANTIC SALMON: U.S. IMPORTS FROM NORWAY, CANADA, CHILE, ICELAND, THE UNITED KINGDOM, IRELAND, THE FAROE ISLANDS, AND ALL OTHER COUNTRIES, 1987-90
2. FRESH ATLANTIC SALMON: APPARENT U.S. CONSUMPTION AND SHARES OF CONSUMPTION SUPPLIED BY NORWAY, ALL OTHER COUNTRIES AND U.S. PRODUCERS, 1987-89, JANURAY-JUNE 1989, AND JANUARY-JUNE 1990
3. FRESH ATLANTIC SALMON: U.S. MONTHLY IMPORTS FROM NORWAY JANUARY 1989- DECEMBER 1990, BY VOLUME AND VALUE
4. LETTERS ADDRESSED TO THE PANEL BY NORWAY AND THE UNITED STATES ON 12 AND 13 NOVEMBER 1992 RESPECTIVELY, AND LETTER BY THE PANEL TO NORWAY DATED 20 NOVEMBER 1992
[Page 159]
1 FRESH ATLANTIC SALMON: U.S. IMPORTS FROM NORWAY, CANADA, CHILE,
ICELAND, THE UNITED KINGDOM, IRELAND, THE FAROE ISLANDS,
AND ALL OTHER COUNTRIES,1 1987-90
(USITC Publication No. 2371, Table 17, p.A-43)
| Source | 19872/ | 19882/ | 1989 | 19903/ |
|---|---|---|---|---|
| Quantity (1,000 kg.) | ||||
| Norway ................................ | 7,610 | 8,895 | 11,396 | 7,699 |
| Canada ................................ | 700 | 1,137 | 2,958 | 4,889 |
| Chile ................................. | 42 | 118 | 557 | 4,077 |
| Iceland ............................... | 78 | 322 | 472 | 1,012 |
| The United Kingdom .................... | 529 | 353 | 1,011 | 901 |
| Ireland ............................... | 47 | 310 | 426 | 333 |
| The Faroe Islands ..................... | - | 35 | 478 | 53 |
| All other countries ................... | 600 | 177 | 207 | 133 |
| Total ............................. | 9,606 | 11,347 | 17,505 | 19,098 |
| Value (1,000 dollars)4/ | ||||
| Norway ................................ | 74,404 | 89,987 | 93,672 | 66,440 |
| Canada ................................ | 5,719 | 10,499 | 22,145 | 36,636 |
| Chile ................................. | 316 | 962 | 3,876 | 27,296 |
| Iceland ............................... | 792 | 3,061 | 3,262 | 7,084 |
| The United Kingdom .................... | 5,588 | 4,122 | 9,167 | 8,288 |
| Ireland ............................... | 471 | 3,058 | 3,486 | 2,887 |
| The Faroe Islands ..................... | - | 349 | 3,472 | 415 |
| All other countries ................... | 5,189 | 1,699 | 1,473 | 1,064 |
| Total ............................. | 92,479 | 113,737 | 140,553 | 150,110 |
| Unit value (dollars per kg.) | ||||
| Norway ................................ | $9.78 | $10.12 | $8.22 | $8.63 |
| Canada ................................ | 8.17 | 9.23 | 7.49 | 7.49 |
| Chile ................................. | 7.58 | 8.19 | 6.95 | 6.70 |
| Iceland ............................... | 10.14 | 9.52 | 6.91 | 7.00 |
| The United Kingdom .................... | 10.57 | 11.69 | 9.07 | 9.20 |
| Ireland ............................... | 10.10 | 9.88 | 8.19 | 8.66 |
| The Faroe Islands ..................... | (5/) | 10.08 | 7.26 | 7.87 |
| All other countries ................... | 8.64 | 9.62 | 7.13 | 7.99 |
| Average ........................... | 9.63 | 10.03 | 8.03 | 7.86 |
1 Includes imports from countries where no Atlantic salmon industry is known to exist. This product is believed to be misreported. ↩
2 1987-88 data were estimated by calculating the ratios of fresh whole Atlantic salmon to all fresh whole salmon as observed in 1989 US import data, and applying those ratios to comparable country-specific 1987 and 1988 quantity and value data for all fresh whole salmon. For Canada and Chile, further adjustments were made using port-of-entry import data and foreign production data, respectively. ↩
3 Includes imports under HTS statistic number 0302.12.0062, "fresh and chilled salmon not elsewhere specified or included", which are believed to be Atlantic salmon. ↩
Source: Compiled from official US import statistics, adjusted as specified.
[Page 160]
2. FRESH ATLANTIC SALMON: APPARENT U.S. CONSUMPTION AND SHARES OF CONSUMPTION SUPPLIED BY NORWAY, ALL OTHER COUNTRIES, AND U.S. PRODUCERS, 1987-89, JANUARY-JUNE 1989, AND JANUARY-JUNE 1990
(USITC Publication No. 2371, Table 18, p.A-45)
| Item | 1987 | 1988 | 1989 | January-June-- | |
|---|---|---|---|---|---|
| 1989 | 1990 | ||||
| Quantity | |||||
| Apparent US consumption (1,000 pounds) ........ | *** | 26,916 | 41,705 | 20,449 | 26,502 |
| Shares of apparent consumption supplied by-- | |||||
| Norway (percent) ................. | *** | 72.9 | 60.2 | 60.1 | 42.2 |
| All other countries (percent) .... | *** | 20.1 | 32.3 | 33.8 | 51.1 |
| All imports (percent) .......... | *** | 92.9 | 92.5 | 93.8 | 93.4 |
| US producers (percent) ........... | *** | 7.1 | 7.5 | 6.2 | 6.6 |
| Total (percent) .............. | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 |
| Value | |||||
| Apparent US consumption (1,000 dollars) ........ | *** | 134,349 | 165,504 | 86,844 | 101,734 |
| Shares of apparent consumption supplied by-- | |||||
| Norway (percent) ................. | *** | 74.0 | 62.5 | 61.7 | 47.0 |
| All other countries (percent) .... | *** | 19.5 | 31.3 | 32.2 | 47.3 |
| All imports (percent) .......... | *** | 93.5 | 93.8 | 94.0 | 94.2 |
| US producers (percent) ........... | *** | 6.5 | 6.2 | 6.0 | 5.8 |
| Total (percent) .............. | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 |
Source: Compiled from data submitted in response to questionnaires of the US International Trade Commission and from official US import statistics.
Note: --Because of rounding, figures may not add to the totals shown.
[Page 161]
3. FRESH ATLANTIC SALMON: U.S. MONTHLY IMPORTS FROM NORWAY
JANUARY 1989-DECEMBER 1990, BY VOLUME AND VALUE
1989 imports from Norway
| Kilograms | $1,000 | |
|---|---|---|
| January | 1,045,47 | 9,634 |
| February | 931,553 | 8,436 |
| March | 905,392 | 8,022 |
| April | 947,617 | 8,117 |
| May | 850,993 | 7,173 |
| June | 890,290 | 7,124 |
| July | 907,416 | 7,069 |
| August | 777,686 | 6,076 |
| September | 931,664 | 7,290 |
| October | 1,042,322 | 8,246 |
| November | 1,016,305 | 7,758 |
| December | 1,148,849 | 8,728 |
| Total | 11,395,566 | 93,672 |
1990 imports from Norway
| Kilograms | $1,000 | |
|---|---|---|
| January | 779,602 | 6,285 |
| February | 743,648 | 6,147 |
| March | 829,449 | 7,075 |
| April | 977,763 | 8,393 |
| May | 916,710 | 8,030 |
| June | 830,847 | 7,302 |
| July | 847,433 | 7,183 |
| August | 650,351 | 5,784 |
| September | 426,714 | 3,794 |
| October | 287,832 | 2,651 |
| November | 230,270 | 2,073 |
| December | 188,646 | 1,723 |
| Total | 7,699,265 | 66,440 |
Source: Data included in the record of the USITC's investigation and provided by the United States to Norway on 8 June 1991.
[Page 162]
4. LETTERS ADDRESSED TO THE PANEL BY NORWAY AND THE
UNITED STATES ON 12 AND 13 NOVEMBER 1992 RESPECTIVELY,
AND LETTER BY THE PANEL TO NORWAY DATED 20 NOVEMBER 1992
Letter from the Delegation of Norway277
12 November 1992
Dear Mr. Chairman,
The Government of Norway is in the process of reviewing the reports of the panels on anti-dumping duties and countervailing duties imposed on imports of fresh and chilled Atlantic salmon from Norway.
The panels appear in general as not having been prepared to question the contents of the information applied by the US authorities in the investigations, nor to take a stand regarding the US' decisions made on the basis of such information. In Norway's view, some aspects of the Panel reports raise questions of principle, and could have ramifications of significance for the international trading system. The Panels seem to have reached conclusions deviating from a number of previous panel recommendations, and they have apparently based themselves on a broad interpretation of the requirements expressed in the General Agreement's Article VI concerning the obligations incumbent upon a party invoking exceptions to the general GATT obligations.
Norway requests that the Panels reconsider the issues raised in this communication. Norway furthermore requests that the reports to the Committees reflect this request for reconsideration before circulating the reports to the members of the Committees, as well as the results of such consideration. Finally, Norway reserves its rights to pursue other aspects of the reports.
Sincerely,
Erik Selmer (signed)
Ambassador
277 The texts contained in this Annex have been circulated to the Committee for the sake of transparency and in response to the request by Norway that the Panel's Report to the Committee reflect its request for reconsideration of certain issues. This Annex does not constitute an integral part of the Report nor should the comments made in the letter by the Panel be seen as in interpretation of the Panel's findings and conclusions ↩
[Page 163]
Initiation Standards
The view of the Panels as stated in the panel reports is that it was reasonable for the DOC to initiate the investigations relying solely upon a statement in the petition concerning support from the US salmon industry, thus implying that the DOC is not required under the Codes to satisfy itself on its own, prior to investigation, that a petition is filed on behalf of the domestic industry.
Norway regards the Panels' view to be unpersuasive in respect of the matter of principle, i.e. the content of the requirement in the AD Code's Article 5:1 and the CVD Code's Article 2:1, respectively. In Norway's view, the Panels' findings are contrary to the Code requirements as expressed in previous panel reports.
Norway notes that in the Swedish Steel case, in which no member of the domestic industry stated any opposition or lack of support (Swedish Steel panel report at paragraph 3.19), the Panel found that the petition did not on its face support the statement in the petition that it was filed on behalf of the domestic industry because it provided no statistical information to support that position, nor did the DOC obtain such information prior to initiation. Swedish Steel panel report at paragraph 5.14. Neither did the petition on Salmon contain any statistical information to support its statement (beyond a number of companies which, by itself, could not indicate any proportion of production); nor did the DOC obtain any statistical information prior to initiation.
The requirements in Article 5:1 of the AD Code concerning initiation were also discussed in the Mexico Cement case (United States - Anti-dumping duties on grey Portland cement and cement clinker from Mexico). The Panel found that Article 5:1 contained a mandatory requirement for the investigating authorities to satisfy themselves, prior to initiation, that a petition was filed by or on behalf of the domestic producers. Mexico Cement panel report at paragraph 5.29. The Panel observed that the information on the extent of the support was not available to the investigating authorities prior to initiation and in fact had not been sought by the DOC or been provided to it by the ITC at any time during the investigation. Mexico Cement at paragraph 5.33. The Panel accordingly concluded that the US initiation of the AD investigation of cement from Mexico was inconsistent with Article 5:1 of the AD Code. Mexico Cement at paragraph 5.34 and 6.1.
The Panels are furthermore of the view that the DOC could continue to rely on the statement in the petition concerning industry support even though one member of the domestic industry had written in to state its opposition to the petition and one of the two Associations of US farmers had withdrawn its original vote of support for the petition. Subsidies Report at paragraph 29 and Anti-Dumping Report at paragraph 362.
In Norway's opinion, it is not reasonable to assume that every individual member of the Washington Fishgrowers Association continued to support the petition once the Association noted that it did not. One cannot assume that the Board of the Association's action in writing a letter stating that it did not support the petition was a unilateral act, not reflecting any change in opinion by any of the Association's members. This is borne out by the fact that in later submissions, there were only 13 members of the petitioning coalition (only 11 of whom were among the original 21 members), none of whom were members of the Association which withdrew support. Norwegian 1st Subsidies and AD Submissions at 10 and Appendix 7.
Finally, Norway notes that in the countervailing duty investigation it would have been futile for Norway to raise the standing issue since the DOC's stated policy is only to consider the issue if raised by a member of the domestic industry. Norwegian 1st Subsidies and AD
[Page 164]
Submissions at 8-9 and Norwegian 2nd Subsidies and AD Submissions at 9. DOC refuses to consult with parties potentially adversely affected by an investigation (e.g., exporters, importers, foreign governments) prior to initiation of an anti-dumping investigation. Indeed, the DOC does not notify anyone of the opportunity to object until initiation. Mexican Cement, paragraph 5.32. Therefore, Norway had no opportunity to raise the issue prior to initiation of the anti-dumping investigation, and it would have been futile to do so in the countervailing duty investigation.
Injury
Norway also requests the Panels to reconsider their views on three aspects of the injury investigation. The first concerns the Panel's determination that the first sentence of Article 6:4 of the Subsidies Code and Article 3:4 of the Anti-Dumping Code requires only the analysis provided for in Articles 6:2 and 6:3 or 3:2 and 3:3, respectively to determine causation. Subsidies Report at paragraph 134 and Anti-Dumping Report at paragraph 571. Such an analysis eliminates separate causation findings from the scheme of the injury investigation. If the analysis suggested by the Panels were correct, then once an investigating authority determined that injury existed in accordance with Article 6:1 or 3:1 of the Codes, based solely on the analysis in Articles 6:2 and 6:3 and Articles 3:2 and 3:3, it would automatically be found that a causal connection existed, since the analysis would be identical with regard to both the AD and CVD case. Norwegian 2nd CVD Submission at 30, 39-40 and Norwegian 2nd AD Submission at 54, 62.
The second aspect of the injury investigation which should be reconsidered is the interpretation of the second sentence of Article 6:4 or Article 3:4 of the Subsidies Code or the Anti-Dumping Code, respectively. The Panel determinations indicate that it is sufficient that the authorities do not ignore other factors rather than applying the Code language that the investigating authority must not attribute injury from other factors to the effects of the subsidies or dumping. Subsidies Panel Report at paragraph 110 and Anti-Dumping Panel Report at paragraph 547. If this analysis were correct, it would eliminate the need for this sentence in its entirety. Such a result is inconsistent with the accepted norms of treaty interpretation, as well as prior GATT panels. Canadian countervailing duties on grain from the United States, SCM, paragraph 5.2.8. The Panels thus endorse the US position in the present cases, i.e. that it is sufficient for a positive injury determination that the imports under investigation were found to be a cause of injury, as long as other possible causes of injury are enumerated. Norway regards the view of the Panels as being contrary to the requirements expressed in the second sentence of Article 6:4 and Article 3:4 of the Subsidies Code and the Anti-Dumping Code, respectively.
Thirdly, the ITC's injury determination was based on effects which occurred in 1989, and the ITC justified its finding of "present" material injury by referring to injury in the form of continuing effects (USITC report at 21). Norway is of the opinion that the Panels in their review of the US injury determination were incorrect in not contesting that the US could disregard the 1990 import records. Inclusion of the 1990 records would result in a finding of additional decline in Norwegian market share, and practically no increase in import volume even in absolute terms. Norway regards the Codes as containing a requirement for the investigating authorities as to consider whether the domestic industry were being injured by the present effects of subsidies or dumping at the time of the injury determination.
The AD Panel's recommendations
Although the AD Panel concluded that the United States had imposed anti-dumping duties inconsistently with its obligations under the AD Code pertaining to certain aspects of the methodology for calculating margins of dumping, it did not recommend a specific remedy as requested by Norway, i.e. that the AD Committee request the United States to revoke the anti-
[Page 165]
dumping duty order and reimburse any duties paid or deposited under this order, as requested by Norway. In Norway's view, the AD Panel should, however, in keeping with previous panel recommendations, have made such a recommendation insofar as the methodology of calculating dumping margins to be applied by the US consistent with the Panel's findings results in a determination that no dumping existed, or to a reduction in the calculated duty margin. New Zealand - imports of electrical transformers from Finland, BISD 32S/70, paragraph 4.11; Canada - imposition of countervailing duties on imports of manufacturing beef from the EEC, SCM/85, paragraph 5.17; United States - imposition of anti-dumping duties on imports of stainless steel hollow products from Sweden, ADP/47, paragraph 5.24; and United States - anti-dumping duties on grey portland cement and cement clinker from Mexico, ADP, paragraph 6.2. The exception to recommending reimbursement was Grain Corn where the complaining party, the United States, did not request reimbursement. Canadian countervailing duties on grain corn from the United States, SCM, paragraphs 3.1.1 and 6.2.
* * * * *
[Page 166]
Letter from the United States Trade Representative
13 November 1992
Dear Mr. Chairman,
My authorities have instructed me to respond to the letter of November 12 from the Government of Norway requesting that the Panels reconsider various issues in their Anti-Dumping and Countervailing Duty reports.
The United States does not share the views expressed by Norway. It believes that the Panels have fully and carefully addressed all of the issues cited in the enclosure to Norway's letter. Accordingly, the United States urges the Panels to direct the secretariat to circulate the reports to signatories of the Anti-Dumping and Subsidies Committees as soon as possible.
If the Panels should choose to reopen their consideration of issues raised by Norway, my authorities ask that they be provided with a full opportunity to present our contrary views. In addition, the United States would request that two additional issues in the Anti-Dumping report be reconsidered. First, the United States does not believe that all provisions of Article 2.6 of the Code apply equally regardless of whether the normal value calculation is based on price or constructed value. Article 2.6 requires that due allowance be made for differences affecting "price comparability". Although constructed value is a proxy for price-based normal value, it is not in fact a price. Further, some conditions which affect price, including the effect of the weight of the fish, have no impact on cost; any attempt to introduce one would yield a distorted comparison.
Second, the conclusion of the Panel that the United States should have considered the effect of using the highest actual cost of six other farms for one farm not providing requested information would undercut the ability of investigating authorities to use facts available under Article 6.8. Using a production cost lower than another producer's actual costs would provide an incentive to refuse to provide information which, if provided, would result in a larger dumping margin. This is prevented only by the ability of the authorities to draw adverse inferences when requested information is not provided. Such ability to draw adverse inferences is particularly important in a case such as this where sampling is used. One would expect that if one producer in a sample of seven did not provide requested information, many more in a larger sample would not. (One cannot conclude that only one producer in a larger sample would fail to provide requested information without rejecting the premise of sampling). Absent the ability to draw adverse inferences, the ability to use sampling would be endangered.
Lest there be any confusion, allow me to repeat that the preference of the United States is not for the Panels to reopen either dispute. We believe that the Panels should direct the secretariat to circulate the reports to signatories of the Anti-Dumping and Subsidies Committees as soon as possible.
Sincerely,
C. Christopher Parlin (signed)
Legal Advisor
[Page 167]
Reply by the Panel to the Delegation of Norway
20 November 1992
Dear Ambassador Selmer,
The Panels in the disputes on anti-dumping and countervailing duties imposed by the United States on imports of salmon have carefully examined your request for a reconsideration of certain issues raised in your letter dated 12 November 1992. The points raised in your letter are virtually identical to arguments presented by Norway in the proceedings before the Panels and have been addressed by the Panels in their findings. Your letter does not identify specific questions of law or of fact which have been overlooked by the Panels. In addition, in a number of places the analysis in your letter seems to be based on a misreading of the Panels' findings. The Panels therefore have decided that the points raised in your letter do not provide a basis for a reconsideration of the Panel's findings. On 23 October, when I informed the parties to the disputes of the Panels' findings and conclusions, I indicated that the full Reports in the two disputes would be circulated to the members of the two Committees unless by 11 November 1992 both parties to the disputes requested an extension of this time period in order to continue their efforts to seek a mutually satisfactory resolution of the disputes. I conclude from your letter dated 12 November and from the letter from the delegation of the United States dated 13 November that there is no agreement between the two parties on such an extension. The Panels therefore have no choice but to direct the GATT secretariat to circulate the full Reports to the two Committees as soon as possible. In the interest of transparency, the Panels will annex to their Reports your letter dated 12 November, the letter received from the United States on 13 November and the Panels' response to your letter. I would like to offer, on behalf of the Panel, the following comments on the points raised in your letter:
1. Initiation of the anti-dumping and countervailing duty investigations
With respect to the initiation of the investigations your letter challenges the Panel's interpretation of the requirements of Article 5:1 of the Anti-Dumping Code and Article 2:1 of the Subsidies Code as being "contrary to the Code requirements as expressed in previous panel reports".
The statements in paragraphs 358-360 of the findings in the dispute on anti-dumping duties clearly indicate that the Panel considers that investigating authorities are required to evaluate, prior to the initiation of an investigation, whether a petition has been filed on behalf of the industry affected, i.e. whether such a petition has been made with the authorization or approval of the domestic industry. In this respect, the Panel's reasoning is entirely consistent with the findings of the Swedish Steel Panel and the Mexican Cement Panel (see paragraph 5.9 of the Swedish Steel Panel Report and paragraph 5.31 of the Mexican Cement Panel Report. There is therefore no basis to argue that with respect to the question of the obligations of investigating authorities to satisfy themselves that a petition has been filed "on behalf of" the domestic industry, the Panel has in any way deviated from past cases. The statements in paragraphs 358-360 also make it clear that the Panel's reasoning in no way implies, as suggested in your letter, "that the DOC is not required under the Code to satisfy itself on its own, prior to investigation, that a petition is filed on behalf of the domestic industry."
While the legal standard articulated by the Panel thus does not differ from the legal standard expressed in other cases, a review of the specific factual circumstances of the case before
[Page 168]
it led the Panel to conclude that the United States had not acted inconsistently with its obligations under Article 5:1 of the Anti-Dumping Code. In paragraph 364, the Panel explicitly stated that "the factual situation presented to it differed significantly from the factual situation presented to the 'Swedish Steel Pipe' panel." In paragraph 361 of its findings the Panel identifies the key factual elements which formed the basis for its conclusion. For example, the Panel notes that the petition was made with a legal certification as to its correctness and completeness; this legal certification also covered the statement in the petition that it was made with the support of twenty-one firms representing well over the majority of all domestic production of Atlantic salmon. Thus, the Department of Commerce had before it a certified statement of industry support; no such certified statement of industry support was before the Department in the case considered by the Swedish Steel Panel.
Norway next takes issue with the Panel's view that it was reasonable for the Department of Commerce to assume that the individual members of the WFGA continued to support the petition after the WFGA had changed its position. However, as reflected in paragraph 355 of the findings of the Panel in the anti-dumping dispute, the letter in which the WFGA withdrew its support stated that the members of this association would be free to express an individual position on the petition. Nothing in the information before the Panel indicated that individual members of the WFGA in fact changed their position with respect to the petition.
Finally, you note that in the countervailing duty case it would have been futile for Norway to raise the standing issue and that in the anti-dumping case Norway did not have an opportunity to raise the standing issue before the initiation of the investigation.
While it is correct that in paragraph 21 of the findings in the countervailing duty dispute the Panel mentions the fact that the Government of Norway apparently had not raised the standing issue in pre-initiation consultations under Article 3:1 of the Subsidies Code, it is evident from paragraphs 28 and 29 of these findings that this element was not of decisive importance to the Panel's conclusions. In the anti-dumping duty dispute the Panel has nowhere in its findings made reference to the fact that the Government of Norway had not raised the issue of standing prior to the initiation of the investigation; this element was simply not among the factual elements upon which the Panel based its conclusion.
2. Determination of the existence of material injury
The first point made in your letter regarding the Panel's findings on the determination of injury concerns the interpretation of the term "through the effects of ..." in Article 3:4 of the Anti-Dumping Code and Article 6:4 of the Subsidies Code. The argument advanced in your letter was made in the course of the proceedings before the Panels and has been dealt with by the Panel in paragraphs 568-569 of the findings in the anti-dumping duty dispute and in paragraphs 131-132 of the findings in the countervailing dispute. In particular, the Panels in these paragraphs explain their view that Articles 3:2 and 3:3 of the Anti-Dumping Code and Articles 6:2 and 6:3 of the Subsidies Code are not limited to an identification of indicia of the extence of material injury but also deal with the causal relationship between the allegedly dumped and subsidized imports and material injury to a domestic industry.
Your second point concerning the Panels' findings on injury pertains to the Panels' interpretation of the second sentence in Article 3:4 of the Anti-Dumping Code and Article 6:4 of the Subsidies Code. You observe in your letter that:
"The Panel determinations indicate that it is sufficient that the authorities do not ignore other factors rather than applying the Code language that the investigating authority
[Page 169]
must not attribute injury from other factors to the effects of the subsidies or dumping. Subsidies Panel Report at paragraph 110 and Anti-Dumping Panel Report at paragraph 547."
This argument seems to be based on a misreading of the rôle in the Panel's analysis of the paragraphs referred to in your letter. These paragraphs simply note that the USITC had acknowledged the possible relevance of other factors as causes of injury but in no way imply that this by itself was sufficient to meet the requirement of Article 3:4 of the Anti-Dumping Code and Article 6:4 of the Subsidies Code (see paragraph 548 of the findings in the anti-dumping duty dispute and paragraph 111 of the findings in the countervailing duty dispute).
In paragraph 555 of its findings in the anti-dumping duty dispute the Panel sets forth its interpretation of the requirement of the second sentence of Article 3:4 of the Anti-Dumping Code. The Panel specifically states that:
"... the USITC was required to conduct an examination sufficient to ensure that in its analysis of the factors set forth in Articles 3:2 and 3:3 it did not find that material injury was caused by imports from Norway when material injury to the domestic industry allegedly caused by imports from Norway was in fact caused by factors other than these imports. The Panel therefore proceeded to consider whether in its investigation the USITC had conducted such an examination". (emphasis added)
In view of this statement, I cannot agree with your view that the Panel's reasoning would eliminate the need for the second sentence of Article 3:4 and would thereby be inconsistent with accepted norms of treaty interpretation. This statement also in no way contradicts the standard reflected in paragraph 5.2.8 of the Report of the Panel in the dispute on countervailing duties by Canada on imports of grain corn from the United States. In paragraphs 556-559 of its findings the Panel examines, on the basis of this interpretation of the second sentence in Article 3:4, the manner in which the USITC treated the alternative causes of injury mentioned by Norway. The standard formulated in paragraph 555 and the Panel's application of this standard to the facts before it in paragraphs 556-559 cannot reasonably be interpreted to mean that in the view of the Panel it is sufficient under Article 3:4 for investigating authorities to simply "enumerate" other possible causes of injury, as suggested on page 5 of your letter.
The third point raised in your letter in respect of the Panel's findings on the injury determination pertains to the alleged failure of the USITC to make a determination that imports from Norway were causing material injury to the domestic industry in the United States at the time of the USITC's determination. Your letter refers in particular to the evolution of the (relative and absolute) volume of imports from Norway during 1990. In this connection I would first like to point out that the Panels' findings do not imply that the USITC could "disregard the 1990 import records", as you suggest in your letter. In paragraph 507 of its findings in the anti-dumping duty dispute the Panel notes the USITC's statement about the limited weight to be accorded to the decline in absolute import volume in 1990, based on the fact that this decline appeared to be largely the result of the filing of the petition and/or the imposition of provisional anti-dumping and countervailing duties. In paragraph 508, the Panel reviews the data provided by Norway on the monthly import volumes in 1989-1990 and concludes that these data are not inconsistent with the explanation offered by the USITC of the decline in the volume of imports in 1990. Paragraph 509 recapitulates the legal standard of an "objective examination" by which the Panel was guided in its review of this aspect of the USITC's determination. This paragraph states quite clearly the Panel's view that the requirement of an "objective examination" means that the USITC was under an obligation to consider the information before it on the decrease in absolute volume of imports and to explain why this information did not detract from a finding of a
[Page 170]
significant increase in the volume of imports. In sum, the Panel's analysis in paragraphs 507-509 clearly indicate that the Panel was not of the opinion that the USITC could "disregard" the data on the evolution of the import volume in 1990. At the same time, however, the Panel found it inappropriate to make its own judgement as to the relative weight to be accorded to the facts before the USITC, as explained in paragraph 494.
The question of "present" material injury is also addressed in paragraphs 575-581 of the findings of the Panel in the anti-dumping dispute and in paragraphs 138-145 of the findings of the Panel in the countervailing duty dispute. As is evident from paragraph 580 of the findings in the anti-dumping duty dispute and paragraph 143 of the findings in the countervailing duty dispute, the Panels have not ignored the fact that the first sentence of Article 3:4 of the Anti-Dumping Code and of Article 6:4 of the Subsidies Code is in the present tense. However, in the view of the Panels, this sentence cannot be interpreted to mean that investigating authorities are required to continue to gather information up to the time of the final determination.
3. Nature of the recommendation of the Panel in the anti-dumping dispute
Let me now turn to your comments on the recommendation in paragraph 597 of the Panel in the anti-dumping dispute. The reasons why the Panel has decided not to make the recommendation requested by Norway are stated in paragraph 596. Leaving aside the question of the precedential value of previous reports, I note that in the four cases to which you refer in your letter the Panels had found that no anti-dumping or countervailing duties should have been levied at all. As explained in paragraph 596, the Panel in the present dispute has not arrived at such a finding. Under these circumstances the Panel did not find it necessary to pronounce itself on the question of reimbursement of the anti-dumping duties. At the same time, the Panel found it appropriate to make a recommendation which is more specific than recommendations generally appearing in GATT Panel Reports and which would require the United States to reconsider those aspects of its determination found by the Panel to be inconsistent with Articles 2:4 and 2:6 of the Anti-Dumping Code. It follows from the last part of paragraph 597 that the steps to be taken by the United States to bring its measures into conformity with its obligations are not limited to a mere reconsideration of the affirmative final determination.
4. General Comments
Finally, allow me to make a comment on some of the general observations in your covering letter. I respectfully disagree with your statement that "the Panels appear in general as not having been prepared to question the contents of the information applied by the United States authorities in the investigations, nor to take a stand regarding the United States decisions made on the basis of such information." With regard to the USITC's injury determinations, the Panels have carefully examined whether these determinations involved a consideration of the factors mandated by the two Codes and were based on positive evidence. Paragraph 494 of the findings in the anti-dumping dispute explicitly notes with regard to the requirement of "positive evidence" that:
"a review of whether in a given case this requirement was met involved an examination of the stated factual basis of the findings made by the investigating authorities in order to determine whether the authorities had correctly identified the appropriate facts, and whether the stated factual basis reasonably supported the findings of the authorities."
For each aspect of these determinations challenged by Norway, the Panels have examined in detail the precise factual basis of the USITC's findings. As you know, where the Panels found it necessary to review confidential information, they have requested the United States to make this
[Page 171]
information available to the Panel. With regard to the determination of dumping, the Panel in the anti-dumping dispute has carried out a detailed examination of whether certain decisions taken by the Department of Commerce were reasonable in light of the information before it. As reflected in the Panel's findings, with respect to three issues the Panel concluded that this was not the case. If, as you suggest, the Panel had not been prepared "... to take a stand regarding the United States decisions made on the basis of such information", it could not have concluded that in these respects the United States had not acted reasonably in light of the information before the Department of Commerce.
I remain, dear Mr. Ambassador,
Yours sincerely,
Janusz Kaczurba (signed)
Chairman
Panels on the Imposition of
Anti-Dumping and Countervailing
Duties on Imports of Fresh and
Chilled Atlantic Salmon from Norway