This Extraordinary Challenge Committee (“ECC”) was convened pursuant to Request for an Extraordinary Challenge Committee (“Request”) pursuant to Article 1904.13 and Annex 1904.13 of the North American Free Trade Agreement (“NAFTA”) and Rule 37(1) of the NAFTA Rules of Procedure for Article 1904 Extraordinary Challenge Committees. The Request was filed on September 24, 2003 by the Office of the United States Trade Representative (“USTR”). The Request asked that an ECC be convened to consider decisions of the binational panel (“Panel”) that reviewed the final results of the full sunset review and remand determinations made by the United States Department of Commerce (“Commerce”) In the Matter of Pure Magnesium from Canada, Secretariat File No. USA/CDA-00-1904-06.

The United States Government (“USG”) filed case briefs in support of the Request and was represented at the hearing by counsel for both the USTR and Commerce. US Magnesium LLC filed briefs in support of the Extraordinary Challenge Request and was represented by counsel at the hearing. The Government of Canada (“GOC”) and the Government of Quebec (“GOQ”) both filed briefs opposing the Request and both were represented by counsel at the hearing. Norsk Hydro Canada Inc. (“NHCI”) also filed a brief opposing the Request but was not represented by counsel at the hearing.

After careful review and considerations of the applicable provisions of NAFTA and other applicable legal provisions and the record and the briefs filed by the parties supporting and opposing the Request, as well as the able arguments of counsel, the ECC makes the following conclusions:

BACKGROUND

1. On July 5, 2000 Commerce issued a Notice of Final Results of Full Sunset Review (“Sunset Review”) in respect of the antidumping order on pure magnesium from Canada originally issued in 1992. The original margin of dumping was 21% but four subsequent administrative reviews found no dumping of pure magnesium. Imports virtually ceased after the antidumping order and then resumed but in much smaller volumes than previously. Pure magnesium exports by NHCI, the sole Canadian producer of pure magnesium, never amounted to more than 10% of their volume prior to the antidumping order. Despite the absence of dumping in four subsequent administrative reviews, Commerce did not revoke the order because it considered that the sales were not in commercial quantities

2. In its Sunset Review, Commerce found that the revocation of the antidumping duty order would likely lead to the continuation or recurrence of dumping. The basis for Commerce’s finding was the drastic reduction of imports of pure magnesium from Canada following the antidumping order coming into effect in 1992. The GOQ and NHCI requested panel review of Commerce’s decision.

3. In its decision of March 27, 2002 (“First Panel Decision”), the Panel did not accept Commerce’s reasons for not considering factors other than the drastic reduction of imports and remanded the matter to Commerce to consider the GOQ’s claims respecting “good cause” to examine other factors.¹

4. On May 28, 2002, Commerce issued a remand determination (“First Remand Determination”) in which Commerce rejected the argument that successive zero dump ing margins in four administrative reviews constituted “good cause” to examine other factors. During the sunset review proceeding, NHCI had argued that it had changed its product mix to become primarily an alloy magnesium as opposed to a pure magnesium producer and cited the existence of long-term contracts for alloy magnesium as a reason that switching back to pure magnesium was not commercially feasible. NHCI had submitted copies of the long-term contracts as attachments to its case brief in the sunset review proceeding but Commerce rejected this information because it had not been submitted within the time period prescribed by Commerce’s regulations.² NHCI had resubmitted its brief without the long-term contracts. In the First Remand Determination, Commerce considered that NHCI had provided no proof that it was primarily a producer of alloy magnesium.

5. On October 15, 2002, the Panel issued its decision (“Second Panel Decision”) respecting the First Remand Determination. The Panel noted that NHCI company officials had certified as accurate statements respecting customer commitments and remanded the First Remand Determination back to Commerce for further consideration of the record evidence. These were in fact factual statements made in briefs filed with the Panel that were verified by an officer of the party filing the brief and were not sworn statements received into evidence. The Panel also instructed Commerce to solicit the views of the parties as to whether the record should be supplemented by production of the long-term contracts and, after considering such views, determining whether the record should be supplemented.

6. In its remand determination of January 28, 2003 (“Second Remand Determination”), Commerce concluded that it was not appropriate to supplement the record with the long-term contracts because NHCI should have been aware of the time limits imposed by the regulation. Nevertheless, Commerce assumed arguendo that NHCI had long-term contracts, and then stated that assuming this fact, it would not alter its conclusion that dumping would resume. Commerce made six findings respecting “additional evidence” (sales by NHCI of pure magnesium in other markets, size of the U.S. pure magnesium market, change of NHCI strategy, ease of NHCI’s switching from alloy to pure magnesium, proposals for NHCI expansion of plant capacity and low volume exports following order) that were sufficient for Commerce to conclude that the revocation of the order likely would lead to the continuance or recurrence of dumping.

7. In its decision of April 28, 2003 (“Third Panel Decision”) respecting the Second Remand Redetermination, the Panel considered Commerce’s failure to explain why the record had not been supplemented an unacceptable disregard of the Panel’s instructions. The Panel stated Commerce’s application of its regulation was “inflexible” and that Commerce had pre-judged the matter by stating that the long-term contract commitments would not alter the outcome. The Panel stated that this position was manifestly incons istent with Commerce’s statutory obligation to conduct a full review based on the evidence on the record. The Panel also took issue with Commerce’s refusal to consider the sworn statements concerning the existence of the long-term contracts as “evidence”, and stated that Commerce should have taken this evidence into account. The Panel analyzed each item of the “additional evidence” and concluded that NHCI had changed its marketing strategy by switching to alloy and had entered into certain long-term contracts in connection with alloy production. The Panel further concluded that the record did not support the finding that the resumption of dumping was probable and remanded with instructions to revoke the antidumping order. In issuing these instructions, the Panel expressly cited the authority provided by the United States Court of International Trade (“CIT”) decision in Nippon Steel Corp. v. United States ³, a case involving an injury determination by the United States International Trade Commission (“Commission”) in which the CIT remanded to the Commission with instructions to revoke the antidumping order.

8. On June 24, 2003, the Panel amended its order to a remand to Commerce to take action consistent with the Panel’s decision.

9. On September 24, 2003 the USTR filed the Request referred to above.

BASIS FOR THE EXTRAORDINARY CHALLENGE

10. The USG Brief alleged that the Panel had manifestly exceeded its powers, authority or jurisdiction and seriously departed from a fundamental rule of procedure and violated the standard of review:

a) by ruling that Commerce’s uncontested decision to exclude untimely filed evidence (the longterm contracts) was arbitrary and contrary to law; and

b) by conducting an impermissible de novo review with regard to the each of Commerce’s six findings supporting its likelihood conclusion.

PRELIMINARY MOTION

11. On October 15, the USG filed a motion under Rule 44 of the Rules of Procedure for Article 1904 Extraordinary Challenges to strike extra-record evidence attached to the Notice of Appearance filed by the GOQ. The document in question was the case brief that NHCI had filed in the sunset review proceeding that Commerce had rejected. The information in question did not form part of the record before either Commerce or the Panel. The Panel can only examine the record before Commerce and the ECC can only examine the record that was before the Panel. Therefore this motion must succeed and the ECC orders the extra-record evidence struck.

FUNCTION AND NATURE OF ECC REVIEW

12. NAFTA Article 1904.13 sets out a three-pronged test that must be satisfied for an extraordinary challenge to succeed:

i) First, the ECC must first find that an action described in subparagraph (a) of Article 1904.13 has occurred. The actions relevant for this extraordinary challenge are those set out in subparagraph (a)(ii), namely that “the panel seriously departed from a fundamental rule of procedure” and in subparagraph (a)(iii), namely that the “panel manifestly exceeded its powers, authority or jurisdiction…, for example by failing to apply the appropriate standard of review.”

ii) Second, the ECC must find that the action described in subparagraph (a) of Article 1904.13 has materially affected the panel’s decision.

iii) Third, the ECC must find that the action threatens the integrity of the binational panel review process.

The three prongs of the test are applied sequentially and all must be satisfied for the extraordinary challenge to succeed.

13. The extraordinary challenge procedure is not an appeal procedure. The purpose is not to correct errors of law or fact by the panel as the use of words such as “gross misconduct” and “seriously departed” and “manifestly exceeded” make clear. Rather it is a safety net to deal with mistakes that are so egregious as to undermine the functioning and acceptance of the entire Chapter 19 of NAFTA. As pointed out by the extraordinary challenge committee in Gray Portland Cement and Clinker from Mexico⁴, the extraordinary challenge process is reserved for extraordinary situations. However, the bar set by NAFTA Article 1904.13 cannot be set so high that an extraordinary challenge can never succeed. Where truly egregious situations arise, the extraordinary challenge procedure must spring into action to serve its purpose of safeguarding the integrity of the binational panel review process.

14. The text of NAFTA Article 1904.13 is almost identical to that of the corresponding provision in the Canada-United States Free Trade Agreement, except for the addition of the words to subparagraph (a)(iii) identifying “by failing to apply the appropriate standard of review” as an example of a panel manifestly exceeding its powers, authority or jurisdiction. This addition is indicative of the importance the NAFTA drafters placed on panels applying the proper standard of review. In accordance with international rules of treaty interpretation ⁵, this additional language must be read in the context in which it appears, namely the balance of subparagraph (a).

15. It must be emphasized that it is the Panel’s decision and not Commerce’s decision that is at issue before the ECC. It is not the role of the ECC to identify errors in Commerce’s decision but, rather, to evaluate how the Panel conducted its ‘judicial review’ of Commerce’s decision and determine whether that judicial review violated Article 1904.13 of the NAFTA.

PANEL CONSIDERING COMMERCE’S EXCLUSION OF UNTIMELY EVIDENCE

16. The US parties argue that the Panel found that “ the GOQ has waived the right to raise [the late evidence] issue because it was not mentioned in the Complaint or in the briefs to the panel”.⁶ Upon making this finding “the issue [was] waived, the Panel never should have reached the question of whether Commerce should consider opening the record. In addressing this question the Panel exceeded its powers and seriously departed from a fundamental rule of procedure.”⁷). The ECC finds this position untenable given that all three U.S. counsel at the hearing, in response to a direct question from the Chairman, stated that a binational panel has the authority to require Commerce to open the record. Given this admission, the ECC does not understand why the USG is alleging that the Panel’s conduct constituted an action described in subparagraph (a) of Article 1904.13 nor accepts this submission.

IMPERMISSIBLE DE NOVO REVIEW

17. The heart of the US parties’ argument concerns the issue of whether the Panel engaged in it a de novo inquiry. To consider this allegation the ECC needs to consider each of the three prongs of the extraordinary challenge test described above.

Subparagraph (a)(iii) - Applying Correct Standard Of Review

18. The USG alleges that the Panel conducted a de novo review, which is the standard of review to be applied by reviewing courts such as the CIT and the U.S. Court of Appeals for the Federal Circuit prohibits. The allegation of de novo review relates to the Panel’s evaluation of Commerce’s six factual findings and conclusion in respect thereof, as set out on pages 12 through 21 of the Third Panel Decision.

19. Much of the Panel’s evaluation of the six factual findings consists of legitimate probing that one would expect a reviewing court to apply when applying the substantial evidence standard that forms the basis for the applicable standard of review. In arriving at its conclusions the Panel also relied on the long-term contracts and on implicit assumptions about the relative profitability of the pure and alloy magnesium businesses.

20. The Panel was aware of the long-term contracts by reason of factual statements made in briefs filed with the Panel that were verified by an officer of the party filing the brief. These were not sworn statements received into evidence and were not part of the factual record before Commerce. The United States argued correctly that the Panel was not correct in treating these statements as evidence. Furthermore, the text of the actual agreements were not on the record, so the Panel could not have been aware of the actual terms of the long-term contracts such as those relating to duration, volumes and termination. The import evidence on the record established that NHCI’s exports to the United States from Canada had shifted from pure magnesium to alloy magnesium following the antidumping order. However, in response to a direct question from ECC Member Getzendanner, counsel for the GOQ conceded that there was no evidence on the record as to the relative profitability of producing pure magnesium versus producing alloy magnesium.

21. In the Second Remand Determination Commerce stated that sales by NHCI of pure magnesium in markets other than the United States indicated that NHCI had not completely redirected its focus from pure to alloy magnesium. ⁸ In the Third Panel Decision, the Panel observed, quite reasonably, that whether it was likely that NHCI would resume dumping and not whether NHCI had completely redirected its worldwide marketing focus was the question that Commerce had to answer, and that Commerce failed to explain how pure magnesium sales in other markets was evidence that NHCI would likely abandon its strategy (substantiated by the import figures) of focusing on sales of alloy magnesium in the United States.⁹ However, the Panel also found that it seemed improbable that NHCI would abandon its alloy strategy and resume dumping of pure magnesium that presumably would be unremunerative, without any evidence of the relative profitability of pure versus alloy magnesium production.¹⁰

22. In the Second Remand Determination, Commerce found that the consumption of pure magnesium in the U.S. market was nearly triple that of alloy and that, given the mix of magnesium products in United States and that U.S. is largest market in world, it appeared likely that NHCI, in the absence of the antidumping order would seek to re-establish itself in the U.S. pure magnesium market. The Panel noted that Commerce had failed to evaluate how the size and product mix of the U.S. magnesium market indicate that NHCI would abandon its alloy strategy. However, the Panel made several references to the long-term contracts as a reason why NHCI would not shift its alloy strategy.¹¹

23. Commerce observed in the Second Remand Determination that evidence on the record suggested that NHCI’s marketing strategy could change very quickly.¹² The Panel observed that nowhere did Commerce evaluate what record evidence made it likely that NHCI would abandon a major shift in U.S. marketing strategy, namely to alloy, which was evident from the import numbers. However, the Panel found that the fact that NHCI had entered long-term contracts supported conclusion that NHCI would likely resume dumping only if that change in strategy would be more profitable than strategy of shifting to alloy that NHCI adopted ten years previously. The Panel stated that it failed to see how low price strategy would cause NHCI to retreat from the alloy market, which presupposes knowledge of the relative profitability of the alloy market.¹³

24. Based on its analysis, the Panel stated that “the evidence on the record leads inexorably to the conclusion that record does not support a finding that the resumption of dumping of pure magnesium is probable if the order were revoked.”¹⁴ Rather than remanding the matter back to Commerce with precise instructions, the Panel remanded with instructions to revoke the antidumping order. The Panel cited the CIT decision in Nippon Steel as authority for this action, but this decision was subsequently overruled by the United States Court of Appeals for the Federal Circuit.¹⁵ The Court of Appeals noted that only the Commission may find facts and determine causation and, ultimately, material injury, and that the CIT had gone beyond its statutorily-assigned role to “review”. Prior to the Nippon Steel decision being overruled, the Panel amended its order to remand to Commerce, “with instructions to take action consistent with the Decision, including the Conclusion and Determination by the Panel, within 15 days of the date of the order”.¹⁶

25. While the Panel’s analysis of Commerce’s six findings consisted in part of legitimate probing of Commerce’s conclusion that there was substantial evidence on the record establishing that dumping would likely resume if the antidumping order were revoked, the Panel clearly based its findings, at least in part, upon speculation respecting the long-term contracts and the relative profitability of pure versus alloy magnesium production.

26. In summary the ECC finds that the Panel by extrapolating from facts, making a number of assumptions, drawing conclusions from assumptions and making findings of likelihood based on such assumptions, extrapolations and conclusions failed to apply the correct standard of review to such a degree that it manifestly exceeded its powers. Consequently this challenge meets the first prong of Article 1904.13.

Subparagraph (b ) - Materially affected the Panel’s decision

27. It is not possible to determine the extent to which the Panel relied upon its speculations, extrapolations, assumptions and assumption-based conclusions respecting the long-term contracts and the extent to which it relied upon its legitimate probing of Commerce’s findings. While the Panel might have reached the same conclusion based on legitimate probing alone, given the frequency and extent of the various speculations, extrapolations, assumptions and assumption-based conclusions occurred in its decision, the ECC has no option but to conclude that the Panel’s failure to apply the correct standard of review materially affected its decision. Consequently the second prong of this challenge test is also met.

Subparagraph (b) - Threaten the integrity of the Binational Panel Review Process

28. In order that the integrity of the binational panel review process be threatened, the action by the Panel must have undermined a fundamental tenet upon which the binational panel review process is based. For example, it is a fundamental tenet that members of binational panels be impartial and not have a personal interest in the outcome of the case or that binational panels cannot conduct their own independent fact-finding inquiries.

29. It is also a fundamental tenet of the binational review process that binational panels apply the domestic antidumping or countervailing duty law of the Party whose investigating authority’s determinations are being challenged. This means a binational panel has to engage in a judicial review of the underlying decision of the competent investigating authority (in this case Commerce) and apply domestic law. It is not the function of panels to make law or evolve law but to apply domestic law as they find it. After all there is no method by which decisions of a panel that fail to apply domestic law can be challenged or appealed. To hold otherwise would allow the formation of two streams of anti-dumping and countervail duty law, one developed by binational panels and one by courts; a result that is clearly antithetical to the whole construct of Chapter 19.

30. The panel in this case justified its finding in the following [Third Panel Decision page 21] summarizing statement: