OPINION

RESTANI, Judge.

This consolidated matter is before the court on cross-motions for judgment on the agency record, pursuant to USCIT Rule 56.2. Corus Staal BV and Corus Steel USA Inc. (collectively "Corus") and Bethlehem Steel Corporation, National Steel Corporation, and United States Steel Corporation (collectively the "petitioners") separately challenge certain aspects of the final determination of the United States Department of Commerce ("Commerce") in Certain Hot-Rolled Steel Flat Products from the Netherlands, 66 Fed. Reg. 50,408 (Dep't Commerce Oct. 3, 2001), as amended by 66 Fed Reg. 55,637 (Dep't Commerce Nov. 2, 2001) (" Final Determination"). In their motion, the petitioners challenge Commerce's decision to reclassify certain Corus sales as export price. In its motion, Corus argues that (1) Commerce's "zeroing" methodology is contrary to law; (2) Commerce erred in determining that Laura Metaal BV was affiliated with Corus; (3) Commerce improperly considered certain non-prime sales; (4) Commerce erred in denying Corus a level of trade adjustment for certain affiliated sales; and (5) Commerce failed to properly instructed the United States Customs Service ("Customs") to cease collection of provision measures after six months.

JURISDICTION STANDARD OF REVIEW

The court has jurisdiction over this antidumping duty proceeding pursuant to 28 U.S.C. § 1581(c) (2000). The court will uphold Commerce's determination in an antidumping investigation unless it is "unsupported by substantial evidence on the record, or otherwise not in accordance with law." 19 U.S.C. § 1516a(b)(1)(B)(i).

FACTUAL PROCEDURAL BACKGROUND

On December 4, 2000, Commerce initiated antidumping investigations to determine whether imports of hot-rolled carbon steel flat products from various countries, including the Netherlands, were being sold in the United States at less-than-fair-value ("LTFV"). See Notice of Initiation of Antidumping Duty Investigations: Certain Hot-Rolled Carbon Steel Flat Products From Argentina, India, Indonesia, Kazakhstan, the Netherlands, the People's Republic of China, Romania, South Africa, Taiwan, and Ukraine, 65 Fed. Reg. 77,568 (Dep't of Commerce Dec. 12, 2000). Corus Staal BV ("CSBV") manufactures subject hot-rolled steel in the Netherlands and was identified as the only producer of hot-rolled steel from the Netherlands. Id. at 77,573. Corus Steel USA Inc. ("CSUSA") is affiliated with CSBV and, among other things, imports the subject merchandise. Both were respondents in the underlying investigation.

On May 3, 2001, Commerce published an affirmative preliminary determination of sales at less than fair vaule ("LTFV") with respect to hot-rolled steel from the Netherlands. Notice of Preliminary Determination of Sales at Less Than Fair Value; Certain Hot-Rolled Carbon Steel Flat Products From the Netherlands, 66 Fed. Reg. 22,146 (Dep't of Commerce May 3, 2001) ("Preliminary Determination"). The Preliminary Determination ordered the suspension of liquidation of all entries of hot-rolled steel from the Netherlands and calculated a weighted-average dumping margin for Corus. Id. at 22,151.

On May 22, 2001, Corus requested a postponement of the final determination pursuant to 19 C.F.R. § 351.210(b)(2)(ii), as permitted by 19 U.S.C. § 1673d(a)(2), including an extension of the provisional measures from a four month period to not more than six months. See Corus' May 22, 2001 Letter to Commerce. On June 4, 2001, Commerce granted the request and postponed the final determination until no later than September 15, 2001, and the suspension of liquidation until October 30, 2001. See Postponement of Final Determination for Antidumping Duty Investigation: Certain Hot-Rolled Carbon Steel Flat Products From the Netherlands, 66 Fed. Reg. 32,600 (Dep't of Commerce June 15, 2001).

Due to the events of September 11, Commerce issued its affirmative final determination on September 21, 2001, finding that hot-rolled steel from the Netherlands was being sold, or likely being sold, in the United States at LTFV. Final Determination, 66 Fed. Reg. at 50,408 (Dep't Commerce Sept. 21, 2001), amended 66 Fed. Reg. at 55,637 (November 2, 2001) (revising the weighted average dumping margin for Corus). In the amended Final Determination, Commerce directed Customs to continue suspending liquidation of subject imports "until further notice." Id. at 55,639. Commerce published its final antidumping order on November 29, 2001. Antidumping Duty Order: Certain Hot rolled Carbon Steel Flat Products From the Netherlands, 66 Fed. Reg. 59,965 (Dep't Commerce Nov. 29, 2001). Corus and the petitioners separately appealed the Final Determination. On June 20, 2002, the court consolidated petitioners' appeal (No. 02-00028) with Corus' appeal (No. 02-00003). These motions followed.

DISCUSSION

As an initial matter, Commerce and the petitioners are in agreement that a remand is necessary to address the United States price treatment of tolled sales (sales to another party for further processing). Corus opposes the remand. In SKF' USA, Inc. v. United States, 254 F.3d 1022, 1028 (Fed. Cir. 2001), the court addressed the issue of voluntary remand which is requested by the governmente when the original determination denying a favorable adjustment to a respondent was not required by statute. The court noted that if "bad faith" is not at issue, remand is appropriate to consider a different statutory interpretation. See id. at 1028-29. Whether or not new adjustments favorable to the petitioners occupy the same position as those affecting a respondent, remand must be on account of appropriate reasons.

The court notes that the Federal Circuit did not discuss the line of cases indicating that mere policy changes should not be allowed to alter final agency determinations. See, e.g., American Trucking Ass'n v. Frisco Transportation Co., 358 U.S. 133, 79 S.Ct. 170, 3 L.Ed.2d 172 (1958) (finding that the Interstate Commerce Commission cannot, without specific statutory authority, reconsider license and certificate decisions because of policy changes); Upjohn Co. v. Pennsylvania Railroad Co., 381 F.2d 4, 5 (6th Cir. 1967) (finding that granting a voluntary remand request based merely upon a change in policy "would result in chaos and uncertainty of action for those who must rely on its findings."); Chapman v. El Paso Natural Gas Co., 204 F.2d 46, 53-54 (D.C. Cir. 1953) ("[A] decision may not be repudiated for the sole purpose of applying some quirk or change in administrative policy."). In any case, concerns for finality do exist and the agency must state its reasons for requesting remand. Further, if only to guard against the "bad faith" requests of concern to the court in SKF, the court must be apprised of the reason for the remand request, whether it be on account of error or merely a change in policy. Commerce's brief does not provide any reason, policy or otherwise, for requesting a remand, although the domestic party makes clear that it believes error occurred. Commerce merely requests remand so that it can "reconsider its decision." This is insufficient to support a voluntary remand.

Commerce's failure, however, does not foreclose completely the petitioners' appeal. The petitioners may demonstrate error. The sole issue raised by the petitioners is whether Commerce properly classified Corus' tolled sales as export price ("EP") transactions in the Final Detemination. Some of Corus' U.S. sales involved tolling operations performed in the United States. The petitioners argue that these sales were actually made in the United States by CSUSA, (CSBV's affiliated importer) and, therefore, should have been classified as constructed export price ("CEP") transactions. The petitioners' challenge stems from a reversal by Commerce on this issue in the course of its investigation.

In the Preliminary Determination, Commerce classified all of Corus' U.S. sales, including tolled sales, as CEP transactions, finding that sales were made in the United States. Preliminary Determination, 66 Fed. Reg. at 22,148. Although Corus did not further challenge this finding, Commerce later reclassified tolled U.S. sales as export price ("EP") transactions following verification. See Final Determination, 66 Fed. Reg. at 55,637-39. "Concerning the tolled merchandise, we verified that the sales in question were between Corus Staal and the U.S. Customer. The record does not show that Corus USA was involved in these transactions." Issues and Decision Memorandum at 15-16, Comment 10.

In AK Steel Corp. v. United States, 226 F.3d 1361 (Fed. Cir. 2000), the court held that, in determining whether sales are EP or CEP transactions, the relevant inquiry is whether the merchandise was sold in the United States. Id. at 1371. In AK Steel, the court found that, under the antidumping statute, merchandise is "sold" when there is "both a transfer of ownership and consideration" and that the "seller" is the "party who transfers property in the contract of sale." See id., 226 F.3d at 1371.

In conducting verification, Commerce reviewed one detailed sales trace of a tolled U.S. sale. See Home Market Verification of the Corus Group's Questionnaire Response in the Antidumping Investigation of Certain Hot-Rolled Carbon Steel Flat Products from the Netherlands, Exhibit 14 ("Sales Verification Report"). In this transaction, the merchandise was shipped from CSBV to an unaffiliated toller in the United States who further processed the material as specified by an unaffiliated U.S. customer. See Sales Verification Report at 7. Corus maintains that the terms of this sale were agreed upon prior to the shipment of the merchandise. In other words, Corus argues that the sale was made by the producer "outside of the United States." There is evidentiary support for this conclusion. CSBV made the requisite shipping arrangements. See Sales Verification Exhibit 14 at C5013-C5014. CSBV incurred the cost of the additional processing. See Sales Verification Exhibit 16. Title transferred directly from CSBV to the U.S. Customer. See Corus' March 16, 2001 Supplemental Section A Response at A-11 ("Supp. Section A Response"). Unlike AK Steel, this case did not involve what are commonly referred to as "back-to-back" sales in which the producer sells the merchandise to its U.S. affiliate, who in turn, sells the merchandise to the unaffiliated U.S. customer. Back-to-back sales are necessarily CEP sales. The direct nature of this transaction supports the conclusion that the parties agreed in advance that the subject merchandise would be further processed, at CSBV's expense, once it arrived in the United States. As such, Commerce could have reasonably concluded that there was a transfer of ownership in the Netherlands.

Corus admits that CSUSA accepted payment for CSBV in the verified transaction (Corus argues that this is the only such instance). The parties, however, do not dispute that CSUSA immediately transferred the full amount to CSBV. See Sales Verification Exhibit 14 at C5034. The petitioners argue that CSUSA's acceptance of payment is the equivalent of consideration, but petitioners have not established that CSUSA benefitted from the transaction in any meaningful way. Based upon this evidence, Commerce could have concluded that CSBV, not CSUSA, received consideration for the sale.

The petitioners argue that, because CSUSA participated in tolled transactions, the sales were necessarily made in the United States. Corus concedes that CSUSA was involved in various aspects of some tolled sales and reported this involvement to Commerce. See. e.g., Corus' Section A Response at A-18 ("CSUSA performs administrative and some limited selling functions on behalf of [CSBV]."). That is presumably why Commerce initially concluded that these tolled sales were "made in the United States by CSUSA on behalf of CSBV" in the Preliminary Determination.

As AK Steel made clear, the focus of the inquiry is on the location of the sale rather than role played by the affiliated importer. See id. at 1369-76 (determining that the traditional test used by Commerce, as articulated in PQ Corp. v. United States, 11 CIT 53, 62-65, 652 F. Supp. 724, 733-35 (CIT 1987), incorrectly focused on the role played by the affiliated importer). The petitioners acknowledge this but, nevertheless, exert substantial energy arguing that the role of CSUSA in these transactions is conclusive. For example, petitioners point out that, on occasion, CSUSA (1) sat in on sales negotiations; (2) acted as importer of record; (3) received payment (from at least one customer and immediately transferred the full amount to CSBV); (4) maintained contact with CSBV's U.S. customers; and (5) performed various other selling functions. See Corus' Supp. Section A. Response at A-11. The court agrees that there is evidence suggesting the CSUSA's role in the transaction was more substantial than Corus admits.

The petitioners argue that Commerce failed to consider this contrary evidence. When asked at oral argument whether Commerce considered CSUSA's role and found nothing in verification, the petitioners replied in the affirmative, stating that Commerce indicated that it found "nothing" in the record. That is not accurate. Commerce did not state that it found "nothing," but instead stated that, upon verification, it concluded that "[t]he record does not show that Corus USA was involved in these transactions." When viewed in the context of the preceding sentence — "we verified that the sales in question were between Corus Staal and the U.S. Customer" — it is clear that. Commerce is explaining that, upon verification, it determined this transaction, the sales contract, was exclusively between CSBV and the customer. During verification, CSBV personnel testified that CSBV was the sole entity with authority to negotiate the terms of sale. See Sales Verification Report at 3-4; see also Supp. Section A Response a A-11. As discussed, the nature of the transaction suggests that CSBV contracted directly with the U.S. customer for the sale of merchandise. Commerce's conclusion is supported by this evidence. It is up to Commerce, not the court, to weigh the impact of' CSUSA's involvement against conflicting evidence that was properly submitted during verification.

Contrary to the petitioners' argument, Commerce may change its position even though Corus did not continue to argue that tolled sales should be considered EP sales after the Preliminary Determination concluded otherwise. That Commerce reversed its position in the absence of such an argument is not, as the petitioners imply, indicative of an arbitrary and uninformed mistake but, rather, suggests that, upon verification, Commerce independently uncovered evidence that contradicted its initial conclusion. The court finds this change compelling.

The petitioners lengthy argument that there is no difference in the record between the nature of tolled sales and non-tolled sales is unavailing. As is frequently pointed out by Commerce and the domestic producers in other antidumping cases, "[e]ven if it is possible to draw two inconsistent conclusions from the evidence contained in the record, this does mean that Commerce's findings are not supported by substantial evidence." Consolo v. Federal Maritime Comm'n, 383 U.S. 607, 620, 86 S.Ct. 1018, 16 L.Ed.2d 131 (1966). Commerce clearly stumbled upon something that caused it to reconsider and reverse its position. It is ultimately Commerce's responsibility to weigh the evidence and the court need only determine whether a determination has proper support. Because there is substantial evidence supporting a direct and exclusive connection between CSBV and the U.S. customer, the court finds that petitioner has not established error with respect to tolled sales. In the absence of a supported request for a remand by the government, no remand is warranted.

II. Corus' Motion

A. Zeroing

In its motion, Corus primarily challenges Commerce's methodology for calculating its weighted-average dumping margin. Commerce calculated the weighted-average dumping margin by aggregating all individual dumping margins. See Issues and Decision Memorandum for the Antidumping Investigation of Certain Hot-Rolled Carbon Steel Flat Products from the Netherlands; Notice of Final Determination of Sales at Less Than Fair Value (A-421-807), from Joseph A. Spetrini to Faryar Shirzad, 66 Fed. Reg. 50,408 (Dep't Commerce October 3, 2001) (" Issues and Decision Memorandum"). Commerce concedes that it discounted those transactions where the U.S. price exceeded normal value ("NV") — i.e., nondumped sales — by assigning a dumping margin of zero. Issues and Decision Memorandum at Comment 1. Corus argues that this methodology, commonly referred to as "zeroing," is an unreasonable interpretation of the statute because it results in a fundamentally unfair comparison that intentionally ignores certain transactions and distorts the final margin.

In both the Final Determination and briefing before the court, Commerce responds that the statute requires zeroing when calculating the dumping margin. See Issues and Decision Memorandum, Comment 1, ¶ 5 ("This methodology is required by U.S. law."). The petitioners agree, arguing that Commerce correctly recognized that zeroing is mandated after amendments to the statute under the to the Uruguay Round Agreements Act ("URAA"), Pub.L. No. 103-465, 108 Stat. 4809 (1994) (effective January 1, 1995). Prior to the URAA, the Tariff Act of 1930, as amended (the "Act"), did not expressly explain how to calculate a dumping margin or weighted-average dumping margin. Under the URAA, the Act was amended in 1994 to define a dumping margin as "the amount by which the normal value exceeds the export price or constructed export price of the subject merchandise." 19 U.S.C. § 1677(35)(A) (emphasis added). Section 1677(35)(B) was also added to define a weighted average dumping margin 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." Id. (emphasis added). Commerce argues that, taken together, these provisions expressly require that it aggregate only those dumping margins where NV exceeds U.S. price (commonly referred to as "positive" dumping margins).

Even if true, the court fails to see how such a directive translates to mandatory zeroing. If, as Commerce suggests, the statute expressly limited Commerce's inquiry to positive margins, there would be no need to zero negative margins at all — nondumped sales could be disregarded entirely. The truth of the matter is that the statute is silent as to the impact of negative margins. The statute neither requires nor prohibits Commerce from considering nondumped sales. The statute certainly does not direct Commerce to manipulate the value of certain transactions. That Commerce has adopted a zeroing methodology while simultaneously arguing that it need not consider negative margins supports the court's conclusion that the statute is ambiguous on this point. The issue, therefore, is whether Commerce's methodology is a reasonable interpretation of the statute. Chevron U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 842-43, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984) (requiring reviewing courts to defer to an agency's reasonable interpretation of an ambiguous statute).

Corus argues that Commerce's zeroing methodology is unreasonable because it violates the WTO Antidumping Agreement. Corus contends that, under the Charming Betsy doctrine, a methodology that violates the Agreement is necessarily an unreasonable interpretation of the underlying statute. Murray v. Charming Betsy, 2 Cranch 64, 6 U.S. 64, 118, 2 L.Ed. 208 (1804) ("An act of Congress ought never to be construed to violate the law of nations if any other possible construction remains. . . ."). Having found that the statute does not clearly require zeroing, the court now turns to whether the WTO Antidumping Agreement clearly prohibits zeroing.

Article 2.4.2 of the Agreement provides that dumping margins shall be established based upon "a comparison of a weighted average normal value with a weighted average of prices of all comparable export transactions. . . ." Agreement on Implementation of Article VI of GATT art. 2.4.2 (Apr. 15, 1994) (emphasis added) ("Antidumping Agreement" or the "Agreement"), reprinted in H.R. Doc. No. 103-316, Vol. 1, at 1455 (1994). Corus argues that zeroing violates Article 2.4.2 because it disregards certain transactions. As explained in Bowe Passat v. United States, nondumped transactions are not entirely ignored but are instead assigned a lesser value (zero) and ultimately combined with the value of the dumped sales for the purpose of calculating the overall margin. 20 CIT 558, 570-71, 926 F. Supp. 1138, 1150 (1996). "This inclusion of value has the effect of diluting the dumping margin, making the margin lower than it would otherwise be if . . . these sales were totally disregarded and were not given any credit at all in the analysis." Id. at 571-72, 926 F. Supp. at 1150. The court therefore cannot find that Commerce completely disregards those transactions.

The court is nevertheless wary of a methodology that intentionally minimizes the impact of nondumped transactions by manipulating the data of potentially equalizing sales. While the court questions whether this manipulation is in the spirit of the Agreement, the Agreement itself does not appear to expressly prohibit this practice. The Agreement merely requires that all sales be considered. As such, zeroing technically appears to comply. In fact, the court suspects that Commerce likely maintains its zeroing methodology, despite its protestations that the statute prohibits consideration of negative margins, to insure that it minimally complies with the agreement by "considering" all sales. Regardless, the court finds that the Agreement, in its present form, does not clearly prohibit zeroing.

Corus urges the court to consider a recent WTO Appellate Body finding that zeroing is inconsistent with its interpretation of the Agreement. In European Communities — Antidumping Duties on Imports of Cotton-Type Bed Linen from India, WT/DS141/AB/R (Mar. 1, 2001) (" Bed Linen"), the WTO Appellate Body ruled that a zeroing methodology employed by the E.C. was inconsistent with Article 2.4.2 because it did not take "fully into account" all transactions and, therefore, the E.C. had not made a "fair comparison." Corus argues that Commerce's methodology is similarly inconsistent with the Agreement and the court should, therefore, find the methodology unreasonable under the Charming Betsy doctrine.

In Timken, the plaintiff challenged zeroing on similar grounds, although in the context of an administrative review. 240 F. Supp.2d 1228 (CIT 2002). Although the court concluded that "[ Bed Linen] does not invalidate Commerce's zeroing practice," the court noted that it could not determine whether the methodology used by the E.C. in Bed Linen was sufficiently comparable to that of Commerce. Timken, 240 F. Supp.2d at 1239. The court pointed out that Bed Linen involved an antidumping investigation comparing weighted averages for export prices and normal value, implicating Article 2.4.2 of the Agreement, whereas Timken involved an administrative review comparing weighted-average normal values to transaction-specific export prices, implicating Article 9.3.1. Reasoning that the two circumstances were significantly different, the court seems to have determined that Bed Linen was inapplicable. This case is more similar to Bed Linen.

Like Bed Linen, this case is an appeal from an antidumping investigation, not an administrative review. Like Bed Linen, Commerce included the use of product groupings within the broader category of subject merchandise. In its margin program, Commerce calculated Corus' weighted-average dumping margin by first comparing the weighted average U.S. price with the weighted-average NV for each product sold in the United States. Dumping margins were calculated for each product group and were determined based on weighted-averages of price comparisons from transactions of products within the group. See Final Determination at 5-6. As discussed, if the product group's margin was negative, the margin was changed to zero. The circumstances and methodology in this case appear to be very similar, if not identical, to the Bed Linen investigation. See Bed Linen, WT/DS141/AB/R ¶ 47.

Despite these similarities, the court concludes that Bed Linen cannot be the basis for striking Commerce's methodology. As this court has frequently recognized, WTO decisions are not binding upon Commerce or the court. See, e.g., Hyundai Elec. Co., Ltd. v. United States, 23 CIT 302, 311, 53 F. Supp.2d 1334, 1343 (1999); Footwear Distributors and Retailers of America v. United States, 18 CIT 391, 852 F. Supp. 1078 (1994). The SAA provides that "[r]eports issued by panels or the Appellate Body under the [WTO Dispute Settlement Understanding] have no binding effect under the law of the United States." SAA at 1032. The SAA goes on to state that "panel reports do not provide legal authority for federal agencies to change their regulations or procedures. . . ." Id.

In addition, it appears that WTO decisions are not binding upon the WTO itself. The common law concept of stare decisis does not expressly apply to WTO rulings. Understanding on Rules and Procedures Governing the Settlement of Disputes, Apr. 15, 1994, Final Act Embodying the Results of the Uruguay Round of Multilateral Trade Negotiations, Legal Instruments — Results of the Uruguay Round, Annex 2, vol. 31, 33 I.L.M. 1226 (1994), art. 3(2). As a result, WTO decisions appear to have very limited precedential value and are binding only upon the particular countries involved. They are not binding upon other signatory countries or future WTO panels.

While this court has acknowledged that WTO decisions may help inform its decisions, Hyundai, 23 CIT at 312, 53 F. Supp.2d at 1343, the court concludes that it cannot solely rely upon a non-binding interpretation of an international agreement as grounds to strike a United States agency interpretation of a statute. When faced with an ambiguous statute and ambiguous international agreement, the court should defer to Commerce's interpretation. See Federal-Mogul Corp. v. United States, 63 F.3d 1572, 1582 (Fed. Cir. 1995) ("Commerce is due judicial deference in part because of its established expertise in administration of the Act, and in part because of `the foreign policy repercussions of a dumping determination.'") (quoting Smith-Corona Group, Consumer Products Div., SCM Corp. v. United States, 713 F.2d 1568, 1571, 1 Fed. Cir. 130, 131 (1983), cert. denied, 465 U.S. 1022, 104 S.Ct. 1274, 79 L.Ed.2d 679 (1984)). After all, zeroing is not new. Congress was presumably aware of the practice when it enacted the URAA. See, e.g., supra note 15 and cases cited therein. Congress could have prohibited zeroing if it so chose. Instead, Congress enacted a statute that, at least, arguably encourages zeroing by referring only to dumping margins where the U.S. price exceeds NV. As a result, the court cannot find that zeroing is an unreasonable application of the statute as it is presently written.

B. Affiliated Sales

Corus next argues that Commerce erroneously found that Laura Metaal Holding BV ("Laura"), a steel service center located in the Netherlands, was affiliated with CSBV. Section 771(33)(F) of the Tariff Act of 1930, as amended (the "Act"), defines "affiliated" as "[t]wo or more persons directly or indirectly controlling, controlled by, or under common control with, any person." 19 U.S.C. § 1677(33)(F). The statute provides that "a person shall be considered to control another person if the person is legally or operationally in a position to exercise restraint or direction over the other person." 19 U.S.C. § 1677(33). "The determination of `control' under the URAA is . . . not dependent on actually exercising control, but rather on the capacity to exercise control." Ferro Union v. United States, 44 F. Supp.2d 1310, 1324 (CIT 1999) (emphasis in original).

Commerce generally finds control to exist where the relationship "has the potential to impact decisions concerning the production, pricing, or cost of the subject merchandise or foreign like product." 19 C.F.R. § 351.102(b). Commerce considers factors such as "corporate or family groupings; franchise or joint venture agreements; debt financing; and close supplier relationships." Id. After reviewing the various ownership interests and structure of Laura's Supervisory Board, Commerce determined that Corus and Laura were affiliated through the common control of Koninklijke Hoogovens NV ("KHNV") and, ultimately Corus Group PLC ("Corus Group"). Issues and Decision Memorandum at Comment 2.

Corus Group is the owner and parent of numerous steel production related companies. See Supp. Section A Response at Ex. A-4. Corus Group was established in October 1999 as a result of a merger between KHNV and British Steel plc. See id. at A-7, n. 4. KHNV is now a holding company for Corus Group and does not engage in the production of steel itself. Id. at A-7. Respondents CSBV and Corus Steel USA Inc. ("CSUSA") are wholly owned subsidiaries of KHNV. Id. at A-8. As discussed, CSBV is the sole producer of subject hot-rolled steel in the Netherlands and is responsible for home market and export operations. Id. at A-7, A-10 to A-11. CSUSA acts as an agent for CSBV in the United States. Id. at A-8 to A-9. Laura and Namascor BV ("Namascor") are two home market service centers related to Corus Group. Id. at A-3 to A-4. Laura is both a consumer of subject merchandise for use in its manufacturing operations and a reseller of subject merchandise through its service center functions. Id. at A3 to A-4. Laura is owned by three companies including KHNV, the parent corporation of CSBV. After reviewing the equity ownership and structure of Laura, Commerce determined that CSBV and Laura were affiliated, concluding that both were under the common control of KHNV and, thus, the Corus Group. Decision Memorandum at Comment 2. Corus does not dispute that CSBV is under the control of the Corus Group by virtue of KHNV's 100 percent ownership of CSBV. See Corus Br. at 26. Corus challenges Commerce's conclusion with respect to Laura, arguing that KHNV does not have a sufficiently significant ownership interest in Laura to exert control.

Corus concedes that a 1992 Shareholder's Agreement revised share ownership for the express purpose of eliminating KHNV's prior majority interest in Laura. See Section A Response at Exhibit A-4 ("Shareholder's Agreement"). As a result, KHNV is now a minority owner, although it continues to own substantial percentage of the shares of Laura. Minority ownership, however, is not dispositive.

Congress amended the affiliation statute as part of the URAA to allow Commerce to consider factors other than equity ownership as indicia of affiliation. The SAA explains that the current law introduces the concept of "control" because "[t]he Administration believes that including control in the definition of `affiliated' will permit a more sophisticated analysis which better reflects the realities of the marketplace." SAA at 838. The SAA goes on to explain that:

The traditional focus on control through stock ownership fails to address adequately modern business arrangements, which often find one firm "operationally in a position to exercise restraint or direction" over another even in the absence of an equity relationship. A company may be in a position to exercise restraint or direction, for example, through corporate or family groupings, franchises or joint venture agreements, debt financing, or close supplier relationships in which the supplier or buyer becomes reliant upon the other.

SAA at 838 (emphasis added).

While not always conclusive, equity ownership remains a highly relevant consideration in determining whether parties are affiliated through common control under Section 771(33)(F) of the Act. See, e.g., Mitsubishi Heavy Industries, Ltd. v. United States, 54 F. Supp.2d 1183, 1192 (CIT 1999) (holding that a minority shareholder may be found to control a company if its ownership interest is significant and puts the shareholder "legally or operationally in a position" to exercise restraint or direction over the company). The court finds that KHNV's [ ] percent ownership interest is sufficient to suggest the capacity for control.

The court further finds that KHNV and the Corus Group also have the potential to exercise control over Laura's operations and management through KHNV's ability to appoint members of Laura's Supervisory Board (the "Board"). Under the Shareholder's Agreement, KHNV has the ability to nominate [ ] voting members on the Supervisory Board. See Verification Exhibits at Exhibit 2, Article 5.1, Corus' Appendix at Exhibit 11. KHNV also nominates one of the two non-voting advisors to the Board. See Section A Response at A-4, Corus' Appendix at Exhibit 3 (Public Version).

Under the Shareholder's Agreement, the Board has ultimate control over the operations and management of Laura. Through its ability to nominate [ ] percent of the voting members and 50 percent of the non-voting advisors to the Board, KHNV may have substantial power to direct the activities of Laura and certainly has the ability to restrain production, pricing, and other decisions by blocking particular votes. Accordingly, the court finds that there is substantial evidence to support Commerce's conclusion that KHNV's significant equity ownership in Laura and its representation on Laura's Board would allow it to exercise restraint or direction over Laura's operations and management. See Issues and Decision Memorandum at Comment 2. The court finds no error in Commerce's determination that Laura and CSBV are affiliated.

C. Non-Prime Sales

Corus next argues that Commerce erred by including certain sales by GalvPro LP ("GalvPro"), a joint venture company owned by Weirton Steel Corporation and Corus Coatings USA. Corus argues that sales of defective hot-rolled steel by GalvPro should have been excluded because those transactions were outside the ordinary course of trade and not representative of GalvPro's other galvanized steel products. Commerce responds that there is no requirement that it exclude sales outside the ordinary course of trade and that it properly compared these U.S. sales to similar sales in the home market. The court agrees.

GalvPro was formed to construct and operate a single-line manufacturing facility for the treatment of cold-rolled full hard steel to produce galvanized steel products. These products were not subject to the underlying investigation. During the POI, GalvPro purchased a limited amount of hot-rolled material from CSBV due to a temporary shortage of cold-rolled steel, the normal input for GalvPro's production. See Corus' January 18, 2001 Letter To Commerce at 2; Corus' February 1, 2001 Section A Response at A-10. After importation, the hot-rolled steel was cold-rolled and galvanized by GalvPro. During this process, some of the resulting galvanized steel was damaged and, consequently, was sold at substantially reduced prices. Corus sought to have these sales excluded from Commerce's margin program on the grounds that they were unrepresentative of Corus' U.S. sales and, therefore, outside of the ordinary course of trade.

"Commerce calculates an antidumping duty by comparing an imported product's price in the United States to its normal value . . . . (i.e., the price of comparable merchandise in the exporting country)." FTC, 41 F. Supp.2d at 323. As previously explained, the U.S. price may be either EP, which is used when the foreign exporter sells directly to an unrelated U.S. customer, or CEP, which is used when the foreign exporter sells through an affiliated party in the United States. Id.; 19 U.S.C. § 1677a(a) and (b). The NV is "the price at which the foreign like product is first sold . . . . for consumption in the exporting country in the usual commercial quantities and in the ordinary course of trade." 19 U.S.C. § 1677b(a)(1)(B)(i) (emphasis added). While the definition for NV requires that Commerce consider the price of sales in the "ordinary course of trade," the definitions of U.S. price contain no such requirement. 19 U.S.C. § 1677a(a) and (b); 19 C.F.R. § 351.402.

The court has previously recognized that the calculation of U.S. price may include sales outside the "ordinary course of trade." In IPSCO, Inc. v. United States, 687 F. Supp. 633 (CIT 1988), rev'd on other grounds, 965 F.2d 1056 (Fed. Cir. 1992), the court concluded:

[T]he court may assume that if Congress intended to require the administering authority to exclude all sales made outside the `ordinary course of trade' from its determination of United States price it could have provided for such an exclusion in the definition of United States price, as it has in the definition of foreign market value. It has not done so.

Where a respondent sells secondary or non-prime merchandise in the United States, it is Commerce's standard practice to include those transactions in its margin calculation by attempting to match these sales to non-prime sales in the home market. See, e.g., Notice of Final Determination of Sales at Less Than Fair Value: Stainless Steel Sheet and Strip in Coils from Italy, 64 Fed. Reg. 30,750, 30,753 (June 8, 1999) (matching non-prime U.S. sales to non-prime sales in the home market in the margin analysis); Final Determination of Sales at Less Than Fair Value: Certain Hot-Rolled Carbon Steel Flat Products, Certain Cold-Rolled Carbon Steel Flat Products, Certain Corrosion-Resistant Carbon Steel Flat Products, and Certain Cut-to-Length Carbon Steel Plate from Korea, 58 Fed. Reg. 37176, 37183 (Dep't of Commerce July 9, 1993)(same). Even where there is no match in the home market, Commerce can assign a constructed value for these sales. 19 U.S.C. § 1677b(a)(4).

This court has previously upheld this practice, stating that "Commerce's interpretation of the United States price provision of the antidumping statute to include all United States sales whether in or out of the ordinary course of trade is a permissible construction of the statute." Bowe Passat, 926 F. Supp. at 1148. Consistent with that position, Commerce attempted to match Galvpro's non-prime sales and included those transactions in the final margin calculation here. Issues and Decision Memorandum at 10 (Comment 3). There is no inherent error in doing so.

While Commerce generally attempts to match non-prime sales, Commerce may, in certain circumstances, disregard U.S. sales where those transactions are unrepresentative and extremely distortive. See, e.g., FAG U.K. Ltd. v. United States, 945 F. Supp. 260, 264, 265 (CIT 1996) (" FAG U.K."); FAG Italia S.p.A. v. United States, 948 F. Supp. 67, 71 (CIT 1996). In FAG U.K., the Court recognized that the statute "does not require that absolutely every U.S. sale of merchandise under investigation be included in every case." FAG U.K., 945 F. Supp. at 265. Citing Ipsco v. United States, 714 F. Supp. 1211, 1217 (1989), rev'd in part, 965 F.2d 1056 (Fed. Cir. 1992). In FAG U.K., the court determined that transactions may be excluded "`in those situations in which [Commerce] finds that inclusion of certain sales which are clearly atypical would undermine the fairness of the comparison of foreign and U.S. sales.'" 945 F. Supp. at 265 (citation omitted). In these cases, however, the court was discussing the limitations upon Commerce's discretion to exclude sales. The court was not creating a requirement that Commerce exclude certain sales. In sum, the court finds that Commerce did not err in comparing GalvPro sales to non-prime sales in the home market.

D. Level of Trade Adjustment

Corus argues that Commerce erred in denying Corus a level of trade ("LOT") adjustment with respect to sales in the home market by two steel service centers. In calculating the margin, the statute provides for an adjustment to NV if the U.S. sale is made at a different LOT from the comparison market sale, and certain factors are satisfied. 19 U.S.C. § 1677b(a)(7)(A); see also Prodotti Alimentari Meridionali, S.R.L. v. United States, slip op. 02-68, 2002 WL 1585672, *3 (CIT July 16, 2002). The regulation provides that "[t]he Secretary will determine that sales are made at different levels of trade if they are made at different marketing stages (or their equivalent)." 19 C.F.R. § 351.412(c)(2) (2002). The SAA provides that, in order to find that two LOTs are different, one requisite factor is "a difference between the actual functions performed by the sellers at the different levels of trade in the two markets." SAA at 829. To qualify for a LOT adjustment, therefore, "a party has the burden to show that the following two conditions have been satisfied: (1) the difference in LOT involves the performance of different selling activities; and (2) the difference affects price comparability." Timken, 240 F. Supp.2d at 1233; see also Torrington Co. v. United States, 146 F. Supp.2d 845, 876 (2001); Koyo Seiko Co., Ltd. v. United States, 8 F. Supp.2d 862, 864 (CIT 1998). The parties here dispute the former.

In conducting its investigation, Commerce requested that Corus report and explain the different LOTs according to the selling function performed and sales services offered to its home market customers. See Commerce's Antidumping Duty Questionnaire (Jan. 4, 2001) at A-6 to A-7. In response, Corus described the selling activities that it and its affiliated service center, Namascor BV, perform for their home market sales. Corus' Section A Response at A-19 to A-33, Exhibit A-8 P.R. 45.

Because Corus initially claimed that Laura was unaffiliated, Corus did not submit complete data for Laura. In a supplemental response issued after Commerce concluded that Laura was an affiliate, Corus stated that Laura's "operations and customer base is such that the activities it performs on behalf of its customers is clearly more similar to that of Namascor than that of Corus." Corus' Response to the Department's Supplemental Sections A-E Questionnaire (Apr. 4, 2001) ("Suppl.A-E") at B-11 Corus' Appendix at Exhibit 6 (Public Version). Because Corus did not present an explanatory narrative with respect to Laura and because the burden is upon Corus to establish a different LOT, the court considers only whether Namascor should have been granted a LOT adjustment.

With respect to Namascor, Corus reported eleven (11) total selling functions in the home market. NPD 11 at A-19 to A-33, Ex. at A-8. After evaluating Corus' description of strategic and economic planning, advertising, freight and delivery arrangements, technical/warranty services, and sales logistics support, Commerce determined that both Namascor and CSBV performed these five (5) selling activities in a similar manner. Issues and Decision Memorandum at 11. Corus itself reported that neither performed three (3) of the remaining selling functions for its home market customers. With respect to the final (3) three selling functions (market research, research and development, and warehousing), Commerce noted some differences between the functions of CSBV and Namascor, but ultimately concluded that these differences were not significant and did not represent sales at separate LOTs. Id. at 12. As a result, Commerce determined that the selling activities of CSBV and Namascor did not differ significantly and, therefore, that both had made sales in the home market at the same LOT. Issues and Decision Memorandum at 11-12.

As it did during the investigation, Corus argues that steel service centers inherently operate at a different LOTs. In Hot-Rolled Steel Products From Argentina, China, India, Indonesia, Kazakhstan, Netherlands, Romania, South Africa, Taiwan, Thailand, and Ukraine, Inv. Nos. 701-TA-404-408, 731-TA-898-908, USITC Pub. 3381 at I-8, and in n. 24 (Jan.2001) (prelim.det.), the International Trade Commission ("ITC") agreed, finding that steel service centers perform two key functions different from manufacturers: (1) "Many service centers maintain extensive inventories of a variety of steel products, providing availability and inventory management services for customers of all sizes, including those with smaller purchasing needs that must place low-volume orders"; and (2) "[s]ome service centers perform a wide range of value-added processing, such as uncoiling, flattening, and cutting flat-rolled products to length or burning hundreds of intricate parts from a single sheet." Id. at I-8 n. 24. Corus argues that, like most service centers, Namascor offers warehousing services and markets a variety of value-adding products and services, such as cutting, decoiling, cutting-to-length, etc., which are significantly different from those available from CSBV. See Corus' February 1, 2001 Section A Response at A-3-4 Ex. A-14; Corus' April 4, 2001 Laura Section B Response at B-1.

Unlike the ITC, Commerce has not explained whether its generally finds that producers and service centers operate at the same LOT — but there is no requirement that it do so. Rather than broadly differentiating between types of sellers, Commerce compares the observations of different selling functions, which is consistent with statute, regulation, and SAA. The court finds no error in applying this quantitative analysis generally. While the court has expressed concerns with a purely quantitative comparison that does not account for qualitative differences in selling functions, Prodotti Alimentari Meridionali, S.R.L. v. U.S, 2002 WL 1585672, *2, Slip op. 02-68 (CIT 2002), Corus does not challenge the Final Determination on this ground. Corus instead attacks Commerce's findings with respect to individual functions. For example, Corus argues that Commerce erred in concluding that CSBV and Namascor perform similar "Technical / Warranty Services." Pls. Br. at 37. Review of the record, however, indicates that, while the services may vary slightly in regard to the size of the technical service department or type of complaints addressed, both CSBV and Namascor offered comparable services. Corus also challenges Commerce's findings with respect to "Sales Logistics Support" and "Warehousing," although Corus does not clearly articulate how Commerce erred in its analysis other than to repeatedly proclaim that Commerce ignored inherent differences. It is worth repeating that Corus bears the burden of establishing that it is entitled to a LOT adjustment. 19 C.F.R. § 351.401(b)(1) (2002); Timken Co. v. United States, et al., 209 F. Supp.2d 1373, 1381-82 (CIT 2002). The court finds that Corus has not met its burden here.

E. Extension of Provisional Measures.

Corus last argues that collection of duties based upon provisional measures after October 30, 2001 was unlawful. Commerce issued its preliminary determination in this matter on May 3, 2001. 66 Fed. Reg. 22,146 (May 3, 2001). Corus subsequently requested an extension of the final determination pursuant to 19 C.F.R. § 353.210(b). In its request, Corus agreed to an extension of provisional measures from a four-month period to not more than six months. See Corus' May 22, 2001 Letter to Commerce. This agreement was required by regulation, 19 C.F.R. § 351.210(e)(2), and reflects the limitation on provisional measures set forth in 19 U.S.C. § 1673b(d) that prohibits the assessment of antidumping duties during the gap period after the expiration of the six month period until issuance of the order. See F.Li De Cecco Di Filippo Fara San Martino S.P.A. v. United States, 21 C.I.T. 1130, 1137-40, 1997 WL 615693 (1997) ("The strong wording in . . . 19 U.S.C. § 1673b(d) and the corresponding sections of the Antidumping Agreement suggests that the GATT signatories meant to put a strict limit on the imposition of provisional measures, particularly because of the harshness of the penalties involved, and that Congress has now dealt specifically with the GATT requirement.").

Commerce granted postponement and stated that it would issue its final determination by September 15, 2001. Postponement of Final Determination for Antidumping Duty Investigation: Certain Hot-Rolled Carbon Steel Flat Products from the Netherlands, 66 Fed. Reg. 32,600 (Dep't Commerce June 15, 2001). Because of the events of September 11, the antidumping order was not issued until November 29, 2001. See Antidumping Duty Order: Certain Hot-rolled Carbon Steel Flat Products From the Netherlands, 66 Fed. Reg. 59565 (November 29, 2001).

Corus argues, and Commerce agrees, that the provisional measures should not have been collected more than six months after the preliminary injunction was issued on May 3, 2001. Commerce admits that it inadvertently excluded appropriate language from the antidumping order that would lift the suspension of liquidation six months after the Preliminary Determination. Commerce requests remand on this issue in order for the agency to insert the appropriate language to correct what it claims is a ministerial error. While the parties agree that remand is appropriate, the parties disagree as to the final date for collection of the provisional measures.

Corus argues that, because Commerce has previously interpreted six months to consist of 180 days, provisional measures should not have been collected after October 30, 2001 (180 days after May 3, 2001). Commerce argues that six months is six months and that the provisional measures were properly collected through November 3 (184 days). Corus argues that there is no standard practice and, when asked at oral argument, Commerce was unable to articulate one. Upon remand, Commerce must explain its practice in such matters and revise the order consistent with that practice.

CONCLUSION

For the foregoing reasons, the court finds that the government's vague request for a remand is insufficient and that Commerce's determination with respect to Corus' tolled sales is otherwise supported by substantial evidence. The court further finds that zeroing is neither required by the URAA nor expressly prohibited by the WTO Antidumping Agreement and is a reasonable interpretation of the statute as it is presently written. The court finds that Commerce's finding that Laura and CSBV are affiliated is supported by substantial evidence and that Corus has not established that a LOT adjustment is warranted for its affiliated steel service centers. The court remands this matter to Commerce for the sole purpose of revising its antidumping order to preclude collection of provisional measures beyond the six month period.

So ordered.