United States Steel Corp. v. United States

219 F. Supp. 3d 1300, 39 I.T.R.D. (BNA) 1071, 2017 Ct. Intl. Trade LEXIS 30

• Court: United States Court of International Trade
• Decided: March 16, 2017
• Docket: Slip Op. 17-28; Consol. Court No. 14-00263
• Status: Published

Opinion

OPINION

Kelly, Judge:

Currently before the court for review is the U.S. Department of Commerce’s (“Department” or “Commerce”) Final Results of Redetermination Pursuant to Remand filed pursuant to the court’s decision in United States Steel Corp. v. United States, 40 CIT —, 179 F.Supp.3d 1114 (2016) (“U.S. Steel”). See Final Results of Redetermination Pursuant to Remand Confidential Version, Aug. 31, 2016, ECF No. 113 (“Remand Results”). The court remanded Commerce’s final determination in its investigation of the antidumping duty (“ADD”) order covering certain oil country tubular goods (“OCTG”) from India for the period July 1, 2012 through June 30, 2013 to reconsider and provide further explanation regarding: (1) its application of the ratio test of its differential pricing analysis; (2) its determination that Jindal SAW, Limited (“Jindal SAW”) is not affiliated with certain of its suppliers of electricity and steel billets; (3) its determination that Jindal SAW’s yield loss data reasonably reflected its costs of production (“COP”); and (4) its assignment of the highest cost data from GVN Fuels, Ltd.’s (“GVN”) production cost database to GVN’s dual-grade OCTG products. See U.S. Steel, 40 CIT at -, 179 F.Supp.3d at 1120; see generally Certain Oil Country Tubular Goods From India, 79 Fed. Reg. 41,981 (Dep’t Commerce July 18, 2014) (final determination of sales at less than fair value and final negative determination of critical circumstances) (“Final Results”) and accompanying Corrected Issues and Decision Memorandum for the Final Affirmative Determination in the Less than Fair Value Investigation of Certain Oil Country Tubular Goods from India, Dec. 19, 2014, ECF No. 22-3 (“Final Decision Memo”). Commerce’s Remand Results adequately address the concerns raised in the court’s decision, and its revised results are supported by substantial evidence. Therefore, the court sustains Commerce’s Remand Results.

BACKGROUND

The court presumes familiarity with the facts as discussed in U.S. Steel. Nevertheless, the court briefly summarizes facts relevant to its discussion here for ease of reference. In its final determination, Commerce applied the mixed alternative methodology of its differential pricing analysis (Le., average-to-transaction (“A-T”) to Jin-dal SAW’S U.S. sales passing the Cohen’s d test) to .calculate a weighted-average dumping margin for mandatory respondent Jindal SAW while Commerce applied its standard average-to-average (“A-A”) methodology to all of GVN’s sales to calculate its margin. See Final Decision Memo at 12. This resulted in an assignment of a weighted average dumping margin of 9.91% to Jindal SAW, 2.05% to GVN, Maharashtra Seamless Limited, and Jindal Pipes Limited,¹ and 5.79% to all other respondents that were not individually investigated. See Final Results, 79 Fed. Reg. at 41,982.

The court held that Commerce reasonably explained why the ratio test of its differential pricing analysis is generally tailored to the statutory purpose under normal circumstances. U.S. Steel, 40 CIT at -, 179 F.Supp.3d at 1125. However, the court held that Commerce failed to adequately explain why the thresholds of the ratio test as applied in this investigation are reasonable where a large value of sales are excluded from the numerator of the ratio test. See Id., 40 CIT at -, 179 F.Supp.3d at 1125-26. The court explained that the breadth of Commerce’s application of its A-T methodology is determined by Commerce’s ratio analysis. Id. Therefore, the court concluded that the results of Commerce’s application of its methodology “has at least the potential to treat the same [pricing] behavior differently” in different proceedings. Id. The court required Commerce on remand to “provide further explanation as to why its thresholds, as applied in this investigation are reasonable or otherwise reconsider the parameters of its differential pricing methodology [where a large value of sales are excluded from by its methodology].” Id., 40 CIT at -, 179 F.Supp.3d at 1126.

Second, the court remanded Commerce’s determination that Jindal SAW and its suppliers of steel billets and electricity are not affiliated to further explain and consider corporate and family relationships among O.P. Jindal family members that held direct and indirect interests in both Jindal SAW and in its suppliers as well as management and board membership overlap of Jindal family members in these supplier entities. Id., 40 CIT at -, 179 F.Supp.3d at 1132. The court also directed Commerce to consider whether close supplier relationships between these entities made Jindal SAW reliant upon its suppliers of steel billets and electricity, or vice versa, in evaluating whether they are under common control, as required by Commerce’s regulation. Id.; see generally 19 C.F.R. § 351.102(b)(3)(2013).²

Third, the court remanded Commerce’s acceptance of Jindal SAW’s reported yield loss COP data because Commerce lacked substantial evidence to conclude that the reported yield loss data reasonably reflected its COP for each specific category of subject merchandise. Id. Specifically, the court concluded that Commerce failed to undertake a comparison of yield losses allocated by physical characteristic or production stage versus the manner in which Jindal SAW allocated-its costs.³ Id., 40 CIT at -, 179 F.Supp.3d at 1134. The court concluded Commerce could not have determined if the difference in Jindal SAW’s reported COP data is distorted and, therefore, accurately reflects its costs of production without comparing yield losses allocated by production stage and the yield loss allocation methodology used by Jindal SAW. Id. The court directed Commerce to explain why Jindal SAW’s “reported yield loss data, which [Jindal SAW concedes] did not track yield losses by production stage or physical characteristics of the merchandise, nonetheless did not distort Jindal SAW’s COP for specific [control'numbers (“CONNUMs”) ] of subject merchandise or reconsider its determination.” Id., 40 CIT at -, 179 F.Supp.3d at 1135.

Fourth, the court remanded Commerce’s assignment of the highest' costs associated with OCTG meeting stricter specifications to GVN’s OCTG products meeting multiple performance specifications (i.e„ dual grade products) because Commerce failed to explain what record evidence supports its decision to select the highest cost data for products within the product grouping meeting higher performance specifications. Id., 40 CIT at -, 179 F.Supp.3d at 1154. The court noted that there was varying cost information for the products with different • physical characteristics meeting higher performance specifications, but Commerce does not explain why GVN’s dual grade products were most similar to the highest cost products that met more stringent product specifications. Id. The court concluded that “without such an explanation, Commerce’s selection of the highest cost information among [higher performance]' products from GVN’s cost database may -only have been the product of an adverse inference' that GVN’s dual grade products were more cost-intensive than any other [such] products.” Id. The court held that Commerce must “either explain why assigning the highest costs for [the products meeting stricter specifications] from GVN’s cost database to its dual use products was reasonable in light of the characteristics of GVN’s dual-use products or explain its application of an adverse inference by satisfying the legal prerequisites for doing so under [Section 776(b) of the Tariff Act of 1930, as amended 19 U.S.C. § 1677e(b) 2012].”⁴ Id.

In its Remand Results, Commerce continues to apply its A-T methodology to Jindal SAW’s U.S. sales passing the Cohen’s d test and its average-to-average (“A-A”) methodology to the remainder of its sales, while continuing to apply the A-A methodology to all of GVN’s sales. See Remand Results 3-6. On affiliation, after reviewing record evidence concerning the direct and indirect ownership interests of members, of the Jindal family in Jindal SAW as well as board memberships, the nature of the supplier relationships, and other indicia of control, Commerce continues to conclude that Jindal SAW and its suppliers of steel billets and electricity are unaffiliated for purposes of 19 U.S.C. § 1677(33)(F). See id. at 7-21. On Jindal SAW’s yield loss COP data, Commerce found that “Jindal SAW’s cost reporting methodology did not allocate yield losses on a basis that reasonably reflected differences in the processing costs for merchandise with differing physical characteristics.” Id at 23. Commerce concludes that the reported yield should be based on yield factors that take physical characteristics, including wall thickness and diameter, into account. Id. at 26. Therefore, Commerce recalculated Jindal SAW’s yield loss calculation to mitigate distortions based on partial adverse facts available (“AFA”).⁵ Id. Lastly, on the issue of assigning costs to GVN’s dual grade products, Commerce determined that it “unintentionally overlooked its standard ‘proxy cost’ methodology by selecting the highest cost of L-80 grade” products in its final determination. See id. at 26-28. Commerce revised the costs assigned to GVN’s dual grade product to that of the most similar product meeting higher performance specification from the cost data associated with such merchandise in GVN’s cost database based on physical characteristics. Id. at 28.

These changes on remand result in an adjusted weighted-average dumping margin of 11.24% for Jindal SAW and a de minimis rate of 1.07% for GVN. Id at 52. Therefore, if the court sustains Commerce’s remand redetermination, Commerce would terminate its investigation with respect to GVN. See id.

JURISDICTION AND STANDARD OF REVIEW

The court continues to have jurisdiction pursuant to 19 U.S.C. § 1516a(a)(2)(B)(i) and 28 U.S.C. § 1581(c) (2012), which grant the court authority to review actions contesting the final determination in an investigation of an antidump-ing duty order. See 19 U.S.C. § 1516a(a)(2)(B)(i); 28 U.S.C. § 1581(c) (2012). “The court shall hold unlawful any determination, finding, or conclusion found ... to be unsupported by substantial evidence on the record, or otherwise not in accordance with law.” 19 U.S.C. § 1516a(b)(1)(B)(i). “The results of a rede-termination pursuant to court remand are also reviewed ‘for compliance with the court’s remand order.’” Xinjiamei Furniture (Zhangzhou) Co. v. United States, 38 CIT -, -, 968 F.Supp.2d 1255, 1259 (2014) (quoting Nakornthai Strip Mill Public Co. v. United States, 32 C.I.T. 1272, 1274, 587 F.Supp.2d 1303, 1306 (2008)).

DISCUSSION

I. Commerce’s Application of the Ratio Test of its Differential Pricing Analysis

In U.S. Steel, the court held that Commerce failed to explain why the thresholds, as applied in the ratio test of its differential pricing analysis are reasonable, where they have the effect of excluding a significant value of sales from being tested for patterns of significant price differences. U.S. Steel, 40 CIT at -, 179 F.Supp.3d at 1125. The court remanded Commerce’s final determination for it to explain why the ratio test, as applied, is reasonable where a significant value of respondents’ sales are excluded. See id. The court explained that, where the value of all respondents sales remains constant, the ratio of the value of sales passing the Cohen’s d component of Commerce’s differential pricing methodology relative to all sales may differ substantially from another investigation where a lesser value of sales is excluded from the Cohen’s d analysis. Id. The court remanded Commerce’s determination because the exclusion of sales has at least the potential to treat the same pricing behavior differently. Id., 40 CIT at -, 179 F.Supp.3d at 1125-26.

Plaintiff contends that Commerce fails to explain how the inclusion of non-tested U.S. sales in the denominator test is reasonable except to assert that it includes all sales “based on an unfounded assumption that they do not contribute to a pattern of significant price differences.” Comments of United States Steel Corporation Final Results of redetermination Pursuant to Court Remand Issued by Dep’t Commerce 13, Sept. 30, 2016, ECF No. 119 (“U.S. Steel Comments”). Defendant responds that Commerce has explained that all sales are included because the weighted-average dumping margin is based on all a respondent’s sales, and to exclude any sales would require an unfounded assumption that the resulting subset of sales is representative of the overall universe of sales. Def.’s Resp. Comments Remand Redeter-mination Proprietary Version 18, Jan. 24, 2017, ECF No. 2017 (“Def.’s Reply”). For the reasons that follow, Commerce has complied with the court’s remand decision, and its determination is reasonable and supported by substantial evidence.

Commerce ordinarily uses an A-A methodology ⁶ to calculate dumping margins in an investigation.⁷ See 19 U.S.C. § 1677f-1(d)(1)(A)-(B). However, Commerce may use an A-T methodology⁸ as an alternative to the default A-A method if:

(i) there is a pattern of export prices ... for comparable merchandise that differ significantly among purchasers, regions, or periods of time, and

(ii) [Commerce] explains why such differences cannot be taken into account using a method described in paragraph (l)(A)(i) [(A-A)] or (ii) [ (transaction-to-transaction) ].⁹

Id. at § 1677f-1(d)(1)(B)(i)-(ii). The statute provides no methodology for how Commerce identifies and measures a pattern of export prices, how significantly those prices must differ among purchasers, regions, or periods of time, or what form of “export prices” Commerce must consider in its pattern analysis. See id. § 1677f-1l(d)(1)(B)(i). Commerce has implemented and continues to develop a practice, which it calls its differential pricing analysis, “for purposes of examining whether to apply an alternative comparison method in this [less than fair value] investigation.”¹⁰ Decision Memorandum for the Preliminary determination in the Less-Than-Fair-Value Investigation of Oil Country Tubular Goods from India at 10, PD 258, bar code 3181829-01 (Feb. 14, 2014) (“Prelim. Decision Memo”).¹¹

The court affords Commerce significant deference in determinations “in-volvpng] complex economic and accounting decisions of a technical nature.” Fujitsu Gen. Ltd. v. United States, 88 F.3d 1034, 1039 (Fed. Cir. 1996). Despite Commerce’s wide discretion, it “must cogently explain why it has exercised its discretion in a given manner,” Motor Vehicle Mfrs. Ass’n of U.S. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 48-49, 103 S.Ct. 2856, 77 L.Ed.2d 443 (1983) (citations omitted). Fashioning a test to evaluate what constitutes a pattern under the statute is sufficiently complex and technical to warrant significant deference. See Fujitsu, 88 F.3d at 1039. Commerce’s methodological approach must nevertheless be a “reasonable means of effectuating the statutory purpose,” and its conclusions must be supported by substantial evidence in order to be afforded deference. Ceramica Regiomontana, S.A. v. United States, 10 C.I.T. 399, 404-05, 636 F.Supp. 961, 966 (1986), aff'd, 810 F.2d 1137, 1139 (Fed. Cir. 1987).

On remand, Commerce explains that its differential pricing methodology proceeds on a presumption that prices do not differ significantly by purchaser, region, or time period; it does not seek to prove that significant price differences do not exist. Remand Results 4. Commerce claims that its methodology does not “exclude” sales from its analysis at all, but rather analyzes all sales and concludes there is no evidence prices differ significantly where there are not two or more sales in the comparison group or the test group.¹² See id. Commerce contends that sales are not deemed “untestable,” but rather are deemed to have failed the Cohen’s d test because there is no evidence of a pattern of sales that differ significantly from prices of all other sales of comparable merchandise. Id.

Commerce reasonably explains that it deems situations where there are not at least two observations in both the comparison and test group not to have demonstrated a pattern of sales that differ significantly from prices of all other sales of comparable merchandise. See id. at 6. Commerce concludes that there can be no pattern of sales that differ significantly from prices of all other sales of comparable merchandise because there are “very few affirmative test results.” See id. Commerce also supports its conclusion that this result is not arbitrary because where a purchaser has many sales of comparable merchandise to given purchasers, regions, or time periods, there can be a “very different pattern[ ] of prices that differ significantly.” See id. Therefore, these sales have not been excluded; they simply demonstrate a different pattern, if any. See id. The statute does not mandate how Commerce should measure whether a pattern of export prices differs significantly or how the A-T methodology may be applied. See 19 U.S.C. § 1677f-1 (d)(1)(B)(i)—(ii). Just as Commerce has discretion under the statute to develop a methodology for measuring the extent of a pattern of prices that differ significantly by purchaser, time period, or region and to measure the extent of a pattern, see id., so it may incorporate into that methodology a reasonable test to determine whether such a pattern exists. U.S. Steel fails to demonstrate that requiring two or more observations in the comparison and test group as applied in this investigation to determine the existence of a pattern of significant price differences is unreasonable or arbitrary.

Likewise, having explained that sales are not excluded from the Cohen’s d test, Commerce reasonably justifies its inclusion of all sales in the denominator of the ratio generated by comparing the value of sales passing the Cohen’s d test to the value of all sales. Remand Results 5. Commerce clarifies that “the ratio test assesses the extent of the significant price differences for all sales.” Remand Results 5 (internal quotations omitted). Commerce further expounds that measuring the extent of price differences for all sales is consistent with the purposes of its differential pricing analysis: to determine whether the A-A methodology, which is generally applied to all U.S. sales, is the appropriate comparison method to calculate a respondent’s weighted-average dumping margin. Id. (citing 19 C.F.R. § 351.414(c)(1)). Because a respondent’s weighted-average dumping margin is based on all of its U.S. sales, so too should the test for evaluating the extent to which prices differ significantly in the U.S. market. Therefore, Commerce has complied with the court’s directions, and it has reasonably explained why its ratio test is reasonable and not arbitrarily applied.

U.S. Steel’s arguments to the contrary are unavailing. U.S. Steel argues that Commerce’s application of its ratio test distorted the differential pricing analysis and masked respondents’ dumping by understating the extent of the pattern of differential pricing by including “untested” sales from its Cohen’s d analysis in the denominator of the ratio. U.S. Steel Comments 14-15. U.S. Steel argues that, because the extent of the pattern is understated, Commerce applies its A-T methodology to fewer sales than it would have had the pattern been accurately measured.¹³ See id. Commerce has adequately explained that it analyzes all sales and concludes that instances with fewer than two observations do not pass the Cohen’s d test because they do not demonstrate the existence of a pattern of prices that differ significantly by purchaser, region, or time period. Remand Results 4. Moreover, U.S. Steel offers no evidence that Commerce’s methodology understates the extent of the pattern because it points to no reason why it is unreasonable to conclude that sales with less than two observations in the comparison group or the test group do not evidence a pattern of significant price differences.¹⁴ Commerce reasonably includes the value of all sales in the denominator of its ratio test calculation to ensure that it measures the extent of any pattern of significant price differences across all sales.

U.S. Steel also argues that Commerce’s inclusion of all sales in the denominator is based on an unsupported assumption that sales that do not pass the Cohen’s d test do not contribute to a pattern of significant price differences. U.S. Steel Comments 16. U.S. Steel’s argument rests on its related misconception that sales not contributing to a pattern of significant price differences are not tested. See id. (calling such sales “non-tested sales”). As already discussed, the latter premise is inaccurate because Commerce subjects all sales to the Cohen’s d test, but concludes that sales do not contribute to a pattern of significant price differences where there are less than two observations in either the test group or the comparison group. See Remand Results 4. After applying the Cohen’s d test to all sales, Commerce determines on the basis of too “few affirmative test results” that the sales do not demonstrate a pattern of prices that differ significantly. See id. As Commerce explains, this conclusion is based upon the number of affirmative test results, not merely on an assumption that sales do not contribute to a pattern. See id.

Lastly, U.S. Steel ai'gues that Commerce’s fails to explain how the results of its ratio test are not arbitrary because Commerce’s Remand Results recognize that the results of its differential pricing analysis may be different for two respondents exhibiting the same pricing behavior if the first has many sales of comparable merchandise and another who has few comparable sales. U.S. Steel Comments 19-20. However, Commerce explains that the pricing behaviors envisioned by U.S. Steel are quite different. Remand Results 34. Commerce notes that, where pricing behavior shows many sales of comparable merchandise to given purchasers, regions, or time periods, it may show a very different pattern than where there are fewer than two sales. Id. at 34. In contrast, where there are less than two observations of comparable sales in each of the comparison and test groups, there can be little or no pattern. Id. Thus, Commerce reasonably treats these two pricing behaviors differently because it concludes that there is a pattern of significant price differences in the former scenario and not in the latter scenario based on the number of affirmative observations of price differences. See id.

II. Commerce’s Determination that Jindal SAW is Unaffiliated with its Suppliers of Inputs

In U.S. Steel, the court held that Commerce failed to adequately explain why Jindal SAW and its suppliers of steel billets and electricity were not under the common control of the Jindal family.¹⁵ U.S. Steel, 40 CIT at -, 179 F.Supp.3d at 1127-28. The court remanded the issue for Commerce to provide further explanation and consideration in view of record evidence of indirect ownership by Jindal family members in Jindal SAW and in its suppliers of steel billets and electricity through promoter group entities, management positions held by the family grouping, and the close supplier relationships between Jindal SAW and those suppliers. Id. For the reasons that follow, Commerce’s determination that Jindal SAW is not affiliated with its suppliers of steel billets and electricity is supported by substantial evidence. The court first discusses Commerce’s findings regarding direct and indirect ownership held through promoter group entities. Next, the court reviews the reasonableness of Commerce’s findings regarding board membership and management positions held by Jindal family members. Lastly, the court evaluates Commerce’s conclusions regarding the significance of the close supplier relationship between Jindal SAW and its suppliers of steel billets and electricity.

A. Direct and Indirect Shareholdings of Jindal Family Members

As in its final determination, Commerce determined in its remand results that the direct holdings of Jindal family members were very small and insufficient to create the potential to exercise legal or operational control. See Remand Results 8. After evaluating the indirect holdings of Jindal families further, Commerce determined that the additional indirect ownership of the Jindal family members that might be added to insignificant direct ownership numbers would be unlikely to rise to the level of ownership necessary to create a potential for control. See id. at 9-14.

U.S. Steel argues that Commerce’s affiliation determination in its Remand Results still fails to account for the significance of the O.P. Jindal family’s indirect stock ownership in Jindal SAW and its suppliers of steel billets and electricity, in particular, the family’s indirect holdmgs through the promoter group entities in which the family has indirect holdings. U.S. Steel Comments 22-27. Defendant counters that Commerce complied with the court’s order by evaluating record evidence surrounding indirect ownership. Def.’s Reply 21. Defendant argues Commerce’s review of this information on indirect ownership supports its determination that the family’s interests in Jindal SAW or any of its subsidiaries and these suppliers was insufficient to demonstrate common control. Id. at 21-31. Commerce’s determination that indirect ownership by the Jindal family is insufficient to demonstrate common control over Jindal SAW and its supplier entities is supported by substantial evidence.

The statute defines affiliated persons through the following categories:

(A) Members of a family, including brothers and sisters (whether by whole or half blood), spouse ancestors, and lineal descendants.

(B) Any officer or director of an organization and such organization.

(C) Partners.

(D) Employer and Employee.

(E) Any person directly or indirectly owning, controlling, or holding with power to vote, 5 percent or more of the outstanding voting stock or shares of any organization and such organization.

(F) Two or more persons directly or indirectly controlling, controlled by, or under common control with, any person.

(G) Any person who controls any other person and such other person.

19 U.S.C. § 1677(33). A person is considered to control another person “if the person is legally or operationally in a position to exercise restraint or direction over the other person.” Id. Commerce’s regulations incorporate the statutory definition of “affiliated persons” and “affiliated parties” and further clarify the non-exhaustive list of considerations Commerce shall take into account in assessing whether control over another person exists as an element of affiliation. 19 C.F.R. § 351.102(b)(3) (2013). In evaluating whether control exists under the statute, Commerce will consider, among other factors, “[corporate or family groupings; franchise or joint ventui-e agreements; debt financing; and close supplier relationships,” Id. However, Commerce “will not find that control exists on the basis of these factors unless the relationship has the potential to impact decisions concerning the production, pricing, or cost of subject merchandise.” Id

Under Commerce’s practice, in assessing common control through family groupings, the potential of the family grouping to impact decisions concerning the firms it is alleged to control are considered for the family grouping as a whole. See, e.g., Issues and Decision Memorandum for the Final Results of the 2001-2002 Administrative Review of the Antidumping Duty Order on Steel Concrete Reinforcing Bar from the Republic of Korea, A-580-844, (Apr. 13, 2004), available at http://ia. ita.doc.gov/frn/summary/korea-south/04-8375-l.pdf (last visited Mar. 13, 2017). However, it does not follow that two companies are under the common control of a family grouping merely because members of a family, who are affiliated with one another, have direct or indirect shareholdings or are involved in the management of the two companies. See 19 U.S.C. § 1677(33)(F).

On the question of indirect ownership, Commerce acknowledged the limitations of the record documents, including financial statements and forms filed with Indian securities regulators, in that neither type of document necessarily provides a complete view of all indirect shareholders.¹⁶ Remand Results at 9. Nevertheless, Commerce was satisfied that the additional indirect ownership of the Jindal family members that might be added to minute direct ownership' numbers would be unlikely to rise to the level of ownership necessary to create a potential for control because Commerce’s review of selected additional information did not reveal inconsistencies with the shareholdings reported by Jindal SAW.¹⁷ Id U.S. Steel points to no evidence to suggest that Commerce’s sampling of record documents revealed discrepancies in the holdings reported by Jindal SAW in its questionnaire responses or that there are discrepancies in the sampling of documents consulted by Commerce. Nor does U.S. Steel refute Commerce’s- contention that there is no way to confirm or calculate a bottom-line figure for indirect ownership based upon documents on the record. The court concludes that Commerce’s determination is based upon reasonable reading of the record because there was no apparent way to confirm or even calculate a bottom-line figure for indirect ownership by consulting record documents and Commerce’s examination of sample documents yielded no discrepancies.¹⁸

U.S. Steel argues.that the O.P. Jindal family has significant direct and indirect ownership interests in Jindal SAW and in its suppliers of steel billets and electricity through the respective promoter groups of Jindal SAW, which collectively have significant interests in each entity.¹⁹ U.S. Steel Comments 22-23. U.S. Steel’s overview of the holdings assumes that the shareholdings of the promoter group is attributable to Jindal family members without sufficient basis in the record. U.S. Steel relies on the notion that a Jindal SAW prospectus provided by U.S. Steel, which defines “promoters” as “Mr. P.R. Jindal, the Jindal Family, and persons or other entities controlled by them,” means that only P.R. Jindal family members make up the promoter groups and own shares in Jindal SAW and its suppliers. See U.S. Steel Comments on Jindal SAW Questionnaire Response at Ex. G, PD 167-171, bar codes 3167500-01-05 (Dec. 6, 2013). As the court noted in U.S. Steel, “ ‘[promoter’ is a -term of art under Indian Securities law.” U.S. Steel, 40 CIT at -, 179 F.Supp.3d at 1129 n.1'6. Commerce considered the potential meaning of this term advanced by U.S. Steel, and it reasonably concluded that the language of this prospectus, which appears in glossary definitions, is not an attempt to define a complete and accurate list of individuals belonging to the promoter group.²⁰ Remand Results at 38. Instead, Commerce reasonably concluded that the glossary definition only highlights the most important or well-known promoters. Id. U.S. Steel points to no record evidence detracting from Commerce’s appraisal of the limited significance of this prospectus.²¹ The court declines to reweigh the evidence and concludes that Commerce adequately examined both direct and indirect ownership to support its conclusion that Jindal family members lack sufficient direct and indirect interests in Jindal SAW to create the potential to impact price, production, and cost of subject merchandise.²²

B. Board Memberships and Management Positions Held By the Jin-dal Family Grouping

U.S. Steel does not challenge Commerce’s determination that the board memberships and management positions held by Jindal family members were insufficient to support its determination that Jindal SAW is not affiliated with its suppliers. See id. at 22-29. The court concludes that Commerce’s determination regarding board memberships and management positions is supported by substantial evidence.

As discussed above, Commerce’s regulations provide that, among other considerations, it will consider “[corporate or family groupings” in evaluating whether or not two entities are affiliated in that they are under the common control under 19 U.S.C. § 1677(33)(F). 19 C.F.R. § 351.102(b)(3). However, as also discussed earlier, Commerce “will not find that control exists on the basis of these factors unless the relationship has the potential to impact decisions concerning the production, pricing, or cost of subject merchandise.” Id.

Although Commerce acknowledged that at least one member of the O.P. Jindal family held directorships in Jindal SAW and in its suppliers of steel billets and electricity, Commerce supported its determination that O.P. Jindal family members do not control Jindal SAW by referencing its findings at verification that only four of its eleven directors are Jindal family members and that two of those four are non-executive members, meaning that they do not work at the company.²³ Id. at 16-17 (quoting Jindal SAW Sales Verification Report at 2-6, CD 318, bar code 3199831-01 (May 6, 2014) (“Jindal SAW Sales Verification Report”)). Commerce also observed that, after examining financial statements, Jindal SAW’s articles of association, and reviewing the responses to determine that none mentioned separate “voting trust agreements” or other such “control” documents giving the Jindal family control of the board other than that provided explicitly in Jindal SAW’s articles of association or Indian company law. Id. at 17 (citing Jindal SAW Sales Verification Report at 2-6). On this basis, Commerce concluded that nothing on the record suggests anything other than “a conventional one-seat-one vote arrangement.” Id. Therefore, the court concludes it is reasonable for Commerce to conclude that board memberships and management positions held by Jindal family members did not create a potential to impact decisions concerning production, pricing, and cost of subject merchandise.

C. Evaluation of Close Supplier Relationships

Lastly, Commerce determined that the supplier relationships between Jindal SAW and its suppliers of steel billets and electricity do not create potential for control because the record does not demonstrate that Jindal SAW’s relationship with its suppliers makes it reliant on those suppliers, or vice versa. See Remand Results 18-19. U.S. Steel argues that Commerce unreasonably concluded the close supplier relationship between Jindal SAW and Jindal Steel and Power Limited (“JSPL”) did not create a significant potential for manipulation of production, pricing and cost of subject merchandise because the extent of Jindal SAW’s supply of steel billets provided by JSPL makes such a determination unreasonable. U.S. Steel Comments at 27-29. Defendant responds that Commerce supported its determination by highlighting: (1) record evidence suggesting Jindal SAW has alternative sources steel billets; (2) the lack of formal or informal documentation requiring Jindal SAW to purchase from JSPL or to limit its purposes from other suppliers; (3) the absence of record evidence indicating JSPL is prohibited from supplying other manufacturers of OCTG; and (4) the absence of leverage one company has of the other. Def.’s Reply 30-31 (citing Remand Results 18-19). Commerce’s determination is supported by substantial evidence.

As already discussed, Commerce’s regulations require it to consider the existence of close supplier relationships in determining whether firms are affiliated under 19 U.S.C. § 1677(33). 19 C.F.R. § 351.102(b)(3). However, in order to find firms that are affiliated on the basis of common control, Commerce’s regulations also require the evidence to show that the close supplier relationships creates the potential to impact pricing, production, or cost of subject merchandise. Id.

Commerce supported its determination that Jindal SAW’s supplier relationship with its suppliers of steel billets and electricity does not create a significant potential to impact the price, production, or cost of subject merchandise by reviewing Jindal SAW’s alternative sources of supply for its inputs as well as record evidence indicating that it can and does avail itself of those alternative sources.²⁴ Remand Results 18. Commerce likewise supported its conclusion that JSPL is not reliant on Jindal SAW as a purchaser of inputs by noting JSPL may sell to other manufacturers OCTG and other downstream steel products. Id. Commerce also observes that record contains no formal or informal contract or other obligation requiring Jindal SAW to purchase from JSPL or limit its purchases from other sources. Id. In addition, Commerce notes that the input at issue is a commodity product readily found elsewhere. Id. Finally, Commerce supported its conclusion that Jindal SAW is unable to extract an unusually low price from JSPL by observing that the inputs actually purchased by Jindal SAW from abroad are not consistently more expensive for Jindal SAW than JSPL’s materials. Id. at 18-19. Based on all of these facts, Commerce’s determination that the supply relationship does not create the potential to impact Jindal SAW’s decisions concerning pricing, production, or costs is reasonable.

U.S. Steel argues that Commerce’s finding that Jindal SAW is not dependent upon JSPL is contradicted by evidence on the record about the extent of inputs Jindal SAW purchased from other suppliers. U.S. Steel Comments 28. However, Commerce supported its determination by reviewing record information reflecting actual purchases from suppliers other than JSPL.²⁵ Remand Results at 44. Further, U.S. Steel argues that Commerce’s determination that there is no indication Jindal SAW is able to extract an unusually low price from JSPL is contradicted by the record. U.S. Steel Comments 29. Commerce considered U.S. Steel’s comparison of prices paid to JSPL versus other suppliers and reasonably concluded that Jindal SAW is unable to extract an unusually low price because the prices placed on the record by U.S. Steel lack information about important determinants of OCTG prices.²⁶ Remand Results 44. U.S. Steel offers no evidence that such details regarding the characteristics of products purchased by Jindal SAW is on the record. Therefore, the court declines to reweigh the evidence, and Commerce’s determination that Jindal SAW is not reliant upon JSPL for its billets is supported by substantial evidence.

D. Conclusion

Commerce reasonably concluded that the collective direct and indirect shareholdings of members of the O.P. Jindal family grouping, including those held through promoter group entities,- did not rise to a level that creates the potential for Jindal SAW to control its suppliers of steel billets and electricity. Commerce’s determination that-the board memberships and management positions held -by Jindal family members as well as the close supplier relationships between Jindal SAW and its suppliers of steel billets and electricity do not create a potential to impact production, pricing, or cost of subject merchandise. Thus,- Commerce’s determination that Jin-dal SAW and its suppliers of these inputs are unaffiliated is supported by substantial evidence.

III. Commerce’s Recalculation of Jindal SAW’s Direct Material Costs

In U.S. Steel, the court held that Commerce inadequately explained its conclusion that the costs for CONNUMs with different physical characteristics would not generate different yield losses, which rendered Commerce’s determination to rely upon Jindal SAW’s reported yield loss data unsupported by substantial evidence.²⁷ See U.S. Steel, 40 CIT at -, 179 F.Supp.3d at 1133-34. The court held that, on remand

Commerce must explain why Jindal SAW’s reported yield loss data, which clearly did not track yield losses by production stage or physical characteristics of the merchandise, nonetheless did not distort Jindal SAW’s COP for specific CONNUMs of subject merchandise or reconsider its determination.

Id., 40 CIT at -, 179 F.Supp.3d at 1135. On remand, Commerce determined that Jindal SAW’s cost reporting methodology does not allocate yield losses on a basis that reasonably reflected differences in processing costs for merchandise with differing physical characteristics and applies partial adverse facts available on remand.²⁸ Remand Results 23. No party challenges Commerce’s determination. For the reasons that follow, Commerce has complied with the court’s directions on remand, and its determination to apply partial AFA to Jindal SAW’s yield loss data is supported by substantial evidence.

Commerce generally “shall consider all available evidence on the proper allocation of costs ... if such allocations have been historically used by the exporter or producer.” 19 U.S.C. § 1677b(f)(1)(A). According to the statute,

[cjosts shall normally be calculated based on the records of the exporter or producer of the merchandise, if such records are kept in accordance with generally accepted accounting principles of the exporting country ... and reasonably reflect the costs associated with the production and sale of the merchandise.

Id. Once Commerce has concluded that a respondent’s COP reporting methodology complies with generally accepted accounting principles, Commerce evaluates whether a respondent’s COP data, as reported, “reasonably reflect[s] the costs associated with the production and sale of the merchandise.” Id.

Although the statute does not define what it means for reported cost information to reasonably reflect that party’s COP, the court noted in U.S. Steel that the Court of Appeals for the Federal Circuit has broadly defined when costs “reasonably reflect the costs associated with the production and sale of the merchandise” to mean that the costs, as reported would not distort the company’s true costs. U.S. Steel, 40 CIT at -, 179 F.Supp.3d at 1133 (citing Am. Silicon Techs. v. United States, 261 F.3d 1371, 1377 (Fed. Cir. 2001)).

On remand, Commerce concludes that, “while Jindal SAW reported certain CON-NUMs with unique direct material costs, it is not apparent that the yields it relied on to calculate these CONNUM costs reflect the unique processing that the CONNUMs received.”²⁹ Id. at 47. Accordingly, Commerce determined that it is necessary to adjust the per-unit direct material costs to take into account varying processing costs for CONNUMs with different physical characteristics, including thickness and diameter. Id. at 48.

No party challenges Commerce’s determinations that Jindal SAW’s yield loss data did not reasonably reflect its COP or its determination to apply partial AFA.³⁰ Commerce has complied with the court’s instructions, and has reasonably supported its determination. See Remand Results 46. Because no party challenges Commerce’s use of partial AFA to fill gaps in the record for Jindal SAW’s reported yield losses any challenge to Commerce’s use of AFA is waived.³¹

IY. Cost Assigned to GVN’s Dual-Grade Merchandise by Commerce

In U.S. Steel, the court held that, while Commerce reasonably assigned GVN’s N/L-80 OCTG, which could meet multiple performance specifications, costs associated with merchandise meeting higher performance specifications (i.e., L-80 grade OCTG), Commerce failed to explain why its decision to assign the highest COP data for products within the L-80 product grouping is reasonable. U.S Steel, 40 CIT at -, 179 F.Supp.3d at 1154. The court reasoned that, since varying cost information is on the record depending upon the products’ physical characteristics, without any effort to match the COP data assigned to GVN’s dual-grade product, Commerce’s decision to assign the highest cost information among L-80 products from GVN’s cost database could only have been the product of an adverse inference. Id. The court held that, on remand, “Commerce must either explain why assigning the highest costs for L-80 products from GVN’s cost database to its dual use products was reasonable in light of the characteristics of GVN’s dual-use products or explain its application of an adverse inference.” Id.

On remand, Commerce found that it had “unintentionally overlooked its standard ‘proxy cost’ methodology by selecting the highest cost of L-80 grade” merchandise. Remand Results 27-28. Commerce revised the cost assigned to the L-80 grade CON-NUMs to conform to its standard proxy cost methodology, which provides that Commerce determines cost values for products for which it lacks COP data by matching COP to the most similar products based on reported physical characteristics where no adverse inference is applied.³² Id. at 28. Commerce supported its determination to assign the costs associated with cost data most similar to L-80 product by matching the reported physical characteristics as part of its application of facts available.³³ Id. Based upon these cost adjustments, Commerce calculated a de minimis AD rate of 1.07 rate for GVN. Id. Commerce has complied with the court’s instructions in U.S. Steel, and its determination is supported by substantial evidence.

According to the statute,

[cjosts shall normally be calculated based on the records of the exporter or producer of the merchandise, if such records are kept in accordance with generally accepted accounting principles of the exporting country ... and reasonably reflect the costs associated with the production and sale of the merchandise.

19 U.S.C. § 1677b(f)(1)(A). If, however, Commerce determines that the records of the respondent cannot properly form an accurate basis upon which to calculate that respondent’s COP, Commerce may use facts otherwise available for calculating a respondent’s COP under certain circumstances. See 19 U.S.C. § 1677e(a).³⁴ If Commerce “finds that an interested party has failed to cooperate by not acting to the best of its ability to comply with a request for information ... [Commerce] may use an inference adverse to the interests of that party in selecting from among the facts otherwise available.” 19 U.S.C. § 1677e(b).

U.S. Steel’s arguments are unavailing. U.S. Steel challenges Commerce’s application of its proxy cost methodology to assign costs other than those associated with the most expensive L-80 grade merchandise as unsupported by substantial evidence because the costs assigned to GVN’s dual-grade product are inconsistent with pricing data on the record for dual-grade product.³⁵ U.S. Steel Comments 5-6. However, Commerce determined that there is inadequate data on the record demonstrating that the costs assigned to GVN’s dual-grade products are inaccurate.³⁶ Remand Results at 49. In the absence of conflicting cost data demonstrating that the costs assigned to GVN’s dual-grade products are distorted, Commerce also reasonably declined to consider pricing data in place of cost data. See id. at 50. Commerce reasonably concluded that pricing factors other than cost, such as customers,' level of trade, timer period, and region, may impact pricing of dual-grade products. See id. at 50. Therefore, Commerce’s determination that its assignment of costs to GVN’s dual-grade products of costs in GVN’s database associated with products whose physical characteristics most closely match the physical characteristics of GVN’s dual-grade products does not distort the COP for such products is supported by substantial evidence.

U.S. Steel also argues that Commerce’s decision not to assign the highest costs associated with dual-grade merchandise is inconsistent with Commerce’s application of its practice in past proceedings.³⁷ U.S. Steel Comments 6, However, U.S. Steel misstates Commerce’s practice. As the court stated in U.S. Steel, “nothing in Commerce’s practice indicates that it selects the highest costs associated with the product with the highest performance specifications where there are multiple CONNUMs within that higher performance product category included with a respondent’s COP database.” U.S. Steel, 40 CIT at -, 179 F.Supp.3d at 1154 (citing Issues and Decision Memorandum for the Eighth Administrative Review of Small Diameter Circular Seamless Carbon and Alloy Steel Standard, Line and Pressure Pipe from Brazil; Final Results of Anti-dumping Duty Administrative Review at 5, A-351-826, (Feb. 4, 2005), available at http://ia.ita.doc.gOv/frn/summary/brazil/E 5-584-l.pdf (last visited Mar. 13, 2017 (“SSSLP from Brazil I & D”)). Rather, Commerce’s practice is to assign costs associated with CONNUMs in the higher performance product category, not the highest costs within that product category. See id. (citing SSSLP from Brazil I&D at 5).

U.S. Steel attempts to limit the application of Commerce’s proxy cost methodology to circumstances where there are no costs for the product because the product is sold but not produced during the POI. U.S. Steel Comments 6-7 (citing SSSLP from Brazil I&D at 5; Issues and Decision Memorandum for the 2002-03 Antidump-ing Duty Administrative Review: Certain Small Diameter Carbon and Alloy Seamless Standard, Line, and Pressure Pipe from Romania at 30, A-485-805, (Feb. 4, 2005), available at http://ia.ita.doc.gov/frn/ summary/romania/E5-586-l.pdf (last visited Mar. 13, 2017); Certain Stilbenic Optical Brightening Agents from Taiwan: Issues and Decision Memorandum for the Final Results of Antidumping Duty Administrative Review; 2013-2014 at 5, A-583-848, (Oct. 5, 2015), available at http:// ia.ita.doc.gov/frn/summary/taiwan/2015-25966-l.pdf (last visited Mar. 13, 2017) (“SOBA from Taiwan I & D”); Issues and Decision for the Final Results of the Anti-dumping Duty New Shipper Review and the Antidumping Duty Administrative. Review of Certain Corrosion-Resistant Carbon Steel Flat Products from the Republic of Korea: Tenth Administrative Review (2002-2003) at 19-20, A-580-816, (Mar. 7, 2005), available at http://ia.ita.doc.gov/frn/ summary/korea-south/E5-1065-l.pdf (last visited Mar. 13, 2017) (“CORE from Korea I & D”)). U.S. Steel correctly points out that, in the proceedings it references, Commerce applied its model matching criteria to determine proxy costs to match sales where there were no identical sales in the respondent’s cost database. See U.S. Steel Comments 7. However, there is no indication Commerce limits its model matching practice to such circumstances. See SOBA from Taiwan I&D at 4-5 (stating that it is Commerce’s practice to rely on its proxy cost where a respondent did not manufacture a product during the reporting period, but not contrasting that scenario to circumstances where such merchandise is produced during the reporting period); CORE from Korea I&D at 20 (similarly not contrasting the use of this practice to circumstances where such merchandise is produced during the reporting period). U.S. Steel’s claim that this practice is limited to such circumstances is belied by Commerce’s application, of its proxy cost methodology in SSSLP from Brazil I&D, where there is no indication that merchandise meeting multiple specifications was not produced during the period of review. See SSSLP from Brazil I&D at 3-5.

U.S. Steel argues it is distortivé for Commerce to apply its proxy cost methodology to assign costs to GVN’s dual-grade products based on L-80 models with similar physical characteristics. U.S. Steel Comments 8. As already discussed, Commerce reasonably concluded that its assignment of costs based upon the products’ physical characteristics does not yield distorted COP for GVN’s N/L-80 products because the price differences highlighted by U.S. Steel may result from factors other than cost differences. See Remand Results 50. Therefore, Commerce reasonably concluded that the price differences between dual-grade products and L-80 grade products may not be the result of differences in cost of producing dual-grade merchandise. See id. Commerce’s practice of applying its model match here is consistent with statutory limitations on Commerce’s authority to apply AFA to a respondent that fails to provide cost information regardless of whether Commerce finds that the respondent has failed to cooperate to the best of its ability. See 19 U.S.C. § 1677e(b).

Lastly, U.S. Steel argues that Commerce acted contrary to law in declining to apply AFA to GVN’s missing COP information for GVN’s dual-grade merchandise. U.S. Steel Comments 10-12. Specifically, U.S. Steel contends Commerce ignored evidence that GVN failed to cooperate during the investigation by failing to provide costs for its dual-grade products. Id. at 12. As an initial matter, the statute gives Commerce discretion to apply an adverse inference where a party fails to cooperate by not acting to the best of its ability. 19 U.S.C. § 1677e(b). Here, Commerce supported its decision not to apply AFA to assign costs to GVN’s dual-grade products because the missing costs for dual grade merchandise did not become evident until the briefing stage of the underlying proceeding after Commerce requested that the dual-grade merchandise be recoded as L-80 merchandise.³⁸ Remand Results 51. Commerce has reasonably exercised its considerable discretion because Commerce explains that GVN’s failure to report COP data for its dual-grade products resulted from Commerce’s requests to recode CONNUMs at the briefing stage. See id. Commerce further justifies its determination that GVN did not fail to cooperate to the best of its ability by explaining that GVN should not have been aware the information was requested earlier. See id. U.S. Steel highlights no record information to support its speculation that GVN failed to provide costs for its dual-grade products to avoid producing cost information for those products. See U.S. Steel Comments at 11. Commerce reasonably grounded its determination that GVN should not have been aware that CONNUM-specific cost information for recoded products was missing in the repeated requests that those products be recoded as L-80 products, and not failure to cooperate. On this record, the court cannot say Commerce’s determination is unreasonable.

CONCLUSION

Therefore, for the reasons discussed, the court sustains the Remand Results, and judgment will enter accordingly.

1 . Commerce collapsed GVN with two of its suppliers, Maharashtra Seamless Limited and Jindal Pipes Limited, because Commerce determined that the companies are affiliated because of common control under the D.P. Jin-dal Group and both GVN and its suppliers possess a level of common ownership, intertwined operations, and shared employees and facilities that create a significant potential for manipulation of price or production. See Final Decision Memo at 24-25; see also 19 C.F.R. § 351.401(f) (2013). The court sustained Commerce’s collapsing determination as supported by substantial evidence. U.S. Steel, 40 CIT at -, 179 F.Supp.3d at 1138-39.

2 . Further references to the Code of Federal Regulations are to the 2013 edition.

3 . Commerce instructed Jindal SAW to report unique cost information in its cost database for each control number ("CONNUM”), which Commerce defines based on the unique physical characteristics identified in subject merchandise. See Final Decision Memo at 36. Yet, Commerce found that Jindal SAW [[]] the costs reported at the production stage for [[]] of OCTG across different CONNÚMs. Id. Jindal SAW reported multiple CONNUMs with [[]] cost information. See Cost of Production and Constructed Value Calculation Adjustments for the Final Determination— Jindal SAW at 1-2, PD 354, bar code 3215359-01 (Jul. 10, 2014).

4 . Further citations to the Tariff Act of 1930, as amended, áre to the relevant provisions of Title 19 of the U.S. Code, 2012 edition.

5 . Although 19 U.S.C. §§ 1677e(a)-(b) and 19 C.F.R. § 351.308(a)-(c) each separately provide for the use of facts otherwise available and the subsequent application of an adverse inference to those facts, Commerce uses the shorthand term "adverse facts available” or "AFA” to refer to Commerce’s use of such facts otherwise available with an adverse inference. See, e.g., Final Decision Memo at 38 (discussing the circumstances where the statute permits Commerce.to apply AFA to uncooperative companies). Specifically, Commerce calculated an adjustment factor representing the absolute difference between the highest yield loss and the simple average yield loss of the three CONNUMs Commerce examined during its cost verification. Remand Results 48.

6 . According to Commerce's regulations, the A-A methodology "involves a comparison of the weighted average of the normal values with the weighted average of the export prices ,.. for comparable merchandise.” 19 C.F.R. § 351.414(b)(1).

7 . Commerce calculates a respondent's dumping margin by determining "the amount by which the normal value exceeds the export price of the subject merchandise.” 19 U.S.C. § 1677(35)(A). The "weighted average dumping margin” is "the percentage determined by dividing the aggregate dumping margins determined for a specific exporter or producer by the aggregate export prices of such exporter or producer.” 19 U.S.C. § 1677(35)(B).

8 . The A-T methodology "involves a comparison of the weighted average of the normal values to the export prices ... of individual transactions for comparable merchandise.” 19 C.F.R. § 351.414(b)(3) (2013).

9 . Commerce’s regulations echo this preference, providing that Commerce will apply AA to calculate dumping margins in investigations unless another method is appropriate in a particular case. See 19 C.F.R. § 351.414(c)(1) (2013). Further citations to the Code of Federal Regulations are to the 2013 edition, unless otherwise noted.

10 . In the first stage of Commerce’s differential pricing analysis applied in this investigation, the Cohen’s d test, Commerce states that it “seeks to determine whether evidence exists demonstrating a pattern of prices that differ significantly.” Remand Results 4; see generally 19 U.S.C. § 167Yf-1 (d)(1)(B)(i). Commerce considers sales to pass the Cohen’s d test if there are two or more sales in both the comparison and test group and the resulting Cohen’s d coefficient is equal to or greater than 0.8. See Remand Results 4. Commerce deems results that satisfy these conditions to be a strong indication of significant price differences. See id Conversely, ”[t]he fact that the sales in the test group have not passed means that there is no evidence that the prices of the sales in the test group differ significantly from the prices of all other sales of the comparable merchandise.” Id.

Commerce states that the second stage, the ratio test, "assesses the extent of the significant price differences for all sales." Id. at 5 (internal quotations omitted); see generally 19 U.S.C. § 1677f-1(d)(1)(B)(i). The purpose of these first two stages of the differential pricing analysis “is to determine whether the [A-A] method, applied to all U.S. sales, is the appropriate comparison method to calculate a respondent’s weighted-average dumping margin.” Remand Results 5 (citing 19 C.F.R. § 351.414(c)(1)).

Commerce states that the third stage, which is not challenged by U.S. Steel, examines "whether using only the [A-A] method can appropriately account for such differences.” Decision Memorandum for the Preliminary Determination in the Less-Than-Fair-Value Investigation of Oil Country Tubular Goods from India at 11, PD 258, bar code 3181829-01 (Feb. 14, 2014); see generally 19 U.S.C. § 1677f-1(d)(1)(B)(ii).

11 . On December 14, 2014, Defendant submitted indices to the public and confidential administrative records for its final results, which identify the documents that comprise the public and confidential administrative records to the Commerce’s final determination. The indi-ces to the public and confidential administrative records to Commerce's final determination can be located at ECF No. 22-1. On September 7, 2016, Defendant submitted indi-ces to the public and confidential administrative records for its Remand Results, which identify documents that comprise the public and confidential administrative records to Commerce’s Remand Results. Those indices can be located at ECF Nos. 116-2 and 116-3, respectively. All further references to the documents from the administrative records to the final results and the remand results are identified by the numbers assigned by Commerce in these administrative records.

12 . Commerce explains that

when sales do not pass the Cohen’s d test, either because the Cohen’s d coefficient is less than 0.8 or because there are not two or more sales in the comparison or test group, the proposition of significant price differences has not been proven for those sales in the test group, and such differences are found not to exist.

Remand Results 4. U.S. Steel does not challenge the reasonableness of Commerce’s determination that sales whose Cohen’s d coefficient is less than 0.8 do not demonstrate the existence of significant price differences. See U.S. Steel Comments 15-20.

13 . Commerce describes its ratio test' as follows:

If the value of sales to purchasers, regions, and time periods that pass the Cohen’s d test accounts for 66 percent or more of the value of total sales, then the identified pattern of [export prices] ... that differ significantly supports the consideration of the application of the [A-T] method to all sales as an alternative to the [A-A] method. If the value of sales to purchasers, regions and time periods that pass the Cohen's d test accounts for more than 33 percent and less than 66 percent of the value of total sales, then the results support consideration of the application of an [A-T] method to those sales identified as passing the Cohen’s d test as an alternative to the [A-A] method, an application of the [A-A] method to those sales identified as not passing the Cohen’s d test. If 33 percent or less of the value of total sales pass the Cohen’s d test, then the results of the Cohen’s d test do not support consideration of an alternative to the [A-A] method.

Prelim. Decision Memo at 11. If the denominator of the ratio of sales passing the Cohen’s d test (i.e., the value of all of respondents’ sales) remains constant, then, where a lesser value of sales passes the Cohen’s d test, the ratio of the value of sales passing the Cohen's d test relative to the value of all .sales would decrease.

14 . U.S. Steel argues that the ratio test, as applied in this investigation, fails to accurately measure the extent of respondents’ sales that contribute to a pattern of differential pricing. U.S, Steel Comments 17-18. Specifically, U.S. Steel argues that Commerce must recognize that any pricing behavior found for a subset of U.S. sales should be applied to all U.S. sales. Id. at 18. Although U.S. Steel contends that Commerce often tests a subset of data and draws inferences regarding an entire universe of data, U.S. Steel points to no reason that Commerce could not reasonably decline to do so where it has reason to believe that such sampling would lead to unrepresentative results. See , U.S. Steel Comments 18. Commerce explains that, absent evidence that including multiple observations in both the test group and the comparison group and a Cohen’s d coefficient equal to or greater than 0.8, it cannot accurately extrapolate a pattern to all sales. See Remand Results 5. Moreover, Commerce explains that extrapolating results from the subset of U.S. sales for which a Cohen’s d coefficient was calculated for all U.S. sales, whether or not a pattern of prices that differ significantly exists, would create a presumption that whatever is found for the subset of U.S. sales should apply to all U.S. sales. Id. at 32. In the absence of record evidence indicating that such a pattern can accurately be extrapolated to all sales, Commerce’s rationale is reasonable.

15 . In the calculation of normal value, the statute permits Commerce to disregard transactions directly or indirectly between affiliated persons in certain circumstances. See 19 U.S.C. § 1677b(f)(2). If the affiliated-party transaction involves the production of a major input to the merchandise, Commerce may determine the value of the major input on the basis of other information on the record regarding the costs of production under certain circumstances. See 19 U.S.C. § 1677b(f)(3).

16 . Commerce also acknowledged that Jindal SAW’s financial statements list direct shareholders but not the shareholders of the corporate entities that are shareholders. Remand Results at 9. Commerce recognized that the forms filed with Indian securities regulators reflect all holdings of individual and corporate shareholders, but they do not indicate the holdings of those individual and corporate shareholders. Id.

17 . Commerce reviewed the shareholding patterns in selected additional forms filed with Indian securities regulators, including that of the single largest shareholder of Jindal SAW. Remand Results at 10. Although Commerce acknowledges that its review method "does not result in a total, ‘bottom-line’ figure for indirect ownership, the Department believes it is a reasonable sampling method by which to confirm Jindal SAW’s own statements that neither itself [[ ]] were controlled by any Jin-dal family grouping.” Id at 10. Commerce explained that it relies on similar sampling methods in other contexts, and it traces additional inaccuracies that appear in the initial sampling to gain a better understanding of the breadth of the inaccuracies where they appear. Id. at 11. Commerce explains that, if its examination of record documents had indicated significant levels of indirect ownership or revealed discrepancies, it would have investigated further. Id. Here, Commerce explains that its investigation yielded no discrepancies and it decided to move on to other issues. Id.

As an illustrative example, Commerce analyzed O.P. Jindal family holdings in one of Jindal SAW’s promoters, [[]], Remand Results at 10. Commerce explains that three O.P. Jindal family members each own less than two percent of the shares of [[ ]], which maltes their ownership in Jindal SAW through the [[ ]] vehicle only [[ ]] (i.e., [[ ]]% x (2% + 2% 2%)) = [[]]%. Id.

18 . U.S. Steel does not challenge Commerce's determination with regard to direct ownership by Jindal family members. See U.S. Steel Comments 22-27. The court did not suggest in its decision that the direct ownership findings in Commerce's final determination were unsupported by substantial evidence, but rather that the statute requires Commerce to also consider indirect ownership, board memberships and management positions held by Jindal family members, and the nature of the supplier relationship in order to assess whether the firms are under the common control of a family grouping. See U.S. Steel, 40 CIT at -, 179 F.Supp.3d at. 1129-31. Therefore, the court did not direct Commerce to revisit the portion of its determination on remand that the direct holdings, of the O.P. Jindal family did not create the potential for control of Jindal SAW and its suppliers is based upon a reasonable reading of all the record evidence.

19 . Specifically, U.S. Steel argues that record documents demonstrate that [[ ]] the promoter groups have the following aggregate interest in each respective entity: (1) [[]] for Jin-dal SAW; (2) [[]] for JSPL; and (3) [[]] for [[]]. U.S. Steel Comments 22-23 (citing Jin-dal SAW Cost and Sales Verification Exhibits at Ex. 4, CD 260-295, bar codes 3190334-01-36' (Mar. 25, 2014); U.S. Steel Corporation Comments re: Responses of Jindal SAW, Limited to Sections A, B, C, and D Questionnaire at Ex. D, CD 108-112, bar codes 3167492-01-05 (Dec. 6, 2013)).

20 . Commerce also noted that this

conclusion is borne out by a nearly identical line on page "i” of a [[]] prospectus, which defines "promoters” as “[[]]” but then includes a second line on page 180 of the prospectus noting "[i]n addition to the Promoters named above [in the glossary], there are a number of companies that form part of the group which constitutes our Promoter Group.”

Remand Results 38 (citing U.S. Steel U.S. Steel Comments on Jindal SAW Supplemental Section D Questionnaire at Ex. C, CD 215-217, bar codes 3181479-01-03 (Feb. 14, 2014)).

21 . Moreover, Commerce reasonably concluded that U.S. Steel’s claim that promoters own significantly more of Jindal SAW than non-promoters was not supported by the record based upon its examination of shareholding breakdown provided in a Jindal SAW prospectus from slightly outside the period of investigation. See Remand Results at 15. After examining a list of top ten shareholders in which Jindal SAW indicated which companies were promoter entities, Commerce concluded that promoter and non-promoter entities are split almost [[]]. IcL This rationale also supports Commerce’s decision not to trace every indirect ownership interest. Commerce reasonably concluded there was no reason to trace these indirect interests because, even if they proved to be 100% Jindal family-owned, they would not be likely to add up to a controlling interest. See id.

U.S. Steel relies on an "Information Memorandum” for Hexa Tradex Limited, which it claims detracts from Commerce’s determination that Jindal SAW is not majority owned through Jindal family holdings in promoter groups. See U.S. Steel Comments 22-24 (citing U.S, Steel Corporation Comments re: Responses of Jindal SAW, Limited to Sections A, B, C, and D Questionnaire at Ex. D, CD 108-112, bar codes 3167492-01-05 (Dec. 6, 2013)). However, Commerce justified its conclusion that the Hexa Information Memorandum, which summarizes information elsewhere on the record, including board members of Jindal SAW, largely confirms information that Jindal SAW is not majority-owned by Jindal family members. Remand Results 20. Commerce found the memorandum indicates 54 percent of Jindal SAW is publicly traded, so whatever the family’s control within the group of promoter shareholders, the promoter group holdings constitute a minority. ⅛ This conclusion is reasonable particularly when viewed together with Commerce’s investigation of the family’s indirect shareholdings in Jindal SAW's largest public shareholders and its finding that the Jindal family held very small interests in the largest public shareholder entities. See Remand Results 9.

Commerce also reasonably concluded that, even if a promoter is an influential member of a corporation’s structure, as U.S. Steel contends, the fact that some promoters belong to one family does not indicate control because a family who holds less than a controlling share of the board through promoter entities, which themselves hold less than a controlling stake, would not necessarily mean the family controls the corporation. See id. at 16.

22 .Commerce correctly notes that, once its determination that Jindal family control over Jindal SAW did not exist, the possibility of finding affiliation between Jindal SAW and any other company through common control by Jindal family members is moot. See Remand Results at 12-13. Nonetheless, Commerce proceeded to analyze the direct and indirect holdings of O.P. Jindal family members in JSPL and [[]]. Id. at 13. Commerce noted that Jindal SAW provided no information contradicting its finding that Savitri Jin-dal is the only Jindal family member with an ownership stake in JSPL, which is [[]]. Id. Commerce also found that no one in the Jin-dal family listed themselves as a director of either JSPL or [[ ]]. IA Commerce found that it could not conduct further analysis into the indirect holdings in either JSPL or [[]] because it found no evidence to contradict Jin-dal SAW's claim that it has no access to information related to the operations of JSPL and [[ ]], which are both privately held companies. Id. The court need not review the reasonableness of Commerce’s findings on Jindal family control over JSPL and [[]] because Commerce’s determination that Jindal SAW is not controlled by the Jindal SAW is supported by substantial evidence,

23 . Commerce notes that at verification it reviewed the directors of Jindal SAW, Jindal Steel and Power Limited and [[]]. Remand Results at 16. However, Commerce notes that [[]], O.P. Jindal's wife, Savitri Jindal. Id. However, Commerce found that there is no record evidence indicating that she or any other Jindal family member is a director of O.IA

24 . Commerce notes that the record shows Jindal SAW has alternative sources to JSPL for purchasing [[ ]], and that it "can and does purchase [[]] from abroad.” Remand Results at 18,

25 . U.S. Steel argues that Commerce's determination that the record shows Jindal SAW has alternative sources of supply for its [[ ]] is unsupported in the record because Jindal SAW purchased [[ ]] billets from- JSPL and the [[]] volumes of purchases from other suppliers were purchases of [[]] that Jindal SAW [[]], not [[]]. U.S. Steel Comments 28 (citing Jindal SAW Second Supplemental Sections A-C Response at 2, CD 248-250, bar codes 3186547-01-02 (Mar. 7, 2014). Further, U.S. Steel argues that other [[]] amounts of billets purchased were for [[]]. See id, (citing Jindal SAW First and Second Supplemental Section D Response at Ex. D-17, CD 117-118, bar codes 3172116-01-02 (Jan. 6, 2014)). Commerce acknowledged that some of the purchases may have been for [[]], but it nonetheless concluded that this fact does not undermine its conclusion that Jindal SAW had alternative sources available. Remand Results 44. Commerce noted that the record contains documents evidencing actual purchases of [[ ]] from several sources other than JSPL. ML (citing Jindal SAW First and Second Supplemental Section D Response at Ex. D-17, CD 117-118, bar codes 3172116-01-02-(Jan. 6, 2014)). Commerce declined to reconsider its determination without actual record evidence that Jin-dal SAW "would experience [[]] disruptions to its operations if JSPL cut off that supply or started charging Jindal SAW [[]]” alone. Id. (quoting U.S. Steel Comments on Draft Remand, Remand CD 5, bar code 3498871-01 (Aug. 18, 2016). Notwithstanding Jindal SAW's [[]] purchases from JSPL, the court cannot say this determination is unreasonable given the evidence of actual purchases from other sources.

26 . U.S. Steel specifically argues that, during the period of investigation, "JSPL provided steel billets to Jindal SAW [[ ]] whereas Jindal SAW’s non-affiliated suppliers provided steel billets to the company [[]]." U.S. Steel Comments at 29. However, Commerce considered this price comparison, and concluded that it does not undermine Commerce’s conclusion that it fails to demonstrate that non-affiliated suppliers charged [[]] prices for similar [[]]. Remand Results 44. Commerce concluded that U.S. Steel’s pricing comparison failed to ' take into account important physical characteristics that Commerce is statutorily required to take into consideration before comparing prices such as the [[]], physical characteris- . tics, level of trade, contemporaneity of -sales, freight adjustments, rebates, direct selling expenses, etc. See id. Without such important information about the characteristics of the billets purchased, the court concludes Commerce’s conclusion is reasonable.

27 . U.S. Steel also argued that Commerce should have applied AFA to Jindal SAW’s reported yield loss cost data because its COP data did not accurately reflect its COP. See Br. Pl. United States Steel Corporation Supp. Mot. J. Agency R. Confidential Version 37-39, Mar. 24, 2015, ECF No. 33. The court deferred decision on Commerce’s decision to decline to apply AFA until after Commerce’s reconsiders its determination to accept Jindal SAW’s yield loss data. U.S. Steel, 40 CIT at -, 179 F.Supp.3dat 1135.

28 . Specifically, Commerce supported its determination by looking at the specific yield loss data reported by three specific CON-NUMs examined at verification. See Remand Results 48-49. Commerce found that, upon further analysis, Jindal SAW calculates yield by [[]] rather than by CONNUM. Remand Results 25. Commerce concluded that, "where a [[]] was used to produce multiple CONNUMs, the CONNUMs have been reported with identical direct material costs regardless of the processing undergone or the physical characteristics of the underlying products. Id. at 25-26. Commerce observed that "there are instances where CONNUMs that received additional processing, such as threading, coupling, upsetting or heat treatment—all physical characteristics defined by the Department—were reported with the same per-until direct material cost as CONNUMs that did not receive additional processing. Id. at 46. Considering the wide range of yields reported by Jindal SAW for the three preselected products reviewed at verification, Commerce concluded “it is not reasonable to assume that this broad mix of products would incur identical yield losses.” IcL

29 . Commerce explained that those direct material costs "may be a weighted-average of the yield loss experience for a multitude of products produced with the same underlying [[]].” Remand Results 47. Commerce noted that, although Jindal SAW acknowledges that additional processing such as threading or heat treatment would impact production costs and reported additional conversion costs for its products, the reported COP do not reflect yield losses reflecting the specific processing each product received. Id. at 48.

30 . Specifically, Commerce calculated and applied an adjustment factor representing "the absolute difference between the highest yield loss and the simple average yield loss of the three preselected CONNUMs” examined at verification to Jindal SAW's yield losses. Remand Results 49.

31 .If Commerce determines that the records of the respondent cannot properly form an accurate basis upon which to calculate that respondent’s COP, then Commerce shall use facts otherwise available in reaching the determination. 19 U.S.C. § 1677e(a). Commerce may apply AFA in selecting from among the facts otherwise available where it "finds that an interested party has failed to cooperate by not acting to the best of its ability to comply with [its] request for information.” 19 U.S.C. § 1677e(b).

32 . In order to assign costs associated with the most similar products, Commerce relies upon the same model matching process used to determine the closest matches between products sold in the home market and in the United States for its dumping margin calculation. Remand Results 28 (citing Certain Stil-benic Optical Brightening Agents from Taiwan: Issues and Decision Memorandum for the Final Results of Antidumping Duty Administrative Review; 2013-2014 at 5, A-583-848, (Oct. 5, 2015), available at http://ia.ita. doc.go v/frn/summary/taiwan/2015-25966-1. pdf (last visited Mar. 13, 2017). Once it identifies the products that are most similar to those for which Commerce lacks cost information on the record, it relies on the cost values reported for those products to replace the missing cost values. Id.

33 . Before matching remaining reported physical characteristics, Commerce limited its examination to L-80 product costs that were for [[]] because the N/L-80 grade products are [[]] products. Remand Results at 28.

34 .Commerce shall "use facts otherwise available in reaching the applicable determination” if:

(1) necessary information is not available on the record, or

(2) an interested party or any other person—

(A) withholds information that has been requested by [Commerce] ...,

(B) fails to provide such information by the deadlines for submission of information or in the form or manner requested ...

(C) significantly impedes a proceeding under this subtitle, or

(D) provides such information but the information cannot be verified ....

19 U.S.C. § 1677e(a).

35 . Specifically, U.S. Steel argues that the record indicates that N/L-80 grade OCTG has [[]] specifications than L-80 grade product, and these stricter specifications make it [[]] to produce N/L-80 OCTG. U.S. Steel Comments 5. U.S. Steel argues that the costs Commerce assigned GVN’s dual grade product are not accurate because the record demonstrates that [[]]. Id. at 5-6. U.S. Steel argues that Commerce’s determination to assign lower costs than the highest costs reported by GVN for L-80 grade OCTG product is inconsistent with this record information. Id. at 6. Commerce explains on remand that its practice is to compare physical characteristics, not price differences because price differences may result from "a number of factors other than cost differences, including customers, level of trade, time period, region, etc.” Remand Results 50. U.S. Steel does not challenge the reasonableness of this practice.

36 . Commerce notes that, although U.S. Steel claims GVN's dual-grade products meet [[]] specifications, which makes them [[]], there is no data on the record to demonstrate that dual-grade products are [[ ]] to produce, only data that such products are sold at [[]]. Remand Results 49-50. Commerce recognizes U.S. Steel’s claim that "[[]].” Id. However, Commerce explains that its practice is to compare physical characteristics, not their price differences because price difference may result from "a number of factors other than cost differences, including customers, level of trade, time period, region, etc.” Id. at 50. The court does not consider it unreasonable for Commerce to consider physical characteristics in matching cost data for the missing cost data to control for pricing variations Commerce does not account for elsewhere in its margin calculations.

37 .U.S. Steel argues that Commerce’s practice is to assign dual-grade products the costs associated with products meeting the strictest requirements. U.S. Steel Comments 6 (citing Issues and Decision Memorandum for the ■ Eighth Administrative Review of Small Diameter Circular Seamless Carbon and Alloy •Steel Standard, Line and Pressure Pipe from Brazil; Final Results of Antidumping Duty Administrative Review at 5, A-351-826, (Feb. 4, 2005), available át http://ia.ita.doc.gov/frn/ ■ summaiy/brazil/E5-5 84-1 .pdf (last visited Mar. 13, 2017).

38 . Commerce explains that GVN submitted a revised U.S. sales database in response to a supplemental questionnaire in certain CON-NUMs were eliminated because they were recoded from dual-grade merchandise to N-80 merchandise. Remand Results 51. Later at verification, after reviewing documentation surrounding dual-grade products, Petitioners requested that the CONNUMs for the dual-grade products be recoded again as 1^80 products. Id. Commerce observed that only after Commerce requested this recoding did it become apparent that the costs for these CONNUMs were missing from the record. Id. Commerce supported its determination by explaining that it was only evident after this second requested recoding that the costs for these CONNUMs were not on the record. Id. Therefore, Commerce concluded that it was reasonable for GVN not to have reported COP data for these CONNUMs earlier. See id.