Timken Co.v.U.S.

United States Court of Appeals, Federal CircuitJan 16, 2004
354 F.3d 1334 (Fed. Cir. 2004)

Nos. 03-1098, 03-1238.

DECIDED: January 16, 2004.

Appeal from the United States Court of International Trade, Donald C. Pogue, J.

William A. Fennell, Stewart and Stewart, of Washington, DC, argued for plaintiff-cross appellant. With him on the brief were Terence P. Stewart and J. Daniel Stirk. Of counsel were Patrick J. McDonough and Sarah V. Stewart.

Neil R. Ellis, Sidley Austin Brown Wood LLP, of Washington, DC, argued for defendants-appellant. With him on the brief was Neil C. Pratt.

Claudia Burke, Attorney, Civil Division, Commercial Litigation Branch, United States Department of Justice, of Washington, DC, argued for defendant-appellee. With her on the brief was David M. Cohen, Director.

Before NEWMAN, BRYSON, and PROST, Circuit Judges.



PROST, Circuit Judge.

Koyo Seiko Co., Ltd. and Koyo Corporation of U.S.A. (collectively, "Koyo") appeal the decision of the United States Court of International Trade in Timken Co. v. United States, 240 F.Supp.2d 1228 (CIT 2002), holding that the United States Department of Commerce ("Commerce") properly "zeroed" any negative dumping margins in calculating the weighted-average dumping margin applied to imports of Koyo's tapered roller bearings ("TRBs") from Japan. The Timken Company ("Timken") cross appeals arguing that Commerce, in calculating Koyo's constructed export price ("CEP"), improperly applied the adverse-facts-available rate to the entered value rather than the sales value of Koyo's TRBs. Because we agree with the Court of International Trade that Commerce properly resolved both issues, we affirm.

I

Commerce issues antidumping duty orders for imported merchandise that is sold in the United States below its fair value and materially injures or threatens to injure a domestic industry. See 19 U.S.C. § 1673e (2000); 19 U.S.C. § 1673d. Specifically, Commerce determines the antidumping duties by first calculating the "dumping margin," which is "the amount by which the normal value exceeds the export price or constructed export price of the subject merchandise." Id. § 1677(35)(A); see also Koyo Seiko Co. v. United States, 258 F.3d 1340, 1342 (Fed. Cir. 2001). Commerce uses a CEP if, "before or after the time of importation, the first sale to an unaffiliated person is made by (or for the account of) the producer or exporter or by a seller in the United States who is affiliated with the producer or exporter." Uruguay Round Agreements Act, Statement of Administrative Action, H.R. Doc. No. 103-826, at 822 (1994), reprinted in 1994 U.S.C.C.A.N. 4040 (" SAA"). Various adjustments may be made to CEP, including reduction by "the cost of any further manufacture or assembly" in the United States. See 19 U.S.C. § 1677a(d)(2).

The SAA is "an authoritative expression by the United States concerning the interpretation and application of the Uruguay Round Agreements and this Act in any judicial proceeding in which a question arises concerning such interpretation or application." 19 U.S.C. § 3512(d).

After calculating the dumping margins on the individual U.S. transactions subject to review, Commerce calculates the weighted-average dumping margin "by dividing the aggregate dumping margins determined for a specific exporter or producer by the aggregate . . . constructed export prices of such exporter or producer." Id. § 1677(35)(B); see also Koyo Seiko, 258 F.3d at 1342-43. When calculating the weighted-average dumping margin, Commerce treats transactions that generate "negative" dumping margins (i.e., a dumping margin with a value less than zero) as if they were zero. See, e.g., Serampore Indus. Pvt. Ltd. v. Dep't of Commerce, 675 F.Supp. 1354, 1360-61 (Ct. Int'l Trade 1987). This practice is referred to as "zeroing." Finally, Commerce uses this weighted-average dumping margin to calculate the duties owed on an entry-by-entry basis. 19 U.S.C. § 1675(a)(2).

In this case Commerce initiated an administrative review of alleged sales of TRBs from Japan at less than fair value during the period of October 1, 1998, through September 30, 1999. Initiation of Antidumping and Countervailing Duty Administrative Reviews, 64 Fed. Reg. 67,846 (Dec. 3, 1999). Based on its review, Commerce calculated preliminary dumping margins of 17.94% on TRBs greater than four inches in diameter and 14.86% on TRBs less than four inches in diameter. 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 of Antidumping Duty Administrative Reviews, 65 Fed. Reg. 66,711 (Nov. 7, 2000) (" Preliminary Results"). Koyo exported both types of merchandise and thus became potentially liable for antidumping duties. Id. Both Koyo and Timken subsequently filed case and rebuttal briefs, with Koyo questioning, inter alia, Commerce's zeroing practice, and Timken questioning, inter alia, Commerce's application of the adverse-facts-available rate to the entered value rather than sales value of Koyo's TRBs.

Commerce provided a response in 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, 66 Fed. Reg. 15,078 (Dep't Commerce Mar. 15, 2001) (" Final Results") and the accompanying Issues and Decision Memorandum, P.R. Doc. No. 141 (Mar. 7, 2001) (" Decision Mem."). With respect to Koyo's challenge, Commerce noted that it properly calculated the weighted-average dumping margin by "zeroing" all negative-dumping-margin transactions. Commerce explained that its calculation methodology was derived from the explicit statutory language, and consistent with international obligations. See Decision Mem., cmt. 1, at 8. Consequently, it finalized Koyo's company-specific, weighted-average dumping margins. Final Results, 66 Fed. Reg. at 15,079. With respect to Timken's challenge to the application of the adverse inference to the entered value, Commerce found that Timken's suggested approach would be unduly punitive. Decision Mem. at 8. Thus, Commerce continued to apply the adverse inference to the entered value rather than the sales value of Koyo's TRBs. Id.

Based on existing information, Commerce, at an earlier stage in this proceeding, determined that the value added to Koyo's TRBs substantially exceeded the value of the subject merchandise. It also concluded that the methods for calculating CEP provided in § 1677a(e) would result in an unreliable surrogate value. Decision Mem. at 5-6. Accordingly, Commerce exercised its discretion to employ "any other reasonable basis" to calculate the margins for further-manufactured merchandise. Id. at 6. It chose to apply its standard methodology, as outlined in § 1677a(d)(2), and required Koyo to file a Section E questionnaire response detailing sales and cost information for its further-manufactured TRBs. Id. at 6-7. Koyo refused to comply, however, because it believed it qualified for application of the approach in 19 U.S.C. § 1677a(e). Id. at 2-4. Consequently, Commerce applied an adverse inference, as authorized by 19 U.S.C. § 1677e(b), and applied the adverse-facts-available rate to the total entered value of Koyo's TRBs. Id. at 8.

The Court of International Trade's review of Commerce's Final Results dealt with, inter alia: (1) Koyo's arguments that Commerce improperly "zeroed" negative dumping margins when calculating Koyo's weighted-average dumping margins, and (2) Timken's arguments that Commerce should have applied the adverse-facts-available rate to the sales value of Koyo's TRBs. See Timken, 240 F.Supp.2d at 1228. As to the first issue, Koyo primarily argued that Commerce's statutory interpretation was unreasonable in light of the World Trade Organization ("WTO") Appellate Body's interpretation of the Antidumping Duty Agreement ("ADA") in European Communities — Antidumping Duties on Imports of Cotton-Type Bed Linen from India, WT/DS/141/AB/R (Mar. 1, 2001) (" EC — Bed Linen"). Timken, 240 F.Supp.2d at 1242. Rejecting the government's threshold argument that § 3512(c) barred Koyo's appeal, the Court of International Trade went on to note that it had previously affirmed zeroing as a reasonable interpretation of 19 U.S.C. § 1673. Id. at 1243 (citing Serampore Indus., 675 F.Supp. at 1360-61 (Ct. Int'l Trade 1987); Bowe Passat Reinigungs-Und Waschereitechnik Gmbh v. United States, 926 F.Supp. 1138, 1150 (Ct. Int'l Trade 1996)). It also distinguished the EC — Bed Linen report because it (1) did not address the U.S. practice of zeroing, and (2) dealt with an antidumping investigation rather than an administrative review. Id. Given that the statute requires Commerce to calculate dumping duties on an entry-by-entry approach and that previous cases determined that zeroing is a reasonable practice under the statute, the Court of International Trade found Commerce's zeroing practice reasonable. Id.

Addressing Timken's arguments that Commerce should have applied the adverse-facts-available rate to the sales value of Koyo's TRBs, the Court of International Trade affirmed Commerce's approach. First, the court found it was consistent with the regulations, see 19 C.F.R. § 351.212(b)(1), and prior court holdings. Timken, 240 F.Supp.2d at 1233 (citing NTN Bearing Corp. of Am. v. United States, 186 F.Supp.2d 1257, 1315 (Ct. Int'l Trade 2002)). Second, according to the court, Commerce properly adhered to the overriding goal of accurately determining dumping margins, rather than creating an overly punitive result. Id. at 1234. Because the value of Koyo's TRBs increased substantially as a result of post-importation manufacturing, and because subjecting this increase to duties would have been contrary to the purpose of the anti-dumping statute, the court found that Commerce reasonably used the entered value. Id. at 1234-35. The Court of International Trade therefore upheld Commerce's application of the adverse-facts-available rate to Koyo's entered value; Commerce's decision was supported by substantial evidence and not contrary to law. Id.

Koyo filed a timely appeal; Timken filed a timely cross appeal. We have jurisdiction pursuant to 28 U.S.C. § 1295(a)(5).

II

In reviewing the Court of International Trade's review of a final determination issued by Commerce, we apply "anew" the statutorily-mandated standard of review. Accordingly, we will "uphold Commerce's determination unless it is `unsupported by substantial evidence on the record, or otherwise not in accordance with law.'" Micron Tech., Inc. v. United States, 117 F.3d 1386, 1393 (Fed. Cir. 1997) (quoting 19 U.S.C. § 1516a(b)(1)(B)(i)).

A

On appeal, Koyo argues that Commerce acted unreasonably in zeroing negative-margin transactions. It contends this practice violates 19 U.S.C. § 1677b(a), which calls for a "fair comparison" of export price ("EP") or CEP and normal value. Koyo relies on the WTO Appellate Body's decision in EC — Bed Linen, holding that the practice of zeroing when calculating dumping margins is not a "fair comparison" between EP or CEP and normal value. Thus, because § 1677b(a) specifically implements the "fair comparison" requirements of Article 2.4 of the ADA, and because Commerce refused to interpret the statute in a manner consistent with U.S. international obligations by ignoring the holding in EC — Bed Linen, Koyo asks that we find that Commerce acted unreasonably in zeroing negative-margin transactions.

Timken and the United States respond that the plain meaning of the antidumping statute calls for Commerce to zero negative-margin transactions, and that the legislative history confirms this reading. Alternatively, even assuming that the statute contains an ambiguous instruction, Timken and the United States argue that Commerce reasonably interpreted the statute and deserves deference. Addressing Koyo's arguments relying on EC — Bed Linen, Timken and the United States reiterate the distinctions between that case and this case articulated by the Court of International Trade. Finally, Timken and the United States argue that 19 U.S.C. § 3512(c) and the SAA bar Koyo from bringing this challenge, and that § 3533 precludes us from addressing the implementation of adverse WTO decisions.

We begin by addressing the government's argument that § 3512(c) bars Koyo from bringing this action. Section 3512(c) bars parties from bringing claims directly against the government on the ground that Commerce acted inconsistently with the Uruguay Round Agreements Act ("URAA"). 19 U.S.C. § 3512(c). As the Court of International Trade noted, however, Koyo brought this action under U.S. law under the assumption that it would be interpreted so as to avoid a conflict with international obligations. Timken, 240 F.Supp.2d at 1242. We agree and find that § 3512(c) does not prevent us from addressing Koyo's appeal.

On the merits of Koyo's appeal, we apply a familiar two-part inquiry to determine whether to sustain an agency's interpretation of the statutory scheme it is charged with administering. See Chevron, U.S.A., Inc. v. Natural Res. Def. Council, Inc., 467 U.S. 837, 842-43, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984). First, we determine "whether Congress has directly spoken to the precise question at issue. If the intent of Congress is clear, that is the end of the matter; for the court, as well as the agency, must give effect to the unambiguously expressed intent of Congress." Id. "[I]f the statute is silent or ambiguous with respect to the specific issue," however, "the question for the court is whether the agency's answer is based on a permissible construction of the statute." Id. at 843, 104 S.Ct. 2778. In this case, therefore, we must determine whether the statute unambiguously requires providing for zeroing negative margin transactions and, if not, whether Commerce reasonably interpreted the statute to so require.

Turning to the words of the statute, 19 U.S.C. § 1677(35)(A) defines "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). Timken and the United States both argue that the use of the word "exceeds" limits the definition of "dumping margin" to positive numbers. In support of their argument, they appropriately cite dictionaries. Pesquera Mares Australes Ltda. v. United States, 266 F.3d 1372, 1381 (Fed. Cir. 2001) ("In order to ascertain the established meaning of a [commonly used] term . . . it is appropriate to consult dictionaries."). These dictionaries define "exceeds" as "to be or go beyond (the given or supposed limit, measure, or quantity)," Webster's New Twentieth Century Dictionary of the English Language Unabridged 636 (2d ed. 1980), or "1. To be greater than; surpass. 2. To go beyond the limits of," The American Heritage College Dictionary 477 (3d 1993).

Recognizing this as a close question, we are reluctant to find these dictionary definitions so clear as to compel a finding that Congress expressly intended to require zeroing. Even using the above "greater than" definitions, the statute does not plainly require consideration of only those dumping margins with a positive value. At least in a mathematical context, "exceeds" does not unambiguously preclude the calculation of a negative dumping margin. We thus disagree with the government's position that Congress deliberately used the word "exceeds" to avoid the calculation of negative dumping margins, instead of using the more open-ended phrase "difference between." Rather, the word "exceeds" could arguably allow for negative dumping margins because it guides the manner in which to set up the mathematical equation — x "exceeds" y = x-y. Similarly, the words "difference between" could arguably be construed as calling for the absolute value of the difference, a positive number — the "difference between" x and y = |x-y|. Finally, reviewing the legislative history as a whole, we disagree with Timken's argument that the relevant statements, or lack thereof, conclusively demonstrate Congress's adoption of definitions specifically requiring zeroing. Accordingly, we conclude that Congress's use of the word "exceeds" does not unambiguously require that dumping margins be positive numbers.

Because we find that the statute does not directly speak to the issue of negative-value dumping margins, we evaluate whether Commerce's interpretation is based on a permissible statutory construction. Under this second step of the Chevron analysis, "[a]ny reasonable construction of the statute is a permissible construction." Torrington v. United States, 82 F.3d 1039, 1044 (Fed. Cir. 1996). "To survive judicial scrutiny, [Commerce's] construction need not be the only reasonable interpretation or even the most reasonable interpretation. . . . Rather, a court must defer to an agency's reasonable interpretation of a statute even if the court might have preferred another." Koyo Seiko Co. v. United States, 36 F.3d 1565, 1570 (Fed. Cir. 1994) (citing Zenith Radio Corp. v. United States, 437 U.S. 443, 450, 98 S.Ct. 2441, 57 L.Ed.2d 337 (1978)). Indeed, we have accorded particular deference to Commerce in antidumping determinations. See Smith-Corona Group v. United States, 713 F.2d 1568, 1571 (Fed. Cir. 1983) ("The Secretary has broad discretion in executing the [anti-dumping] law.").

We conclude Commerce based its zeroing practice on a reasonable interpretation of the statute. First, while the statutory definitions do not unambiguously preclude the existence of negative dumping margins, they do at a minimum allow for Commerce's construction. Basically, one number "exceeds" another if it is "greater than" the other, meaning it falls to the right of it on the number line. Here, because Commerce's zeroing practice is a reasonable interpretation of the statutory language, we do not question it in light of other reasonable possibilities.

Second, Commerce's methodology for calculating dumping margins makes practical sense. Commerce calculates dumping duties on an entry-by-entry basis. 19 U.S.C. § 1675(a)(2). Its practice of zeroing negative dumping margins comports with this approach. Borrowing Timken's example, suppose a foreign exporter sells the same product to two U.S. customers. The product has a normal value of $0.90, and is sold to the first customer for $1.00 and the second customer for $0.70. Calculated in accordance with § 1677(35)(A), the dumping margin for the first customer is zeroed (0.90-1.00 = -0.10 0) and for the second customer is 0.20 (0.90-0.70 = 0.20). Assuming sales of 1000 units to each customer, the first customer would not have to pay any dumping duties because it paid a price above normal value, and the second customer would have to pay $200 (1000 transactions × 0.20 dumping margin/transaction = 200) because it paid a price below normal value. This approach makes sense; it neutralizes dumped sales and has no effect on fair-value sales. On the other hand, the approach urged by Koyo, whereby Commerce could not zero negative transactions, would essentially require Commerce to grant the first customer a credit. In the absence of offsetting sales below fair market value, however, Commerce could potentially owe the first customer a payment — a result clearly not contemplated by the statutory scheme.

Third, we note that the Court of International Trade has specifically addressed Commerce's zeroing practice and found it reasonable and in accordance with the law. See Serampore, 675 F.Supp. at 1360-61; Bowe Passat, 926 F.Supp. at 1150. We find these decisions instructive for purposes of our analysis. Notably, both of these pre-URAA cases found zeroing to be a reasonable statutory interpretation given that it legitimately combats the problem of masked dumping, wherein certain profitable sales serve to "mask" sales at less than fair value. Serampore, 675 F.Supp. at 1360-61; Bowe Passat, 926 F.Supp. at 1150. Although Koyo argues that these decisions addressed the law as it existed prior to the adoption of the URAA's "fair comparison" requirements, we find this argument unpersuasive. Neither the pre-URAA nor the current definitions unambiguously address the practice of zeroing. The URAA-incorporated "fair comparison" requirement, which simply applies to the calculation of normal value under § 1677b(a), does not change this result. The § 1677(35) definitions do, however, lend support to Commerce's statutory construction (as discussed above), even when considered in light of this new "fair comparison" requirement. According Commerce its proper deference, we hold that it reasonably interpreted § 1677(35)(A) to allow for zeroing.

We turn now to Koyo's argument that Commerce unreasonably interpreted § 1677(35) to allow for zeroing, beginning with a review of the WTO's EC — Bed Linen report upon which Koyo so heavily relies. EC — Bed Linen dealt with a trade dispute between the European Community ("EC") and India. EC — Bed Linen, WT/DS141/AB/R ¶ 1. The EC, in investigating and levying duties on the importation of bed sheets from India, engaged in the practice of zeroing negative-margin transactions. See id. ¶ 8. India brought an action at the WTO, claiming that the EC practice of zeroing violated Article 2.4.2 of the ADA. See id. ¶ 1. The WTO Appellate Body evaluated the EC's practice in light of both this article and Article 2.4, which states that one must make a "fair comparison" between the EP or CEP and the normal value when calculating dumping margins. Id. ¶¶ 46-66. It took the "fair comparison" language to mean that all transactions must be accounted for, whether creating positive or negative dumping margins. Id. ¶ 55. Thus, because it was unfair to exclude negative-margin transactions, the WTO Appellate Body concluded that the EC practice of zeroing during an antidumping investigation was impermissible under the ADA. Id. ¶ 66.

In light of the decision in EC — Bed Linen, Koyo asks us to interpret the "fair comparison" language in the U.S. antidumping statute in a manner consistent with U.S. international obligations, thereby adopting the holding in EC — Bed Linen and finding Commerce's zeroing practice an unreasonable statutory interpretation. The crux of its argument hinges on the Charming Betsy canon of claim construction, according to which courts should interpret U.S. law, whenever possible, in a manner consistent with U.S. international obligations. Murray v. Charming Betsy, 6 U.S. (2 Cranch) 64, 118, 2 L.Ed. 208 (1804); see also Fed.-Mogul Corp. v. United States, 63 F.3d 1572, 1581 (Fed. Cir. 1995) (noting that trade laws are not exempt from the Charming Betsy principle); Luigi Bormioli Corp. v. United States, 304 F.3d 1362, 1368 (Fed. Cir. 2002) ("[T]he statute must be interpreted to be consistent with [international] obligations, absent contrary indications in the statutory language or its legislative history."). Koyo secondarily relies on its claim that the EC and United States have functionally identical practices that demand similar treatment. Because Commerce has the authority to bring its practice into conformity with U.S. international obligations, Koyo contends, we should find that it was unreasonable for Commerce to continue interpreting the antidumping statute in a non-conforming manner by allowing zeroing.

Koyo's reliance on the "fair comparison" language is misplaced. Section 1677b(a) requires a "fair comparison" as follows: "a fair comparison shall be made between the export price or constructed export price and normal value." 19 U.S.C. § 1677b(a). Thus, "in order to achieve a fair comparison with the export price or constructed export price," id., the statute particularly sets out how to calculate "normal value." See id. § 1677b(a)(1)-(8). We agree with the government's position that this is an exhaustive list, and that the "fair comparison" requirement upon which Koyo now relies is specifically defined in the normal-value-calculation instructions. As such, the "fair comparison" requirement of § 1677b(a) does not impose any requirements for calculating normal value beyond those explicitly established in the statute and does not carry over to create additional limitations on the calculation of dumping margins. The SAA supports our conclusion. SAA, supra, at 820 ("To achieve such a [fair] comparison, section 773 [§ 1677b] provides for the selection and adjustment of normal value to avoid or adjust for differences between sales which affect price comparability."). This court has also previously recognized that the explicit statutory adjustments help make a "fair, `apples-to-apples' comparison" between normal value and EP or CEP. Micron Tech., Inc. v. United States, 243 F.3d 1301, 1313 (Fed. Cir. 2001) (quoting Torrington Co. v. United States, 68 F.3d 1347, 1352 (Fed. Cir. 1995)).

Further, even assuming that the "fair comparison" language in § 1677b(a) did not limit the analysis to calculating normal value using the statutorily-prescribed adjustments, we would nonetheless find Commerce's continued practice of zeroing reasonable in light of EC — Bed Linen. As Koyo acknowledges, the decision is not binding on the United States, much less this court. While Koyo relies on EC — Bed Linen for its persuasive value in an effort to convince us of the unreasonableness of Commerce's zeroing practice, we do not find it sufficiently persuasive to find Commerce's practice unreasonable. In light of the fact that Commerce's "longstanding and consistent administrative interpretation is entitled to considerable weight," Zenith, 437 U.S. at 450, 98 S.Ct. 2441, we refuse to overturn the zeroing practice based on EC — Bed Linen.

Finally, we disagree with Koyo's assertion that the Court of International Trade erroneously distinguished EC — Bed Linen. As discussed above, the court distinguished EC — Bed Linen for two reasons: (1) it did not involve the United States, and (2) it dealt with an antidumping investigation as opposed to an administrative review of dumping. We agree with the Court of International Trade and find the absence of the United States as a party to be an important distinction.

As to the difference between an investigation and an administrative review, Koyo contends that the distinction is irrelevant — Article 2.4.2 and Article 9.3.1, dealing with investigations and administrative reviews respectively, should be read together. In addition, Koyo argues, the "fair comparison" requirements of § 1677b(a) do not differ depending on the nature of the action. We disagree and find that the Court of International Trade properly distinguished the current case over EC — Bed Linen.

Because we find Commerce's zeroing practice to be a reasonable interpretation of the statute, even in light of the decision in EC — Bed Linen, we do not address the government's or Timken's arguments that this court is statutorily precluded from reconciling the two under 19 U.S.C § 3533.

B

On cross appeal, Timken contends that Commerce should have applied the adverse-facts-available rate to the sales value of Koyo's TRBs rather than the entered value. According to Timken, Commerce failed to effectuate the statutory purpose — to provide an incentive to comply — by applying an insufficiently adverse inference. Thus, Timken argues, relying on Kawasaki Steel Corp. v. United States, 110 F.Supp.2d 1029, 1041 (Ct. Int'l Trade 2000), that Koyo improperly gained a more favorable result by refusing to supply the Section E response.

In response, Koyo and the United States argue that Commerce reasonably applied the adverse-facts-available rate to the entered value of Koyo's TRBs. According to Koyo, using the entered value was reasonable because of the significant amount of value added in the United States — on average approximately 25% of the sales value consisted of the entered value — for which Koyo should not have to pay dumping duties. By choosing the entered value as the more reasonable surrogate, Koyo claims, Commerce properly balanced the statutory objectives of finding an accurate dumping margin and inducing compliance.

Again we begin with the relevant law. Commerce has stated that it "normally will calculate the assessment rate by dividing the dumping margin found on the subject merchandise examined by the entered value of such merchandise for normal customs duty purposes." 19 C.F.R. § 351.212(b)(1) (emphasis added). In addition, the statute gives Commerce the discretion to apply an adverse inference if, as here, a party fails to comply with requests for information. 19 U.S.C. § 1677e(b); see also SAA, supra, at 870 ("Where a party has not cooperated, Commerce . . . may employ adverse inferences about the missing information to ensure that the party does not obtain a more favorable result by failing to cooperate than if it had cooperated fully."). In keeping with these directives, the Court of International Trade has previously held that CEP can be calculated by applying adverse facts available to a party's entered value when there is further manufacturing. See NTN Bearing Corp. of Am. v. United States, 186 F.Supp.2d 1257, 1315 (2002). Finally, in selecting a reasonably adverse facts-available rate, Commerce must balance the statutory objectives of finding an accurate dumping margin and inducing compliance, rather than creating an overly punitive result. See F.LLI De Cecco Di Filippo Fara S. Martino S.p.A. v. United States, 216 F.3d 1027, 1032 (Fed. Cir. 2000) (" De Cecco").

We conclude that Commerce's decision to use the entered rather than sales value was supported by substantial evidence and supported by law. First, we note that Commerce's methodology was consistent with the regulation, 19 C.F.R. § 351.212(b)(1), as well as the Court of International Trade's decision in NTN Bearing. Moreover, under the law, Commerce had no reason to alter its methodology to induce Koyo's compliance.

Second, Commerce has discretion to apply adverse inferences. As both Commerce and the Court of International Trade noted, Commerce is not statutorily required to select a method that is the "most" or "more" reasonably adverse. Decision Mem. at 8; Timken, 240 F.Supp.2d at 1234. Here, Commerce applied the highest adverse-facts-available rate ever applied in an investigation of Koyo's dumping — 41.04%. By using this high rate, Commerce reasonably attempted to induce Koyo's compliance with the Section E request. But in applying this rate to the entered value rather than the sales value, Commerce appropriately balanced the goal of accuracy against the threat of an unduly punitive result. See Decision Mem. at 8. Contrary to Timken's arguments, this demonstrates Commerce's adherence to the principles in Kawasaki. Given the significant amount of post-importation manufacturing that took place — the entered value represented on average 25% of the sales value — we find that Commerce's decision to use the entry value was in keeping with the statutory goals highlighted in De Cecco. Moreover, as the Court of International Trade noted, this value increase should not be subject to duties because it occurs after importation. To do so would be contrary to the inherent purpose of the antidumping statute — to prevent below-fair-market-value imports.

Timken challenges Commerce's conclusion that Timken's approach would be unduly punitive. In particular, Timken argues that Commerce erroneously found the substantial increase in value attributed to post-importation manufacturing necessarily implies that applying the adverse-facts-available rate to the sales value creates a punitive result. According to Timken, the information regarding entered and sales values did not in fact provide Commerce with sufficient information about production costs or dumping margins to support that conclusion.

We disagree, and defer to Commerce's analysis in light of its extensive experience with and knowledge of Koyo's further-manufactured sales and the value added in the United States following importation. See Decision Mem. at 6. Despite Koyo's failure to comply with the Section E request, Commerce had other information from Koyo demonstrating that the value added in the United States substantially exceeded the entry value. As the Court of International Trade also noted, Commerce is "in the best position, based on its expert knowledge of the market and the individual respondent, to select adverse facts that will create the proper deterrent to non-cooperation with its investigations and assure a reasonable margin." De Cecco, 216 F.3d at 1032. We similarly find that Commerce's expertise with and knowledge of Koyo's further manufacturing allowed it to select adverse facts to effectuate the statutory goals.

Based on the above, we find that Commerce reasonably applied the adverse-facts-available rate to the entered value. We therefore uphold the Court of International Trade's finding that "Commerce's application of adverse facts available to Koyo's entered value as supported by substantial evidence and in accordance with law." Timken, 240 F.Supp.2d at 1235.

CONCLUSION

For the foregoing reasons, we affirm the Court of International Trade's conclusion that Commerce's final results are supported by substantial evidence and otherwise in accordance with law.

AFFIRMED.