Koyo Seiko Co., Ltd. v. United States , 26 Ct. Int'l Trade 170 ( 2002 )

186 F. Supp. 2d 1332 (2002)

KOYO SEIKO CO., LTD. and Koyo Corporation of USA; NSK Ltd. and NSK Corporation; NTN Bearing Corporation of America, American NTN Bearing Manufacturing Corporation and NTN Corporation; and the Timken Company, Plaintiffs and Defendant-Intervenors,
v.
UNITED STATES, Defendant.

Slip Op. 02-11, Court No. 98-06-02274.

United States Court of International Trade.

February 1, 2002.

Powell, Goldstein, Frazer & Murphy LLP (Peter O. Suchman, Neil R. Ellis and Elizabeth C. Hafner), Washington, DC, for Koyo, plaintiff and defendant-intervenor.

Lipstein, Jaffe & Lawson, L.L.P. (Robert A. Lipstein, Matthew P. Jaffe and Grace W. Lawson), Washington, DC, for NSK, plaintiff and defendant-intervenor.

Barnes, Richardson & Colburn (Donald J. Unger, Kazumune V. Kano, David G. Forgue and Clarice K.M. McCauley), Chicago, IL, for NTN, plaintiff and defendant-intervenor.

Stewart and Stewart (Terence P. Stewart, William A. Fennell and Patrick J. McDonough), Washington, DC, for Timken, plaintiff and defendant-intervenor.

Robert D. McCallum, Jr., Assistant Attorney General; David M. Cohen, Director, Commercial Litigation Branch, Civil Division, United States Department of Justice (Lucius B. Lau); Joan L. Mackenzie and Barbara Campbell Potter, Office of the Chief Counsel for Import Administration, United States Department of Commerce, Washington, DC, for defendant, of counsel.

OPINION

TSOUCALAS, Senior Judge.

This consolidated action concerns the claims raised by: Koyo Seiko Co., Ltd. and Koyo Corporation of USA (collectively "Koyo"); NSK Ltd. and NSK Corporation (collectively "NSK"); NTN Bearing Corporation of America, American NTN Bearing Manufacturing Corporation and NTN Corporation (collectively "NTN"); and The Timken Company ("Timken"), plaintiffs and defendant-intervenors, that move pursuant to USCIT R. 56.2 for judgment upon the agency record challenging various aspects of the United States Department of Commerce, International Trade Administration's ("Commerce") final results of administrative review of: (1) tapered roller bearings and parts thereof, finished and unfinished, from Japan; and (2) tapered roller bearings, four inches or less in outside diameter, and components thereof, from Japan, entitled Final Results of Antidumping Duty Administrative Reviews and Termination in Part of Tapered Roller Bearings and Parts Thereof, Finished and Unfinished, From Japan, and Tapered Roller Bearings, Four Inches or Less in Outside Diameter, and Components Thereof, From Japan ("Final Results"), 63 Fed.Reg. 20,585 (April 27, 1998).

Specifically, Koyo argues that Commerce: (1) failed to calculate constructed value profit so that home market movement expenses are excluded from the gross unit price; (2) erred in Commerce's decision to use the entered value of the subject merchandise to determine assessment rates; (3) erred in Commerce's calculation of marine insurance charges; (4) erred in Commerce's calculation of certain constructed value commissions and direct selling expenses; and (5) wrongly used Koyo's product nomenclature in Commerce's computer program. NSK asserts that: (1) Commerce's model matching program is not supported by substantial evidence; and (2) Commerce erred in determining NSK's general and administrative expenses factor in the cost of production calculation. NTN alleges that Commerce erred in: (1) determining that NTN's sample and small-quantity home market sales are within the ordinary course of trade; (2) denying an adjustment to NTN's United States indirect selling expenses for expenses purportedly related to the financing of antidumping duty cash deposits; (3) disallowing an adjustment to foreign market value for NTN's home market discounts; and (4) reallocating NTN's selling expenses without regard to level of trade and denying a level of trade adjustment. Timken contends that Commerce: (1) erred by failing to adjust Koyo's further-manufactured import prices to reflect inventory carrying costs associated with further-manufacturing in the United States; and (2) committed a clerical error in calculating NTN's indirect selling expenses for United States constructed export price sales.

BACKGROUND

The administrative review at issue arose from two antidumping orders: Antidumping Duty Order on Tapered Roller Bearings, Four Inches or Less in Outside Diameter, and Components Thereof, From Japan, 41 Fed.Reg. 34,974 (Aug. 18, 1976),^[1] and Antidumping Duty Order on Tapered Roller Bearings and Parts Thereof, Finished and Unfinished, From Japan, 52 Fed.Reg. 37,352 (Oct. 6, 1987). The reviews for the period 1992-93 were initiated on November 17, 1993. See Initiation of Antidumping and Countervailing Duty Administrative Reviews, 58 Fed. Reg. 60,600 (Nov. 17, 1993). The reviews for the period 1993-94 were initiated on November 14, 1994. See Initiation of Antidumping and Countervailing Duty Administrative Reviews, 59 Fed.Reg. 56,459 (Nov. 14, 1994).^[2] The preliminary results of the reviews were published on May 20, 1996. See Preliminary Results of Antidumping Duty Administrative Reviews and Termination in Part of Tapered Roller Bearings and Parts Thereof, Finished and Unfinished, From Japan, and Tapered Roller Bearings, Four Inches or Less in Outside Diameter, and Components Thereof, From Japan ("Preliminary Results"), 61 Fed.Reg. 25,200. The final results of the reviews were published on April 27, 1998. See Final Results, 63 Fed. Reg. 20,585.

JURISDICTION

The Court has jurisdiction over this matter pursuant to 19 U.S.C. § 1516a(a) (1994) and 28 U.S.C. § 1581(c) (1994).

STANDARD OF REVIEW

In reviewing a challenge to Commerce's final determination in an antidumping administrative review, the Court will uphold Commerce's determination 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) (1994).

I. SUBSTANTIAL EVIDENCE TEST

Substantial evidence is "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, 71 S. Ct. 456, 95 L. Ed. 456 (1951) (quoting Consolidated Edison Co. v. NLRB, 305 U.S. 197, 229, 59 S. Ct. 206, 83 L. Ed. 126 (1938)). Substantial evidence "is something less than the weight of the 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, 86 S. Ct. 1018, 16 L. Ed. 2d 131 (1966) (citations omitted). Moreover, "[t]he court may not substitute its judgment for that of the [agency] when the choice is ``between two fairly conflicting views, even though the court would justifiably have made a different choice had the matter been before it de novo.'" American Spring Wire Corp. v. United States, 8 CIT 20, 22, 590 F. Supp. 1273, 1276 (1984) (quoting Penntech Papers, Inc. v. NLRB, 706 F.2d 18, 22-23 (1st Cir.1983) (quoting, in turn, Universal Camera, 340 U.S. at 488, 71 S. Ct. 456)).

II. CHEVRON TWO-STEP ANALYSIS

To determine whether Commerce's interpretation and application of the antidumping statute is "in accordance with law," the Court must undertake the two-step analysis prescribed by Chevron U.S.A. Inc. v. Natural Resources Defense Council, Inc. ("Chevron"), 467 U.S. 837, 104 S. Ct. 2778, 81 L. Ed. 2d 694 (1984). Under the first step, the Court reviews Commerce's construction of a statutory provision to determine whether "Congress has directly spoken to the precise question at issue." Id. at 842, 104 S. Ct. 2778. "To ascertain whether Congress had an intention on the precise question at issue, [the Court] employ[s] the ``traditional tools of statutory construction.'" Timex V.I., Inc. v. United States, 157 F.3d 879, 882 (Fed. Cir.1998) (citing Chevron, 467 U.S. at 843 n. 9, 104 S. Ct. 2778). "The first and foremost ``tool' to be used is the statute's text, giving it its plain meaning. Because a statute's text is Congress's final expression of its intent, if the text answers the question, that is the end of the matter." Id. (citations omitted). Beyond the statute's text, the tools of statutory construction "include the statute's structure, canons of statutory construction, and legislative history." Id. (citations omitted); but see Floral Trade Council v. United States, 23 CIT ____, ____ n. 6, 41 F. Supp. 2d 319, 323 n. 6 (1999) (noting that "[n]ot all rules of statutory construction rise to the level of a canon, however") (citation omitted).

If, after employing the first prong of Chevron, the Court determines that the statute is silent or ambiguous with respect to the specific issue, the question for the Court becomes whether Commerce's construction of the statute is permissible. See Chevron, 467 U.S. at 843, 104 S. Ct. 2778. Essentially, this is an inquiry into the reasonableness of Commerce's interpretation. See Fujitsu Gen. Ltd. v. United States, 88 F.3d 1034, 1038 (Fed.Cir.1996). Provided Commerce has acted rationally, the Court may not substitute its judgment for the agency's. See Koyo Seiko Co. v. United States, 36 F.3d 1565, 1570 (Fed.Cir.1994) (holding that "a court must defer to an agency's reasonable interpretation of a statute even if the court might have preferred another"); see also IPSCO, Inc. v. United States, 965 F.2d 1056, 1061 (Fed. Cir.1992). The "[C]ourt will sustain the determination if it is reasonable and supported by the record as a whole, including whatever fairly detracts from the substantiality of the evidence." Negev Phosphates, Ltd. v. United States, 12 CIT 1074, 1077, 699 F. Supp. 938, 942 (1988) (citations omitted). In determining whether Commerce's interpretation is reasonable, the Court considers the following non-exclusive list of factors: the express terms of the provisions at issue, the objectives of those provisions and the objectives of the antidumping scheme as a whole. See Mitsubishi Heavy Indus. v. United States, 22 CIT 541, 545, 15 F. Supp. 2d 807, 813 (1998).

DISCUSSION

I. COMMERCE'S DECISION TO USE ENTERED VALUE OF THE SUBJECT MERCHANDISE TO DETERMINE ASSESSMENT RATES

As a preliminary matter, Commerce raises two affirmative defenses to the particular claim. Specifically, Commerce asserts that the plaintiff on the claim, Koyo: (1) has failed to exhaust plaintiff's administrative remedies, see Def.'s Resp. Partial Opp'n Mem. P. & A. Supp. Mot. Pls. Koyo J. Agency R. ("Def.'s Resp. Koyo") at 11; and (2) is precluded from litigating the issue by the Court's ruling in Koyo Seiko Co. v. United States, 16 CIT 539, 796 F. Supp. 1526 (1992) under the doctrine of collateral estoppel. See Def.'s Resp. Koyo at 11-12.

A. Exhaustion of Administrative Remedies

1. Background

The exhaustion doctrine requires a party to present its claims to the relevant administrative agency for the agency's consideration before raising these claims to the Court. See Unemployment Compensation Comm'n of Alaska v. Aragon, 329 U.S. 143, 155, 67 S. Ct. 245, 91 L. Ed. 136 (1946) ("A reviewing court usurps the agency's function when it sets aside the administrative determination upon a ground not theretofore presented and deprives the [agency] of an opportunity to consider the matter, make its ruling, and state the reasons for its action"). There is, however, no absolute requirement of exhaustion in the Court of International Trade in non-classification cases. See Alhambra Foundry Co. v. United States, 12 CIT 343, 346-47, 685 F. Supp. 1252, 1255-56 (1988). Section 2637(d) of Title 28 directs that "the Court of International Trade shall, where appropriate, require the exhaustion of administrative remedies." By its use of the phrase "where appropriate," Congress vested discretion in the Court to determine the circumstances under which it shall require the exhaustion of administrative remedies. See Cemex, S.A. v. United States, 133 F.3d 897, 905 (Fed.Cir.1998). Therefore, because of "judicial discretion in not requiring litigants to exhaust administrative remedies," the Court is authorized to determine proper exceptions to the doctrine of exhaustion. Alhambra Foundry, 12 CIT at 347, 685 F. Supp. at 1256 (citing Timken Co. v. United States, 10 CIT 86, 93, 630 F. Supp. 1327, 1334 (1986), rev'd in part on other grounds, Koyo Seiko Co. v. United States, 20 F.3d 1156 (Fed.Cir. 1994)).

In the past, the Court has exercised its discretion to obviate exhaustion where: (1) requiring it would be futile, see Rhone Poulenc, S.A. v. United States, 7 CIT 133, 135, 583 F. Supp. 607, 610 (1984) (in those cases when "it appears that it would have been futile for plaintiffs to argue that the agency should not apply its own regulation"), or would be "inequitable and an insistence of a useless formality" as in the case where "there is no relief which plaintiff may be granted at the administrative level," United States Cane Sugar Refiners' Ass'n v. Block, 3 CIT 196, 201, 544 F. Supp. 883, 887 (1982); (2) a subsequent court decision has interpreted existing law after the administrative determination at issue was published, and the new decision might have materially affected the agency's actions, see Timken, 10 CIT at 93, 630 F. Supp. at 1334; (3) the question is one of law and does not require further factual development and, therefore, the court does not invade the province of the agency by considering the question, see id.; R.R. Yardmasters of Am. v. Harris, 721 F.2d 1332, 1337-39 (D.C.Cir.1983); and (4) the plaintiff had no reason to suspect that the agency would refuse to adhere to clearly applicable precedent. See Philipp Bros., Inc. v. United States, 10 CIT 76, 79-80, 630 F. Supp. 1317, 1321 (1986).

2. Contentions of the Parties

Commerce asserts that Koyo failed to exhaust its administrative remedies. Pointing to the Preliminary Results, 61 Fed.Reg. at 25,205, that "set forth [Commerce's] assessment rate methodology" as well as Commerce's "briefing and hearing schedule," Commerce contends that parties to the review had to "raise[] [their] concerns about [Commerce's] assessment rate methodology" prior to Commerce's issuance of the Final Results, 63 Fed.Reg. 20,585. Def.'s Resp. Koyo at 11 (citing Final Results, 63 Fed.Reg. at 20,590-91 and relying on 28 U.S.C. § 2637(d) ("the Court ... shall, where appropriate, require the exhaustion of administrative remedies")) and McCarthy v. Madigan, 503 U.S. 140, 145, 112 S. Ct. 1081, 117 L. Ed. 2d 291 (1992) ("Exhaustion concerns apply with particular force when the action under review involves exercise of the agency's discretionary power or when the agency proceedings in question allow the agency to apply its special expertise" (citations omitted)).

Timken supports Commerce and points out that Commerce's method of calculating assessment rates was in accordance with law. See Timken's Resp. Mot. J. Agency R. Filed by Koyo, NTN and NSK ("Timken's Resp.") at 6-7.

Koyo maintains that, under the "futility" exception, "[t]he exhaustion doctrine does not[] ... preclude Koyo from raising this issue." Koyo's Reply Br. Supp. Mot. J. Agency R. ("Koyo's Reply") at 3-4 (citing Asociacion Colombiana de Exportadores de Flores v. United States, 916 F.2d 1571, 1575 (Fed.Cir.1990) ("A party need not exhaust his administrative remedies where invoking such remedies would be futile") and Koyo Seiko Co., 16 CIT at 544, 796 F. Supp. at 1531). Specifically, Koyo alleges that it would be futile for Koyo to raise the issue before Commerce because of: (1) Commerce's practice of using the methodology challenged by Koyo; and (2) the Court's ruling on the issue in Koyo Seiko Co., 16 CIT 539, 796 F. Supp. 1526.

3. Analysis

While Koyo errs in a part of its reasoning,^[3] Koyo is correct in its conclusion. Indeed the very fact that Commerce had employed the methodology at issue during the past reviews notwithstanding analogous challenges to the methodology submitted during those reviews, see, e.g., Final Results of Antidumping Duty Administrative Reviews of Tapered Roller Hearings and Parts Thereof, Finished and Unfinished, From Japan and Tapered Roller Bearings, Four Inches or Less in Outside Diameter, and Components Thereof, From Japan, 58 Fed.Reg. 64,720, 64,731 (Dec. 9, 1993), rendered a challenge by Koyo during the review at issue futile. See Von Hoffburg v. Alexander, 615 F.2d 633, 638 (1980) (stating that the exhaustion is futile if an agency: (1) consistently applies the challenged policy or methodology; (2) issues rules, regulations or bulletins promulgating such policy or methodology; and (3) rejects similar challenges); see also Rhone Poulenc, S.A., 7 CIT at 135, 583 F. Supp. at 610; United States Cane Sugar Refiners' Ass'n, 3 CIT at 201, 544 F. Supp. at 887. Therefore, Koyo is not barred from raising this issue before the Court.

B. Collateral Estoppel

1. Background

Collateral estoppel doctrine provides that "an issue of ultimate fact ... determined by a valid judgment, ... cannot be again litigated between the same parties [or their privies] in future litigation." BLACK'S LAW DICTIONARY 261 (6th ed.1990) (citing City of St. Joseph v. Johnson, 539 S.W.2d 784, 785 (Mo.Ct.App.1976)). In other words,

Collateral estoppel scans the first action and takes note of each issue decided. Then if the second action, although based on a different cause of action, attempts to reintroduce the same issue, collateral estoppel intervenes to preclude its relitigation and to bind the party, against whom the doctrine is being invoked, to the way the issue was decided in the first action.

DAVID D. SIEGEL, NEW YORK PRACTICE 715-16 (3d ed.1999).

For example, if a plaintiff enters a contract with several defendants who, upon the plaintiff's suit for one installment under the contract, are deemed jointly liable to the plaintiff, such defendants are collaterally estopped from claiming several liability when the plaintiff sues them for the next installment under the contract. See id. at 716 (referring to Schuylkill Fuel Corp. v. B. & C. Nieberg Realty Corp., 250 N.Y. 304, 165 N.E. 456 (1929)).

In contrast, the doctrine of stare decisis, "which makes the common law what it is," id. at 724, provides that "[once the] court has ... laid down a principle of law as applicable to a certain state of facts, it will adhere to that principle, and apply it to all future cases, where facts are substantially the same; regardless of whether the parties and property are the same." BLACK's LAW DICTIONARY 1406 (6th ed.1990) (citing Horne v. Moody, 146 S.W.2d 505, 509-10 (Tex.Civ.App.1940)).

2. Contentions of the Parties

Commerce asserts that the doctrine of collateral estoppel applies to the issue because

"(i) the issue previously adjudicated [in Koyo Seiko Co., 16 CIT 539, 796 F. Supp. 1526] is identical with that now presented, (ii) that issue was ``actually litigated' in the prior case, (iii) the previous determination of that issue was necessary to the end-decision then made, and (iv) the party precluded was fully represented in the prior action."

Def.'s Reply Koyo at 12 (quoting PPG Indus., Inc. v. United States, 978 F.2d 1232, 1239 (Fed.Cir.1992), citation omitted, and relying on RESTATEMENT (SECOND) OF JUDGMENTS § 27 (1982)).

Koyo contends that Commerce's reliance on the doctrine of collateral estoppel is: (1) misplaced because "[p]arties ... frequently continue to contest issues previously ruled on by [a] [c]ourt [of original jurisdiction] until the matter has been ultimately resolved by [a][c]ourt of [a]ppel[late jurisdiction]," Koyo's Reply at 4-5; and (2) "ironic [because Commerce] ... itself has made a regular practice of disregarding decisions of [the][c]ourt until the matter at issue is resolved by [t]he [c]ourt of [a]ppeals." Id. at 5.

3. Analysis

The Court is equally amused with the arguments offered by both parties. Koyo's innovative test effectively makes the doctrine of collateral estoppel: (1) dependent on an affirmation by an appellate court (thus, automatically placing all original jurisdiction cases out of the realm of the doctrine); and (2) inapplicable under some concept of "validation of illegal acts through vigilante violations" that is unknown to the Court.

A more conventional test offered by Commerce lists four elements of collateral estoppel: "(i) the issue previously adjudicated is identical with that presented[;] (ii) that issue was ``actually litigated' in the prior case[;] (iii) the previous determination of that issue was necessary to the end-decision then made[;] and (iv) the party precluded was fully represented in the prior action." Def.'s Reply Koyo at 12. The Court agrees with Commerce that these elements are indeed indispensable to any analysis of collateral estoppel doctrine. The list offered by Commerce is, however, not exhaustive: There are two other elements that must be satisfied to trigger the application of the doctrine. One of these elements provides that the party to be estopped in the second action must either be the same party that lost in the first action or someone "in strict privity" with the losing party. In the given case, this element is satisfied because Koyo's motion against Commerce was denied in Koyo Seiko Co., 16 CIT 539, 796 F. Supp. 1526.

The final element is best summarized by Judge Cardozo, who stated that

[a] judgment in one action is conclusive in a later one ... when the two causes of action have such a measure of identity that a different judgment in the second would destroy or impair rights or interests established by the first....

Schuylkill Fuel Corp., 250 N.Y. at 306, 165 N.E. 456 (emphasis supplied).

This final element is not present in the case at bar. Indeed, if, in this case, the Court holds differently from its decision in Koyo Seiko Co., 16 CIT 539, 796 F. Supp. 1526, such holding would not and could not destroy or impair any rights or interests established in Koyo Seiko Co., 16 CIT 539, 796 F. Supp. 1526 (that dealt with Commerce's determination in Final Results of Antidumping Duty Administrative Reviews of Antifriction Bearings (Other Than Tapered Roller Bearings) and Parts Thereof From Japan, 56 Fed.Reg. 31,754 (July 11, 1991), a determination entirely different from that made in the Final Results, 63 Fed.Reg. 20,585). While indeed Koyo Seiko Co., 16 CIT 539, 796 F. Supp. 1526, created a precedent on the issue under the doctrine of stare decisis, it clearly cannot collaterally prevent Koyo's challenge in this case under the doctrine of collateral estoppel. Commerce simply fails to appreciate the distinction between the doctrines.^[4]

C. Use of Entered Value in Order to Determine Assessment Rates

Commerce's use of entered value for the purpose of determining assessment rate was approved by the Court in Koyo Seiko Co., 16 CIT 539, 796 F. Supp. 1526, and Koyo Seiko, Co. v. United States, 24 CIT ___, 110 F. Supp. 2d 934 (2000), aff'd, Koyo Seiko, Co. v. United States, 258 F.3d 1340 (Fed.Cir.2001). Because the arguments at issue in this case (same as the parties) are practically identical to those presented in Koyo Seiko Co., 16 CIT 539, 796 F. Supp. 1526, and Koyo Seiko, Co., 24 CIT ___, 110 F. Supp. 2d 934, the Court adheres to its reasoning as it is stated in these cases.

II. COMMERCE'S INCLUSION OF HOME MARKET MOVEMENT EXPENSES IN THE CALCULATION OF CONSTRUCTED VALUE PROFIT

A. Background

In the Final Results, 63 Fed.Reg. at 20,608, Commerce, while recalculating Koyo's reported constructed value ("CV") profit, included movement expenses in the per-unit profit amount. In order to calculate the per-unit CV profit amount, Commerce defined the term "profit" as a difference between adjusted gross unit price and Koyo's reported cost of production ("COP"). See id. Then, for each model of the merchandise, Commerce multiplied the per-unit profit amount and COP by the total quantity sold in order to calculate model-specific profit and COP amounts. See id. The ratio of total profit for all models to the total COP for all models was used by Commerce as the profit ratio for the purpose of calculating CV. See id.

As Koyo observes, see Koyo's Mem. at 9-10, and Commerce acknowledges, see Def.'s Resp. Koyo at 8, Commerce did not take into account the fact that movement expenses were included in the gross-unit price but were not included in the COP. Koyo's reported gross-unit price included various selling expenses (including movement expenses, that is, both home market pre-sale inland freight and home market post-sale inland freight), whereas Koyo's reported COP did not include movement expenses, even though it did include the cost of manufacture ("COM"), selling, general and administrative ("SG & A") expenses, interest and packing. See id. Therefore, Commerce omitted to consider the fact that all the expenses (except for the movement expenses) included in Koyo's gross-unit price were mirrored in Koyo's COP calculation. Consequently, the amount reached by Commerce that represented the difference between the gross-unit price and COP and intended by Commerce to be Koyo's CV profit was, in fact, the CV profit plus Koyo's movement expenses.

Koyo contends that movement expenses should be deducted from "the net home market price to ensure an accurate CV profit calculation." Koyo's Mem. at 11. Commerce concurs with Koyo. See Def.'s Resp. Koyo at 8. As both parties correctly observed, a nearly identical issue was dealt by the Court in Fag Kugelfischer Georg Schafer Ag v. United States ("FAG Kugelfischer"), 19 CIT 634, 1995 WL 283834 (1995). Indeed, following FAG Kugelfischer, 19 CIT 634, 1995 WL 283834, Commerce's practice is to "deduct certain home market expenses (which are not included in the total ... COP) from net unit price" for the purpose of calculating CV profit. Therefore, the issue is remanded to Commerce to recalculate Koyo's CV profit so that Koyo's home market movement expenses are deducted from Koyo's net home market price.

III. COMMERCE ERROR IN CALCULATION OF MARINE INSURANCE CHARGES

Koyo contends that Commerce relied upon an erroneous marine insurance factor in the Final Results, 63 Fed.Reg. at 20, 610-11. See Koyo Mem. at 14. Specifically, Koyo argues that Commerce calculated an amount for marine insurance charges for sales of further-processed merchandise by doing the following: (1) subtracting the custom duties from landed cost; and (2) multiplying the result by a certain factor, the use of which was inappropriate for the calculation. See id. Commerce agrees with Koyo's observation. See Def.'s Resp. Koyo at 16. Therefore, the issue is remanded to Commerce to recalculate Koyo's marine insurance charges using the correct factor indicated by Koyo. Accord Koyo's Mem. at 14 (proprietary version).

IV. COMMERCE'S CALCULATION OF CONSTRUCTED VALUE RELYING UPON BEST INFORMATION AVAILABLE

During the administrative review, Commerce calculated CV for a particular part of the merchandise imported by Koyo relying upon best information available, operating under the impression that Koyo did not provide Commerce with CV commission factor. Final Results, 63 Fed. Reg. at 20,596-98. Koyo, however, reported CV commission factor. See Koyo's Mem. at 15-16; Def.'s Resp. Koyo at 16. It follows that Commerce erred in using an incorrect commission expense factor in order to calculate CV for the merchandise at issue. Therefore, the issue is remanded to Commerce to recalculate Koyo's CV using commission factor provided by Koyo.

V. COMMERCE'S CALCULATION OF CONSTRUCTED VALUE DIRECT SELLING EXPENSES

During the administrative review, Commerce applied a certain direct selling expense ratio in order to calculate the CV of particular merchandise imported by Koyo. While the factor used by Commerce should have been the sum of warranty and imputed credit expenses that were indicated in Koyo's questionnaire response, Commerce used a factor inappropriate for the calculation. See id. Therefore, the issue is remanded to Commerce to recalculate CV direct selling expenses relying on the factor indicated in Koyo's questionnaire response.

VI. COMMERCE'S USE OF COMPUTER PROGRAM CONCERNING PRODUCT NOMENCLATURE

In the Final Results, Commerce corrected an input error that Koyo made in Koyo's reported data for the nomenclature of a particular product. See Def.'s Resp. Koyo at 17. While making this correction, however, Commerce did the following: (1) improperly transcribed the product nomenclature for two types of entries by erroneously entering a certain letter in place of a certain number; and (2) entering a double slash symbol instead of a single slash symbol between certain denominations that transcribed the nomenclature of a particular product. See id. Moreover, Commerce entered a certain correction language too early in the computer program, consequently instructing the computer to enter the correction for a certain model before a necessary variable came to exist in the program. See id. at 17-18. As a result of placing the correction language prior to the importation of the necessary data, the program produced erroneous results. See id. Therefore, this issue is remanded to Commerce to enter necessary corrections and recalculate pertinent parts of Commerce's determination with respect to Koyo's imports.

VII. COMMERCE'S CALCULATION OF INVENTORY CARRYING COSTS FOR FURTHER-MANUFACTURED UNITED STATES SALES

A. Background

During the review at issue, Commerce calculated two separate inventory carrying costs for Koyo's further-manufactured United States sales: (1) an inventory carrying cost that reflected the amount of time unfinished product spent in inventory in Koyo's home market; and (2) an inventory carrying cost that reflected the amount of time the finished product at issue spent in inventory in the United States. See Final Results, 63 Fed.Reg. at 20,600. Commerce did not calculate an inventory carrying cost from the time that Koyo's subsidiary purchased the unfinished product until the time that the unfinished product was processed into the finished product at issue See Def.'s Resp. Partial Opp'n Timken's Mem. Supp. Rule 56.2 Mot. J. Agency R. ("Def.'s Resp. Timken") at 12-13.

B. Contentions of the Parties

Relying on 19 U.S.C. § 1677a(e)(2) (1988) and Koyo Seiko Co. v. United States, 36 F.3d 1565 (Fed.Cir.1994), and pointing out that "[i]nventory carrying cost measures the imputed cost incurred by a company for storing merchandise in inventory," Commerce asserts that it acted properly when Commerce calculated the inventory carrying costs in the above-stated manner. Id. at 12 (quoting Thai Pineapple Pub. Co. v. United States, 20 CIT 1312, 1329, 946 F. Supp. 11, 26 (1996), rev'd, Thai Pineapple Pub. Co. v. United States, 187 F.3d 1362 (1999)). Specifically, Commerce maintains that because the product in question was further-manufactured, separate inventory periods existed. See id. at 14. Commerce further explains that its policy is "to make an [inventory carrying costs] adjustment to [United States price] for only finished goods in inventory because unfinished goods represent production expenses rather than [United States] selling expenses." Id. (quoting Final Results, 63 Fed.Reg. at 20,600, and relying on Final Results of Antidumping Duty Administrative Review of Dynamic Random Access Memory Semiconductors of One Megabit or Above From the Republic of Korea, 61 Fed.Reg. 20,216, 20,221 (May 6, 1996)). Commerce notes that Timken's assertion that Commerce "deducted only the inventory carrying cost attributable to the imported portion of the merchandise, not the [United States] portion" is only partially correct and does not render Commerce's calculations wrongful. Id. at 14-15. In sum, Commerce: (1) reads Timken's challenge as a statement that the whole of further-manufactured merchandise is kept in inventory pending sale, not just the imported parts; and (2) points out that the "inquiry, however, is not whether [merchandise] physically remains in inventory. Rather, the pertinent question is whether time spent in inventory represents a selling expense or a production expense." Id. at 15.

Koyo supports Commerce's assertions and points out that, while "Timken's argument ... is brief and rather cryptic," Koyo's Mem. Resp. Timken's Mot. J. Agency R. ("Koyo Resp. Timken") at 7, the two readings of Timken's argument that Koyo envisions, namely: (1) the argument asserting that Koyo's inventory carrying costs should be calculated for the time period in which the unfinished product was held by Koyo's subsidiary as raw material inventory; or (2) the argument that Koyo's inventory carrying costs for the time period during which the finished product was held in Koyo's inventory prior to sale should be included among the expenses deducted from Koyo's United States price, see id. at 8-9, are erroneous because: (1) "[i]n the first case, the proposed calculation would double count the expense[;] while [(2)] in the second case[,] [Commerce] has already" included among the expenses deducted from Koyo's United States price the amount of Koyo's inventory carrying costs for the time period during which the finished product was held in Koyo's inventory. Id. at 10.

Timken, however, states that Timken "is not arguing that Commerce erred in [the above discussed] respect." Timken's Reply Resps. Timken's Mot. J. Agency R. ("Timken's Reply Commerce") at 3 n. 1. "The crux of Timken's argument ... is ... that the method of calculation of the [United States inventory carrying costs] identified by Koyo" itself was not consistently implemented in the way corresponding to Koyo's questionnaire response. Id. at 2 (proprietary version).

C. Analysis

The Court agrees with Timken's argument. While Koyo's formulae satisfy the requirements of the statute, factual data reveals that Koyo's application of the formulae suffers of inconsistency. See id. at 3-4 (proprietary version, 1992/93 and 1993/94 tables). Therefore, the issue is remanded to Commerce to recalculate Koyo's United States inventory carrying costs for final product by applying the appropriate inventory carrying costs factor reported by Koyo to the landed cost for the 1992/93 period of review ("POR") and to the COM for the 1993/94 POR.

VIII. COMMERCE'S MODEL MATCHING PROGRAM

A. Background

During the review at issue, Commerce relied upon the "sum-of-the-deviations" ("SUMDEV") methodology to rank NSK's similar home market models of the merchandise as potential matches to United States models of the merchandise. See Def.'s Resp. Partial Opp'n Mem. P. & A. Supp. NSK's Mot. J. Agency R. ("Def.'s Resp. NSK") at 5; NSK's Reply Mem. Supp. Mot. J. Agency R. ("NSK's Reply") at 2-3. Using the SUMDEV methodology, the model with the smallest SUMDEV is deemed the most similar. See Koyo Seiko Co. v. United States, 66 F.3d 1204, 1207 (Fed.Cir.1995). In cases where the SUMDEV of two models is the same, the program determines the most-similar model based upon a series of "tie breakers": (1) level of trade; then (2) the deviation in cost of production ("COSTDEV"); and then (3) NSK's alpha-numeric product nomenclature. See Def.'s Resp. NSK at 5-6; NSK's Reply at 3-4.

The review at issue presented Commerce with a model, the matching process of which reached the last tier of the "tie breaking" process, namely, the comparison of NSK's alpha-numeric product nomenclature. NSK's designation of the model to be matched is comprised of fourteen various symbols (the one like ABCDEFGHIJKLMN), inclusive of but not limited to numeric and letter symbols. See id.

In the process of matching, Commerce's computer executed two operations. See id. First, the computer created a list of models to be matched with the model at issue. The list created included two models, one having nine various symbols (the one like ABCDEFGHI) and another having eleven symbols, with identical first nine symbols (the one like ABCDEFGHIJ *). See id. Commerce's computer, however, listed the eleven-symbols model prior to the nine-symbols model, effectively creating a "backwards" list that: (1) deviated from all other lists Commerce's computer created for all other NSK's models; and (2) presented a list opposite to those commonly used in listing compilations, e.g., dictionaries that list such a word as "work" prior to the word "worker." See id.

Second, Commerce's computer matched the model at issue (the one like ABCDEFGHIJKLMN) to the eleven-symbols model (the one like ABCDEFGHIJ *) that was, as stated above, listed prior to the nine-symbols model (the one like ABCDEFGHI). See id. The model at issue had the same first ten symbols as the eleven-symbols model (and, obviously, the very same first nine symbols as the nine-symbols one).

B. Contentions of the Parties

NSK argues that Commerce's computer program made an error by matching the model at issue to the eleven-symbols one rather than to the nine-symbols one. See NSK's Reply at 3-4. Specifically, NSK contends that, because the nine-symbols model (the one like ABCDEFGHI) should have been listed prior to the eleven-symbols one (the one like ABCDEFGHIJ*), the computer program should have matched the model at issue (the one like ABCDEFGHIJKLMN) to the model that should have been the first on the list (the one like ABCDEFGHI). See id. In support of its contention, NSK points to all other lists created by Commerce's computer program that correctly listed the models with less symbols (the one like 12345) prior to those models that were designated first with the symbols identical to the less-symbols models and then, in addition, with a number of following symbols (the ones like 12345ABC). See id. at 4.

Commerce asserts that Commerce's matching program was not in error. Commerce points out that the "program matched identical nomenclature out to the tenth place ..., rather than the ninth place ... [because the eleven-symbols] home market model [chosen by the computer] is more similar to [the model at issue] based upon model names." Def.'s Resp. NSK at 6-7.

C. Analysis

The Court agrees with Commerce. While NSK is correct in its contention that the nine-symbols model (the one like ABCDEFGHI) should have been listed by the program prior to the eleven-symbols model (the one like ABCDEFGHIJ*), this error is not dispositive with regard to the outcome of the inquiry. The gist of the inquiry is to find models that match to the highest degree. The model at issue (the one like ABCDEFGHIJKLMN) is certainly more like the eleven-symbols model (the one like ABCDEFGHIJ*) than the nine-symbols model (the one like ABCDEFGHI). The mere fact that the initial comparative list was created in a "backwards" order does not detract from the fact that the model at issue and the eleven-symbols model have the first ten symbols identical, while the model at issue and the nine-symbols model have only nine identical symbols. Because NSK failed to demonstrate the error in the conclusion reached by Commerce's computer program, the Court holds that Commerce's conclusion was correct.

IX. COMMERCE'S CALCULATION OF A REVISED COST OF PRODUCTION

In order to calculate NSK's COP, Commerce revised NSK's general and administrative ("G & A") expenses to account for certain non-operating gains and losses. Following the revision, Commerce: (1) decreased NSK's G & A factor accordingly; and (2) implemented the change in Commerce's calculation of NSK's COP. See NSK's Mem. P. & A. Supp. Mot. J. Agency R. ("NSK's Mem.") at 4. Commerce, however, erred while implementing the latter changes. See id.; see also Def.'s Resp. NSK at 7. The issue is, therefore, remanded to Commerce to apply the correct G & A expenses factor in Commerce's calculation of NSK's COP.

X. COMMERCE'S DETERMINATION THAT SAMPLE AND SMALL-QUANTITY HOME MARKET SALES ARE WITHIN THE ORDINARY COURSE OF TRADE

A. Background

Under the pre-URAA law, fair market value ("FMV") was defined as "the price ... at which such or similar merchandise is sold or, in the absence of sales, offered for sale in the principal markets of the country from which exported, in the usual commercial quantities and in the ordinary course of trade for home consumption. ..." 19 U.S.C. § 1677b(a)(1) (1988) (emphasis supplied). The term "ordinary course of trade" was, in turn, defined as

the conditions and practices which, for a reasonable time prior to the exportation of the merchandise ..., have been normal in the trade under consideration with respect to merchandise of the same class or kind.

19 U.S.C. § 1677(15) (1988).

While the language of § 1677(15) indicates that various types of sales could be considered outside the ordinary course of trade,^[5] Commerce's regulations (of the pertinent period of time) did not provide any further clarification of the phrase "ordinary course of trade." See 19 C.F.R. § 353.46(b) (1994) (providing a definition that mirrored the statutory one).

In the Final Results, 63 Fed.Reg. at 20,588, Commerce determined that NTN's sample and small-quantity home market sales were within the ordinary course of trade. Consequently, Commerce used these sales for purposes of determining FMV. Commerce's decision was based on Commerce's determination that NTN failed to provide Commerce with any information other than a general description of those sales that, according to NTN, fell outside the ordinary course of trade. See Final Results, 63 Fed.Reg. at 20,588. Specifically, Commerce concluded that there was "no evidence supporting the notion that NTN's sample sales were sold only for the purpose of allowing the customer to make a decision to buy" as well as "no evidence supporting NTN's categorization of its ``small-quantity' sales as abnormal, other than the fact that they were small-quantity...." Id. at 20,588 (emphasis supplied).

B. Contentions of the Parties

NTN contends that Commerce erred in Commerce's determination that NTN's sample and small-quantity home market sales were within the ordinary course of trade. See NTN's Mot. Mem. Supp. J. Agency R. ("NTN's Mem.") at 5-7; NTN's Reply Mem. Def.'s and Def.-Intervenor's May 28, 1999 Resp. Mem. Mot. J. Agency R. ("NTN's Reply") at 2-4. Specifically, NTN asserts that: (1) "[t]here is more than adequate information on the record to establish [that] these ... sales were outside the ordinary course of trade"; and (2) Commerce erred in Commerce's decision to assess "whether [NTN's] sale is ``abnormal or aberrational.'" NTN's Resp. at 2-4.

Commerce maintains that its determination that NTN's sales at issue were within the ordinary course of trade was supported by substantial evidence since NTN provided Commerce with nothing but a general description of the sales. See Def.'s Resp. Opp'n Pl. NTN's Mot. Mem. Supp. J. Agency R. ("Def.'s Resp. NTN") at 21-26. Timken supports Commerce's assertion and states that NTN failed to satisfy NTN's burden to demonstrate that NTN's sample sales and small-quantity sales were outside the ordinary course of trade. See Timken's Resp. at 9-11.

C. Analysis

The purpose of the ordinary course of trade provision "is to prevent dumping margins from being based on sales which are not representative...." Monsanto Co. v. United States, 12 CIT 937, 940, 698 F. Supp. 275, 278 (1988). Accordingly, "Commerce must evaluate not just one factor taken in isolation but rather ... all the circumstances particular to the sales in question,'" CEMEX, S.A., 133 F.3d at 900 (citation omitted), and the burden rests with the plaintiff to provide Commerce with sufficient evidence showing that the sales used in Commerce's calculations are outside the ordinary course of trade. See Nachi-Fujikoshi Corp. v. United States, 16 CIT 606, 608, 798 F. Supp. 716, 718 (1992). In the absence of adequate evidence to the contrary, Commerce considers sales within the ordinary course of trade. See Torrington Co. v. United States, 127 F.3d 1077, 1081 (Fed.Cir.1997). Thus, a determination of whether a sale or transaction is outside the ordinary course of trade is a question of fact,^[6] and Commerce has the discretion to interpret § 1677(15) and to determine which sales are outside the ordinary course of trade. See Mitsubishi Heavy Indus. v. United States ("Mitsubishi"), 22 CIT 541, 568, 15 F. Supp. 2d 807, 830 (1998) ("Congress granted Commerce discretion to decide under what circumstances ... sales would be considered to be outside of the ordinary course of trade."); cf. Koenig & Bauer-Albert AG v. United States, 22 CIT 574, 589 n. 8, 15 F. Supp. 2d 834, 850 n. 8 (1998) (noting that although Commerce has the discretion to decide under what circumstances highly profitable sales are outside of the ordinary course of trade, "Commerce may not impose this requirement arbitrarily, ... nor may Commerce impose impossible burdens of proof on claimants" and citing NEC Home Elecs. v. United States, 54 F.3d 736, 745 (Fed.Cir.1995) "(holding that burden imposed to prove a level of trade adjustment was unreasonable because claimant could, under no practical circumstances, meet the burden)"). In making this determination, Commerce considers not just "one factor taken in isolation but rather ... all the circumstances particular to the sales in question." Murata Mfg. Co. v. United States, 17 CIT 259, 264, 820 F. Supp. 603, 607 (1993) (citation omitted). Commerce's methodology for deciding when sales are outside the "ordinary course of trade" has been to examine, on a case-by-case basis, the totality of the circumstances surrounding the sale or transaction in question to determine whether the sale or transaction is extraordinary.

NTN alleges that "there is ample evidence establishing that NTN's sales were outside the ordinary course of trade." NTN's Reply at 2-3. Citing to its questionnaire response, NTN states that sample sales at issue were made to enable a customer to decide whether or not to make a decision to buy. See id. at 3. NTN also provided Commerce with information on NTN's procedure to track small sales. See id. NTN, however, provided Commerce with no evidence of statements and representations made by NTN to the entities obtaining samples or small-quantity purchases to enable Commerce to establish that sample sales were "not a normal condition or practice in the trade under consideration." Def.'s Resp. NTN at 23-24.

While NTN is correct in noting that "[t]he [statutory] standard ... is not [specifically testing] whether a sale is ``abnormal or aberrational,'" see NTN's Reply at 3, Commerce's choice of terms "abnormal" or "aberrational" with regard to Commerce's inquiry does not violate the gist of the statutory mandate. Accord H.R. Doc. 103-316, at 834, reprinted in 1994 U.S.C.C.A.N. 4171. The term "abnormal" means "[d]eviating from normal condition," while the term "aberrational" stands for an act characterized by "deviation ... from the natural state, or from a normal type." WEBSTER'S DICTIONARY 4, 6 (New Int'l, 2d ed.1948). Thus, these terms are in accord with the definition provided by 19 U.S.C. § 1677(15), and Commerce was: (1) entitled to request evidence supporting NTN's assertion about the irregularity of the sales at issue; and (2) refusing to accept NTN's claim that sample sales are "by their very nature" outside the ordinary course of trade. Accord Koyo Seiko Co., 16 CIT at 543, 796 F. Supp. at 1530. Commerce is correct in pointing out that "[s]ample sales may well be a type of ``condition or practice' that is ``normal in the trade under consideration,'" especially if such sales are done on a regular and continuous basis.^[7] Def.'s Resp. NTN. at 24-25 (relying on Final Results of Antidumping Duty Administrative Reviews of Tapered Roller Bearings and Parts Thereof, Finished and Unfinished, From Japan and Tapered Roller Bearings, Four Inches or Less in Outside Diameter, and Components Thereof, From Japan, 58 Fed.Reg. at 64,732, and Final Results of Antidumping Duty Administrative Reviews and Revocation in Part of an Antidumping Finding on Tapered Roller Bearings and Parts Thereof, Finished and Unfinished, From Japan and Tapered Roller Bearings, Four Inches or Less in Outside Diameter, and Components Thereof, From Japan, 61 Fed.Reg. 57,629, 57,639 (Nov. 7, 1996)); see Torrington Co. v. United States, 25 CIT ___, 146 F. Supp. 2d 845 (2001); NTN Bearing Corp. of Am. v. United States, 20 CIT 508, 924 F. Supp. 200 (1996); NSK Ltd. v. United States, 17 CIT 590, 825 F. Supp. 315 (1993). Therefore, Commerce's decision to consider the sales at issue within the ordinary course of trade is not unreasonable, provided Commerce did not receive any specific evidence to the contrary. While NTN claims that "there is ample evidence," the actual evidence NTN refers to is nothing but the very same NTN's questionnaire response discussing NTN's system of order placement and NTN's tracking procedures. See NTN's Reply at 2-3. The mere reference to procedures is, however, insufficient to persuade the Court that Commerce's conclusion was unreasonable. Therefore, the Court finds Commerce's determination that NTN's sample and small-quantity home market sales were within the ordinary course of trade in accordance with law and supported by substantial evidence.

XI. DENIAL OF AN ADJUSTMENT TO UNITED STATES INDIRECT SELLING EXPENSES FOR INTEREST ALLEGEDLY INCURRED IN FINANCING CASH DEPOSITS FOR ANTIDUMPING DUTIES

A. Background

Following the review at issue, Commerce denied an adjustment to NTN's United States indirect selling expenses to exclude expenses that NTN claimed were related to the financing of antidumping duty cash deposits. See Final Results, 63 Fed.Reg. at 20,595-96.

Under the pre-URAA law, Commerce struggled with the issue whether financing expenses for antidumping cash deposits were: (1) selling expenses within the meaning of 19 U.S.C. § 1677a(e)(2) and, thus, deductible from exporter's sales price ("ESP"); or (2) analogous, e.g., to legal fees that fall outside 19 U.S.C. § 1677a(e)(2) and should be subject to an offset against the selling expenses deducted from ESP.

In one of the previous reviews, specifically, Final Results of Antidumping Duty Administrative Reviews and Revocation in Part of an Antidumping Duty Order, 58 Fed.Reg. 39,729, 39,749 (July 26, 1993), Commerce: (1) granted an offset to NTN's United States selling expenses to account for imputed interest expenses on antidumping cash deposits; and (2) upon remand, explained Commerce's belief that the interest expenses were analogous to antidumping legal fees (because they were incurred solely as a result of the antidumping duty order) that could not be categorized as selling expenses. The Court sustained Commerce's grant of the offset. See Federal-Mogul Corp. v. United States, 20 CIT 1438, 1440, 950 F. Supp. 1179, 1182 (1996); see also Timken Co. v. United States, 22 CIT 621, 622-24, 16 F. Supp. 2d 1102, 1104-05 (1998) (approving analogous offset for Final Results of Antidumping Duty Administrative Reviews and Termination in Part of Tapered Roller Bearings and Parts Thereof, Finished and Unfinished, From Japan and Tapered Roller Bearings, Four Inches or Less in Outside Diameter, and Components Thereof, From Japan, 62 Fed.Reg. 11,825 (March 13, 1997)).

B. Contentions of the Parties

NTN asserts that Commerce erroneously rejected NTN's adjustment to NTN's indirect selling expenses incurred in regard to antidumping duty cash deposits. See NTN's Mem. at 8-11. Specifically, NTN points out that imputed interest expenses cannot be (and previously were not) categorized as selling expenses. See id. at 9-10 (relying on Federal-Mogul Corp., 20 CIT at 1440, 950 F. Supp. at 1182).

Commerce maintains that Commerce's denial of an adjustment to NTN's United States indirect selling expenses for the expenses related to the financing of antidumping duty cash deposits reflected Commerce's reasonable reading and application of the statutory mandate. See Def.'s Resp. NTN at 26-33. Timken supports Commerce's contention and points out that: (1) the purpose of the statutory provision for interest on over and under deposits of duties would be defeated by allowing an expense reduction for interest on cash deposits, see Timken's Resp. at 14-15; and (2) NTN failed to demonstrate that it actually incurred interest expenses attributable to financing payment of antidumping duty cash deposits. Id. at 15.^[8]

C. Analysis

Because financing expenses on antidumping duty cash deposits do not fall squarely within the "selling expenses" category and do not imitate perfectly the expenses that typically fall outside the reach of Section 1677a(e)(2), Commerce is, effectively, obligated to find the best default category to house these expenses. In the Final Results at issue, Commerce: (1) reconsidered Commerce's previous position and concluded that expenses pertaining to the financing of antidumping duty cash deposits would be more properly treated as indirect selling expenses; and (2) based its decision on the fact that the statute does not contain a precise definition of what constitutes a "selling expense." See 63 Fed.Reg. at 20,595; accord Torrington Co. v. United States, 44 F.3d 1572, 1579-81 (Fed.Cir.1995) (holding that Commerce's treatment of inventory carrying costs as an indirect selling expense is a reasonable interpretation of the statute).

In its efforts to identify a selling expense, Commerce attempts to distinguish "between business expenses that arise from economic activities in the United States and business expenses that are direct, inevitable consequences of an antidumping duty order." Final Results, 63 Fed.Reg. at 20,595-96. Antidumping duties, cash deposits of antidumping duties, and legal fees associated with participation in an antidumping case are all expenses that fall within the category of "business expenses that are direct, inevitable consequences of an antidumping duty order" because they are incurred solely as a result of the existence of an antidumping duty order. Def.'s Resp. NTN at 29 (quoting Final Results at 20,596). Therefore, Commerce does not consider such expenses to be selling expenses deductible pursuant to 19 U.S.C. § 1677a(e)(2). See Daewoo Elecs. Co. v. United States, 13 CIT 253, 269, 712 F. Supp. 931, 947 (1989), rev'd on other grounds, Daewoo Elecs. Co. v. Int'l Union of Elec., Tech., Salaried and Machine Workers, AFL-CIO, 6 F.3d 1511 (Fed.Cir.1993), cert. denied, 512 U.S. 1204, 114 S. Ct. 2672, 129 L. Ed. 2d 808 (1994).

Following its change in statutory interpretation, Commerce deems any financing expenses associated with antidumping duty cash deposits to be a type of selling expense properly deductible pursuant to 19 U.S.C. § 1677a(e)(2). In the Final Results, Commerce explained that: (1) money is fungible because "[i]f an importer acquires a loan to cover one operating cost, that may simply mean that it will not be necessary to borrow money to cover a different operating cost"; and (2) "[c]ompanies may choose to meet obligations for cash deposits in a variety of ways that rely on existing capital resources or that require raising new resources through debt or equity." 63 Fed.Reg. at 20,596. Thus, Commerce concluded that there is nothing inevitable about a company incurring financing expenses to meet the company's obligations for cash deposits. Therefore, these particular expenses could: (1) fall within the realm of selling expenses, direct and indirect; and (2) be reasonably qualified as selling expenses for purposes of 19 U.S.C. § 1677a(e)(2).

NTN argues that Commerce's statutory interpretation is in error because, regardless of how a company posts cash deposits, there is an "opportunity cost," that is, that "money will cease to be available for other company uses." NTN Mem. at 8. Commerce, however, is correct in noting that "it is within Commerce's authority to deduct opportunity costs as selling expenses." Def.'s Resp. NTN at 30; see also Torrington Co., 44 F.3d at 1580 (sustaining Commerce's decision to treat inventory carrying costs as indirect selling expenses). Conversely, NTN is incorrect in its assertion that since "interest expenses for antidumping duty cash deposits are not incurred in the course of selling merchandise in the United States," there is no basis for deducting these expenses from United States price. NTN Mem. at 10. In the Final Results, Commerce implicitly recognized that the selling expenses referred to in 19 U.S.C. § 1677a(e)(2) is a broad category encompassing numerous expenses whereas antidumping duties, cash deposits, and legal fees represent "a limited exemption." 63 Fed.Reg. at 20,596. While the plain language of 19 U.S.C. § 1677a(e)(2) does not provide the precise meaning of the term "expenses," "[t]he purpose of deducting selling expenses from exporter's sales price was to estimate the ``net amount returned to the foreign exporter' ... of the merchandise by removing all of the expenses incurred after importation by the related-party importer." Koyo Seiko Co. v. United States, 36 F.3d 1565, 1571-72 (Fed.Cir.1994) (quoting Emergency Tariff and Antidumping: Hearings on H.R. 2435 Before the Senate Comm. on Finance, 67th Cong., 1st Sess. 10-12 (1921)); accord 19 U.S.C. § 1673g(b)(4) (1988); 19 C.F.R. § 353.26(a) (1994). Accordingly, Commerce could reasonably deduct financial expenses associated with cash deposits from United States price in order to arrive at the ESP, that is, the net amount returned to the exporter.

The fact that Commerce has changed its interpretation of 19 U.S.C. § 1677a(e)(2) over time does not detract from the reasonableness of Commerce's current statutory interpretation.^[9]See Consolo, 383 U.S. at 620, 86 S. Ct. 1018 (noting 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"); AK Steel Corp. v. United States, 22 CIT 1070, 1082, 34 F. Supp. 2d 756, 766 (1998) (stating that an initial interpretation by an agency is not "carved in stone" and citing Chevron, 467 U.S. at 863, 104 S. Ct. 2778).

XII. DENIAL OF ADJUSTMENT TO FOREIGN MARKET VALUE FOR HOME MARKET DISCOUNTS

A. Background

The pre-URAA statute defined FMV as "the price ... at which such or similar merchandise is sold or, in the absence of sales, offered for sale in the principal markets of the country from which exported, in the usual commercial quantities and in the ordinary course of trade for home consumption. ..." 19 U.S.C. § 1677b(a)(1)(A). Commerce interpreted the term "price" used in the definition as the price for the foreign like product after any adjustments for post-sale price adjustments ("PSPAs"), where PSAPs were direct adjustments to price. See Final Results, 63 Fed.Reg. at 20,602 ("As a general matter, ... [Commerce] only accepts claims for discounts, rebates, or other PSPAs as direct adjustments to price if actual amounts are reported for each transaction"). Pursuant to Commerce's statutory interpretation, a respondent may demonstrate entitlement to a PSPA if the PSPA is: (1) "reported on a transactionspecific basis"; or (2) "allocated ... if the respondent demonstrates that the PSPA was granted as a fixed and constant percentage of the sales price of all transactions for which it was reported and to which it was allocated." Id.

In the review at issue, Commerce disallowed an adjustment to FMV for NTN's home market discounts, see Final Results, 63 Fed.Reg. at 20,602-03, because NTN: (1) "did not report these discounts on a transaction-specific basis, but, rather, reported the adjustments on a product-[specific] or customer-specific basis"; and (2) "did not grant and report these discounts as a fixed and constant percentage of sales." Id. at 20,603.

B. Contentions of the Parties

NTN asserts that Commerce erred in disallowing any adjustments to FMV for NTN's discounts. Specifically, NTN argues that: (1) "[b]ecause the discounts at issue ... did vary with the quantity sold, and were related to particular sales, [Commerce] should have treated them as direct selling expenses and granted NTN an adjustment to FMV to account for the expenses"; and (2) because the discounts at issue "constitute a discount which is a fixed and constant percentage of sales," Commerce should have allowed the adjustments for these discounts. NTN Mem. at 12-13 (relying on Torrington Co. v. United States ("Torrington"), 82 F.3d 1039 (Fed. Cir.1996)).

Commerce contends that its decision to disallow any adjustments to FMV for NTN's discounts was in accordance with law and a reasonable interpretation of 19 U.S.C. § 1677b(a)(1)(A). Def.'s Resp. NTN at 33-37. Timken supports Commerce's assertion and points out that Commerce's conclusion was reasonable in view of the fact that NTN failed to report NTN's home market discounts in a manner required. See Timken's Resp. at 17-20.

C. Analysis

NTN's reading of the statutory mandate and case law is incorrect. NTN errs in its reliance on Torrington, 82 F.3d 1039, when it argues that the fact that discounts at issue did vary obligates Commerce to treat these discounts as direct selling expenses subject to an adjustment. See NTN Mem. at 12-13. Torrington, 82 F.3d 1039, does not provide a set of two propositions where from each one the other could be automatically inferred. While Torrington recognized that "direct selling expenses ``are expenses which vary with the quantity sold,'" 82 F.3d at 1050 (quoting Zenith Elecs. Corp. v. United States, 77 F.3d 426, 431 (Fed.Cir.1996) (internal quotation omitted)), it does not necessarily mandate that each expense that could vary with the quantity sold should be deemed a direct selling expense. See id. Conversely, Torrington recognized that "[t]he allocation of expenses in a manner different from the calculation of the expenses, however, does not alter the relationship between the expenses and the sales under consideration." 82 F.3d at 1051 (citing Smith-Corona Group, Consumer Prods. Div., SCM Corp. v. United States, 713 F.2d 1568 (Fed.Cir.1983), cert. denied, 465 U.S. 1022, 104 S. Ct. 1274, 79 L. Ed. 2d 679 (1984)). The court in Torrington ruled that Commerce erroneously granted a respondent an ESP offset pursuant to 19 C.F.R. § 353.56(b)(2) (1994) because that regulation limited the offset to "all selling expenses except direct selling expenses," while, in fact, the respondent's PSPAs were direct selling expenses. 82 F.3d at 1051 (emphasis omitted). Torrington, however, does not stand for the proposition that Commerce must grant a PSPA for all discounts or even each selling expense that might be qualified as a direct selling expense.

While the underlying discounts reported by NTN may indeed vary with the quantity sold, NTN did not report this information to Commerce in the required form, namely, on a transaction-specific basis and as a fixed and constant percentage of sales. Consequently, Commerce reasonably could deny an adjustment if NTN has failed to properly allocate or report expenses.

NTN's second argument is equally unpersuasive. NTN states that an adjustment should be allowed because the discounts at issue "constitute a discount which is a fixed and constant percentage of sales." See NTN Mem. at 13. As Commerce correctly observes, every discount granted

may be expressed as a "fixed and constant" percentage of sales by mere mathematical operation (i.e., dividing the discount by the number of sales). In focusing upon the "fixed and constant" percentage of sales, Commerce is not inquiring whether a party has the ability to perform this simple calculation, but, rather, whether the party is granting the discount itself as a fixed and constant percentage of the sales to which it pertains. As explained by [Commerce], "[i]f a respondent grants and reports a PSPA as a fixed percentage of the sales to which it pertains, the fact that this pool of sales may include non-scope merchandise does not distort the amount of the expense the respondent granted and reported on sales of subject merchandise because the same adjustment percentage applied to both scope and non-scope merchandise."

Def.'s Resp. NTN at 36 (quoting Final Results, 63 Fed.Reg. at 20,603, and relying on Final Results of Antidumping Duty Administrative Reviews and Revocation in Part of an Antidumping Finding on Tapered Roller Bearings and Parts Thereof, Finished and Unfinished, From Japan and Tapered Roller Bearings, Four Inches or Less in Outside Diameter, and Components Thereof, From Japan, 61 Fed.Reg. at 57,642 (emphasis omitted)).

Because NTN's reporting method was not equivalent to transaction-specific reporting, "allocated price adjustments [may] have the effect of distorting individual prices by diluting the discounts or rebates received on some sales, inflating them on other sales, and attributing them to still other sales that did not actually receive any at all." NSK Ltd. v. Koyo Seiko Co., 190 F.3d 1321, 1329 (Fed.Cir. 1999) (internal quotation omitted). Therefore, the Court holds that Commerce acted reasonably when it disallowed any adjustments to FMV for NTN's discounts under the mandate of 19 U.S.C. § 1677b(a)(1)(A).

XIII. COMMERCE'S REALLOCATION OF SELLING EXPENSES WITHOUT REGARD TO LEVEL OF TRADE AND DENIAL OF A LEVEL OF TRADE ADJUSTMENT

A. Background

The pre-URAA law did not contain a specific provision concerning adjustments for differences in levels of trade ("LOTs"). See, e.g., NEC Home Elecs., 54 F.3d at 739. Commerce, however, provided that it would normally calculate FMV and United States price at the same commercial LOT. See 19 C.F.R. § 353.58 (1994). If such sales were insufficient in number to permit an adequate comparison, Commerce would calculate FMV based upon such or similar sales at the most comparable LOT to the LOT in the United States market "and make appropriate adjustments for differences affecting price comparability." Id.

In the Preliminary Results, Commerce: (1) accepted NTN's proposed allocation of selling expenses by LOT; and (2) granted NTN an LOT adjustment. See 61 Fed. Reg. at 25,204. However, in the Final Results, 63 Fed.Reg. at 20,608-09, Commerce reexamined the record, reallocated NTN's selling expenses without regard to LOT, and denied NTN an LOT adjustment, operating under the precedent set by Timken Co. v. United States ("Timken"), 20 CIT 645, 930 F. Supp. 621 (1996).

B. Contentions of the Parties

NTN contends that Commerce erroneously reallocated NTN's selling expenses without regard to LOT and eliminated the LOT adjustment granted to NTN in the Preliminary Results, 61 Fed.Reg. at 25,204. See NTN's Mem. at 14-16. NTN maintains that it provided "sufficient information to [Commerce] ... to convince [Commerce] that NTN's LOT-specific reporting methodology was accurate." NTN's Reply at 8.

Commerce asserts that it acted reasonably and properly when it followed the precedent set by Timken, 20 CIT 645, 930 F. Supp. 621. Timken supports Commerce's assertion and points out that: (1) Commerce properly disallowed a home market LOT adjustment because NTN failed to demonstrate that NTN incurred different selling expenses at the different LOTs in the home market due to the differences in selling to the different LOTs, see Timken's Resp. at 21-23; (2) Commerce properly rejected NTN's LOT-based expense allocation method and reasonably reallocated NTN's home-market selling expenses without regard to LOT, see id. at 23-25; and (3) Commerce properly reallocated NTN's United States selling expenses without regard to LOT. See id. at 25-26.

C. Analysis

In the Final Results, Commerce examined Timken, 20 CIT 645, 930 F. Supp. 621,^[10] and stated that the fact that a respondent merely makes allocations according to LOT does not demonstrate that the relevant expenses demonstrably varied based upon LOT. See 63 Fed.Reg. at 20,608. Commerce, therefore, concluded that "in order to determine if a respondent's expenses demonstrably varied according to LOT, additional narrative and quantitative evidence must exist which demonstrates that the respondent either performed different activities/functions or performed activities/functions to a different degree when selling to each LOT[,] such that the amount of expenses incurred for the sale of the identical merchandise to different LOTs would vary." Final Results, 63 Fed.Reg. at 20,608.

Examining the record, Commerce determined that NTN did not provide the necessary narrative and quantitative evidence. See id. Commerce concluded that "NTN's sole support for [NTN's] LOT-specific allocations is the [very fact of the] allocations themselves." Id. Because the mere fact of these allocations did not give Commerce sufficient evidence that NTN's home market selling expenses varied by LOT, Commerce reallocated these expenses without regard to LOT. Id.

Commerce's decision to re[]allocate NTN's selling expenses was not based upon inaccuracies discovered at verification but, rather, upon Commerce's determination that mere allocation of expenses by level of trade does not demonstrate that [NTN] actually performed different selling functions at different levels of trade.^[11]

Def.'s Resp. NTN at 40.

NTN's failure to demonstrate that these different functions existed provided reasonable grounds for Commerce to proceed with the reallocation. The mere fact of Commerce's reallocation of expenses does not make Commerce's decision in the Final Results, 63 Fed.Reg. at 20,608-09, invalid. See NSK Ltd. v. United States, 21 CIT 617, 635-36, 969 F. Supp. 34, 54 (1997) ("The evidence on the record does not establish differences in selling expenses at [different LOTs] because NTN's allocation methodology ... does not reasonably quantify the expenses incurred at each [LOT]"); Timken, 20 CIT 645, 930 F. Supp. 621; NTN Bearing Corp. of Am. v. United States, 19 CIT 1221, 1234, 905 F. Supp. 1083, 1094-95 (1995) ("Although NTN purports to show that it incurred different selling expenses for different trade levels, the record demonstrates that NTN's allocation methodology does not reasonably quantify the expenses incurred at each level of trade" (citation omitted)).

After having determined that reallocation was necessary, Commerce properly denied NTN an LOT adjustment. Commerce's preliminary decision to grant the adjustment was premised upon Commerce's preliminary acceptance of NTN's allocations, see Preliminary Result, 61 Fed.Reg. at 25,204. Once Commerce has determined that NTN's allocations were inapplicable, the basis for an LOT adjustment was not present. See Final Results, 63 Fed.Reg. at 20,609; accord 19 C.F.R. § 353.54 (1994) ("Any interested party that claims an adjustment [under a particular federal regulation] must establish the claim to the satisfaction of [Commerce]"). Because NTN claimed an adjustment pursuant to the particular regulation, namely, the provision authorizing LOT adjustments, NTN was obligated to provide Commerce with sufficient information in support of NTN's claim. The Court concludes that NTN's inability to do so provided Commerce with a reasonable basis to deny NTN an LOT adjustment.

XIV. COMMERCE'S APPLICATION OF REVISED INDIRECT SELLING RATIO TO PURCHASE PRICE SALES

Timken argues that, notwithstanding Commerce's statement in Commerce's NTN analysis that Commerce intended to apply the revised indirect selling ratio only to NTN's purchase price sales, Commerce's computer program erroneously applied this change to all of NTN's sales in the United States sales database (purchase price and ESP). See Timken's Reply Commerce at 5. Commerce states that: (1) while it was Commerce's intent to apply the revised indirect selling expense ratio only to NTN's purchase sales, Commerce incorrectly entered certain language outside the proper location; and (2) the revised ratio was incorrectly applied to all of NTN's sales. See Def.'s Resp. Timken at 17. Therefore, this issue is remanded to Commerce to correct the computer program error with respect to NTN's sales.

CONCLUSION

The case is remanded to Commerce to: (1) recalculate Koyo's CV profit so that Koyo's home market movement expenses are deducted from Koyo's net home market price; (2) recalculate Koyo's marine insurance charges using the correct factor indicated by Koyo; (3) recalculate Koyo's CV using commission factor provided by Koyo; (4) recalculate CV direct selling expenses relying on the factor indicated in Koyo's questionnaire response; (5) enter necessary corrections and recalculate pertinent parts of Commerce's determination with respect to Koyo's imports; (6) recalculate Koyo's United States inventory carrying costs for final product by applying the appropriate inventory carrying costs factor reported by Koyo to the landed cost for the 1992/93 POR and to the COM for the 1993/94 POR; (7) apply the correct G & A expenses factor in Commerce's calculation of NSK's COP; and (8) correct the computer program error with respect to NTN's sales. Commerce's final determination is affirmed in all other respects.

NOTES

[1] The name of the document is given in accordance with the name provided by the parties in their briefs. The document is: (1) omitted from the print in Vol. 41 of the Federal Register; and (2) unavailable in electronic legal databases.

[2] Since the administrative reviews at issue were initiated before January 1, 1995, the applicable law is the antidumping statute as it existed prior to the amendments made by the Uruguay Round Agreements Act, Pub.L. No. 103-465, 108 Stat. 4809 (1994). See Torrington Co. v. United States, 68 F.3d 1347, 1352 (Fed.Cir.1995).

[3] The Court's ruling on the issue in Koyo Seiko Co., 16 CIT 539, 796 F. Supp. 1526, is irrelevant to the futility exception to administrative exhaustion. "As a general rule, courts may ``refuse to require administrative exhaustion when resort to the administrative remedy would be futile....'" Koyo Seiko, Co., 16 CIT at 544, 796 F. Supp. at 1531 (quoting Asociacion Colombiana de Exportadores de Flores, 916 F.2d at 1575, quoting, in turn, Bendure v. United States, 213 Ct. Cl. 633, 554 F.2d 427, 431 (1977), emphasis supplied); see also United States Cane Sugar Refiners' Ass'n, 3 CIT at 201, 544 F. Supp. at 887.

[4] If Commerce's circumcised test for collateral estoppel was operable, the judicial docket would indeed be greatly relieved, although to a detriment of parties stripped from their constitutional right to a day in court.

[5] A later clarification came in the Statement of Administrative Action ("SAA"), accompanying the URAA. SAA provides that, aside from §§ 1677b(b)(1) and (f)(2) (1994) transactions,

Commerce may consider other types of sales or transactions to be outside the ordinary course of trade when such sales or transactions have characteristics that are not ordinary as compared to sales or transactions generally made in the same market. Examples of such sales or transactions include merchandise produced according to unusual product specifications, merchandise sold at aberrational prices, or merchandise sold pursuant to unusual terms of sale. As under existing law, [the pertinent] section ... does not establish an exhaustive list, but the Administration intends that Commerce will interpret [this] section ... in a manner which will avoid basing normal value on sales which are extraordinary for the market in question, particularly when the use of such sales would lead to irrational or unrepresentative results.

H.R. Doc. 103-316, at 834 (1994), reprinted in 1994 U.S.C.C.A.N. 4171 (emphasis supplied).

[6] Because neither the statutory language nor the legislative history explicitly establishes what is considered to be outside the "ordinary course of trade," the Court assesses whether Commerce's interpretation or application is reasonable and in accordance with the legislative purpose on a case-by-case basis. See Chevron, 467 U.S. at 843, 104 S. Ct. 2778. "In determining whether Commerce's interpretation is reasonable, the Court considers, among other factors, the express terms of the provisions at issue, the objectives of those provisions and the objectives of the antidumping scheme as a whole." Mitsubishi, 22 CIT at 545, 15 F. Supp. 2d at 813. The purpose of the ordinary course of trade provision is "to prevent dumping margins from being based on sales which are not representative" of the home market. Monsanto Co., 12 CIT at 940, 698 F. Supp. at 278.

[7] Indeed, while the mere fact that NTN has made sample sales to customers for a number of years does not, alone, prove that the sales were in the ordinary course of trade, the Court finds the fact relevant because a customer rarely needs to be familiarized and re-familiarized and yet re-familiarized with the product, especially if the customer has already made a purchase at least once and found the product to be suitable for the customer's needs.

[8] The Court disagrees with Timken's contentions that these two points could be dispositive of the issue. Timken asserts that

[a]llowing selling expenses to be reduced (with consequent increase in [United States] prices and reduction of margins of dumping) by amounts of alleged interest on antidumping duty deposits would provide an incentive to respondents to prolong litigation over entries so as to avoid actual payment of duties.

Timken's Resp. at 15.

The Court is not convinced by Timken's argument. A defeat in litigation implies the necessity of eventual payment of the duties due, and the mere possibility of "opportunity use," possibly resulting in collection of interest on the funds available calls for an argument seeking collection of duties together with a prevailing interest rate rather than for the "anti-incentive" argument fostered by Timken.

Next, not only the record contains NTN's claim for the amount of imputed interest attributable to NTN's antidumping duty deposits (the claim that, under the administrative scheme, is subject to verification by Commerce rather than Timken), but also the factual inquiry of whether NTN actually incurred interest expenses attributable to financing payment is secondary to the threshold legal inquiry if an adjustment should be allowed for such expenses.

[9] While "``an agency does not act rationally when it chooses and implements one policy and decides to consider the merits of a potentially inconsistent policy in the very near future,'" Transcom, Inc. v. United States, 24 CIT ___, ___, 123 F. Supp. 2d 1372, 1381 (2000) (quoting ITT World Communications, Inc. v. FCC, 725 F.2d 732, 754 (D.C.Cir. 1984)), Commerce should be able to rely on its "unique expertise and policy-making prerogatives." Southern Cal. Edison Co. v. United States, 226 F.3d 1349, 1357 (Fed.Cir.2000). "``The power of an administrative agency to administer a congressionally created ... program necessarily requires the formulation of policy....'" Chevron, 467 U.S. at 843, 104 S. Ct. 2778 (quoting Morton v. Ruiz, 415 U.S. 199, 231, 94 S. Ct. 1055, 39 L. Ed. 2d 270 (1974)).

An agency decision involving the meaning or reach of a statute that reconciles conflicting policies "``represents a reasonable accommodation of conflicting policies that were committed to the agency's care by the statute, [and a reviewing court] should not disturb [the agency decision] unless it appears from the statute or its legislative history that the accommodation is not one that Congress would have sanctioned.'" Id. at 845, 104 S. Ct. 2778 (quoting United States v. Shimer, 367 U.S. 374, 382-83, 81 S. Ct. 1554, 6 L. Ed. 2d 908 (1961)). Furthermore, an agency must be allowed to assess the wisdom of its policy on a continuing basis. Under the Chevron regime, agency discretion to reconsider policies is inalienable. See id. at 843, 104 S. Ct. 2778. Any assumption that Congress intended to freeze an administrative interpretation of a statute would be entirely contrary to the concept of Chevron which assumes and approves of the ability of administrative agencies to change their interpretations. See, e.g., Maier, P.E. v. United States EPA, 114 F.3d 1032, 1043 (10th Cir.1997), J.L. v. Social Sec. Admin., 971 F.2d 260, 265 (9th Cir.1992), Saco Defense Sys. Div., Maremont Corp. v. Weinberger, 606 F. Supp. 446, 450-51 (D.Me.1985). In sum, underlying agency interpretative policies "are given controlling weight unless they are arbitrary, capricious, or manifestly contrary to the statute." Chevron, 467 U.S. at 844, 104 S. Ct. 2778.

[10] In Timken, after reviewing the pertinent statutory and regulatory criteria governing LOT adjustments, the Court noted that the statute conferred upon Commerce broad discretion in determining whether an LOT adjustment is warranted, but that such discretion is not without limits "as the statute requires a direct relationship between the expenses and the relevant sales." 20 CIT at 653, 930 F.Supp. at 628 (citing Smith-Corona Group, 713 F.2d at 1575, rev'd on other grounds, Consumer Prods. Div., SCM Corp. v. Silver Reed America, 753 F.2d 1033 (Fed. Cir.1985)). "The issue, therefore, is whether the reported expenses demonstrably vary according to levels of trade." Id. (emphasis supplied).

[11] Commerce stated that it did "not consider NTN's [relevant] exhibit [on] home market indirect selling expense differentials as a reliable basis for a[n] LOT adjustment." Final Results, 63 Fed.Reg. at 20,609. NTN points out that Commerce, however, "verified NTN's data, and had the opportunity to verify all of the information on [the particular] exhibit." NTN's Reply at 8; see also NTN's Mem. at 14. NTN misses the point fostered by Commerce. In the Final Results, 63 Fed.Reg. at 20,608-09, Commerce did not assert that the verified information suddenly became wrongful, rather Commerce deemed it to be insufficient for the lack of "necessary narrative and quantitative evidence," Def.'s Resp. NTN at 39, that is, the exhibit was deemed unreliable because of insufficiency of its content and not because of the invalidity of data.