Psc Vsmpo-Avismo Corp. v. United States

688 F.3d 751, 2012 WL 3055876, 34 I.T.R.D. (BNA) 1737, 2012 U.S. App. LEXIS 15638

• Court: Court of Appeals for the Federal Circuit
• Decided: July 27, 2012
• Docket: 2011-1370, 2011-1395
• Status: Published

Opinion

SCHALL, Circuit Judge.

This is an international trade case. It arises out of the Department of Commerce’s (“Commerce”) 2006/2007 administrative review of the antidumping duty order on magnesium metal from the Russian Federation. See Notice of Antidumping Duty Order: Magnesium Metal from the Russian Federation, 70 Fed. Reg. 19,-930 (Dep’t of Commerce Apr. 15, 2005) (“Antidumping Order”). Following the review, Commerce imposed an antidumping duty of 15.77 percent on U.S. imports of magnesium metal exported to the United States from the Russian Federation by PSC VSMPO-Avisma Corporation (“Avis-ma”), a manufacturer of magnesium and titanium sponge. Magnesium Metal from the Russian Federation: Final Results of Antidumping Duty Administrative Review, 73 Fed. Reg. 52,642 (Dep’t of Commerce Sept. 10, 2008) (“Final Results ”); Issues and Decision Memorandum for the Anti-dumping Duty Administrative Review of Magnesium Metal from the Russian Federation (Sept. 2, 2008) (“Final Results Memo ”). Commerce imposed this anti-dumping margin after calculating the normal value of Avisma’s sales of magnesium based upon constructed value.

Avisma and its U.S. subsidiary, VSMPO Tirus, U.S. Inc. (“Tirus”), challenged the Final Results in the United States Court of International Trade (“Trade Court”). 1 In doing so, Avisma also challenged Commerce’s decision during the administrative review to exclude as untimely, and not consider, the affidavit of its accounting expert, Dr. George Foster (“Foster Affidavit ”). US Magnesium LLC (“US Magnesium”), a U.S. producer of magnesium, intervened in the proceedings as a defendant. On October 20, 2009, without reaching the merits, the Trade Court remanded the case to Commerce with the instruction that Commerce consider the Foster Affidavit. PSC VSMPO-AVISMA Corp. v. United States, No. 08-00321, 2009 WL 3423021 (Ct. Int’l Trade Oct. 20, 2009) (“Avisma I”). On remand, Commerce considered the Foster Affidavit but declined to alter the determination in the Final Results. Results of Redetermination Pursuant to Remand, PSC VSMPO-Avisma Corp. v. United States, (Mar. 30, 2010) (“First Remand Determination”). On August 17, 2010, the Trade Court held that the Final Results and First Remand Determination were not in accordance with law and not supported in the record. PSC VSMPO-Avisma Corp. v. United States, 724 F.Supp.2d 1308 (Ct. Int’l Trade 2010) (“Avisma II”). The court based its decision on its conclusion that, when determining Avisma’s magnesium production costs for purposes of calculating the constructed value of Avisma’s magnesium, Commerce was required to take into account Avisma’s entire production process, which includes titanium, as well as magnesium. Id. at 1316. In the Final Results and First Remand Determination, Commerce only focused on Avisma’s magnesium production process. The court remanded to Commerce for further proceedings. Id.

In its second remand determination, in accordance with Avisma II, Commerce determined the constructed value of Avisma’s magnesium by taking into account Avis-ma’s entire production process. Results of Redetermination Pursuant to Remand, PSC VSMPO-Avisma Corp. v. United States, (Nov. 22, 2010) (“Second Remand Determination ”). 2 This resulted in an antidumping duty of 8.51 percent for Avisma’s magnesium. Id. at 15. On March 1, 2011, the Trade Court sustained Commerce’s Second Remand Determination and issued final judgment accordingly. PSC VSMPO-AVISMA Corp. v. United States, No. 08-00321, 2011 WL 769998 (Ct. Int’l Trade Mar. 1, 2011) (“Avisma III ”).

US Magnesium now appeals the decision of the Trade Court. It contends that the court erred (1) in requiring Commerce to consider the Foster Affidavit; and (2) in rejecting the Final Results. For its part, Avisma cross-appeals part of the decision in Avisma III; it contends that the Trade Court erred in affirming Commerce’s selection of the cost database it used in the Second Remand Determination. The United States urges affirmance of the final judgment in Avisma III.

For the reasons set forth below, we hold that the Trade Court erred in requiring Commerce to consider the Foster Affidavit. We also hold that the court erred in rejecting the Final Results. Accordingly, we reverse Avisma III and remand the case to the Trade Court, which is instructed to enter judgment reinstating the Final Restilts. Because of our disposition of the case, it is not necessary for us to reach Avisma’s cross-appeal.

Background

I.

A brief description of Avisma’s magnesium and titanium sponge production processes is necessary in order to understand the issues presented in this appeal. At its facility located in the city of Berezniki in Perm Krai, Russia, Avisma produces magnesium and titanium sponge, along with other products. Final Results Memo at 2. In the magnesium production process, enriched earnallite is first refined through dehydration and then through electrolysis to produce raw magnesium and chlorine gas. 3 Id. Most of the resultant raw magnesium is then processed into pure and alloyed magnesium, which is the subject merchandise of the antidumping order. Id. However, a portion of the raw magnesium is used in the titanium sponge production process. Id. Some of the chlorine gas produced in the earnallite refinement process also is used in the production of titanium sponge. Id.

The titanium sponge production process, which takes place at OPU-3, uses ilmenite ore as the major input. Id. After the ilmenite ore is reduced to a slag consisting of iron, titanium oxide, and carbon dioxide, the chlorine gas generated in the magnesium production process is used to separate titanium from the titanium oxide, forming titanium tetrachloride. Id. The raw magnesium from the magnesium production process is then used to separate the chlorine gas from the titanium tetrachloride, which produces titanium and magnesium dichloride. Id. The titanium is further refined to produce titanium powder and titanium sponge. Finally, chlorine gas resulting from the magnesium production process and not used in the production of titanium sponge is either recycled back into the earnallite refinement process or used to produce calcium chloride, a de-icer. Id.

II.

On February 27, 2004, a coalition of domestic magnesium producers, including U.S. Magnesium, filed an antidumping duty petition. Notice of Initiation of Antidumping Duty Investigations: Magnesium Metal from the People’s Republic of China and the Russian Federation, 69 Fed. Reg. 15,293, 15,293 (Dep’t of Commerce Mar. 25, 2004). The petition related to imports of pure and alloy magnesium metal from the Russian Federation and the People’s Republic of China. Id. at 15293-94. In 2005, Commerce issued an antidumping order on magnesium metal from the Russian Federation. Antidumping Order. Pursuant to that order, Com merce imposed an antidumping duty of 21.71 percent on magnesium metal imported by Avisma. Id., 70 Fed. Reg. at 19,931. Commerce subsequently instituted a second administrative review of the antidumping order with respect to Avis-ma, among others, for the period from April 1, 2006, through March 31, 2007. Initiation of Antidumping and Countervailing Duty Administrative Reviews and Request for Revocation in Part, 72 Fed. Reg. 29,968 (Dep’t of Commerce May 30, 2007). As noted, after conducting the administrative review, Commerce established an antidumping duty rate of 15.77 percent ad valorem on Avisma’s U.S. imports of magnesium made during the period of review.

An antidumping duty represents the amount by which the normal value of subject merchandise exceeds its export price. 19 U.S.C. § 1673. Normal value is the price at which the merchandise is sold for consumption within the exporting country. Id. § 1677b(a)(l). Export price is the price at which the merchandise is sold for importation into the United States. Id. § 1677a(a). When Commerce cannot determine normal value based on actual sales of the subject merchandise in the home market, it may base normal value on a constructed value as a proxy for the sale price in the home market. Id. § 1677b(a)(4). Constructed value for merchandise represents the sum of (1) the cost of materials and fabrication or other processing; (2) selling, general and administrative expenses, and profit; and (3) the cost. of containers and coverings. Id. § 1677b(e). In the Final Results, Commerce based its determination of the normal value of Avisma’s magnesium on constructed value.

III.

As described above, Avisma’s production process does not yield magnesium as a distinct product until after enriched carnallite has been refined through dehydration at OPU-1 and then electrolysis at OPU-2. At that point, raw magnesium and chlorine gas emerge as joint products— two dissimilar end products that are produced from a single production process. See Robert A. Anthony & James S. Reece, Accounting Principles 442 (5th ed. 1983). During the administrative review, in its preliminary results, in arriving at a constructed value for magnesium, Commerce adopted a methodology whereby magnesium and chlorine gas were treated as joint main products, with the costs of production being allocated between them at OPU-2, the split-off point where they become distinct products. Commerce performed the allocation based upon each product’s net realizable value (“NRV”) through OPU-2. See Magnesium Metal from the Russian Federation: Preliminary Results of Anti-dumping Duty Administrative Review, 73 Fed. Reg. 24,541, 24,544-45 (Dep’t of Commerce May 5, 2008) (‘Preliminary Results ”).

NRV is the selling price of a product less any costs necessary to complete and sell it. Anthony & Reece at 442. Accordingly, when costs are allocated to joint products based upon NRV in order to determine constructed value, they are allocated to each product in proportion to the amount of revenue contributed by that product. Consequently, the allocated costs, and thus the constructed value, of each joint product depend not only upon that product’s NRV but also the NRV of the other joint product. That is because any increase in the NRV of a joint product results in a greater proportion of the costs being allocated to that joint product and a reduced proportion of the costs being allocated to the other joint product. In its constructed value calculation for magnesi um, Commerce thus was required to fix an NRV for both magnesium and chlorine. In its preliminary results, Commerce determined the NRV for magnesium based upon magnesium prices from before the period of review. 4 Preliminary Results, 73 Fed. Reg. at 24,545. Commerce determined that because chlorine is not sold by Avisma and because its main benefit to Avisma is its use in Avisma’s titanium production process, the NRV of chlorine should be based upon what Avisma would have to spend to purchase the chlorine necessary for its titanium production process. Final Results Memo at 4.

Avisma argued that Commerce’s approach misunderstood Avisma’s production process. First, Avisma contended that magnesium should be treated as a byproduct of its titanium production rather than as a main product. 5 Id. Second, Avis-ma urged that if Commerce was going to treat magnesium as a main product, then the NRV of chlorine should not be based upon the market price of chlorine but, rather, upon the sale prices of products downstream from chlorine, Avisma’s titanium products. Id. at 10. In presenting these arguments, Avisma urged that Commerce’s approach in the Preliminary Results failed to “incorporate] the economic reality of AVISMA’s entire production facility.” Id. at 11. Importantly, Avisma’s approach resulted in a lower constructed value for magnesium and thus a lower antidumping margin because its approach increased the NRV of chlorine and thus reduced the amount of costs allocated to magnesium through OPU-2.

Avisma and U.S. Magnesium each submitted case briefs addressing changes they believed should be made in the preliminary results. Avisma I, 2009 WL 3423021, at *2. Commerce, however, rejected Avisma’s first brief and a portion of its second on the ground that both contained new factual information in the form of the attached Foster Affidavit, in which Dr. Foster advocated Avisma’s approach for determining constructed value. Commerce took the position that because the affidavit constituted an expert opinion submitted to provide further evidentiary support for Avisma’s arguments, it was new factual information and was therefore untimely pursuant to 19 C.F.R. § 351.301(b)(2). Under that regulation, the deadline for the submission of factual information is 140 days after the last day of the anniversary month of the publication of the Antidumping Order. In this case, the deadline for the submission of factual information was September 17, 2007. Avisma did not submit the Foster Affidavit until June 11, 2008, however. Letter from Laurie Parkhill, Dep’t of Commerce, to John M. Gurley (June 20, 2008) at 1 (“Parkhill Letter ”).

In September of 2008, Commerce formally adopted the methodology set forth in the preliminary results. Final Results; Final Results Memo. In doing so, Commerce rejected Avisma’s assertion that magnesium should be treated as a byproduct of the titanium process. Final Results Memo at 5-10. Commerce also rejected Avisma’s contention that the NRV of chlorine should be based upon the sale price of titanium products, stating that “it is illogical to create a value for chlorine using an NRV method that -does not result in a value that reflects a real world-market price of chlorine.... ” Id. at 13-16. On the basis of its constructed value methodology, Commerce imposed an antidumping margin of 15.77 percent on Avisma’s magnesium metal.

IV.

Avisma appealed Commerce’s determination to the Trade Court. Avisma I, 2009 WL 3423021. Thereafter, in due course, Avisma and U.S. Magnesium each moved for judgment on the agency record. Addressing the motions, the court turned first to the issue of the Foster Affidavit. The court noted that “[l]ong-established principles of administrative law imbue agencies with ample discretion to craft their own rules and procedures, including the authority to establish and enforce time limits concerning the submission of written information and data.” Id. at *5 (internal quotations omitted). Next, the court held that the Foster Affidavit “unambiguously” fell into the category of factual information. Id. at *6. The court found, however, that the circumstances of the case were not typical because the question of constructed value in the case presented an issue of first impression, and because a leading accounting expert (Dr. Foster) deemed Commerce’s approach inappropriate. Id. Based on these circumstances and “the mandate for accuracy in antidumping determinations,” the Trade Court remanded the case for further proceedings with the instructions that Commerce admit and consider the Foster Affidavit. Id. The court reserved for future proceedings the question of whether Commerce’s or Avis-ma’s methodology was proper. Id. at *6-*7.

Pursuant to the Trade Court’s order, Commerce undertook a redetermination in light of the Foster Affidavit, submitting the results in the First Remand Determination. In that determination, Commerce decided that the methodology applied in the Final Results was reasonable and appropriate. First Remand Determination at 3^1. Consequently, it did not recalculate the antidumping margin for Avisma’s magnesium metal. Id. at 30.

Avisma and U.S. Magnesium both appealed Commerce’s determination to the Trade Court. Avisma II, 724 F.Supp.2d 1308. Avisma contested Commerce’s chlorine gas valuation methodology, arguing that the approach “inappropriately truncates the production process at Avisma’s facilities and thereby ignores the intertwined nature of Avisma’s operations.” Id. at 1312. Despite agreeing “in large part” with Commerce’s determination, U.S. Magnesium disagreed with Commerce’s method for constructing the value of chlorine gas. Id. In reviewing Commerce’s methodology, the Trade Court interpreted the statutory framework as requiring Commerce “to take into account ‘the cost of materials and fabrication or other processing of any kind employed in producing the merchandise ... in the ordinary course of business.’ ” Id. at 1313 (quoting 19 U.S.C. § 1677b(e)(l)) (emphasis in original). The court determined that Commerce’s methodology was not in accordance with the statute because it failed to take into account Avisma’s ordinary course of business; that is, Commerce’s methodology did not reflect Avisma’s status as primarily a producer of titanium sponge. Id. at 1313-16. The Trade Court thus remanded the case with instructions to recalculate the NRV of chlorine by focusing on Avisma’s entire production process, including the titanium production process. Id. at 1316.

On remand, Commerce recalculated the antidumping margin pursuant to the methodology advocated by Avisma and endorsed by the Trade Court. Second Remand Determination. In doing so, Commerce relied on a cost database supplied by Avisma called COP-1.2. Id. at 8. Commerce rejected Avisma’s contention that the COP-1.2 database contained errors and that another cost database supplied by Avisma, COP-1.1, should form the basis for Commerce’s calculations. Id. at 8-12.

Both Avisma and U.S. Magnesium appealed aspects of the Second Remand Determination. Avisma III, 2011 WL 769998. Avisma appealed Commerce’s determination that COP-1.2 was the proper cost database, while U.S. Magnesium sought reconsideration of the legal conclusions in Avisma II. 6 Id. at *1. Despite the parties’ arguments, the Trade Court sustained the Second Remand Determination. Id. This appeal followed. We have jurisdiction pursuant to 28 U.S.C. § 1295(a)(5).

Discussion

“In reviewing a decision by the Court of International Trade, this court applies anew the statutory standard of review applied by that court to the agency’s decision.” Thai Pineapple Pub. Co. v. United States, 187 F.3d 1362, 1365 (Fed. Cir.1999) (citing PPG Indus., Inc. v. United States, 978 F.2d 1232, 1236 (Fed.Cir.1992)). “In doing so, we 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)).

US Magnesium appeals to this court on the grounds that the Foster Affidavit was properly excluded and that the Final Results were in accordance with law. Avis-ma cross-appeals on the ground that Commerce’s reliance on the COP-1.2 database was improper. Commerce urges affirmance of Avisma III. We begin our analysis with the issue of the Foster Affidavit.

A.

US Magnesium contends that the Trade Court erred first in requiring Commerce to admit and consider the Foster Affidavit. It argues that, pursuant to 19 C.F.R. § 351.102(b)(21), the affidavit constituted new factual information, a conclusion with which the Trade Court agreed. Def.’s Reply Br. at 20. Because the record was closed to new factual information at the time the Foster Affidavit was submitted, U.S. Magnesium claims that Commerce was within its discretion in rejecting it. Def.’s Br. at 33-34. US Magnesium states that none of the grounds identified by the Trade Court merited reversing Commerce’s discretion to enforce its deadlines. Id. at 34-35.

Avisma responds by arguing that Commerce’s exclusion of the Foster Affidavit was improper. Avisma contends that the affidavit was not new factual information, but simply corroboration of claims and data previously submitted. Pis.’ Br. at 26-28. Alternatively, Avisma argues that the Trade Court did not err in requiring Commerce to consider the affidavit because the matter before Commerce was one of first impression that would have significant impact on future antidumping determinations. Id. at 22-23. Lastly, Avisma contends that it submitted the Foster Affidavit when it first learned that Commerce had decided to use an accounting methodology different from the one Avis-ma had proposed and that therefore refusal to accept the affidavit constituted a denial of due process. Id. at 26-29.

B.

We start from the premise that, “[ajbsent constitutional constraints or extremely compelling circumstances the administrative agencies should be free to fashion their own rules of procedure and to pursue methods of inquiry capable of permitting them to discharge their multitudinous duties.” Vt. Yankee Nuclear Power Corp. v. Natural Res. Def. Council, Inc., 435 U.S. 519, 543-44, 98 S.Ct. 1197, 1211, 55 L.Ed.2d 460, 479-80 (1978) (internal quotations omitted). Accordingly, absent such constraints or circumstances, courts will defer to the judgment of an agency regarding the development of the agency record. To do otherwise would “run[ ] the risk of ‘propelling] the eourt[s] into the domain which Congress has set aside exclusively for the administrative agency.’ ” FPC v. Transcon. Gas Pipe Line Corp., 423 U.S. 326, 333, 96 S.Ct. 579, 583, 46 L.Ed.2d 533, 540 (1976) (quoting SEC v. Chenery Corp., 332 U.S. 194, 196, 67 S.Ct. 1575, 1577, 91 L.Ed. 1995, 1999 (1947)).

Commerce rejected the Foster Affidavit on the ground, that as expert opinion submitted to provide further evidentiary support for Avisma’s arguments, the affidavit constituted new factual information and was therefore untimely pursuant to 19 C.F.R. § 351.301(b). Parkhill Letter at 2. Pursuant to Commerce’s regulations, factual information includes (1) initial and supplemental questionnaire responses, (2) data or statements of fact in support of allegations, (3) other data or statements of fact, and (4) documentary evidence. 19 C.F.R. § 351.102(b)(21). Commerce determined that the Foster Affidavit constituted new factual information because “it provide[d] further evidentiary support for [AVISMA’s] arguments” and “inelude[d] detailed analysis regarding AVISMA’s products, production facilities, and production processes as well as discussion regarding preferred approaches to allocate joint costs.” Parkhill Letter at 2. Commerce contrasted the affidavit with “arguments included in case and rebuttal briefs which are intended as persuasive argumentation rather than evidence.” Id.

We agree with both Commerce’s and the Trade Court’s characterization of the Foster Affidavit as factual information. The weight accorded the Foster Affidavit arose not from the underlying data submitted therein, but rather from Dr. Foster’s analysis of that data. Thus, although the underlying data might in fact have been cumulative, the importance of the affidavit rests in Dr. Foster’s status as a third-party expert. Accordingly, as the Trade Court stated, “expert opinion analyzing reported information ‘clearly assumes the weight of evidence’ and, as such, amounts to ‘[d]ata or statements of fact in support of allegations,’ i.e., factual information.” Avisma I, 2009 WL 3423021 at *6 (citing Coal, for the Pres, of Am. Brake Drum & Rotor Aftermkt. Mfrs. v. United States, 44 F.Supp.2d 229, 240-41 & n. 19 (Ct. Int’l Trade 1999)).

Under 19 C.F.R. § 351.301(b)(2), Avisma had until September 17, 2007, to submit factual information to Commerce to be used in the Final Results. Avisma I, 2009 WL 3423021, at *5. Despite that deadline, Avisma did not submit the Foster Affidavit until June 11, 2008. Parkhill Letter at 1. Thus, the Foster Affidavit was untimely and properly rejected by Commerce. 7

Although the Trade Court correctly determined that the Foster Affidavit was untimely-submitted factual information, the court erred when, in spite of this determination, it ordered Commerce to admit the affidavit into the record because of circumstances the court described as “not typical.” Avisma I, 2009 WL 3423021, at *6. The court expressed concern that the case presented “an issue of first impression for the agency” that “may establish methodological precedent for future similar investigations.” Id. Because of this concern, it ruled that Commerce should consider the Foster Affidavit “to ensure the intent of the antidumping laws is upheld.” Id.

In our view, the Trade Court improperly intruded upon Commerce’s power to apply its own procedures for the timely resolution of antidumping reviews. The role of judicial review is limited to determining whether the record is adequate to support the administrative action. A court cannot set aside application of a proper administrative procedure because it believes that properly excluded evidence would yield a more accurate result if the evidence were considered. Indeed, the pursuit of “what the court perceives to be the ‘best’ or ‘correct’ result” would render judicial review “totally unpredictable.” Vt. Yankee, 435 U.S. at 546, 98 S.Ct. at 1213, 55 L.Ed.2d at 481. That an antidumping case may present an issue of first impression does not serve to give the Trade Court and this court license to usurp the role that Congress has delegated to Commerce.

Our conclusion is supported by our recent decision in Essar Steel Ltd. v. United States, 678 F.3d 1268 (Fed.Cir.2012). There, the plaintiff had appealed Commerce’s application of adverse facts during an antidumping investigation based upon the plaintiff having provided false information to Commerce regarding one of the its foreign facilities during the investigation. Id. at 1271. Relying on information which had arisen in a separate investigation and which contradicted the adverse facts, the Trade Court remanded the case to Commerce with instructions to consider the information despite the court’s recognition that the plaintiff had not acted to the best of its ability in supplying information to Commerce. Id. at 1276-77. We reversed the Trade Court’s ruling on the issue, noting that “the trial court’s order usurps agency power, undermines Commerce’s ability to administer the statute entrusted to it, contradicts important principles of finality, and discourages compliance.” Id. at 1278. Underlying both the holding in Essar Steel and our holding today is a recognition that courts must not improperly intrude upon an agency’s power to implement and enforce proper procedures for constructing an agency record.

Avisma’s complaint that exclusion of the Foster Affidavit violated its due process rights lacks merit. The due process right to which Avisma was entitled was “the right to notice and a meaningful opportunity to be heard.” LaChance v. Erickson, 522 U.S. 262, 266, 118 S.Ct. 753, 756, 139 L.Ed.2d 695, 700 (1998). Avisma only sought the opinion of Dr. Foster after its proposed methodology was rejected by Commerce. Pis.’ Br. at 22. Avisma was well aware throughout the review, however, that Commerce was considering using the accounting methodology ultimately adopted in the Final Results and of the applicable dates for submission of factual information. Commerce employed this methodology during the first administrative review and, indeed, this methodology was even used by Avisma in its internal records until January 1, 2007. Final Results Memo at 7. Avisma cannot claim to have been unaware that Commerce might opt to reject its proposed new methodology. Avisma had the opportunity to put forth evidence supporting its proposed accounting methodology but failed to do so. Avisma was not deprived of due process.

US Magnesium also contends that the Trade Court erred in rejecting the Final Results. US Magnesium argues that 19 U.S.C. § 1677b(f) sets forth how to calculate costs of production and constructed value and that Commerce’s original methodology comported with the statute, as the methodology was consistent with both Avisma’s historical accounting practice and generally accepted accounting principles. 8 Def.’s Br. at 39-48. Alternatively, U.S. Magnesium argues that Commerce’s methodology was a reasonable interpretation of the statute and thus was entitled to Chevron deference. Id. at 44r- 49.

According to U.S. Magnesium, in its analysis the Trade Court misapplied 19 U.S.C. § 1677b(e)(l). Id. at 49-54. Section 1677b(e)(l) provides that one component of constructed value is “the cost of materials and fabrication or other processing of any kind employed in producing the merchandise, during a period which would ordinarily permit the production of the merchandise in the ordinary course of business.” US Magnesium contends that the “ordinary course of business” language within the statute merely required Commerce to construct the value of magnesium from the records of Avisma during a period of normal business operation. Id. at 49-53. Thus, U.S. Magnesium urges that the Trade Court erred when it held that the “ordinary course of business” language required Commerce to adopt a methodology that encompassed Avisma’s titanium production process.

Avisma contends that the Trade Court was correct in ordering the use of Avis-ma’s proposed methodology. Noting the opinion of the Trade Court, Avisma contends that the “ordinary course of business” language in the statute requires Commerce to take into account Avisma’s entire business, which includes the production of titanium. Pis.’ Br. at 35-36.

B.

We first address the Trade Court’s determination that the language of 19 U.S.C. § 1677b(e)(l) mandates the adoption of Avisma’s facility-wide approach in order “to take into account Avisma’s ordinary course of business.” Avisma II, 724 F.Supp.2d at 1313. Based on its reading of the statute, the court reversed the Final Results as not in accordance with law. Id. 9 The plain language of § 1677b(e)(l), however, does not support the imposition of such a requirement. Accordingly, the Trial Court erred when it rejected the Final Results on this basis.

Section 1677b(e)(l) requires that constructed value be calculated, in part, from “the cost of materials and fabrication or other processing of any kind employed in producing the merchandise, during a period which would ordinarily permit the production of the merchandise in the ordinary course of business.” The Trade Court construed the phrase “in the ordinary course of business” as requiring Commerce to include in its constructed value methodology costs involved in Avis-ma’s entire production process. This was error. The plain language of the statute indicates that the phrase “during a period which would ordinarily permit the production of the merchandise in the ordinary course of business” aims to ensure that constructed value is based on the costs of production incurred when the facility is operating in the usual and ordinary course of business, so as to guard against aberrant costs not typically incurred in the normal course of producing the subject merchandise. The phrase does not dictate that Commerce employ a specific cost accounting methodology that would include Avisma’s entire production process.

Although we hold the Trade Court erred in its interpretation of the “ordinary course of business” requirement, we still must review Commerce’s determination to ensure that it is supported by substantial evidence and otherwise in accordance with law. 19 U.S.C. § 1516a(b)(l)(B)(i). In reviewing whether Commerce’s interpretation of a statute is in accordance with law, we review the issue de novo under the Chevron framework. See Ningbo Dafa Chem. Fiber Co. v. United States, 580 F.3d 1247, 1253 (Fed.Cir.2009). The first step of the Chevron analysis is to 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.” Chevron U.S.A. Inc. v. Natural Res. Def. Council, Inc., 467 U.S. 837, 842-43, 104 S.Ct. 2778, 2781, 81 L.Ed.2d 694, 702-03 (1984). If, however, “Congress has not directly addressed the precise question at issue,” under step two of Chevron, if an agency’s interpretation is “reasonable,” it is “given controlling weight unless [it is] arbitrary, capricious, or manifestly contrary to statute.” Id. at 843-44, 104 S.Ct. at 2781-82, 81 L.Ed.2d at 703. Because in this case Congress has not directly spoken to the precise question at issue (what methodology should be used when allocating costs between products produced jointly for purposes of determining constructed value for one of the products), our review centers on whether Commerce’s chosen methodology in the Final Results rests upon a reasonable interpretation of the statutes and regulations it administers. Thai Pineapple, 187 F.3d at 1365 (citing Chevron, 467 U.S. at 844, 104 S.Ct. at 2782, 81 L.Ed.2d at 703). In examining Commerce’s approach, we must be mindful that as the “master of anti-dumping law,” id. (quoting Daewoo Elec. Co. v. Int’l Union, 6 F.3d 1511, 1516 (Fed. Cir.1993)), Commerce is entitled to substantial deference in its choice of accounting methodology:

This court has recognized that the anti-dumping statute reveals tremendous deference to the expertise of the Secretary of Commerce in administering the antidumping law. Antidumping and countervailing duty determinations involve complex economic and accounting decisions of a technical nature, for which agencies possess far greater expertise than courts. This deference is both greater than and distinct from that accorded the agency in interpreting the statutes it administers, because it is based on Commerce’s technical expertise in identifying, selecting and applying methodologies to implement the dictates set forth in the governing statute, as opposed to interpreting the meaning of the statute itself where ambiguous.

Fujitsu Gen. Ltd. v. United States, 88 F.3d 1034, 1039 (Fed.Cir.1996) (internal quotations and citations omitted). Accordingly, “[t]he methodologies relied upon by Commerce in making its determinations are presumptively correct.” Thai Pineapple, 187 F.3d at 1365.

As seen above, Commerce treated magnesium and chlorine gas as joint products, thus rejecting Avisma’s characterization of magnesium as a by-product of the titanium production process. Final Results Memo at 5-10. Additionally, Commerce calculated the NRV of chlorine based upon what Avisma would have to spend to purchase the chlorine necessary for its titanium production process. In doing so, Commerce also rejected Avisma’s contention that the NRV of chlorine should be based upon the sale price of titanium products. Id. at 13-16.

As Avisma acknowledges, the statute does not expressly provide for any specific allocation methodology for goods produced jointly. Pis.’ Br. at 34. In view of the “tremendous deference to the expertise of the Secretary of Commerce,” Fujitsu, 88 F.3d at 1039, Commerce’s choice of accounting methodology in the Final Results is supported by substantial evidence. Commerce supported its decision to treat chlorine and magnesium as main products and to allocate costs at the OPU-2 split-off point by stating that “chlorine and magnesium can be identified and valued objectively before they enter the process for producing titanium.” Final Results Memo at 6. Moreover, Commerce highlighted the following reasons further supporting its chosen approach:

the relative values at the split-off point are significant; both products have been treated as main products in AVTSMA’s books and records during some period of the [Period of Review]; management undertakes the production of both products intentionally either for resale or for use as inputs in titanium; both products are inputs to, not unavoidable consequences of, titanium production; management further-processes the raw-magnesium joint product intentionally.

Id. at 10. Lastly, noting that Avisma had used two separate accounting methodologies during the period of review, Commerce declined to find that Avisma’s own records provided clear support for treating magnesium as a by-product. Id. at 7.

Commerce also provided ample support for its decision to determine the price of chlorine based upon the cost to Avisma to purchase chlorine necessary for use in its titanium-production unit. In adopting that approach, Commerce focused on “the benefits received from the joint products as of the split-off point” and kept the value of chlorine tied to a real world price. Id. at 15. Furthermore, Commerce noted that its approach was supported by the pertinent accounting literature. 10 Id. Finally, Commerce determined that Avisma’s proposed approach created an inflated value of chlorine that creates “distortion [by] tying the value of chlorine to the profits earned on titanium.” Id. at 14-15.

Commerce’s choice of methodology and the resulting determination of the anti-dumping margin are both supported by substantial evidence and are therefore reasonable. In holding that the statutory language required the use of Avisma’s proposed accounting methodology, the Trade Court erred. Accordingly, we reverse the decision of the Trade Court setting aside the Final Results and remand for entry of judgment reinstating the Final Results.

In light of our decision instructing the Trade Court to reinstate the Final Results, Avisma’s cross-appeal is rendered moot. Avisma’s cross-appeal involves Commerce’s use of the COP-1.2 database in the Second Remand Determination and an alleged error in the database related to the pre-split-off costs of titanium products. Because the pre-split-off costs of titanium products were not used in the calculation of an antidumping margin in the Final Results, any alleged discrepancy would have no effect on the Final Results, a point upon which both Commerce and Avisma agree. See Pis.’ Br. at 48; Commerce’s Br. at 6. Because we hold that the Trade Court erred in setting aside the Final Results, it is unnecessary to reach Avisma’s cross-appeal.

Conclusion

For the foregoing reasons, we reverse the decision of the Trade Court. We remand the case to the Trade Court with the instruction that it enter judgment reinstating the Final Results.

Each party shall bear its own costs.

REVERSED and REMANDED.

1 . When discussing proceedings in the Trade Court and this court, we refer to Avisma and Tirus collectively as "Avisma.” Otherwise, all references are to Avisma alone.

2 . Although Commerce complied with the Trade Court's rulings in Avisma I and Avisma II, it expressed disagreement with the rulings in its remand determinations. First Remand Determination at 2 ("The Department disagrees, respectfully, with the Court's conclusion. ...”); Second Remand Determination at 3-5 ("At the outset, the Department respectfully disagrees with the Court’s Remand Order.”).

3 . Dehydration takes place at OPU-1; electrolysis takes place at OPU-2. OPU-1, OPU-2, and OPU-3 are production points in the Berezniki facility. See Final Results Memo at 3.

4 . “When determining the NRV of the subject merchandise for a value allocation, [Commerce] typically looks to prices prior to the period of investigation or review in order to avoid the problem of circularity. The circularity problem occurs when an allegation of dumping calls into question the reliability of using dumped prices to allocate costs to subject merchandise in the process of determining whether dumping occurred during the period." Final Results Memo at 4 n.3.

5 . The classification of a joint product as either a main product or a by-product affects the constructed value determination. Byproducts, a type of joint product, are products that result from a process whose main objective is the production of another product and not the byproduct itself. Anthony & Reece at 443. By-products are treated such that no profit is reported for them. Id. Thus, all profits are attributed to the main product. Id. Consequently, the constructed value of a product is lower if it is classified as a byproduct rather than a main product because the constructed value of a by-product does not include an amount for profit.

6 . Specifically, U.S. Magnesium asked the Trade Court to reconsider the holding of Avis-ma II and reinstate the First Remand Determination. Avisma III, 2011 WL 769998, at *1.

7 . "Commerce generally does not consider untimely filed factual information.” Essar Steel Ltd. v. United States, 678 F.3d 1268, 1278 (Fed.Cir.2012) (citing 19 C.F.R. § 351.302(d)(1) (“[T]he Secretary will not consider or retain in the official record of the proceeding ... untimely filed factual information. ...”)).

8 . Section 1677b(£)(l)(A) provides that:

Costs shall normally be calculated based on the records of the exporter or producer of the merchandise, if such records are kept in accordance with the generally accepted accounting principles of the exporting country (or the producing country, where appropriate) and reasonably reflect the costs associated with the production and sale of the merchandise. The administering authority shall consider all available evidence on the proper allocation of costs, including that which is made available by the exporter or producer on a timely basis, if such allocations have been historically used by the exporter or producer, in particular for establishing appropriate amortization and depreciation periods, and allowances for capital expenditures and other development costs.

9 . The Trade Court explained its reasoning as follows:

When Commerce uses a constructed value as the normal value of the subject merchandise, Congress instructs the Department to take into account "the cost of materials and fabrication or other processing of any kind employed in producing the merchandise ... in the ordinary course of business.” 19 U.S.C. § 1677b(e)(l). The “ordinary course of business” means “[t]he transaction of business according to the common usages and customs of ... the particular individual whose acts are under consideration.” Black's Law Dictionary 1098 (6th ed. 1990); see also Black's Law Dictionary 404, 1209 (9th ed. 2009). From a thorough examination of the record, the court finds that the Department's chosen chlorine gas valuation methodology in the Rede termination Results fails to take into account Avis-ma’s ordinary course of business and, therefore, does not accord with law.

Avisma II, 724 F.Supp.2d at 1313 (emphasis in original).

10 . Commerce quoted the following passage from Charles T. Horngren, Srikant M. Datar & George Foster, Cost Accounting: A Managerial Emphasis 547-48 (9th ed. 1997):

The sales at split-off method allocates joint costs based on the relative sales value at the split off point of the total production in the accounting period.... The sales value at split-off point method exemplifies the benefits-received criterion of cost allocation. Costs are allocated to products in proportion to their ability to contribute revenue. This method is both straightforward and intuitive.

Final Results Memo at 15 n.5.