Mid Continent Steel & Wire v. United States

United States Court of Appeals for the Federal Circuit ______________________ MID CONTINENT STEEL & WIRE, INC., Plaintiff-Appellee v. UNITED STATES, Defendant-Appellee v. OMAN FASTENERS, LLC, Defendant-Appellant ______________________ 2018-1296 ______________________ Appeal from the United States Court of International Trade in Nos. 1:15-cv-00214-RWG, 1:15-cv-00228-RWG, Senior Judge Richard W. Goldberg. ______________________ Decided: October 17, 2019 ______________________ ADAM H. GORDON, The Bristol Group PLLC, Washing- ton, DC, argued for plaintiff-appellee. Also represented by PING GONG. MIKKI COTTET, Appellate Staff, Civil Division, United States Department of Justice, Washington, DC, argued for defendant-appellee. Also represented by JOSEPH H. HUNT, 2 MID CONTINENT STEEL & WIRE v. UNITED STATES JEANNE DAVIDSON, PATRICIA M. MCCARTHY. MICHAEL PAUL HOUSE, Perkins Coie, LLP, Washington, DC, argued for defendant-appellant. Also represented by ANDREW CARIDAS, SHUAIQI YUAN. ______________________ Before DYK, LINN, and TARANTO, Circuit Judges. TARANTO, Circuit Judge. The United States Department of Commerce deter- mined that Oman Fasteners, LLC, a foreign producer and exporter of steel nails, was selling its products into the United States at less than fair value as judged by those nails’ “normal value” in the home country (or, in certain circumstances, a relevant third country) under the control- ling statute. Because the company did not sell a significant volume of nails in its home market, Commerce, to assess the normal value, calculated a “constructed value” of the nails through use of one of four methods provided by the governing statute. Oman Fasteners (“OF”) challenges sev- eral aspects of Commerce’s calculation of constructed value: Commerce’s initial choice of method; Commerce’s se- lection of certain information as an input into the calcula- tion required by the chosen method; and Commerce’s conclusion that it could not calculate a “cap” limiting the profit component of the constructed value. We reject OF’s challenge to the basic choice of method and the profit-cap ruling. As to Commerce’s information selection when ap- plying the chosen method, we partly reject OF’s challenge, but we remand to secure further explanation from Com- merce about one ground of this challenge—Commerce’s re- fusal to consider the effect of subsidies on whether the information it selected was accurate for the relevant statu- tory purpose. MID CONTINENT STEEL & WIRE v. UNITED STATES 3 I In June 2014, acting on a petition filed by Mid Conti- nent Steel & Wire, Inc., Commerce initiated an antidump- ing-duty investigation under 19 U.S.C. §§ 1673–1673h into steel nail products from Oman and other countries. See Certain Steel Nails from India, the Republic of Korea, Ma- laysia, the Sultanate of Oman, Taiwan, the Republic of Tur- key, and the Socialist Republic of Vietnam: Initiation of Less-Than-Fair-Value-Investigations, 79 Fed. Reg. 36019 (Dep’t of Commerce June 25, 2014) (Initiation Decision). In July 2014, Commerce separated the Omani investigation into its own proceeding and designated OF a mandatory respondent for investigation. Antidumping Duty Investi- gation of Certain Steel Nails from the Sultanate of Oman Respondent Selection (issued July 28, 2014) (Selection Mem.); J.A. 770–75. OF is the cross-appellant before us. The statute directs Commerce to impose an antidump- ing duty on foreign merchandise if the “merchandise is be- ing, or is likely to be, sold in the United States at less than its fair value.” 19 U.S.C. § 1673(1). The statutory language governing this dispute originated in the Uruguay Round Agreements Act (URAA), Pub. L. No. 103-465, 108 Stat. 4809 (1994), which implemented certain aspects of the Uruguay Round of negotiations establishing the World Trade Organization. To determine whether merchandise is being sold at less than fair value, Commerce must deter- mine the difference “between the export price or con- structed export price and normal value.” 19 U.S.C. § 1677b(a). Normal value is based on the price at which the merchandise is sold in the exporting country (the home- market) or, in the alternative, the price at which the mer- chandise is sold in a third country that is not the United States. See id., § 1677b(a)(1)(B). But if the “aggregate quantity” of merchandise sold in either the exporting coun- try or the third country is less than five percent of the quantity sold in the United States, Commerce must instead 4 MID CONTINENT STEEL & WIRE v. UNITED STATES calculate a “constructed value” of the merchandise. See id., § 1677b(a)(1)(B)(ii)(II), (1)(C)(ii), (4). In response to Commerce’s initial questionnaire, OF noted that its volume of sales in Oman, as well as in each third country that it operated in, was less than five percent of its U.S. sales and could not be the basis for the normal value calculation. Certain Steel Nails from Oman; AD In- vestigation; Section A Response (sent Aug. 26, 2014) (Ques- tionnaire Response); J.A. 954. Accordingly, Commerce’s task in this matter was to calculate the constructed value to establish the normal value. The statute identifies four methods for calculating con- structed value: one preferred method and three alternative methods among which there is no hierarchy of preference. SKF USA Inc. v. United States, 263 F.3d 1369, 1374 (Fed. Cir. 2001). All four methods require Commerce to look at the company’s costs of producing and packaging the mer- chandise. 19 U.S.C. § 1677b(e)(1), (3). The preferred method directs Commerce to look at the company’s “actual amounts” of profits, and selling, general, and administra- tive (SG&A) expenses, “in connection with the production and sale of a foreign like product, in the ordinary course of trade, for consumption in” the company’s home market. Id., § 1677b(e)(2)(A). But if “actual data are not available with respect to the[se] amounts,” Commerce can select one of the three alternative methods. Id., § 1677b(e)(2)(B). Each of the three alternative methods, like the pre- ferred method, calls for consideration of profits and SG&A expenses—though each method specifies a different source for that data. The first alternative method focuses on the data associated with the respondent company’s other prod- ucts “in the same general category of products as the sub- ject merchandise.” Id., § 1677b(e)(2)(B)(i). The second focuses on the data of other respondents to the investiga- tion. Id., § 1677b(e)(2)(B)(ii). The third allows Commerce MID CONTINENT STEEL & WIRE v. UNITED STATES 5 to use “any other reasonable method,” subject to what the parties here call a “profit cap”: the amount allowed for profit may not exceed the amount normally realized by exporters or produc- ers (other than the [specific respondent at issue] in connection with the sale, for consumption in [the specific respondent’s home market], of merchan- dise that is in the same general category of prod- ucts as the subject merchandise. Id., § 1677b(e)(2)(B)(iii); see SKF USA, 263 F.3d at 1372– 74. In this matter, Commerce determined that there was insufficient data to support use of the preferred method be- cause OF did not have “viable home or third country mar- kets.” Antidumping Duty Investigation of Certain Steel Nails from Oman: Request for Constructed Value Profit and Selling Expenses Comments and Information (issued Oct. 17, 2014) (Request for Comments and Info.); J.A. 1532. Two weeks before the Preliminary Determination, Com- merce asked OF and Mid Continent to submit, by October 31, 2014, data relevant to use of the alternative methods. Id. OF submitted the financial statements of several Omani companies that sold steel products for various in- dustries: civil construction, power transmission, mining, oil and gas, and packaging. OF also provided, to corroborate the profit rates reflected in its primary submissions, a par- tially translated financial statement of L.S. Industry Co., Ltd. (LSI), a Thai producer of steel nails. Mid Continent, for its part, submitted the partially translated financial statement of an Indian producer of steel nails, the partially translated statements of two Taiwanese producers of steel nails, and the fully translated statement of Hitech Fas- tener Manufacture (Thailand) Co., Ltd. (Hitech), a Thai producer of steel screws. In its Preliminary Determination, Commerce con- firmed its earlier decision not to use the preferred method, 6 MID CONTINENT STEEL & WIRE v. UNITED STATES selected Hitech’s financial statement for use in the third alternative method, and found that OF had been dumping steel nails during the Period of Investigation. Certain Steel Nails From the Sultanate of Oman: Affirmative Prelimi- nary Determination of Sales at Less Than Fair Value and Postponement of Final Determination, 79 Fed. Reg. at 78,035 (Dep’t of Commerce Dec. 29, 2014) (Preliminary De- termination); Decision Memorandum for the Preliminary Determination in the Antidumping Duty Investigation: Cer- tain Steel Nails from the Sultanate of Oman, 79 ITADOC 78,034 (issued Dec. 17, 2014) (Preliminary Determination Mem.). Commerce also determined that there was insuffi- cient data to quantify a profit cap under 19 U.S.C. § 1677b(e)(2)(B)(iii). Commerce then conducted its full in- vestigation and analysis, including verification of key fac- tual submissions. While its investigation was under way, Commerce cor- responded with OF to clarify certain aspects of the con- structed value calculation. Shortly after the Preliminary Determination, OF filed a motion contending that Com- merce had erred in refusing to consider the partially trans- lated LSI statement. OF’s primary contention was that Commerce had accepted the LSI statement in concurrent proceedings dealing with steel nails from China (China Nails) and thus was bound to use the LSI statement in the Oman proceeding. Commerce denied the motion, citing a Department policy that requires fully translated docu- ments and noting that OF had submitted LSI only as sup- plementary evidence to support its primary submissions based on information about home-country (Omani) produc- ers of industrial materials. Preliminary Determination in the Less-Than-Fair-Value Investigation on Certain Steel Nails from Oman: Allegation of Ministerial Error (issued Jan. 28, 2015) (Ministerial Error Decision); J.A. 3292–94. Just before the Final Determination, Commerce reiterated that it could not use the preferred method to calculate con- structed value. Cost of Production and Constructed Value MID CONTINENT STEEL & WIRE v. UNITED STATES 7 Calculation Adjustments for the Final Determination— Oman Fasteners LLC (issued May 13, 2015) (Calculation Adjustments); J.A. 4877. In its Final Determination, Commerce continued to rely on the Hitech statement and found no profit cap avail- able to limit the profits calculated for Hitech. See Certain Steel Nails from the Sultanate of Oman: Final Determina- tion of Sales at Less Than Fair Value, 80 Fed. Reg. 28,972 (Dep’t of Commerce May 20, 2015) (Final Determination); Certain Steel Nails from the Sultanate of Oman: Issues and Decision Memorandum for the Final Determination of Sales at Less Than Fair Value, 80 ITADOC 28,972, at 12– 19 (issued May 13, 2015) (Issues and Decision Mem.). Commerce rejected each of OF’s challenges, including OF’s challenge that Commerce was required to use the LSI statement. Issues and Decision Mem. at 17–19. Commerce imposed an antidumping duty on OF. Final Determina- tion, 80 Fed. Reg. at 28,972. Mid Continent filed an action in the Court of Interna- tional Trade (Trade Court) challenging Commerce’s deter- mination in a respect not before us in the present appeal. 1 OF intervened, raising several challenges to aspects of Commerce’s determination of constructed value. The Trade Court sustained Commerce’s conclusions on several of the matters raised by OF, but it remanded for further explanation or reconsideration by Commerce of its reliance on Hitech and refusal to calculate a profit cap. Mid Conti- nent Steel & Wire, Inc. v. United States, 203 F. Supp. 3d 1295 (Ct. Int’l Trade 2017). On remand, Commerce reaf- firmed its previous determinations and provided further explanation on both of the remanded issues. Final Results 1 The Trade Court rejected Mid Continent’s chal- lenge, and we affirmed. Mid Continent Steel & Wire Inc. v. United States, No. 18-1250, 2019 WL 4316996, --- F. App’x --- (Fed. Cir. Sept. 12, 2019). 8 MID CONTINENT STEEL & WIRE v. UNITED STATES of Redetermination Pursuant to Court Order (issued Jan. 26, 2017) (Redetermination). The Trade Court subse- quently entered judgment sustaining Commerce’s Final Determination. J.A. 59. OF timely appealed to this court. We have jurisdiction pursuant to 28 U.S.C. §§ 1295(a)(5) and 2645(c). II OF appeals Commerce’s decision not to use the pre- ferred method, Commerce’s selection of Hitech over LSI and use of the Hitech profit data, and Commerce’s decision not to calculate a profit cap. We review Commerce’s decision using the same stand- ard of review applied by the Trade Court, while carefully considering that court’s analysis. Diamond Sawblades Mfrs. Coal. v. United States, 866 F.3d 1304, 1310 (Fed. Cir. 2017). We decide legal issues de novo and uphold factual determinations if they are supported by substantial evi- dence. 19 U.S.C. § 1516a(b)(1)(B)(i); see Diamond Saw- blades, 866 F.3d at 1310; Dupont Teijin Films USA, LP v. United States, 407 F.3d 1211, 1215 (Fed. Cir. 2005). For factual determinations, substantial evidence is “such relevant evidence as a reasonable mind might accept to support a conclusion” considering the record as a whole. See Novartis AG v. Torrent Pharm. Ltd., 853 F.3d 1316, 1324 (Fed. Cir. 2017);Universal Camera Corp. v. NLRB, 340 U.S. 474, 487–88 (1951). For legal determinations, Commerce, in carrying out its statutorily assigned tasks, must make reasonable choices within statutory con- straints. See, e.g., Nucor Corp. v. United States, 927 F.3d 1243, 1248–49 (Fed. Cir. 2019); Apex Frozen Foods Private Ltd. v. United States, 862 F.3d 1337, 1350–51 (Fed. Cir. 2017); see also Utility Air Regulatory Group v. EPA, 573 U.S. 302, 315, 321 (2014) (summarizing principles); City of Arlington v. FCC, 569 U.S. 290, 297 (2013) (same). Related principles govern the interpretation of regulations by an MID CONTINENT STEEL & WIRE v. UNITED STATES 9 agency. See Kisor v. Wilkie, 139 S. Ct. 2400, 2414–18 (2019). Commerce must provide an explanation that is ade- quate to enable the court to determine whether its choices are actually reasonable, including as to calculation meth- ods. See CS Wind Vietnam Co., Ltd. v. United States, 832 F.3d 1367, 1376–77 (Fed. Cir. 2016); SKF USA, 263 F.3d at 1383. We insist that Commerce “examine the record and articulate a satisfactory explanation for its action.” Yang- zhou Bestpak Gifts & Crafts Co., Ltd. v. United States, 716 F.3d 1370, 1378 (Fed. Cir. 2013). Although we uphold “a decision of less than ideal clarity if the agency’s path may reasonably be discerned,” Bowman Transp., Inc. v. Arkan- sas-Best Freight Sys., Inc., 419 U.S. 281, 286, (1974), the required explanation must reasonably tie the determina- tion under review to the governing statutory standard and to the record evidence by indicating what statutory inter- pretations the agency is adopting and what facts the agency is finding. “[A]n agency’s statement of what it ‘nor- mally’ does or has done before . . . is not, by itself, an expla- nation of ‘why its methodology comports with the statute.’ SKF USA, 263 F.3d at 1383. Whether it does so in a par- ticular agency decision or in a cited earlier decision, the agency must ground such a normal or past practice in the statutory standard.” CS Wind Vietnam, 832 F.3d at 1377 (record citation omitted where ellipsis appears). We reject OF’s challenge in part, but we vacate the de- cision of the Trade Court on Commerce’s decision not to an- alyze Hitech’s subsidies. We remand for that court to remand to Commerce for further explanation. III We begin with Commerce’s decision not to use the stat- ute’s preferred source of the profit and S, G & A compo- nents of constructed value. The statute directs Commerce generally to use 10 MID CONTINENT STEEL & WIRE v. UNITED STATES the actual amounts incurred and realized by the specific exporter or producer being examined in the investigation or review for selling, general, and ad- ministrative expenses, and for profits, in connec- tion with the production and sale of a foreign like product, in the ordinary course of trade, for con- sumption in the foreign country. 19 U.S.C. § 1677b(e)(2)(A); see SKF USA, 263 F.3d at 1374. But if “actual data are not available with respect to the amounts described in subparagraph (A),” Commerce must use one of three alternative methods. 19 U.S.C. § 1677b(e)(2)(B)(i)–(iii). Commerce rejected use of the preferred-method data upon finding that “because [OF] did not have a viable home or third-country market, its volume of home market sales during the [period of investigation] is too insignificant to reflect a meaningful home market profit rate.” Issues and Decision Mem. at 13. Commerce made the same point in slightly different language in its preliminary determina- tion, finding that OF did not have enough sales in the ordi- nary course of trade, either in the home market or in the third-country markets to which OF pointed, for those sales to be a “viable” basis for the calculation. Preliminary De- termination Mem. at 9–10. In determining that there were not enough sales for a proper comparison, Commerce relied in part on Congress’s quantity-focus embodied in a closely related provision. See id. at 9. Specifically, § 1677b(a)(1)(C) allows use of certain third-country sales if the aggregate quantity or value of home-country sales “is insufficient to permit a proper comparison with the sales of the subject merchandise to the United States,” and it adds that the quantity “shall normally be considered to be insuf- ficient” if it is less than five percent of sales of the merchan- dise to the U.S. 19 U.S.C. § 1677b(a)(1)(C). OF contends that Commerce acted contrary to the un- ambiguous meaning of the statute when it rejected the MID CONTINENT STEEL & WIRE v. UNITED STATES 11 preferred method on the ground that the volume of identi- fied sales and transactions was too insignificant. OF’s con- tention is that the specified home-country sales data is “available,” and therefore must be used, no matter how lit- tle of it there may be. OF does not argue that, even if the statutory language leaves room for implementation choices, Commerce’s choice was unreasonable. We reject OF’s argument. Although we do not think that the five-percent standard applies here, we conclude that the statute does not exclude Commerce’s practical, function-based understanding of “available,” as applied to the data identified in § 1677b(e)(2)(A). The statute in- cludes no definition of “available,” but the ordinary under- standing of the term in this setting is that the data at issue be “[p]resent and ready for use,” American Heritage Dic- tionary 127 (3d ed. 1992), in Commerce’s task of calculating a profit value. Availability, in ordinary language and in law, can depend on utility for the relevant function. Cf. Ross v. Blake, 136 S. Ct. 1850, 1859 (2016) (holding that “an administrative procedure,” even if formally in exist- ence, “is unavailable when (despite what regulations or guidance materials may promise) it operates as a simple dead end”). Because “accuracy and fairness must be Com- merce’s primary objectives” in this task, Albemarle Corp. & Subsidiaries v. United States, 821 F.3d 1345, 1354 (Fed. Cir. 2016), the statutory language allows Commerce to con- sider whether the information suffices for use in making accurate calculations—which may well depend on the vol- ume of home-country sales and how that volume affects the reliability of the information used for Commerce’s accom- plishment of its assigned task. On the purely legal question of ambiguity in the stat- ute—an inquiry that precedes any issue as to which Com- merce must exercise its own judgment—we may consider other relevant statutory provisions, helping us to better un- derstand the statutory terms at issue “in their context and with a view to their place in the overall statutory scheme.” 12 MID CONTINENT STEEL & WIRE v. UNITED STATES Roberts v. Sea-Land Servs., Inc., 566 U.S. 93, 101 (2012) (quoting Davis v. Mich. Dep’t of Treasury, 489 U.S. 803, 809 (1989)). As the Trade Court concluded, Mid Continent, 203 F. Supp. 3d at 1307–08, Commerce’s approach fits with other statutory language. What must be available, for the preferred method to apply, is not just any sales data, but data about sales in the “ordinary course of trade.” 19 U.S.C. § 1677b(e)(2)(A). Congress defined that phrase to mean “the conditions and practices which, for a reasonable time prior to the exportation of the subject merchandise, have been normal in the trade under consideration with re- spect to merchandise of the same class or kind.” 19 U.S.C. § 1677(15). The qualifiers “reasonable time” and “normal,” on their face, allow for judgment, keyed to the statutory function, about what sales count in determining whether the preferred method applies. So, too, does the directive of Congress to consider, as “outside the ordinary course of trade,” “[s]ituations in which [Commerce] determines that the particular market situation prevents a proper compar- ison with the export price or constructed export price.” Id., § 1677(15)(C). Relatedly, our statutory conclusion is reinforced by Congress’s Statement of Administrative Action for the URAA, H.R. Doc. No. 103–316, vol. 1 (1994), reprinted in 1994 U.S.C.C.A.N. 4040, which Congress declared to be “an authoritative expression by the United States concerning the interpretation and application of the [URAA] in any ju- dicial proceeding in which a question arises concerning such interpretation or application.” 19 U.S.C. § 3512(d); see also Nucor Corp. v. United States, 927 F.3d 1243, 1252 (Fed. Cir. 2019) (the Statement of Administrative Action provides “interpretive guidance”). The Statement of Ad- ministrative Action explains that 19 U.S.C. § 1677b(e)(2)(A) replaces preexisting “statutory minimums for profit and SG&A expenses” with a new “general rule that Commerce will base amounts for SG&A expenses and profit only on amounts incurred and realized in connection MID CONTINENT STEEL & WIRE v. UNITED STATES 13 with sales in the ordinary course of trade of the particular merchandise in question (foreign like product).” Statement of Administrative Action, H.R. Doc. No. 103–316, vol. 1 at 839–40 (1994), reprinted in 1994 U.S.C.C.A.N. 4040, 4175. For those reasons, we answer the statutory question before us the same way as the Trade Court. We conclude that the statutory language at issue, enacted in 1994, does not unambiguously forbid Commerce, in making its “avail- able” determination, to consider whether the volume of home-market sales suffices for Commerce accurately to cal- culate a constructed value and ultimately to assess the presence of dumping. We add that we do not draw a differ- ent conclusion because of a 2017 panel decision within the World Trade Organization that OF, in one paragraph, re- lies on to assert the absence of a de minimis exception in Article 2.2.2 of the Agreement on Implementation of Article VI of the General Agreement on Tariffs and Trade 1994. When approving that agreement and others in 1994, Con- gress declared that such agreements have no effect to the extent there is inconsistency with United States law. 19 U.S.C. § 3512(a)(1). For those reasons, we affirm the Trade Court with re- spect to OF’s challenge to Commerce’s rejection of the method set forth in 19 U.S.C. § 1677b(e)(2)(A). IV OF challenges Commerce’s rejection of certain LSI fi- nancial statements. And it challenges Commerce’s adop- tion of Hitech over LSI and other sources of profit information. We consider OF’s challenges in turn. A OF submitted an LSI financial statement that had sub- stantial portions untranslated, as corroboration for its con- tention that Commerce should use the sales of Omani companies in its calculation. Commerce rejected the sub- mission. Later, after the relevant deadline for submission 14 MID CONTINENT STEEL & WIRE v. UNITED STATES of evidence in the proceeding had passed, OF submitted a fully translated version of the LSI statement. Commerce rejected that late filing. OF challenges both actions. Like the Trade Court, Mid Continent, 203 F. Supp. 3d at 1311– 15, we reject the challenges. An applicable regulation provides: “A document sub- mitted in a foreign language must be accompanied by an English translation of the entire document or of only perti- nent portions, where appropriate, unless the Secretary waives this requirement.” 19 C.F.R. § 351.303(e). It adds: “A party must obtain the Department’s approval for sub- mission of an English translation of only portions of a doc- ument prior to submission to the Department.” Id. OF did not obtain approval before it submitted its partial transla- tion. Thus, OF’s initial submission was contrary to the reg- ulation. We see no ground for disturbing Commerce’s enforcement of the regulation in this case. Commerce did not single out OF; it rejected other par- tially translated financial statements in this proceeding and explained these decisions in detail. Issues and Deci- sion Mem. at 16–17. It explained that in general, the “ab- sence of complete translations precludes the Department from fully evaluating the appropriateness of the financial information set forth.” Id. at 16. With respect to LSI, Com- merce found that the entire audit report, several pages of financial statements, and all but one footnote were left un- translated. Id. Noting that footnotes and disclosures are included pursuant to a country’s generally accepted ac- counting principles, Commerce explained that each one of these should be considered vital information. Id. And left untranslated, this vital information might as well have been left out entirely. Id. To the extent that OF contends that Commerce had discretion to accept the partial translation despite the lack of pre-submission approval, we see no abuse of discretion in Commerce’s refusal to accept the partial translation. MID CONTINENT STEEL & WIRE v. UNITED STATES 15 Commerce’s explanation fulfills its obligation to examine the governing legal standard and the facts and to articulate a satisfactory explanation for its decision. Commerce iden- tified the regulation on untranslated or partially trans- lated documents, noted past decisions where it had applied this regulation, and explained why the rejected documents at issue in this case were especially unreliable. This treat- ment accords with the regulation, is supported by substan- tial evidence, and is reasonable. OF presents essentially one argument for a contrary conclusion—that Commerce had to rule otherwise because it had just recently accepted the partially translated LSI statement in the China Nails proceeding. We do not find that acceptance to render Commerce’s contrary action in this matter unreasonable. The China Nails acceptance did not purport to change the regulation, which provided OF clear notice of the precondition for submission of any par- tially translated document. Moreover, even if the single action by Commerce is viewed as a new agency “position,” Commerce fulfilled the general agency obligation to “dis- play awareness that it is changing position” and “show that there are good reasons for the new policy.” FCC v. Fox Tel- evision Stations, Inc., 556 U.S. 502, 515 (2009). Commerce acknowledged the difference in treatment and gave a good reason: its action in China Nails “was contrary to [its] es- tablished practice,” and Commerce was “not obligated to accept an incorrect methodology and perpetuate a mis- take.” Issues and Decisions Mem. at 17. This explanation, combined with Commerce’s explanation of the unreliability of partially translated documents like the LSI document at issue here, satisfies Commerce’s obligation to display awareness of the change and provide a reasonable justifi- cation for the new approach. Commerce also offered a simple and reasonable expla- nation for refusing to accept the fully translated version of the LSI statement when OF eventually offered it. By that time, the expressly stated deadline for submission of this 16 MID CONTINENT STEEL & WIRE v. UNITED STATES evidence—stated in a case-specific communication from Commerce and in a regulation—had come and gone. OF cannot claim lack of awareness of the deadlines. Id. at 17. We see no basis for disturbing Commerce’s enforce- ment of its deadlines. We have said that “[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 consid- ered.” PSC VSMPO-Avisma Corp. v. United States, 688 F.3d 751, 761 (Fed. Cir. 2012). “Absent constitutional con- straints or extremely compelling circumstances[,] the ad- ministrative agencies should be free to fashion their own rules of procedure and to pursue methods of inquiry capa- ble of permitting them to discharge their multitudinous du- ties.” Vt. Yankee Nuclear Power Corp. v. Natural Res. Def. Council, Inc., 435 U.S. 519, 543 (1978) (internal quotation marks and citation omitted). In particular, deadlines are important in proceedings like those at issue here: “[i]n or- der for Commerce to fulfill its mandate to administer the antidumping duty law, including its obligation to calculate accurate dumping margins, it must be permitted to enforce the time frame provided in its regulations.” Dongtai Peak Honey Industry Co., Ltd. v. United States, 777 F.3d 1343, 1351 (Fed. Cir. 2015). Because OF had clear notice of the deadlines, it easily could have submitted the fully trans- lated material in time; and Commerce could reasonably de- termine that the China Nails acceptance of the materials was not a good enough reason, in the face of an explicit reg- ulation and established practice, to excuse OF’s late sub- mission. For those reasons, we affirm the Trade Court with re- spect to OF’s challenge to Commerce’s rejection of the LSI materials. B With the LSI financial information properly excluded from the proceeding, Commerce had to choose what source MID CONTINENT STEEL & WIRE v. UNITED STATES 17 of information about profits to use in calculating the con- structed value of OF’s nails. It chose Hitech. OF chal- lenges that choice as unreasonable. The third alternative method for calculating con- structed value allows Commerce to use “any other reason- able method” to calculate profits and SG&A expenses. 19 U.S.C. § 1677b(e)(2)(B)(iii). The third alternative method is the broadest option because “[u]nlike the first alterna- tive, there is no limitation that data be for the specific ex- porter or producer, and, unlike the second alternative, there is no limitation that the data relate to foreign like products.” Thai I-Mei Frozen Foods Co. v. United States, 616 F.3d 1300, 1308 (Fed. Cir. 2010). Nevertheless, Com- merce’s choices must be reasonable ones within the dual constraints of the statute and the record. The objective is to find a good proxy (or surrogate) for the profits that the respondent can fairly be expected to build into a fair sales price for the particular merchandise. SKF USA, 263 F.3d at 1373 (concluding that “‘constructed value serves as a proxy for a sales price’ of the subject mer- chandise in the home market) (quoting Statement of Ad- ministrative Action, H.R. Doc. 103–316, at 839 (1994), reprinted in 1994 U.S.C.C.A.N. 3773, 4175); CS Wind, 832 F.3d at 1377 (noting general idea that “expenses should be included in calculating normal value for the merchandise at issue only to the extent one would expect a fair sales price for that merchandise to be set to recoup such ex- penses”); Mid Continent, 203 F. Supp. 3d at 1310 (“The goal in calculating [constructed value] profit is to approximate the home market profit experience of the respondents.”); Issues & Decision Mem. at 14 (Commerce recognizing treating Hitech as a “surrogate” for OF). And “accuracy and fairness must be Commerce’s primary objectives.” Al- bermarle, 821 F.3d at 1354; Yangzhou Bestpak Gifts & Crafts Co. v. United States, 716 F.3d 1370, 1379 (Fed. Cir. 2013) (“An overriding purpose of Commerce’s administra- tion of antidumping laws is to calculate dumping margins 18 MID CONTINENT STEEL & WIRE v. UNITED STATES as accurately as possible.”); Rhone Poulenc, Inc. v. United States, 899 F.2d 1185, 1191 (Fed. Cir. 1990) (the “basic pur- pose of the statute” is to “determin[e] current margins as accurately as possible”). Commerce had to choose reason- ably, given the record, in adopting information from a sur- rogate company. Applying those standards, we discuss OF’s challenges to Commerce’s adoption of Hitech’s profits for use in the constructed value for OF. 1 OF argues that Commerce acted unreasonably in se- lecting Hitech, a Thai seller of screws, over LSI, a seller of nails, and over certain Omani companies not in the nails business. Putting aside for now the issue of subsidies dis- cussed in the next subsection of this opinion, we reject this argument. Commerce began its analysis by laying out a frame- work based on four criteria: (1) the similarity of the surro- gate company’s business and products to the respondent’s business and products; (2) the extent to which the surro- gate company’s sales reflect sales in the respondent’s home market; (3) the contemporaneity of the data; and (4) the similarity of the surrogate company’s customer base to the respondent’s customer base. Issues and Decisions Mem. at 14. Commerce determined that the ideal surrogate would be an Omani producer of steel nails. Commerce first con- sidered the statements of the Omani companies on the rec- ord. Commerce determined that none of the Omani companies on the record sold products “identical or compa- rable to” steel nails and excluded these companies from consideration. Id. at 15. With the financial statements of the Omani companies eliminated, Commerce noted that while it would prefer to use statements “of a producer of steel nails that primarily produces and sells steel nails in Oman, such information is not available on the record.” Id. That was so because Commerce had excluded the LSI MID CONTINENT STEEL & WIRE v. UNITED STATES 19 financial statements for the reasons we have discussed— the lateness of the fully translated version and the unreli- ability of the partly translated version. Accordingly, Com- merce turned to statements of companies from outside Oman and determined that because Hitech sold compara- ble merchandise and was the only fully translated state- ment available, Hitech was the only potential option. Id. at 16. Commerce recognized that nails and screws are not identical merchandise, but it reasoned: Both nails and screws share a similar production process whereby wire is drawn, heat treated and galvanized. Further, both may undergo threading and both may be collated for use in gun applica- tions. Nails and screws are both within the same family of fasteners, and they both can be used in the same applications, i.e., fastening surfaces to- gether. Id. at 18. After the Trade Court remanded for further ex- planation, Commerce bolstered this conclusion, explaining that Oman and Hitech “use the same or similar type of plant facilities, machinery, and equipment. Accordingly, they are both subject to similar levels of capital expendi- tures and are also subject to similar market conditions when purchasing or replacing machinery.” Redetermina- tion at 5; J.A. 4940. Again putting aside the subsidies issue discussed next, we conclude that Commerce’s application of the statute was reasonable on the record here. OF has not shown a lack of substantial evidence for the factual determinations in Commerce’s analysis. And we see no legal error. The statute expressly anticipates that Commerce might use sales that do not precisely match the preferred method. As discussed above, Commerce’s method must be “reasona- ble,” 19 U.S.C. § 1677b(e)(2)(B)(iii), and that assessment is tied to the goal of achieving accuracy. Commerce’s expla- nation shows that it “addressed th[e] issue seriously and 20 MID CONTINENT STEEL & WIRE v. UNITED STATES carefully, providing reasons in support of its position and responding to the principal alternative advanced.” F.E.R.C. v. Elec. Power Supply Ass’n, 136 S. Ct. 760, 784 (2016). With the exception about to be discussed, that ex- planation adequately confirms that Commerce’s determi- nation is a permissible application of the statutory standard and supported by substantial evidence. 2 Separately, OF argues that Commerce erred in refus- ing to consider the effect on Hitech’s profits of subsidies OF indicated Hitech was receiving from the Thai government. Proper consideration, OF suggests, might lead the balance to shift away from Hitech to other sources of profit infor- mation or, in any event, require reduction of the amount of Hitech’s profit to be borrowed for calculating the con- structed value of OF’s nails. On this issue, we agree with OF; Commerce’s explanation is wanting. We do not decide that Commerce must reject Hitech or lower the profit fig- ure to be borrowed for OF, but we remand for further con- sideration and explanation. Commerce undisputedly refused to consider whether Hitech was receiving subsidies. Issues and Decision Mem. at 18. In its one-paragraph explanation, Commerce ob- served that, for proceedings involving nonmarket econo- mies, Congress expressly authorized Commerce to disregard price or cost data if it determines that “broadly available export subsidies existed or particular instances of subsidization occurred with respect to those [data].” 19 U.S.C. § 1677b(c)(5). Commerce then said that because the statute mentions subsidies in provisions dealing with non- market economies, and not in the statute’s general provi- sions, the “Department’s practice with regard to financial statements with evidence of countervailable subsidies re- lates solely to [nonmarket economy] proceedings.” Issues and Decisions Mem. at 18. MID CONTINENT STEEL & WIRE v. UNITED STATES 21 Commerce’s reasoning is insufficient. First, all this statement does is declare what Commerce “‘normally’ does or has done before,” but such a declaration is “not, by itself, an explanation of ‘why its methodology comports with the statute.’” CS Wind, 832 F.3d at 1376 (quoting SKF USA, 263 F.3d at 1383). Second, although the statute explicitly provides for the removal of subsidies in nonmarket-econ- omy proceedings, it does not follow from that focused au- thorization in one context that Commerce is free to ignore subsidies in market-economy proceedings when doing so would be an unreasonable exercise of its authority in those proceedings. Third, Commerce has not provided any expla- nation at all of why, either generally or in this case, it is reasonable to decline to consider government subsidies to the company whose profits Commerce is borrowing for use in calculating a constructed value. A company that receives government subsidies to pro- duce certain merchandise may have “expenses separately recouped by income other than receipts from selling that merchandise.” Id. Subsidies may increase recorded profit, for example, if they are included on the receipt side of a profit calculation or if they take the form of artificially lower input costs. Thus, government subsidies are pre- cisely the kind of factor that could distort the accuracy of a surrogate company’s information. As a logical matter, Hitech would be a weaker surro- gate for constructed value if government subsidies heavily distort its profits. It might be so much weaker that Com- merce would no longer have a sound reason to choose Hitech over another proposed source of profit information, whether from OF’s home country or elsewhere. The size of any subsidies would obviously be relevant, as would the comparative deficiencies of the alternative sources. Choice of a different source is not the only possible response to a determination of the existence, likelihood, or magnitude of subsidies that artificially increase a surrogate’s profit. If the statute permits, Commerce might determine the 22 MID CONTINENT STEEL & WIRE v. UNITED STATES amount of subsidies and adjust its calculation of con- structed value downward to eliminate the effect of the sub- sidies. Or Commerce might decide that the subsidies are so insignificant that no change needs to be made at all. Practical considerations might play a role in the rea- sonableness of Commerce’s choice. It might be reasonable to avoid methods that demand information that cannot practically be obtained in reliable form. On the other hand, it can be unreasonable for an agency to refuse to obtain readily available, highly relevant information. See id. at 1380 & n.7 (“We note that the Supreme Court has made clear that an agency’s ‘failure to adduce empirical data that can readily be obtained’ can sometimes require setting aside an agency’s decision under the Administrative Proce- dure Act.” (citing FCC v. Fox Television Stations, 556 U.S. at 519)). We do not prescribe an ultimate result. Rather, we re- quire that Commerce reconsider an important aspect of its determination. Whatever result it reaches upon such re- consideration, it must articulate an explanation of why its result is a reasonable one, given relevant statutory duties, including the broad duty to strive for accuracy. We remand to give Commerce an opportunity to conduct this analysis and provide this explanation. V OF’s final challenge is to Commerce’s conclusion that it could not calculate a “profit cap” to limit the use of Hitech’s profits in calculating the constructed value for OF. We re- ject this challenge. As quoted above, Congress has provided that, when Commerce uses the third alternative method to determine the profit component of a constructed value, “the amount allowed for profit may not exceed the amount normally re- alized by exporters or producers (other than the exporter or producer described in clause (i)) in connection with the MID CONTINENT STEEL & WIRE v. UNITED STATES 23 sale, for consumption in the foreign country, of merchan- dise that is in the same general category of products as the subject merchandise.” 19 U.S.C. § 1677b(e)(2)(B)(iii). This “profit cap” prevents the “various possible calculation methods from yielding anomalous results that stray be- yond the ‘amount normally realized’ from sales of merchan- dise in the same general category.” Atar S.R.L. v. United States, 730 F.3d 1320, 1327 (Fed. Cir. 2013). But the State- ment of Administrative Action, which Congress has deemed authoritative, anticipated that there would be sce- narios in which the expressly identified information is not available: [W]here, due to the absence of data, Commerce can- not determine amounts for profit under alterna- tives (1) and (2) or a “profit cap” under alternative (3), it might have to apply alternative (3) on the ba- sis of “the facts available.” This ensures that Com- merce can use alternative (3) when it cannot calculate the profit normally realized by other com- panies on sales of the same general category of products. H.R. Doc. No. 103–316, vol. 1 at 841 (1994), reprinted in 1994 U.S.C.C.A.N. 4040, 4177. In the present proceeding, Commerce noted that “Con- gress intended the profit cap to be: (1) based on home mar- ket sales information of the same general category of products as the subject merchandise, (2) non-aberrational to the industry under consideration, (i.e., ‘the amount nor- mally realized’), and (3) not based on the data of the re- spondent.” Issues and Decisions Mem. at 19. Commerce determined that because no other Omani company sold steel nails or comparable products, “there is no viable do- mestic market in the exporting country for merchandise that is in the same general category of products as the sub- ject merchandise.” Id. at 18. Thus, the statutorily specified information was not available to calculate a profit cap. 24 MID CONTINENT STEEL & WIRE v. UNITED STATES Later, on remand from the Trade Court, Commerce added that there were no other “facts available” to calcu- late a profit cap that would limit use of the Hitech results, writing that it had “examined the evidence presented by all the parties to determine whether there is any source on the record of this proceeding to serve as a suitable facts avail- able profit cap.” Redetermination at 11; J.A. 4946. Com- merce explained that the statements from companies other than Hitech “suffer[ed] from significant flaws that render them unusable.” Id. at 12–13; J.A. 4947–48. In particular, each statement lacked an auditor’s report, and the submis- sions were missing a “majority of the data disclosures re- quired under each home country’s [generally accepted accounting principles].” Id. at 13; J.A. 4948. Commerce effectively found Hitech’s own data to be the only facts available; therefore, there were no facts on the record that could limit use of Hitech’s data. This discussion satisfies Commerce’s burden to articu- late a reasonable justification for its decision, tied to the record in the proceeding. In arguing otherwise, OF largely repeats the challenges it makes to the other aspects of Commerce’s calculation of constructed value, asserting that Commerce could have used the LSI statement or OF’s home market profits. Given the reasonable determination not to turn to those sources of information for other aspects of the proceeding, we see no statutory impediment to Com- merce’s treatment of the profit-cap provision on the facts of this case. We therefore affirm the Trade Court’s rejection of OF’s challenge on the profit-cap determination. J.A. 59- 9 to 59-12. VI As discussed above, we affirm the Trade Court’s judg- ment with one exception. We vacate the Trade Court’s judgment upholding Commerce’s refusal to consider subsi- dies in assessing Hitech as a reliable surrogate for deter- mining the profit component of the constructed value. We MID CONTINENT STEEL & WIRE v. UNITED STATES 25 remand the case to that court for it to remand to Commerce for further proceedings on that issue. The parties shall bear their own costs. AFFIRMED IN PART, VACATED AND REMANDED IN PART