CONCLUSION

Appearances :

Dewey Ballantine LLP (Bradford L. Ward, John A. Ragosta, Linda Andros, David A Bentley, Monica Welt, Rory Quirk) for Coalition for Fair Lumber Imports Executive Committee

United States Department of Commerce, Office of the Chief Counsel for Import Administration (John D. McInerney, Berniece A. Browne, Robert J. Heilferty, D. Michael Stroud, Jr., William G. Isasi, Scott D. McBride, James K. Lockett, Christine J. Sohar, Linda Chang, Mark Barnett) for United States Department of Commerce, International Trade Administration

Arnold & Porter (Michael T. Shor) for Abitibi-Consolidated, Inc. And Scieres Saguenay Ltee.

Baker & Hostetler, LLP (Elliot J. Feldman, John J. Burke, Arland M. DiGirolamo, Michael S. Snarr), for Tembec, Inc.

Baker & McKenzie (B. Thomas Peele III, Kevin M. O'Brien, Lisa A. Murray) for Slocan Forest Products, Ltd.

Piper Rudnick LLP (John E. Corette, III, David M. Rubinstein) and Howrey Simon Arnold & White LLP (Michael A. Hertzberg, Juliana M. Cofrancesco) for The Provinces of New Brunswick, Nova Scotia, Prince Edward Island, Newfoundland, The Maritime Lumber Bureau, and Lumber Producers in the Maritime Provinces

Kaye Scholer LLP (Donald B. Cameron, Julie C. Mendoza) for Canfor Corporation

Wilmer Cutler & Pickering (Robert C. Cassidy, Jr., John M. Greenwald, Jack A. Levy) for the Quebec Lumber Manufacturers Association

Steptoe & Johnson LLP (W. George Grandison, Matthew Frumin, Lesley Stone) and Farris, Vaughan, Wills & Murphy (Brian R. Canfield) for British Columbia Lumber Trade Council and its Constituent Associations

Gibson, Dunn & Crutcher LLP (Daniel J. Plaine, Gracia M. Berg, Lisa A. Murray, Gregory C. Gerdes) for West Fraser Mills Ltd.

Miller & Chevalier (Matthew M. Nolan, James B. Altman, Sydney Mintzer, Sylwia A. Lis) for Weyerhaeuser Corporation

Law Offices of George R. Tuttle (Steven S. Spraitzar) for Anderson Wholesale, Inc.

Patterson, Betts & Mines, P.S (Livingston Wernecke) for Fred Tebb & Sons, Inc.

This Panel was constituted under Article 1904 of the North American Free Trade Agreement (NAFTA) to review the amended Final Determination by the United States Department of Commerce that certain softwood lumber was exported from Canada to the United States during the period April 1, 2000 to March 31, 2001 at prices that were less than fair value (LTFV). Notice of this amended determination was published in the Federal Register on May 22, 2002.²

Article 1904 of the NAFTA states:

1. As provided in this Article, each Party shall replace judicial review of final antidumping and countervailing duty determinations with binational panel review.

2. An involved Party may request that a panel review, based on the administrative record, a final antidumping or countervailing duty determination of a competent investigating authority of an importing Party to determine whether such determination was in accordance with the antidumping or countervailing duty law of the importing Party. For this purpose, the antidumping or countervailing duty law consists of the relevant statutes, legislative history, regulations, administrative practice and judicial precedents to the extent that a court of the importing Party would rely on such materials in reviewing a final determination of the competent investigating authority. Solely for purposes of the panel review provided for in this Article, the antidumping and countervailing duty statutes of the Parties, as those statutes may be amended from time to time, are incorporated into and made a part of this Agreement.

3. The panel shall apply the standard of review set out in Annex 1911 and the general legal principles that a court of the importing Party otherwise would apply to a review of a determination of the competent investigating authority.

Article 1911 of the NAFTA provides the standard of review to be applied by this Panel:

standard of review means the following standards, as may be amended from time to time by the relevant Party: ...

(b) in the case of the United States,

(i) the standard set out in section 516A(b)(l)(B) of the Tariff Act of 1930, as amended, ....

The referenced statute provides that a reviewing court “shall hold unlawful any determination, finding, or conclusion found . . . in an action brought under paragraph (2) of subsection (a) of this section, to be unsupported by substantial evidence on the record, or otherwise not in accordance with law." 19 U.S.C. § 1516a(b)(1)(B).

In summary, this Panel, in reviewing a decision of the United States Department of Commerce, must be guided by the same principles and rules as would the United States Court of International Trade. Particular principles and rules will be discussed in this decision as relevant to the several issues raised by the parties who filed complaints against the Final Determination.
 

II. PROCEDURAL HISTORY
 

On April 23, 2002, the United States Department of Commerce, International Trade Administration (hereinafter, “Commerce”) initiated an investigation under Section 731 of the Tariff Act of 1930, as amended [19 U.S.C. §1673] to determine whether Certain Softwood Lumber from Canada was being sold for exportation to the United States at less than fair value (LTFV) prices. The investigation was commenced following the receipt of an antidumping petition filed by the Coalition for Fair Lumber Imports Executive Committee, the United Brotherhood of Carpenters and Joiners, and the Paper, Allied-Industrial, Chemical and Energy Workers International Union, all the foregoing representing the domestic industry engaged in manufacture of a like product. See, Notice of Initiation of Antidumping Duty Investigation: Certain Softwood Lumber Products From Canada, 66 Fed. Reg. 21,328, April 30, 2001 (“Initiation Notice”), Public Record (P.R.) Doc. 86.

The goods subject to the investigation were described as consisting of softwood lumber products defined generally as dimensional lumber, flooring and siding and other products, classified under subheadings 4407.1000, 4409.1090, and 4409.1020 of the Harmonized Tariff Schedule of the United States (“HTSUS”). Id.

On April 25, 2001, in advance of issuing antidumping questionnaires, Commerce issued a letter to interested parties, including the petitioners and the 15 largest known producers/exporters of softwood lumber from Canada, soliciting comments on issues of Respondent selection, fair value comparison methodology, and possible limitation of reporting of sales and cost data. See, P.R. Doc. 60.

On May 18, 2001, the United States International Trade Commission (ITC) preliminarily determined that a reasonable indication existed that an industry in the United States is threatened with material injury by reason of Canadian softwood lumber imports.

Thereafter, on May 25, 2001, Commerce selected as mandatory respondents the six largest Canadian producers and exporters of subject merchandise: Abitibi-Consolidated Inc. (Abitibi); Canfor Corporation (Canfor); Slocan Forest Products Ltd. (Slocan); Tembec Inc. (Tembec); West Fraser Mills Ltd. (West Fraser), and Weyerhaeuser Company (Weyerhaeuser). Following this selection, Commerce issued antidumping questionnaires to the six mandatory respondents. The period of investigation (POI) covered by the questionnaires was April 1, 2000 through March 31, 2001.

On November 6, 2001, Commerce published notice of its preliminary determination that Canadian softwood products were being sold for exportation to the United States at LTFV prices. Preliminary Determination of Sales at Less Than Fair Value and Postponement of Final Determination: Certain Softwood Lumber from Canada, 66 Fed. Reg. 56,062 (Nov. 6, 2001)³. From December 2001 through February 2002, Commerce verified the responses submitted by the six mandatory respondents in the investigation. During January and February 2002, Commerce issued its verification reports.

On February 25, 2002, Commerce held a public hearing on all issues in the investigation except for scope-related issues. On March 19, 2002, Commerce held a public hearing on scope-related issues that had been analyzed during the course of the investigation.

On March 21, 2002, Commerce issued a final affirmative LTFV determination, which was published in the Federal Register on April 2, 2002. See, Notice of Final Determination of Sales at Less Than Fair Value: Certain Softwood Lumber from Canada, 67 Fed. Reg. 15,539 (April 2, 2002). The Final Determination was accompanied by the publication of an Issues and Decision Memorandum (IDM), which described the basis for Commerce’s Final Determination in detail.

On May 16, 2002, the International Trade Commission (ITC) notified Commerce of its final determination that the industry in the United States producing softwood lumber products is threatened with material injury by reason of imports of the subject merchandise from Canada.

On May 22, 2002, Commerce published its Notice of Amended Final Determination of Sales at Less than Fair Value and Antidumping Duty Order: Certain Softwood Lumber from Canada, 67 Fed. Reg. 36, 068 (May 22, 2002). Commerce also published a final Antidumping Duty Order.

Following receipt by the NAFTA Secretariat of requests for binational panel review of various aspects of the Final Determination, the instant Panel was constituted.
 

III. STANDARD OF REVIEW
 

Article 1904(3) of the NAFTA requires that this Panel apply the standard of review and “general legal principles”⁴ that a U.S. court would apply in its review of a decision of the competent investigating authority.⁵ The standard of review that a reviewing court and, consequently, this Panel, must apply is dictated by section 516A(b)(1)(B) of the Tariff Act of 1930, as amended (19 U.S.C. § 1516a(b)(1)(B), which requires the Panel to “hold unlawful any determination, finding, or conclusion found … to be unsupported by substantial evidence on the record, or otherwise not in accordance with law.” Id.⁶

Substantial evidence has been defined by the Supreme Court as “more than a mere scintilla. It means such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.” Universal Camera Corp. v. NLRB, 340 U.S. 474, 477 (1951) (quoting Consolidated Edison Co. v. NLRB, 305 U.S. 197, 229 (1938)); see also Matsushita Elec. Indus. Co. v. United States, 750 F.2d 927, 933 (Fed. Cir. 1984). Substantial evidence “is something less than the weight of evidence, and the possibility of drawing two inconsistent conclusions from the evidence does not prevent an administrative agency’s finding from being supported by substantial evidence.” Consolo v. Federal Maritime Comm’n, 383 U.S. 607, 620 (1966). Nevertheless, an agency determination must be supported by the administrative record as a whole, including evidence that detracts from the substantiality of the evidence relied upon by the agency. See, Universal Camera, 340 U.S. at 477.

Under the substantial evidence standard the reviewing Panel must not reweigh the evidence, or substitute its judgment for that of the agency. Fresh, Chilled and Frozen Pork from Canada, USA-89-1904-11, at 8 (Aug. 24, 1990); see also Metallverken Nederland B.V. v. United States, 728 F. Supp. 730, 734 (Ct. Int’l Trade 1989). It is well settled that “the possibility of drawing two inconsistent conclusions from the evidence does not prevent an administrative agency’s finding from being supported by substantial evidence.” Consolo, 383 U.S. at 620; see also Matsushita Elec. Indus. Co., 750 F.2d at 933 (“The [agency’s] decision does not depend on the ‘weight’ of the evidence, but rather on the expert judgment of the [agency] based on the evidence of record.”) The reviewing authority therefore may not “displace the [agency’s] choice between two fairly conflicting views, even though [it] would justifiably have made a different choice had the matter been before it de novo.” Universal Camera Corp., 340 U.S. at 488; accord American Spring Wire Corp. v. United States, 590 F. Supp. 1273, 1276 (Ct. Int’l Trade 1984), aff’d sub nom. Armco, Inc. v. United States, 760 F.2d 249 (Fed. Cir. 1985).

The substantial evidence standard generally requires the reviewing authority to accord deference to an agency’s factual findings and to the methodologies selected and applied by the agency. See, e.g., American Silicon Techs. v. United States, 2003 U.S. App. LEXIS 13506 (Fed. Cir. 2003); Micron Tech., Inc. v. United States, 117 F.3d 1386, 1394 (Fed. Cir. 1997); Hercules, Inc. v. United States, 673 F. Supp. 454, 463 (Ct. Int’l Trade 1987)(agencies have “broad discretion in the enforcement of trade laws.”)(quoting Manufacturas Industriales de Nogales, S.A. v. United States, 666 F. Supp. 1562, 1567 (Ct. Int’l Trade 1987)); see also Brother Industries, Ltd. v. United States, 771 F. Supp. 374, 381 (Ct. Int’l Trade 1991) (“Methodology is the means by which an agency carries out its statutory mandate and, as such, is generally regarded as within its discretion.”). Agency determinations are presumed to be correct, and the burden of demonstrating otherwise is on the party challenging a determination. 28 U.S.C. § 2639(a)(1); see, Hannibal Industries, Inc. v. United States, 710 F. Supp. 332, 337 (Ct. Int’l Trade 1989).

Although review under the substantial evidence standard is by definition limited, the Panel nonetheless must conduct a meaningful review of Commerce’s determination. It is well established, for instance, that an agency’s determination must have a reasoned basis. See, American Lamb Co. v. United States, 785 F.2d 994, 1004 (Fed. Cir. 1986)(citing S. Rep. No. 249, 96^th Cong., 1^st Sess. 252 (1979), reprinted in 1979 U.S.C.C.A.N. 381, 638). The reviewing authority may not defer to an agency determination premised on inadequate analysis or reasoning. See, USX Corp. v. United States, 655 F. Supp. 487, 492 (Ct. Int’l Trade 1987).

Furthermore, there must be a rational connection between the facts found and the choice made by the agency. See, Bando Chem. Indus., Ltd. v. United States, 787 F. Supp. 224, 227 (Ct. Int’l Trade 1992) (citing Bowman Transportation, Inc. v. Arkansas-Best Freight System, Inc., 419 U.S. 281, 285 (1974) and Burlington Truck Lines, Inc. v. United States, 371 U.S. 156, 168 (1962)). There must be an adequate explanation of the bases for the agency’s decision in order for the reviewing authority meaningfully to assess whether the decision is supported by substantial evidence on the record. The agency therefore must clearly articulate the reasons for its conclusions. See, e.g., Mitsubishi Materials Corp. v. United States, 820 F. Supp. 608, 623-24 (Ct. Int’l Trade 1993).

The agency determination must be grounded in the record evidence; substantial evidence must consist of facts that support the agency’s findings. See, Baltimore & O.R. Co. v. Aberdeen & R.R. Co., 393 U.S. 87, 91-92 (1968). Where there is conflicting evidence, there must be “some justification, supported by substantial evidence in the record,” for relying on one set of facts over another. Timken Co. v. United States, 894 F.2d 385, 388-389 (Fed. Cir. 1990). The Panel determines the existence of substantial evidence by considering the record as a whole, including evidence that supports as well as evidence that “fairly detracts from the substantiality of the evidence.” Atlantic Sugar, Ltd. v. United States, 744 F.2d 1556, 1562 (Fed. Cir. 1984).

Finally, in reviewing an agency’s construction of a statute it administers, the Panel will defer to the agency’s reasonable interpretation of the statute, if not contrary to an unambiguous legislative intent as expressed in the words of the statute. Chevron U.S.A. Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 843-44 (1984); Timex, V.I., Inc. v. United States, 157 F.3d 879, 881-882 (Fed. Cir. 1998). The Panel will also accord substantial deference to the agency’s interpretations of its own regulations. See, Torrington Co. Inc. v. United States, 156 F.3d 1361, 1363-64 (Fed. Cir. 1998).
 

IV. ANALYSIS
 

In briefing and at oral argument, the parties to this Proceeding organized the contested issues into three basic categories: (A) “general” issues, whose determination is likely to affect multiple parties to the proceeding; (B) “company specific” issues, whose determination is likely to affect individual companies; and (C) “scope” issues, involving the determination of the “class or kind” of merchandise subject to Commerce’s investigation and contested determination. For good order’s sake, the Panel’s decision follows this organization in dealing with the issues raised by the parties.

A. GENERAL ISSUES
 

1. Commerce Was Not Required to Rescind its Notice of Initiation

Several Respondents⁷ contend that Commerce’s investigation was initiated contrary to the requirements of antidumping law. Their primary contention is that the antidumping petition unreasonably omitted information regarding Canadian prices that was reasonably available to Petitioners, that the information was material to Commerce’s decision to reject Canadian prices in determining the prices of softwood lumber, and that had Petitioners disclosed such material information, Commerce reasonably would have relied on Canadian softwood lumber prices instead of extrapolating prices on the basis of U.S. data.

In support of these contentions, the Respondents allege that International Paper, a member of the Petitioner coalition, unreasonably withheld information regarding Canadian prices that was known by its wholly owned Canadian subsidiary, Weldwood. They contend that, by virtue of the failure to disclose this information, the antidumping petition was inherently defective and that Commerce’s decision to initiate an investigation of the claims contained in the petition ought to be rescinded. In arguing that Petitioners acted deliberately, Respondents also allege that International Paper purposefully deleted its relationship to Weldwood from the “relationship list” submitted to Commerce.⁸

Respondents allege further that the Petitioners clearly were aware of Canadian prices, given their purchase of softwood lumber from at least three of the Respondents.

Relying on the decision of the United States Court of International Trade (CIT) in Gilmore Steel Corp. v. United States, 585 F. Supp. 620 (1984) to support their contentions, Respondents ask the Panel to hold that Commerce erred in failing to rescind the notice of investigation in this case. They contend, in their Joint Brief, that the "Petitioner’s decision to omit the Weldwood data from the Petition, and to misrepresent to Commerce the availability of those data, rendered the Petition legally inadequate and deprived Commerce of the statutory authority to initiate the antidumping investigation pursuant to 19 USC §1673a(b)."⁹

Commerce responds that sufficient information was made available to it in order to decide whether to proceed with an antidumping investigation, that such information reasonably established that some dumping of Canadian softwood lumber in the United States had occurred, and that Commerce’s use of such information was reasonable in reaching its decision to initiate and conduct an antidumping investigation, regardless of whether there was other available or new pricing data, such as Respondents attribute to Petitioners.

Commerce maintains, further, that it is not required to take account of transaction specific data, and that it is entitled to rely on observed and constructed data in determining whether or not to proceed with a antidumping investigation following initiation. The agency states that it was reasonably satisfied throughout the process of initiation that sufficient evidence was available to it to demonstrate that dumping had occurred, that the agency has discretion to decide whether or not to proceed to an antidumping investigation, and that it reasonably exercised that discretion in this case. Moreover, Commerce argues that, if the Petitioners had data available to them in regard to Canadian softwood lumber prices that did not demonstrate the presence of dumping, Petitioners were not required to provide all that data in the petition.

Commerce distinguishes Gilmore, supra, noting that there was a lack of industry support for the antidumping petition in that case, depriving petitioners of legal standing, and that this is not so in the instant proceeding.¹⁰

Petitioners contend that they provided Commerce with information that was both reasonably available to them and accurate. They assert further that the Petition need merely provide a reason to investigate, rather than all information that may be available.¹¹ Petitioners also deny that they deliberately withheld information from Commerce, or intended to mislead it about the Canadian prices of softwood lumber. They maintain that they received no communication from Weldwood in regard to Canadian prices, that Weldwood was bound by a duty of confidentiality to other Canadian producers and that International Paper was not privy to Weldwood prices on that account. Petitioners also maintain that Commerce is not restricted to the Petition in deciding whether to proceed with an antidumping investigation.

The Panel determines that Commerce acted within its lawful authority in proceeding with an antidumping investigation following initiation. Commerce also acted reasonably in determining the extent to which it would rely on the Petition in deciding to proceed with the investigation. In so deciding, the Panel is cognizant of the applicable standard of review of reasonableness and the deference that ought to be accorded to Commerce in determining whether the information available to it is reasonably adequate to initiate a dumping investigation.

Commerce’s regulations [19 C.F.R. §353.203] contain guidelines for determining the sufficiency of an antidumping petition filed with the agency. The regulation directs the Secretary of Commerce to conduct this review “on the basis of sources readily available to the Secretary”, and to examine the accuracy and adequacy of the evidence provided in the petition in determining whether to initiate an antidumping investigation. Neither the statute nor the regulations explicitly require that a petition contain all information publicly or privately available concerning the allegations made in the petition. Furthermore, it is not anticipated that every allegation contained in a petition will necessarily be verified as true.

In Luciano Pisoni Fabbrica Accessori Instrument Musicali v. United States, 640 F. Supp. 255 (Ct. Int’l Trade 1986), Respondent in an antidumping investigation contended that the petition lacked up-to-date pricing information, and that the investigation commenced following review of the petition should be rescinded. The Court of International Trade disagreed, noting that:

Commerce has some discretion in deciding whether to initiate an investigation.
Commerce is permitted to assess the sufficiency of a petition “in light of its own
knowledge and expertise and facts capable of judicial notice”.
United States v. Roses, Inc., 706 F.2d 1563, 1568-69 (1983).

640 F. Supp. at 257. The court also noted that “the statutory scheme offers no basis for plaintiff’s position that Commerce is required to rescind a notice of initiation upon discovering inaccuracies in a petition”. Id. at 258.

Moreover, since the antidumping statute requires that data relied upon in making a final determination be verified, the fact that information contained in a petition may subsequently be found to be inaccurate, outdated or incomplete is of no moment. As the court held in Luciano Pisoni, supra:

In making a final determination in an investigation, Commerce must verify all information on which it relies. Corrections to petitioners’ data are the very point of verification procedures. It is therefore proper for Commerce to complete an investigation commenced after a petition is filed by an interested party even when a petition is determined to contain inaccuracies. Since Commerce is authorized to commence an antidumping duty petition sua sponte whenever it determines that an investigation is warranted, it would be unreasonable to require that Commerce terminate an investigation commenced after the filing of a petition by an interested party when, despite inaccuracies contained in the petition, it finds evidence of sales at less than fair value.

Id. at 258 (citations omitted).

The Panel notes further that Gilmore is materially different from the instant case since the lack of industry support for an antidumping investigation there deprived the would-be petitioners of standing to request an antidumping investigation -- a situation not present in the instant case.

Commerce therefore did not err by initiating and continuing the investigation.
 

2. Commerce Did Not Err in its Consideration of the Impact of the Softwood Lumber Agreement (SLA) on Softwood Lumber Prices in Canada During the Period of Investigation.

Respondents contend that Commerce failed to consider the effects of the Softwood Lumber Agreement¹² on prices in Canada during the period of investigation (POI) in this case. The gravamen of Respondents' complaint is stated as follows:

In comparing prices for export sales to the United States with sales in the home market, Commerce has discretion to disregard home market sales that are below cost as calculated by the Department on the grounds they are not “in the ordinary course of trade.” The statute recognizes that some sales below cost will occur even in normal circumstances. See 19 U.S.C. § 1677b(b)(1). For example, the Department may not disregard below cost sales if they are less than 20 percent of home market sales. See 19 U.S.C. § 1677b(b)(2)(c)(i).

In making its sales comparisons in this case, Commerce ignored the effect of the SLA and mechanically disregarded below cost home market sales of a particular product whenever they exceeded 20 percent of such sales. Because lower lumber prices in the Canadian home market were an inevitable result of the SLA, such low priced sales were in the ordinary course of trade during the five years covered by the Agreement, which included the period covered by Commerce’s investigation. Commerce had an obligation to consider the effect of the SLA on Canadian lumber prices when exercising its discretion to determine what home market sales were in the ordinary course of trade. If Commerce had treated such sales as within the ordinary course of trade - as they were during the period of the SLA - it would have found no dumping¹³.

Respondents assert that during the POI prices in Canada were significantly lower than prices at which lumber was sold into the US market:

Canadian producers operating under the SLA ... had an incentive to sell quantities of lumber that exceeded their allotted SLA quotas to customers in the Canadian domestic market. This incentive persisted until the sales volume added to Canadian supply had suppressed Canadian prices to the point where the added cost of selling such lumber to the United States and paying the SLA export tax was offset by the difference between higher U.S. and lower Canadian prices. As a result, the SLA caused lumber prices in Canada to decline and prices in the United States to increase as it restricted the flow of exports to the United States and diverted that lumber into the Canadian market. Since the Canadian market for softwood lumber is less than one-sixth the size of the U.S. market, the price effects of the SLA were far more pronounced in the Canadian market than in the United States¹⁴ [footnote omitted].

Based on that state of the market in Canada, Respondents argue that Commerce should have exercised its discretion to regard sales below Cost of Production (COP) as having been made in the ordinary course of trade, rather than disregard those sales. Essentially, Respondents' complaint is that Commerce, by neglecting the effect of the SLA on prices in Canada, failed to achieve a fair comparison between normal value and export price, as required by 19 U.S.C. § 1677b(a)(1)(B).¹⁵

Commerce dealt with the issue in the Issues and Decision Memorandum (“IDM”) which accompanied the Final Detemination as follows:

As to whether {the} SLA has distorted Canadian and U.S. timber prices, we find that the likely distortion is at most minimal. Moreover, removing all “distortion” would be nearly impossible.

A border measure, such as the SLA, could affect a market-oriented timber industry across North America by driving down stumpage prices in Canada and driving up stumpage prices in the United States. However, we find that this theory is not applicable to the facts of this case. Moreover, throughout the duration of the SLA, Canadian lumber exports to the United States consistently exceeded the specified volume thresholds that triggered the export fees. Therefore, it is unclear to what extent, if any, the SLA actually constrained the quantity of lumber exported to the United States. For all these reasons, the SLA likely did not cause any significant distortion of lumber prices in the United States during the POI. However, even if the SLA had caused a substantive distortion, it is unclear whether, how, and under what circumstances we could account for any distortion.

IDM at Comment 2 [emphasis as added in Respondents’ Joint Brief].

Regarding the statement quoted above, Respondents argue that the references to timber and lumber prices in the United States indicate that Commerce failed to consider the effect of the SLA on lumber prices in Canada, referring to 19 U.S.C. § 1677f(i)(3)(A), which requires that Commerce’s final determination provide “an explanation of the basis for its determination that addresses relevant arguments, made by interested parties . . . , concerning the establishment of dumping . . . .”

Commerce responds that Respondents placed "little emphasis on the 'distortive' effects of the SLA in their case briefs. Certainly no party advanced any substantive proposal for how Commerce should quantify the 'distortive' effect of the SLA or how such a distortion should be accounted for in its dumping analysis under U.S. law."¹⁶ Commerce then argues that Respondents are presenting a new argument before the Panel that should be rejected for failure to exhaust administrative remedies. Moreover, argues Commerce,

... the new arguments should be taken with a large helping of salt.

For example, Respondents state that “[n]owhere does Commerce actually address the question of the SLA’s impact on prices for lumber in the Canadian market and whether the resulting lower priced Canadian sales were in the ordinary course of trade for purposes of calculating normal value.” Joint Brief 107. Given that this theory was never advanced before Commerce, it is unreasonable to expect the agency to have addressed it. Even were Commerce to entertain such an argument, Commerce has rarely found sales that fail the statutory cost test to be in the ordinary course of trade, as generally sales sold in substantial quantities over an extended period of time which do not permit the recovery of all costs within a reasonable period of time are, by their very nature, outside the ordinary course of trade.

Finally, Respondents improperly attempt to support their new theory with information that was not part of the administrative record. See Joint Brief at 107 (supporting its argument with findings by the ITC made several weeks after the date of Commerce’s determination.)

For these reasons, Commerce acted reasonably and in accordance with law when it addressed the minimal arguments related to the alleged (and never quantified) distortions caused by the SLA.¹⁷

For their part, Petitioners contend that the effects of the SLA are legally irrelevant in this case. They submit that Commerce is required to examine whether and to what extent dumping has occurred as opposed to why the dumping occurred. In support they assert:¹⁸

The statute plainly states that "[i]f the administering authority determines that sales made at less than the cost of production -- (A) have been made within an extended period of time in substantial quantities, and (B) were not at prices which permit recovery of all costs within a reasonable period of time, such sales may be disregarded in the determination of normal value."¹⁹ The statute further makes clear that sales disregarded as below cost "shall" be considered to be outside the ordinary course of trade.²⁰ Those conditions were plainly met in this case.

Canadian Parties' claims to the contrary notwithstanding, the statute does not obligate the Department to find below-cost sales made over an extended period of time in substantial quantities to be within the ordinary course of trade. Plainly, having the discretion under law to do something and being obligated under the law to do something are quite different. Not only is the Department not required to find sales below cost to be within the ordinary course of trade, but its practice has been consistent in treating sales failing the statutory cost test to be outside the ordinary course of trade.²¹

Further, the URAA SAA supports the Department's determination. It states, "[i]f home market . . . sales are below-cost and all of the criteria of section 773(b) are satisfied, Commerce may exclude such sales for purposes of determining normal value. The Administration intends that Commerce will disregard sales when the conditions in the law are met."²² There is no question in this case that the conditions in the law were met, as the Department determined: Canadian Parties sold lumber below cost in substantial quantities over an extended period of time. Accordingly, the Department was consistent with the statute, its precedent, and the plainly expressed intent of Congress when it excluded from the calculation of normal value all sales that failed the statutory cost test.

The URAA SAA does state that "in some cases, below-cost sales may be used to determine normal value if those sales are of obsolete or end-of-model-year merchandise" if exports of that merchandise to the United States are "similarly obsolete or end-of-model year."²³ However, it is important to note two elements of the exception to the general rule. First, the exception provides the Department with the discretion to include certain types of sales failing the statutory cost test but does not require their inclusion. Second, the exception is expressly linked to certain types of products such as "obsolete" or "end-of-model year" merchandise and particularly perishable agricultural products (such as flowers) not at issue here.²⁴ Clearly, lumber does not fall into these exceptions. As such, the Department had no reason to contravene its standard practice of excluding sales failing the statutory cost test.

In reply, Respondents note that Commerce first asserts in the IDM that the issue was not raised by Respondents during the investigations, then proceeds to assert that it adequately responded to the Respondents' argument in the IDM. Respondents' citations to the record indicate that the matter was raised by at least some of the Respondents.²⁵ The issue was thus placed before Commerce and needed to be addressed by Commerce.

On the substantive issue, Respondents base their case on the requirement of U.S.C. §1677b(a) that a “fair comparison shall be made between export price or constructed export price and normal value” in determining whether subject merchandise has been dumped into the U.S. market. (emphasis added). Respondents assert:²⁶

Commerce failed to take into account the effect of the SLA in determining which sales were in the ordinary course of trade in Canada. As a result, in nearly all the comparisons where Commerce found dumping, the average normal value calculated by Commerce was artificially inflated because lower-priced Canadian sales were disregarded as outside the ordinary course of trade. Jt. Initial Br. (Vol. I) at 105. Indeed, well over one-third of the dumping margin found for the Canadian Respondents was based on less than two percent of the relevant Canadian home market sales. As a consequence of Commerce’s failure to take into account the effect of the SLA in Canada, the dumping found in this case was not based on fair comparisons.

Commerce has recognized that it “must administer the antidumping law in a manner which reflects economic reality and is consistent with the basic purpose of the Act,” and that “{i}n determining when to exclude below-cost sales from computation of fair value, one must interpret the language of the statute in light of the normal business practice of the industry subject to the investigation.” Certain Fresh Winter Vegetables from Mexico, 45 Fed. Reg. 20,512, 20,514 and 20,515 (Mar. 28, 1980), aff’d, 584 F. Supp. 10 (Ct. Int’l Trade 1984) (“Winter Vegetables”). In Winter Vegetables, after analyzing business practices in the North American market for winter vegetables during the period of investigation, Commerce found that the governing statute provided “sufficient flexibility” to allow it to take the economic circumstances into account in order to implement the purpose of the law. Id. at 20,516.

Respondents thus assert that Commerce failed to take the effect of the SLA into account in its dumping analysis, and failed to meet its obligation to “include in a final determination . . . an explanation of the basis for its determination that addresses relevant arguments, made by interested parties who are parties to the investigation . . . , concerning the establishment of dumping . . .” 19 U.S.C. § 1677f(i)(3)(A). Respondents argue that the economic reality of the market in Canada while the SLA was in effect was that sales below fully allocated costs were normal during the period investigated. Respondents refute the assertion by Petitioners that a profit maximizing company would not sell "below cost" by drawing attention to the well-established practice of selling at a price to equal to or exceeding marginal cost, thus contributing to the liquidation of fixed costs.

The antidumping law does not deal directly with the rationality of selling below fully allocated costs but in excess of marginal cost. Rather, the statute requires that consideration of sales in the exporter's home market begin (as relevant to this analysis) with a determination of whether any sales are below cost of production (i.e., fully allocated costs as defined by § 1677b(b)(3)), and, if so, then consider whether they are to be disregarded in determining normal value. The statute permits Commerce to disregard such sales, but does not require that it do so. The word "may" confers authority to exercise judgment as to when such sales should be disregarded and when not. Any reasonable exercise of judgment by Commerce will satisfy its responsibility. The exercise of judgment requires that the issue be addressed and considered in the light of the evidence on the record.

Commerce applied its longstanding practice of excluding sales below cost of production, as that concept is defined in the antidumping law of the United States, when determining the normal value of subject merchandise. The practice is approved by the Statement of Administrative Action relating to the Uruguay Round Agreements Act, which statement is to be given effect.²⁷ Cases in which sales below cost of production have been found to be in the ordinary course of trade are rare, as reflected in the Statement of Administrative Action and illustrated by Southwest Florida Winter Vegetable Growers Assn. v. United States, 584 F. Supp. 10 (Ct. Int’l Trade 1984) While ambiguous,²⁸ the Statement of Administrative Action authoritatively indicates the narrow scope for exceptions.²⁹ Respondents therefore faced a substantial challenge to show that lumber prices in Canada during the period of investigation were in the ordinary course of trade. Although the treatment of the issue in the IDM is perfunctory and dismissive of Respondents’ arguments, Respondents have not provided an analysis of the Canadian market during the period of investigation (the fifth year of the operation of the SLA) that would quantify the effect of the SLA and that would satisfy the burden of showing that Commerce’ determination on the matter is not in accordance with the law. Accordingly, the Final Determination is affirmed with respect to the issue about the effect of the SLA on softwood lumber prices in Canada during the period of investigation.
 

3. Commerce Did Not Err in Deducting from Export Prices the Export Taxes Assessed by Canada under the Softwood Lumber Agreement.

Respondents argue that Commerce’s decision to reduce U.S. export prices by the amount of the export taxes paid under the Softwood Lumber Agreement (SLA) is not supported by substantial evidence or otherwise in accordance with law.

As a result of negotiations between Canada and the United States, the SLA (applicable from April 1, 1996 to March 31, 2001) subjected lumber from certain Canadian provinces to an export tax on shipments to the United States, when shipments of such lumber exceeded 14.7 billion board feet per year. Following expiry of the SLA, Petitioners immediately requested new antidumping and countervailing duty investigations. In this antidumping investigation, Commerce, in calculating export prices and constructed export prices (CEP) during the POI, adjusted the prices to deduct therefrom the SLA export taxes paid by Respondents³⁰ and used values relating to shipments that included the SLA taxes. In a countervailing duty case launched at the same time as this antidumping investigation, the agency found that countervailable subsidies were bestowed on lumber exports from Canada.³¹

Respondents allege that Commerce’s decision to deduct the annualized SLA tax for shipments of subject merchandise to the U.S. was contrary to the intent of Congress. In this connection, they maintain that the SLA taxes were imposed in lieu of a countervailing duty to offset a subsidy found by the agency on the subject goods, that these amounts were reflected in the prices charged to U.S. importers, and that deducting these amounts from the export prices effectively increases the margin of dumping by an equivalent amount for goods subject to this investigation - in effect, asking firms to pay the tax again in the form of antidumping duties. Respondents cite 19 U.S.C. § 1677(6)(C)³² in arguing that this tax qualifies for the exemption to the requirement set out in 19 U.S.C. §1677a(c)(2)(B)³³ regarding the deduction of an export tax when calculating export price. Further, they claim that the wording of 19 U.S.C. §1677(6)(C) makes clear that Congress intended that taxes specifically intended to offset a subsidy should be taken into account in Commerce’s calculations. Alternatively, Respondents note that the SLA tax involves elements comparable to those covered in suspension agreements, to the extent that the governments of Canada and the United States entered into a voluntary arrangement specifically designed to address subsidy practices and the avoidance of U.S. countervailing duties. Respondents argue that the agency’s decision to deduct the SLA tax in calculating export and constructed export price is not in accordance with the law.

Commerce’s asserts that 19 U.S.C. §1677(6)(C) is unambiguous in requiring that, to qualify for this exemption, the tax must be “specifically intended to offset the countervailable subsidy received.” The agency notes that the intent of the SLA was to ensure that there was no material injury or threat thereof to an industry in the United States due to imports of softwood lumber from Canada, that the SLA was not negotiated under the countervailing duty or suspension agreement provisions of the Act, and that there is no reference to a countervailable subsidy in the SLA. Commerce also notes that the tax provided for in the SLA, unlike a countervailing duty, only applied to exports above a certain quantity (14.7 billion board feet). No tax was imposed on quantities below this volume. Accordingly, the agency determined that this particular tax did not meet the definition as set out in section 1677(6)(C) for the exemption described in section 1677(c)(2)(B). Petitioner concurs with the decision of the agency.

The issue in this case is whether the deduction of the SLA export tax from the export price or constructed export price to United States purchasers was appropriate in the context of establishing whether dumping was taking place and, if so, the margins of dumping involved. In considering this matter it is important to note that (1) the prices used by Commerce in its calculations relate to sales made during a period when the SLA was operational (the POI), (2) the SLA tax only applied to sales above a certain level and (3) the agency annualized the tax for each Respondent to derive an export tax expense per MBF sold during the POI. At the present time, cash deposits are being paid based on the estimated amount of the dumping duties as calculated by the agency. On the anniversary of the Order companies can request an administrative review based on the actual prices that were involved on the specific transactions.³⁴

Record evidence establishes that the SLA tax was put in place in an attempt to reach agreement on a mechanism that would replace countervailing duties that might otherwise have been imposed by the agency if an investigation resulted in a determination that certain Canadian practices constituted countervailable subsidies. The relationship between a received subsidy practice and the initiation of countervailing investigations is evident. It is also clear from the record that there was no countervailing duty investigation underway nor was a countervailing duty in place at the time the SLA was agreed. At the same time, section 1677(6)(C) is explicit in requiring that the export tax be specifically intended to offset the countervailable subsidy received. In this instance, while the SLA tax was designed to alleviate a situation in which countervailing duties might have been imposed, the tax itself did not offset a countervailing duty. Neither did it offset the full amount of any subsidy that might have applied on individual shipments of subject merchandise because (a) it only applied to exports that exceeded an agreed volume and (b) the tax paid by producers to the Canadian government was annualized over all shipments during the POI so that the actual amount factored into the per unit export price to U.S. customers was somewhat less than the SLA tax that was paid on any specific shipment.

Essentially, Respondents are arguing that the POI chosen to establish the export price already includes a factor to offset any countervailing duty that might have been applicable. Indeed, the parties put forth various arguments concerning whether the U.S. government had determined whether a subsidy was paid on subject merchandise and whether Canada accepted the conclusions reached in this regard. However, regardless of the merits of these arguments, the issue involved here is the calculation of export price or constructed export price for the purposes of an antidumping investigation, not a countervailing duty investigation. The scheme of the Act as it relates to antidumping involves a requirement to establish a normal value and export price of subject merchandise. In order to ensure a fair comparison, certain adjustments are prescribed in the statute. As noted above, one of the adjustments required is the deduction from the export price for any amount included in such price for export taxes, duties or other charges imposed by the exporting country (19 U.S.C. §1677(c)(2)(B)). The only exception from this requirement is that provided for in 19 U.S.C. §1677(6)(C) that allows Commerce to deduct from the gross countervailable subsidy an amount for export taxes specifically intended to offset the countervailable subsidy received. In this case, no countervailable subsidy was established for the purposes of the SLA nor was the SLA tax specifically intended to offset a countervailing subsidy received.

Respondents also referred to the comparability of the SLA to Suspension Agreements. These are provided for in 19 U.S.C. §1671c(a)(b), which permits acceptance of Suspension Agreements from the government of an exporting country to eliminate or offset a countervailing subsidy or to cease exporting the subject merchandise to the United States. This provision parallels Article 18 of the World Trade Organization Agreement on Subsidies and Countervailing Measures and both contain a variety of provisions circumscribing the use of Undertakings or Suspension Agreements. For example, a Suspension Agreement may only be accepted in the context of an ongoing investigation, it may only be accepted following the making of a preliminary determination and an Agreement only suspends rather than terminates an investigation. Clearly, the SLA was not negotiated with a view to putting a Suspension Agreement in place, especially since the two governments could not agree on the existence of a subsidy practice. Moreover, one of the provisions of the SLA limited the United States from taking countervailing measures during the life of the Agreement.

Given the foregoing, the Panel finds that Commerce’s decision concerning the calculation of export price and constructed export price is in accordance with law.
 

4. Commerce's Methodology for Calculating A Constructed Value Profit  Has Not Been Adequately Explained

Where possible, Commerce calculated the Normal Value (NV) of imported softwood lumber products on the basis of home market sales prices for the corresponding A foreign like product ³⁵. In making price-based comparisons, Commerce defined the A foreign like product  in accordance with 19 U.S.C. §1677(16)³⁶, and identified several different "A like products" . However, in cases where there were no home market sales of matching products, or where sales were below cost of production (COP), Commerce determined Normal Value on the basis of "Constructed Value" in accordance with 19 U.S.C. §1677b(a)(4). "Constructed Value" (CV) is not an actual price at which merchandise is sold, or offered for sale. As the Statement of Administrative Action accompanying the Uruguay Round Agreements Act, P.L. 103-465, 108 Stat. 4809, makes clear, "constructed value serves as a proxy for a sales price" in the country of exportation. H.R. Doc 103-316, at 839 (1994), reprinted in 1994 U.S.C.C.A.N. 3773, 4175. CV is defined by statute as the sum of (1) the "total cost of materials and fabrication" incurred in producing the merchandise, (2) the "cost of all containers and coverings" and other costs incidental to placing the merchandise is condition packed ready for exportation to the United States, and (3) the "actual amounts incurred and realized . . . for selling, general and administrative expenses, and for profits" earned on home market sales in the ordinary course of trade. See19 U.S.C. §1677b(e)(1-3).

In cases where Normal Value was determined on the basis of CV, Commerce did not attempt to calculate a separate "CV profit" for each type of softwood lumber product being examined. Rather, Commerce calculated CV profit on the basis of profits realized on each Respondent's  aggregate home market sales of softwood lumber products, treating such aggregate sales as the " foreign like product" . Commerce calculated CV profit by multiplying the weighted average profit rate calculated on each Respondent's home market sales made in the ordinary course of trade by the cost of production of the like product. Commerce asserts that, in so doing, it was applying the "preferred" method for calculating CV profit, as set out in 19 U.S.C. §1677b(e)(2)(A).

Respondents assert that Commerce's calculation of CV profit was arbitrary and unlawful, since the agency defined "foreign like product" for purposes of the CV profit calculation differently than for purposes of its price-based calculations of normal value. Respondents also allege that Commerce's  definition of "foreign like product" for purposes of the CV profit calculation is unlawful, since it is inconsistent with the statutory definition set out at 19 U.S.C. §1677(16). Commerce does not deny that it used different definitions of the term for purposes of its price-based comparisons and its CV profit computation.

The Court of Appeals for the Federal Circuit has ruled that, as a general matter, it must be presumed that Congress intended the term "foreign like product" to be defined the same way in different sections of the antidumping statute. However, the presumption is rebuttable. As noted in SKF USA, Inc. v. United States, 263 F.3d 1369, 1382 (Fed. Cir. 2001):

In the statute Congress has used the term "foreign like product" in various sections and has specifically defined it in 19 U.S.C. §1677(16). We therefore presume that Congress intended that the term have the same meaning in each of the pertinent subsections of the statute and we presume that Congress intended that Commerce, in defining the term, would define it consistently. Without explanation sufficient to rebut this presumption, Commerce cannot give the term "foreign like product" a different definition (at least in the same proceeding) when making the price determination and in making the constructed value determination. This is particularly so because the two provisions are directed to the same calculation, namely, the computation of normal value (or its proxy, constructed value) of the subject merchandise.

See also RHP Bearings, Inc. v. United States, 288 F.3d 1334 (Fed. Cir. 2002)(applying the same presumption to different definitions of A foreign like product”).

In the instant case, Commerce has provided reasons why it was necessary to define the term "foreign like product" differently for purposes of price based calculations of normal value and for purposes of calculating CV profit. However, Commerce has not provided an explanation sufficient to allow the Panel to determine whether the specific definition of like product” adopted by the agency in calculating CV profit in this case is reasonable and lawful. For that reason, the Panel remands this issue to Commerce, with instructions for the agency to provide a reasoned explanation of why it elected to define "foreign like product" for purposes of the CV profit calculation with reference to an aggregate of each Respondent's  total home market sales of softwood lumber products.

The antidumping statute contains a "preferred" method of calculating CV profit, and also specifies three non-hierarchical "alternative" methods for calculating CV profit in cases where the preferred method cannot be used.

The "preferred" method requires Commerce to calculate CV profit based upon:

…the actual amounts incurred and realized by the specific exporter or producer being examined in the . . . review for . . . profits, in connection with the production and sale of a foreign like product, in the ordinary course of trade, for consumption in the foreign country.

19 U.S.C. § 1677b(e)(2)(A). Significantly, the "preferred" method requires Commerce to determine profit amounts in connection with sales "in the ordinary course of trade" ; this requires the agency to eliminate from its calculation any sales which are made below cost, and which are thus not in the "ordinary course" of trade.

By contrast, the three "alternative" methods, specified in 19 U.S.C. § 1677b(e)(2)(B) are not limited to sales made in the "ordinary course" of trade. The law providing these alternative methods states CV profit may be calculated as follows:

(B) if actual data are not available with respect to the amounts described in subparagraph (A), then B

(I) the actual amounts incurred and realized by the specific exporter or producer being examined in the . . . review . . . for profits, in connection with the production and sale, for consumption in the foreign country, of merchandise, that is in the same general category of products as the subject merchandise;

(II) the weighted average of the actual amounts incurred and realized by exporters or producers that are subject to the . . . review (other than the exporter or producer described in clause (I)) for . . . profits, in connection with the production and sale of a foreign like product, in the ordinary course of trade, for consumption in the foreign country, or

(III) the amounts incurred and realized for . . . profits, based on any other reasonable method, except that the amount allowed for profit may not exceed the amount normally realized by exporters or producers (other than the exporter or producer described in clause (I)) in connection with the sale, for consumption in the foreign country, of merchandise that is in the same general category or products as the subject merchandise . . . .

It is to be expected that Respondents in an antidumping investigation might prefer the use of the alternative methods of determining CV profit, particularly those set out in 19 U.S.C. §1677b(e)(2)(B)(I) and (III), since those methods do not require the exclusion from consideration of sales which are not in the "ordinary course of trade". In this case, Respondents assert that, had Commerce applied the same definition of "foreign like product" to CV profit as it applied to the determination of price-based Normal Value, it would have to conclude that, for sales subject to CV analysis, there were no home market sales in the "ordinary course of trade" and the CV profit could not be determined using the "preferred" method of 19 U.S.C. §1677b(e)(2)(A). Presumably, these alternative methods would yield a lower CV profit, a lower CV, and lower margins of less than fair value sales.

As noted above, the Federal Circuit's  SKF decision holds that, absent a reasoned explanation, Commerce's  application of two different interpretations to the term "foreign like product" in the context of a single antidumping calculation is presumably arbitrary and unlawful. See also Transactive Corp. v. United States, 91 F.3d 232, 237 (D.C. Cir. 1996). Aware of the Federal Circuit's  SKF decision, issued shortly before the publication of the final determination at issue in this case, Commerce sought in its determination to explain its decision to use different definitions of "foreign like product":

The SAA establishes a general rule or preferred methodology for calculating the amounts for SG&A and profits in the CV calculation. For the preferred methodology to be applicable, there must be sales of a foreign like product in the ordinary course of trade (i.e., sales made at prices above cost). However, the statute and SAA also establish when normal value is to be based upon CV, stating that A only if there are no above cost sales in the ordinary course of trade in the foreign market under consideration will Commerce resort to constructed value. Thus, if the Department were required to interpret and apply the term "foreign like product" in precisely the same manner in the CV profit context as in the price context there would be no sales of foreign like product on which to base the CV profit calculation. Accordingly, the preferred method of calculating CV profit established by Congress would become an inoperative provision of the statute. Final Determination, Comment 6.

This statement does explain why the term "foreign like product" need not be interpreted identically for purposes of price-based calculation and for purposes of determining CV profit. Indeed, in RHP Bearings Ltd. et al. v. United States, 2003 Ct. Intl. Trade LEXIS 11, Slip Op. 03-10 (January 28, 2003), the United States Court of International Trade, evaluating a Commerce remand determination addressed to this issue rendered in the context of an antidumping proceeding involving bearings, held that a similar explanation by Commerce provided "sufficient explanation to rebut the presumption that Commerce cannot use differing definitions for an identical term in the same proceeding", and held that  "Commerce adequately explained why the differing use of the same term is necessary to establish NV and CV profit in the same antidumping proceeding" . In similar fashion, the explanation appearing at Comment 6 of the IDM in this case is sufficient, in the Panel's  view, to rebut the presumption that the term "foreign like product" must be given an identical definition wherever it appears in the antidumping statute.

Recently, in FAG Kugelfischer Georg Schafer AG et al. v. United States, 2003 U.S. App. LEXIS 11607, No. 02-1500, -1538 (June 11, 2003), the Court of Appeals for the Federal Circuit ruled that a decision by Commerce to calculate CV profit based upon the  "preferred" basis set out in 19 U.S.C. §1677b(e)(2)(A) was reasonable and not overbroad, in connection with an antidumping review of certain antifriction bearings. The Circuit court there ruled:

Section 1677(16) . . . offers three alternative definitions for foreign like product, which increase in the scope of products that may be included. See 19 U.S.C. §1677(16). The first available category, with which differing determinations may satisfactorily be made, is to be applied. Id. There is no restriction that Commerce use just one subsection per proceeding. Id. Accordingly, we believe that Commerce reasonably explained that the determinations for the variables at issue require different sets of foreign like product data. The bearing market, with its wide disparity in products, necessitates that direct price comparisons be done on a model-by-model basis. Therefore, the use of price comparisons requires the identical model and product family data of sections 1677(A) and (B). And CV profit may be based on a broader scope of products because use of aggregate data, as described in section 1677(16)(C), results in a practical measure of profit that can be applied consistently and with administrative ease over the range of included products. The Federal Circuit also rejected the notion that Commerce must work its way through the hierarchy of definitions set out in 19 U.S.C. §1677(16), a methodology which would allow the agency to take below-cost or non-contemporaneous sales into consideration.  "This logic fails, however, because calculating constructed value under section 1677b(e)(2)(A) requires that the sales of foreign like product occur within the ordinary course of trade. And the definition of ordinary course of trade requires that the sales used must not be below cost, id., §1677b(b)(1)(disregarding below cost sales that meet the requirements of subsections (A) and (B)), and must be contemporaneous to the exportation of the subject merchandise, see id. §1677(15)."

FAG Kugelfischer Georg Schafer AG, supra, at 9.

Thus, Commerce's  decision to define "like product" for purposes of calculating CV profit on the basis of the definition set out at 19 U.S.C. § 1677b(e)(2)(A) is neither unlawful, nor based on incorrect methodology. That does not end the inquiry, however. That Commerce need not define the term "foreign like product" identically in all phases of an antidumping determination does not establish that the particular definition applied in this case in determining CV profit B defining each Respondent's  aggregate home market sales of as sales of a "foreign like product" B is reasonable. Indeed, the Final Determination in this case merely asserted that:

In [RHP Bearings, Inc. v. United States, 2001 Ct. Intl. Trade LEXIS 109, Slip Op. 01-106 (2001)], the CIT affirmed the Department's  calculation of CV profit of the class or kind of merchandise, which encompassed all foreign like products under consideration, because the use of such data matched the criteria of section 771(16)( C) of the Act (i.e., the same general class or kind of merchandise). We believe a method based on varied groupings of foreign like products, each defined by a minimum set of matching criteria shared with a particular model of the subject merchandise, would add an additional layer of complexity and uncertainty to the antidumping proceedings without generating more accurate results.

IDM, Comment 6.

However, the CIT determination in RHP Bearings, on which the Final Determination relies, was vacated by the Court of Appeals for the Federal Circuit, and remanded to Commerce for a more complete explanation of Commerce's  CV profit methodology. RHP Bearings, Inc. v. United States, 288 F.3d 1334 (2002). In particular, the Federal Circuit vacated the lower court's  decision completely, including its decision with respect to Commerce's  finding in that case that "the use of aggregate data results in a reasonable and practical measure of profit that we can apply consistently in each case" . Following remand proceedings, the CIT upheld Commerce's  remand determination not only because it found that a satisfactory explanation had been provided for the agency's  decision to apply different definitions to the term "foreign like product" , but also based upon Commerce's  description of the "factual background of its calculations", and Commerce's  own admission in that case that the use of different definitions will depend on the specific facts of each case. "Commerce further explains that differing categories of merchandise can satisfy the meaning of the term "foreign like product" , depending on the specific facts of each antidumping proceeding . . .". RHP Bearings Ltd v. United States, Slip Op. 03-10, at p. 11 (emphasis added). The Panel does not have the benefit of a similar explanation of the factual background of Commerce's  decision in this case.

While the courts have held that "like product" can be defined differently in the NV and “CV profit” contexts, the courts have not upheld a blanket rule that the use of a Respondent's  aggregate home market sales for calculating CV profit will be an acceptable method in all cases. While the Final Determination asserts that the use of a model-match comparison in determining CV profit "would add an additional layer of complexity and uncertainty to antidumping proceedings without generating more accurate results" , Commerce furnishes no explanation for this assertion. Had Congress intended that Commerce exclusively determine CV profit on the basis of a Respondent's total aggregate sales, it would have so provided in the statute. Indeed, certain Respondents have suggested that the use of such an aggregate may frustrate the antidumping law's  goals of a fair "apples-to-apples" comparison in the case of softwood lumber products, since certain products are generally sold at low or no profit, and the application of an aggregate CV profit calculation might artificially create or inflate LTFV margins.³⁷ Absent Commerce's  explanation of why the use of the aggregate method in this case is appropriate, and not arbitrary, the Panel is not in a position to decide whether the agency’s specific definition of “like product” is reasonable. That the method used is administratively convenient does not demonstrate that it represents a reasonable interpretation and application of the statute.

Accordingly, the Panel remands this issue to Commerce with directions that the agency provide an explanation of why, in this case, the agency's  decision to define "foreign like product" for purposes of calculating CV profit as each Respondent's aggregate sales of subject merchandise is reasonable and in accordance with law.
 

5. Commerce Erred in Failing to Take Dimensional Differences Into Account in Allocating Joint Costs

Respondents contend that Commerce, in its Final Determination, used an improper methodology for allocating joint production costs. Respondents contend, specifically, that the appropriate method for allocating joint production costs must be sensitive to value differences associated with particular product characteristics. Commerce used value-based cost allocations for different grades of lumber, but relied on average-cost allocations for different lumber dimensions. As a result, within a particular grade, lumber products that differed by thickness, width and/or length were all assigned identical production costs on a volumetric (million board feet or MBF) basis. Commerce’s failure to allocate joint production costs on the basis of relative values corresponding to dimensional differences resulted in a variety of alleged distortions in crucial calculations, including the below-cost test, the price-to-price comparisons that Commerce allowed, DIFMER adjustments and the CV calculation.³⁸ These distortions resulted in a cost allocation that did not “reasonably reflect the costs associated with the production and sale” of individual products.³⁹

To support their contention, Respondents argue that because all softwood lumber products are processed jointly and indistinguishably, it is impossible to differentiate between different thicknesses, widths, lengths, and grades of lumber until the end of the sawmill process. As a result, the costs of stumpage, harvesting, hauling, storing, debarking and sawing whole logs are not directly traceable to particular lumber products. Thus, the entire logging and sawmill operation is a quintessential example of joint production.

Respondents further argue that size affects price. All parties advocated the inclusion of width, thickness and length among the product characteristics that Commerce should rely on in order to match U.S. and home market products for price comparisons. Commerce agreed and included those criteria.

Commerce responds that it only departed from the companies’ normal records for the calculation of log and saw mill costs, which are joint costs associated with wood grades. Commerce stated that it relied on the normal books and records for all other costs, as required by statute. Since there are no measurable differences in the inputs used or the processing required to produce lumber of various dimensions within each grade, Commerce claims it is reasonable to conclude that there are no actual cost differences associated with dimension. Thus, it was reasonable for Commerce to limit the value-based cost allocation to grades.

Commerce also argues that while the differing grade bands are inherent in the log, the differing dimensions into which the log is cut are controlled almost entirely by management. This results in each mill shifting production to the dimensions that are commanding higher prices, which creates extra supply that fills any shortages and then brings down prices. Commerce contends costs associated with dimension should not be allocated based on sales prices. Further, Commerce argues, pricing patterns of specific lumber dimensions do not follow a reliable pattern to justify the use of dimensional prices to allocate costs. In fact, the relationships between dimensions vary significantly between companies.

The antidumping law requires that Commerce calculate costs “based on the records of the exporter or producer of the merchandise, if such records are kept in accordance with the generally accepted accounting principles of the exporting country . . . and reasonably reflect the costs associated with the production and sale of the merchandise.”⁴⁰ Where a Respondent’s books and records do not meet these criteria, Commerce must employ an alternative method to calculate costs. In this case, Commerce determined that the average joint costs used by Respondents in their accounting records did not reasonably reflect the production costs of the individual products under investigation. As a result, Commerce devised a methodology by which it used a value-based methodology for allocating joint wood and sawmill costs between different grades of lumber. Commerce, however, continued to use a volume-based average cost methodology to allocate joint costs between different sizes of lumber within a particular grade. The Panel holds that Commerce erred by failing to use the same value-based methodology to allocate joint costs between different sizes of lumber as well.

The Panel is aware of the deference to be accorded to Commerce’s determinations on the methodologies to be utilized in cost allocation determinations; however, the methodologies selected must be sufficiently reasonable.⁴¹ The logic of applying a value-based cost allocation to joint products is as compelling with respect to dimensional differences as it is with respect to grade differences: (1) the products are produced simultaneously through a joint production process and (2) the products produced have significantly different values. Although Commerce argues in its brief that softwood lumber of varying dimensions do not meet the criteria of “joint products”,⁴² Commerce acknowledged in its Final Determination that “lumber production is the result of a joint production process”⁴³ and that downstream processing (e.g., planing, kiln drying) marks the point where the process is moved outside of a joint product scenario.⁴⁴ As described by the parties, the production process for lumber is such that various sizes and grades of lumber are all produced simultaneously; a single input goes in to the production process and multiple products come out⁴⁵ - a quintessential example of a joint production process.

Whether or not the mix of various lumber products produced result from management decision, the economics of joint production remain the same. This mix is likely to include products of varying values whether determined intrinsically (such as maximum lengths set by the size of the log) or by design. Production is rational so long as total joint production costs are recovered through sales of the total mix of joint products produced. Disregarding value differences in joint products will necessarily result in some products displaying extraordinarily high profit margins and other products low (or negative) profit margins. A finding of less than fair value sales is an almost inevitable methodological artifact of a value-insensitive cost allocation approach.

Any cost allocation method applied to joint products has an element of arbitrariness. In the end, what are of economic substance in joint products (in making production decisions) are total costs. So long as total production costs are recovered (through the sale of the various joint products) it is arguable whether a finding of below cost sales (for dumping purposes) is warranted. A value-insensitive cost allocation methodology that mechanically generates significant below-cost sales on some joint products and extraordinary high-profit sales on others cannot be said to be reasonable.

With regard to the second criteria for the use of value-based cost, Commerce argues that the price differences between lumber dimensions do not provide a reliable and consistent pattern to show that there are distinct differences between the inherent values of the products. Commerce’s argument is somewhat surprising in light of the fact that Commerce included price differences between dimensions in the product matching criteria. Regardless, there is at least some data to suggest that dimension matters.⁴⁶ Ignoring this data necessarily introduces distortions into the calculations.⁴⁷

The Panel holds that disregarding value differences corresponding to dimensional distinctions for purposes of cost allocation is unreasonable and remands this determination to Commerce with instructions to apply a value-based allocation methodology to make cost allocations in respect of dimension as well as grade.
 

6. Commerce Erred in Failing to Make An Adjustment to Account for Dimensional Differences in the Merchandise Being Compared

Respondents argue that the antidumping law (19 U.S.C. 1677b(a)(6)(c)(ii))⁴⁸ requires that Commerce calculate a “difference in merchandise” (DIFMER) adjustment when comparing prices of products that differ in physical characteristics. In this instance, the agency did not calculate a DIFMER adjustment for grade or dimension differences in its Preliminary Determination. However, in the Final Determination, as a result of having made a value based cost allocation for wood and sawmill costs, Commerce made a DIFMER adjustment for differences in grade, but determined that no DIFMER adjustment for dimension was appropriate since the record did not establish that there were differences in cost or value attributable to differences in dimension.⁴⁹

There is no disagreement between the parties that lumber size has an impact on price and that physical differences can result in different market values. In this context, Respondents point to the parallel between this and the cost allocation issue discussed above, but stress that the DIFMER issue is narrower as it is comparison specific, i.e. it relates only to a determination of whether a difference in price is in whole or in part due to physical differences in the products compared. They argue that if there is a price difference due to physical characteristics, Commerce must make a DIFMER adjustment. They claim that in accordance with 19 U.S.C. §1677-1(a)(2) and 19 C.F.R. §351.413, the agency can only disregard such an adjustment as being insignificant if its impact on the dumping margin is less than 0.33 percent and that it cannot ignore below cost sales in the home market data base in calculating DIFMER adjustments. Respondents maintain the record demonstrates that (a) there are significant price differences between all similar product comparisons used, (b) there is no alternative explanation for the price differences other than differences in size and (c) there is no analysis by Commerce with respect to specific product comparisons.⁵⁰

Commerce explained that when it does not have identical home market sales within the ordinary course of trade, normal value is based on sales of the most similar product adjusted for physical differences if the amount of any price difference attributable to such physical differences has been established to its satisfaction. The agency notes that, while the statute grants it discretion to determine a suitable method to calculate a DIFMER, it has rarely been able to determine the direct price effect of differences in distinct physical characteristics of merchandise. As a result, the preferred method adopted is to base allowances almost exclusively on the differences in variable costs of manufacturing between the products compared. Commerce has found that the result achieved by this methodology approximates the effect that such differences have in products. However, it retained, as an option, the use of market value where appropriate.⁵¹

Petitioners argued that a DIFMER adjustment was made for all price-to-price matches involving non-identical merchandise and, where Commerce found no difference in variable costs between the products, the adjustment equaled zero. Petitioners take the position that while Commerce may calculate value-based DIFMERS, it is not required to do this when cost-based adjustments are possible. In this instance, they maintain that the agency calculated cost-based adjustments that equaled zero.⁵²

The issue before the Panel is whether or not Commerce should have made an adjustment for dimension differences and, if so, whether there was sufficient information on the record to establish to its satisfaction that the price differences between the goods compared were attributable to different physical characteristics. There is no dispute between the parties that 19 U.S.C. §1677b(a)(6)(c)(ii) provides for an adjustment to normal value for differences in physical characteristics between the products being compared. The governing regulation, 19 C.F.R.§351.411(a) also makes clear that while the Secretary has discretion to the extent that he “may” determine that the merchandise sold in the United States does not have the same physical characteristics as the merchandise sold in the foreign market, and that the difference has an effect on price, the Panel considers that the use of the word “will” in the last sentence of this subsection of the regulation imposes an obligation to grant a reasonable allowance when the physical differences between compared products has an effect on prices⁵³. (Emphasis added) The record indicates Commerce recognized that dimension could result in differences in market value but it did not grant an allowance for dimensional differences because it determined that, since it could not use home market sales that were below cost of production, there was no information on the record to enable it to make a DIFMER calculation.⁵⁴

The law requires that a fair comparison be made between the United States price and the foreign market price adjusted to account for physical differences when similar products are compared. In this case, Commerce stated that it did not have a cost basis for calculating a DIFMER adjustment. However, the record contains a considerable amount of data and charts concerning market prices for these products in Canada. In this connection, the agency noted that to the extent that it compared products having different dimensions, the differences were small, prices fluctuated in relation to each other during the POI and there appeared to be little, if any, difference in home market prices that was attributable to differences in dimensions of the products compared, especially where dimensional differences were minor.⁵⁵

It is relevant to note that while Commerce obviously made some price comparisons, it did not indicate that it had not been provided with sufficient data to enable it to establish to its satisfaction that price differences were attributable to physical differences. As well, it would be unreasonable for the agency to require constant prices or price consistency for and between products in a dynamic commercial market. Moreover, there does not appear to be any evidence on the record to substantiate the agency’s judgment concerning the effects of dimension on market values and prices. While 19 U.S.C. §1677f-1(a)(2) provides discretion to the administering authority to “decline to take into account adjustments which are insignificant in relation to the price or value of the merchandise”, the record does not provide an analysis to support the conclusion that “there is little, if any, difference in home market prices that is attributable to dimension.”

In determining that no value-based DIFMER could be calculated, Commerce decided that it would be inappropriate to use the Respondents’ home market prices where there were home market sales of certain products outside the ordinary course of trade, i.e. sales below cost of production during the POI. The logic of this approach, as stated by the agency, is that to do so would result in adjusting normal values for the similar merchandise back to the prices for the like merchandise that had already been determined to be outside the ordinary course of trade.⁵⁶ While the law may require the agency to disregard below cost sales for the purpose of calculating constructed value, the Panel does not accept that they should be disregarded when comparing similar merchandise for the sole purpose of establishing market value differentials due to physical differences.

In this instance, the objective in comparing Canadian domestic sales of the like product sold to the United States and Canadian sales of a similar product is to measure the market value of the differences between two physically different products. Clearly, the market will normally attach a different value to two pieces of lumber, one measuring 2x4x8’ and the other 2x4x16’. The only issue here is how to measure the price difference or ratio that the market is prepared to pay for one size over the other.

Under normal circumstances, producers’ selling price patterns, over a given period of time should reflect the market price differentials between the various sizes and grades of lumber. Mathematically it follows that, all other things being equal, if a producer’s Canadian price of the like product is below the cost of production, adjusting a profitable Canadian price for a similar product used for comparison purposes by the ratio between these two selling prices would result in a figure that is below the cost of production for the similar product. However, when the similar product is sold in various dimensions in Canada at a profit, it should be possible to establish a dimension differential that would reflect the relative market price for various sizes of the like and similar products. For example, if SPF 2x4x8’ lumber sells for $10.00 and SPF 2x4x16 sells at a profitable figure of 25.00, the cost per foot for the two would be $1.25 and $1.56 respectively; that is, a normal value of $12.48 for an 8 foot length rather than $10.00. Another possible approach might be to establish average Canadian market prices for similar products based on the selling prices of all producers in the Canadian market and establish price ratios between the different dimensions for use in calculating DIFMERs for the similar products being compared to the goods sold to the United States.

With regard to Petitioner’s contention, at the hearing its counsel noted that because home market prices were judged unreliable, the variable costs of production for the products concerned were entered into the agency’s computer program and the mechanical result that emerged for some of the products was a DIFMER equal to zero.⁵⁷ The Panel notes that Commerce has not claimed that it calculated a DIFMER for dimension but did state categorically “there was no basis for calculating a DIFMER for dimensions based on value or cost.”⁵⁸ Given the circumstances, the Panel cannot accept that a DIFMER was calculated when the agency itself declares that it was not.

The Panel concurs with Respondents that the law requires that a DIFMER adjustment be made where similar products are compared in the establishment of normal values based on record data relating to pricing of similar merchandise in Canada. Accordingly, this matter is remanded to Commerce for the calculation of a DIFMER for dimension.
 

7. Commerce Did Not Err in Employing a Practice of “Zeroing” When Determining Weighted Average Margins of Dumping

The Respondents contend that Commerce’s practice of “zeroing” is contrary to law. “Zeroing” involves the disregarding of the quantum by which export price (or its equivalent) exceeds normal value for certain products under investigation - these products display so-called “negative dumping margins.” In determining the weighted-average dumping margin across all products under investigation, Commerce sets the dumping margin at zero for each product with a negative dumping margin. The resultant weighted-average dumping margin is greater than would be the case were the amount of “negative dumping margins” included in the calculation.

Respondents argue that zeroing creates artificial margins by resetting all negative margins to zero and that to make a fair comparison, these differences should be part of the averaging. A “fair comparison”, Respondents continue, is required to be made “between the export price or constructed export price and normal value” under U.S. law, as amended by the Uruguay Round Agreements Act (URAA).⁵⁹ According to Respondents, for the comparisons to be fair, they must include all the weight-averaged values.⁶⁰

Further, Respondents argue that the practice of zeroing is illegal under international law. To support their claim, Respondents cite the report of the WTO Appellate Body in European Communities - Bed Linen from India⁶¹ (“EC-Bed Linens”) as establishing that the practice of zeroing is inconsistent with the WTO Antidumping Agreement.

Commerce responds that the statute directs Commerce to calculate the weighted average dumping margin for each Respondent and to consider “the percentage determined by dividing the aggregate dumping margins determined … by the aggregate export prices and constructed export prices …”⁶² Nowhere, Commerce contends, is the agency directed to include the amount by which sales are above fair value; indeed that would contravene the explicit terms of the statute. Although the statute is silent on how to account for non-dumped sales for purposes of calculating a dumping margin, Commerce continues, the United States Court of International Trade has upheld the practice as reasonable and legitimate.⁶³ Further, argues Commerce, including negative margins in its margin analysis would cancel out margins on dumped sales, effectively eviscerating the purpose of Commerce’s antidumping efforts. The agency also contends that the EC Bed Linens report does not provide support for Respondents’ case as EC Bed Linens did not comment on U.S. practices.

Recently, the Court of International Trade found that the statute, as amended by the URAA, “is silent as to the impact of negative margins” and the “statute neither requires nor prohibits Commerce from considering nondumped sales.”⁶⁴ Applying the deference accorded under Chevron,⁶⁵ the Court upheld Commerce’s practice of zeroing.

The post-URAA statute defines dumping margin as “the amount by which the normal value exceeds the export price or constructed export price of the subject merchandise.”⁶⁶ Weighted average dumping margin is defined as “the percentage determined by dividing the aggregate dumping margins determined for a specific exporter or producer by the aggregate export prices and constructed export prices of such exporter or producer.”⁶⁷ Commerce interprets these two sections as directing it to aggregate all individual dumping margins, each of which is determined by the amount by which normal value exceeds export price or constructed export price, and to divide this amount by the value of all sales. Therefore, Commerce, in calculating the dumping rate, included dumped and non-dumped sales in the denominator, but only dumped sales in the numerator.

Respondents’ main argument -- that the URAA amended U.S. law such that it requires a fair comparison be made between the export price or constructed export price and normal value which, in turn, requires that all the weight-averaged values must be included in the dumping margin calculation -- is not persuasive. The language “fair comparison” is used in 19 U.S.C. § 1677b(a) in the context of comparing export price or constructed export price and normal value in determining whether subject merchandise is being sold at less than fair value; it is not used in the statutory language for calculating a dumping margin. Further, Respondents do not provide persuasive evidence for their leap in reasoning that a “fair comparison” requires that the dumping margin calculation include all the weight-averaged values. While the Tariff Act of 1930 clearly has undergone substantial modifications by virtue of the URAA, this Panel does not find in it a clear prohibition of zeroing. The obligation to make a “fair comparison” contained at 19 U.S.C. 1677b does not seem to be a general source for attack against all practices in an antidumping investigation that arguably have a distorting effect.

Respondents’ argument that zeroing is inconsistent with the WTO Antidumping Agreement also is unpersuasive. The EC-Bed Linens report can at most be merely suggestive that Commerce’s practices are in conflict with WTO Antidumping Agreement obligations.⁶⁸ WTO decisions are not binding upon Commerce or on this Panel.

The Panel finds that Commerce’s practice of zeroing is a permissible application of the statute - one which has been upheld by various reviewing courts⁶⁹ -- and therefore is affirmed.
 

2. COMPANY-SPECIFIC ISSUES

1. Commerce’s Decision to Treat Abitibi-Consolidated Inc. and its Affiliate, Scieries Saguenay Ltee. As a Single Producer is Unsupported by Substantial Evidence on the Record.

Respondent Abitibi asserts that the determination by Commerce to treat Scieries Saguenay Ltée ("SSL") and Abitibi as one producer is not in accordance with law and not supported by substantial evidence. In the result, Abitibi is seeking to remove SSL from the scope of the dumping margin found for Abitibi (12.44%) to the "all others" rate of 8.43%.

Abitibi owns fifty (50) percent of the outstanding shares of SSL and has the right to appoint two directors of SSL's supervisory board of four directors who meet once a year to approve the prior year's financial statements and the budget for the current year. Abitibi also has the right to appoint a vice-president of SSL, and receives monthly financial statements (balance sheet, income statement, and cash-flow statement). Abitibi purchases wood chips from SSL and dries and planes some of SSL's lumber production on a contract basis. No lumber produced by SSL is sold through or by Abitibi, nor is any lumber produced by Abitibi sold through or by SSL. Abitibi is not provided with information about SSL's customers, prices for individual products or product mix. SSL does not consult with Abitibi regarding products, customers or prices.⁷⁰ During the POI, Abitibi's Senior Vice-President, Woodlands and Sawmills, was a vice president of SSL. He submitted a declaration on the administrative record stating that:

... he did not even know he had this title until the issue arose in the context of this antidumping proceeding. He stated that he had no management responsibilities at SSL, that no employees of SSL reported to him, and that he received no remuneration of any kind from SSL. His only role at SSL was to monitor its overall financial performance so as to safeguard Abitibi’s investment.⁷¹

The evidence is not challenged.

Abitibi notes that the antidumping statute contains no provision expressly authorizing Commerce to disregard the separate legal existence of distinct companies, or to collapse multiple companies. To the contrary, asserts Abitibi, the statute requires Commerce to compute a margin of dumping for each individual exporter or producer it investigates, i.e., each separate legal entity. In support of its position, Abitibi reviews certain judicial discussions of the issue in the context of antidumping law, and challenges the validity of 19 C.F.R. §351.401(f) that sets out the policy of Commerce. As well, Abitibi refers to the common law with regard to the treatment of separate legal entities, and concludes its arguments with the assertion that, in any event, Commerce has failed to apply its own criteria properly. These contentions will be dealt with below.

Commerce responds essentially as follows:⁷²

The Department has a long-standing practice whereby it treats two or more affiliated producers as one entity if it is shown that those companies have production facilities for similar or identical products that would not require substantial retooling of either facility in order to restructure manufacturing priorities and there is a significant potential for the manipulation of price and/or production. See Koenig & Bauer-Albert AG, et al. v. United States, 90 F. Supp. 2d 1284, 1288 (Ct. Int’l Trade 2000) (“Koenig & Bauer”) (citing Certain Pasta From Italy, 61 Fed. Reg. 30,326, 30,351 (Dep’t of Commerce 1996) (final determ.); Certain Hot-Rolled Carbon Steel Products from Canada, 58 Fed. Reg. 37,099, 37,107 (Dep’t of Commerce 1993) (final determ.); Certain Granite Products from Spain, 53 Fed. Reg. 24,335, 24,337 (Dep’t of Commerce 1988) (final determ.). When these factors are met the Department “collapses” these producers into one entity and assigns them a single dumping margin. id. This collapsing practice helps to “ensure that {the Department} reviews the entire producer or reseller, not merely a part of it.” Queen’s Flowers De Colombia, et. al. v. United States, 981 F. Supp. 617, 622 (Ct. Int’l Trade 1997) (“Flowers De Colombia”). Most importantly, by treating as one entity affiliated companies that could produce the same product and are under sufficient common control such that price and/or production could be manipulated, the Department is able to more effectively fulfill one of its basic responsibilities, preventing evasion of the Antidumping Law. Flowers De Columbia, 981 F. Supp. at 622.

The Department has established a detailed collapsing test in its regulations. 19 C.F.R. 351.401(f). In order to collapse two producers, the Department must first find that they are affiliated. 19 C.F.R. 351.401(f). “Affiliated” is defined within the Act and includes, inter alia, a company that owns more than 5 percent of the voting stock or shares of another company. 19 U.S.C. § 1677(33). In the instant investigation it was uncontested that Abitibi owned 50 percent of SSL’s stock. Abitibi Brief at pg. 60. Therefore, the Department properly found that Abitibi and SSL were affiliated.

Next the Department had to determine whether Abitibi and SSL had “. . . production facilities for similar or identical products that would not require substantial retooling of either facility in order to restructure manufacturing priorities.” 19 C.F.R. 351.401(f)(1). The Department found that both companies produced softwood lumber products falling within the scope of the investigation and, as such, the Department properly determined that both companies produced identical or similar products that would not require substantial retooling of either facility in order to restructure manufacturing priorities. Memorandum Re: Collapsing of Respondent Abitibi Consolidated Inc with Scieries Saguenay Ltee. from Amber Musser to Bernard T. Carreau PR Doc. 467 (July 18, 2001). (“Abitibi Collapsing Memo”)

Finally, under the Department’s collapsing test, the Department must determine whether there is a significant potential for the manipulation of price and/or production. 19 C.F.R. 351.401(f)(2). The factors that the Department will analyze include (1) the level of common ownership, (2) the extent to which managerial employees sit on the board of directors of an affiliated firm, and (3) whether operations between the two companies are intertwined. 19 C.F.R. 351.401(f)(2