Deacero S.A.P.I. De C v. v. US

Case: 20-1918 Document: 62-2 Page: 1 Filed: 04/13/2021 United States Court of Appeals for the Federal Circuit ______________________ DEACERO S.A.P.I. DE C.V., DEACERO USA, INC., Plaintiffs-Appellants v. UNITED STATES, NUCOR CORPORATION, Defendants-Appellees ______________________ 2020-1918 ______________________ Appeal from the United States Court of International Trade in No. 1:17-cv-00183-CRK, Judge Claire R. Kelly. ______________________ SEALED OPINION ISSUED: April 13, 2021 PUBLIC OPINION ISSUED: April 30, 2021 * ______________________ SONALI DOHALE, Greenberg Traurig, LLP, Washington, DC, argued for plaintiffs-appellants. Also represented by ROSA JEONG, FRANCHINY MANUEL OVALLE. KELLY A. KRYSTYNIAK, Commercial Litigation Branch, Civil Division, United States Department of Justice, ar- gued for defendant-appellee United States. Also repre- sented by BRIAN M. BOYNTON, JEANNE DAVIDSON, TARA K. * This opinion was originally filed under seal and has been unsealed in full. Case: 20-1918 Document: 62-2 Page: 2 Filed: 04/13/2021 2 DEACERO S.A.P.I. DE C.V. v. US HOGAN, KARA WESTERCAMP; LESLIE MAE LEWIS, Office of the Chief Counsel for Trade Enforcement & Compliance, United States Department of Commerce, Capitol Heights, MD. DERICK HOLT, Wiley Rein, LLP, argued for defendant- appellee Nucor Corporation. Also represented by TESSA V. CAPELOTO, DANIEL B. PICKARD, ALAN H. PRICE, MAUREEN E. THORSON. ______________________ Before WALLACH, CHEN, and HUGHES, Circuit Judges. WALLACH, Circuit Judge. Appellants, Deacero S.A.P.I. de C.V. and Deacero USA, Inc. (together, “Deacero”), filed suit against Appellee, the United States (“Government”), in the U.S. Court of Inter- national Trade (“CIT”), challenging the U.S. Department of Commerce’s (“Commerce”) final results in the 2014–2015 administrative review of the antidumping (“antidumping” or “AD”) duty order covering carbon and certain alloy steel wire rod from Mexico. See Carbon and Certain Alloy Steel Wire Rod from Mexico: Final Results of Antidumping Duty Administrative Review and Final Determination of No Shipments; 2014–2015 (“Final Results”), 82 Fed. Reg. 23,190 (May 22, 2017). 1 Appellee, Nucor Corporation (“Nucor”), participated as a defendant-intervenor. The CIT “sustain[ed] [Commerce’s] determination to apply total facts available with an adverse inference (‘AFA’),” Deacero I, 353 F. Supp. 3d at 1306, but remanded to Commerce 1 Deacero S.A.P.I. de C.V. is a Mexican producer and exporter to the United States of the subject merchandise, carbon and certain alloy steel wire rod; Deacero USA, Inc. is its affiliated importer in the United States. J.A. 698; see Deacero S.A.P.I. de C.V. v. United States (Deacero I), 353 F. Supp. 3d 1303, 1305 (Ct. Int’l Trade 2018). Case: 20-1918 Document: 62-2 Page: 3 Filed: 04/13/2021 DEACERO S.A.P.I. DE C.V. v. US 3 twice for “further explanation or reconsideration” of “Com- merce’s selection of 40.52 [percent] as the AFA rate,” id.; see Deacero S.A.P.I. de C.V. v. United States (Deacero II), 393 F. Supp. 3d 1280, 1281 (Ct. Int’l Trade 2019) (conclud- ing that “Commerce’s [first] [r]emand [r]esults d[id] not comply with the [CIT’s] remand order in Deacero I and its decision to apply the 40.52 [percent] AFA-rate to Deacero continues to be unsupported by substantial evidence”); J.A. 1644–60 (First Remand Results). After Commerce placed additional information on the record corroborating the 40.52 percent rate, the CIT sustained Commerce’s sec- ond remand results. See Deacero S.A.P.I. de C.V. v. United States (Deacero III), 456 F. Supp. 3d 1263, 1265 (Ct. Int’l Trade 2020); J.A. 68–69 (Judgment), 4960–80 (Second Re- mand Results). Deacero appeals. We have jurisdiction pursuant to 28 U.S.C. § 1295(a)(5). We affirm. BACKGROUND I. Legal Framework Antidumping duties may be imposed on “foreign mer- chandise [that] is being, or is likely to be, sold in the United States at less than its fair value.” 19 U.S.C. § 1673(1). 2 Domestic industries may seek “relief from [such] imports,” Allegheny Ludlum Corp. v. United States, 287 F.3d 1365, 1368 (Fed. Cir. 2002), by filing a petition with Commerce and the U.S. International Trade Commission (“ITC”) to 2 In June 2015, Congress amended the statutes con- taining the antidumping provisions. See Trade Preferences Extension Act of 2015 (“TPEA”), Pub. L. No. 114-27, §§ 501–07, 129 Stat. 362, 383–87. We review the Final Re- sults in accordance with the TPEA because they issued af- ter the TPEA became effective. See Ad Hoc Shrimp Trade Action Comm. v. United States, 802 F.3d 1339, 1348–52 (Fed. Cir. 2015). Case: 20-1918 Document: 62-2 Page: 4 Filed: 04/13/2021 4 DEACERO S.A.P.I. DE C.V. v. US initiate an antidumping duty investigation, see 19 U.S.C. §§ 1673a(b), 1677(9)(C). Following investigation, if Com- merce determines that imported merchandise “is being, or is likely to be, sold in the United States at less than its fair value,” id. § 1673(1), and the ITC determines that the im- portation or sale of that merchandise has “materially in- jured” or “threaten[s]” to “materially injur[e]” “an industry in the United States,” id. § 1673(2), then Commerce will “publish an antidumping duty order . . . direct[ing] [U.S. Customs and Border Protection] to assess . . . antidumping dut[ies]” on subject merchandise, id. § 1673e(a)(1). Each year after the order is published, “if [Commerce receives] a request for . . . review” of that order from an interested party, Commerce will “review[] and determine . . . the amount of any antidumping duty” under the order. Id. § 1675(a)(1)(B); see 19 C.F.R. § 351.213(b) (providing for the “[a]dministrative review of orders” on the request of “an interested party”). 3 In the course of an investigation or review, Commerce “determine[s] the estimated weighted average dumping margin for each exporter and producer individually investigated” or reviewed and “the estimated all-others rate for all exporters and producers not individually investigated” or reviewed. 19 U.S.C. § 1673d(c)(1)(B)(i); see id. § 1677f-1(c) (providing for the “[d]etermination of dumping margin[s]” in investigations and reviews for “a reasonable number of exporters or producers”). A dumping 3 An “interested party” includes: “a foreign manu- facturer, producer, or exporter, or the United States im- porter, of subject merchandise”; “the government of a country in which such merchandise is produced or manu- factured or from which such merchandise is exported”; and “a manufacturer, producer, or wholesaler in the United States of a domestic like product.” 19 U.S.C. § 1677(9)(A)– (C). Case: 20-1918 Document: 62-2 Page: 5 Filed: 04/13/2021 DEACERO S.A.P.I. DE C.V. v. US 5 margin reflects the amount by which the “‘normal value’ (the price a producer charges in its home market) exceeds the ‘export price’ (the price of the product in the United States) or ‘constructed export price.’” U.S. Steel Corp. v. United States, 621 F.3d 1351, 1353 (Fed. Cir. 2010) (footnote omitted) (citing 19 U.S.C. § 1677(35)(A)); see 19 U.S.C. §§ 1677b(a)(1) (defining “normal value” as “the price at which the [merchandise] is first sold . . . for consumption” in the home country or a third country), 1677a(b) (defining “constructed export price” as “the price at which the subject merchandise is first sold . . . in the United States” to “a purchaser not affiliated with the producer or exporter”). In the course of an investigation or review, if Commerce determines that “necessary information is not available on the record,” 19 U.S.C. § 1677e(a)(1), or “an interested party or any other person . . . withholds information that has been requested by [Commerce],” “fails to provide such information by the deadlines . . . or in the form and manner requested,” “significantly impede[d] a proceeding,” or “provides such information but the information cannot be verified,” id. § 1677e(a)(2)(A)–(D), then Commerce “shall, subject to [19 U.S.C. § 1677m(d)], use facts otherwise available” in making its determination, id. § 1677e(a); see 19 C.F.R. § 351.308(a) (similar); see also 19 U.S.C. § 1677m(d) (providing that if Commerce “determines that a response to a request for information . . . does not comply with the request,” Commerce “shall promptly inform” the respondent “of the deficiency and shall, to the extent practicable, provide that [respondent] with an opportunity to remedy or explain the deficiency”). Further, 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,” then Commerce “may use an inference that is adverse to the interests of that party in selecting from among the facts otherwise Case: 20-1918 Document: 62-2 Page: 6 Filed: 04/13/2021 6 DEACERO S.A.P.I. DE C.V. v. US available.” 19 U.S.C. § 1677e(b)(1)(A). In making this determination, Commerce may consider “information derived from”: “the petition,” the “final determination in the investigation,” “any previous [administrative] review,” or “any other information placed on the record.” 19 U.S.C. § 1677e(b)(2). When Commerce “relies on secondary information rather than on information obtained in the course of an investigation or review,” it “shall, to the extent practicable, corroborate that information from independent sources that are reasonably at [its] disposal.” 19 U.S.C. § 1677e(c)(1); see Nan Ya Plastics Corp. v. United States, 810 F.3d 1333, 1338 (Fed. Cir. 2016) (“Secondary information does not include information obtained from the subject [review], which is known as ‘primary information.’”). II. Procedural History A. The Investigation and AD Duty Order In August 2001, three domestic producers (“Petition- ers”) petitioned Commerce, “alleg[ing] that imports of car- bon and certain alloy steel wire rod” from Mexico, were being, or were likely to be “sold in the United States at less than fair value” and “that such imports [were] materially injuring, or [were] threatening to materially injure, an in- dustry in the United States.” Notice of Initiation of Anti- dumping Duty Investigations: Carbon and Certain Alloy Steel Wire Rod From Brazil, Canada, Egypt, Germany, In- donesia, Mexico, Moldova, South Africa, Trinidad and To- bago, Ukraine, and Venezuela (“Investigation Initiation Notice”), 66 Fed. Reg. 50,164, 50,164 (Oct. 2, 2001); see J.A. 1695–1759 (Petition). Petitioners alleged that “[p]ric- ing and cost information available to [P]etitioners con- firm[ed] that carbon and certain steel wire rod . . . from Mexico [wa]s being sold, or offered for sale, in the United States at less than fair value.” J.A. 1698. Following the Petition, the Petitioners submitted various supporting doc- uments and research (“the Petition Supplements”) to Case: 20-1918 Document: 62-2 Page: 7 Filed: 04/13/2021 DEACERO S.A.P.I. DE C.V. v. US 7 support their dumping allegations. J.A. 1695–1759 (Sup- plement 1.A), 1760–67 (Supplement 1.B), 1768–1855 (Sup- plement 1.C), 1856–1919 (Supplement 1.D), 1920–21 (Supplement 1.E), 1922–23 (Supplement 1.F). “Where the [P]etitioners obtained data from foreign market research, [Commerce] contacted the researchers to establish their credentials and to confirm the validity of the information being provided.” Investigation Initiation Notice, 66 Fed. Reg. at 50,165. Based on the Petition, Commerce calculated a potential dumping margin of 29.63 to 40.52 percent for Mexican pro- ducers. Id.; see id. (“The sources of data for the deductions and adjustments relating to home market price, U.S. price, constructed value . . . and factors of production . . . are de- tailed in the Initiation Checklist.”). Commerce concluded that the “country-wide import statistics for the anticipated period of investigation . . . and price quotes based on mar- ket research used to calculate the estimated margin[] for [Mexico] [were] sufficient for purposes of initiation,” and, in October 2001, initiated an antidumping duty investiga- tion. Id. at 50,164–65. In October 2002, following affirmative determinations of less-than-fair-value imports from Mexico and resulting material injury to domestic industries, Commerce issued an antidumping duty order on carbon and certain alloy steel wire rod from Mexico. Notice of Antidumping Duty Orders: Carbon and Certain Alloy Steel Wire Rod from Brazil, Indonesia, Mexico, Moldova, Trinidad and Tobago, and Ukraine (“AD Order”), 67 Fed. Reg. 65,945, 65,946 (Oct. 29, 2002). The AD Order covers “certain hot-rolled products of carbon steel and alloy steel, in coils, of approx- imately round cross section, 5.00 mm or more, but less than 19.00 mm, in solid cross-sectional diameter.” Id. Deacero was not individually investigated. See generally id. at 65,945–47. Case: 20-1918 Document: 62-2 Page: 8 Filed: 04/13/2021 8 DEACERO S.A.P.I. DE C.V. v. US In 2011, Commerce determined that Deacero had cir- cumvented the AD Order by importing into the United States “steel wire rod within a diameter of 4.75 mm, 0.25 mm smaller than the steel wire rod subject to the duty or- der.” Deacero S.A. de C.V. v. United States, 817 F.3d 1332, 1335 (Fed. Cir. 2016); see id. at 1339 (sustaining Com- merce’s “minor alteration anti-circumvention affirmative determination” against Deacero as “in accordance with law and supported by substantial evidence”); see also Carbon and Certain Alloy Steel Wire Rod from Mexico: Affirmative Final Determination of Circumvention of the [AD] Order (“Circumvention Determination”), 77 Fed. Reg. 59,892, 59,892 (Oct. 1, 2012) (concluding that Deacero made “ship- ments of wire rod with an actual diameter of 4.75 mm to 5.00 mm . . . constitut[ing] merchandise altered in form or appearance in such minor respects that it should be in- cluded within the scope of the [AD] [O]rder”). Deacero’s imports were then subject to the all others rate under the AD Order, 20.11 percent. See Deacero, 817 F.3d at 1335; Circumvention Determination, 77 Fed. Reg. at 59,893. Thereafter, Deacero was a mandatory respondent in the 2010–2011, 2012–2013, and 2013–2014 administrative re- views. See Carbon and Certain Alloy Steel Wire Rod From Mexico: Final Results of Antidumping Duty Administra- tive Review; 2010–2011, 78 Fed. Reg. 28,190, 28,191 (May 14, 2013) (calculating a margin of 12.08 percent for Deac- ero, as the sole mandatory respondent, for the period Octo- ber 1, 2010, through September 30, 2011); Carbon and Certain Alloy Steel Wire Rod From Mexico: Final Results of Antidumping Duty Administrative Review; 2012–2013, 80 Fed. Reg. 27,147, 27,148 (May 12, 2015) (calculating a margin of 2.13 percent for Deacero, as the sole mandatory respondent, for the period October 1, 2012, through Sep- tember 30, 2013); Carbon and Certain Alloy Steel Wire Rod From Mexico: Final Results of Antidumping Duty Admin- istrative Review; 2013–2014, 81 Fed. Reg. 31,592, 31,593 (May 19, 2016) (calculating a margin of 1.54 percent for Case: 20-1918 Document: 62-2 Page: 9 Filed: 04/13/2021 DEACERO S.A.P.I. DE C.V. v. US 9 Deacero, as one of two mandatory respondents, for the pe- riod October 1, 2013, through September 30, 2014). B. The 2014–2015 Administrative Review In November 2015, both Nucor and Deacero requested administrative review of Deacero’s AD duty rate. J.A. 695– 96; see J.A. 695–708 (Preliminary I&D Mem.); see also 19 U.S.C. § 1675(a)(1); 19 C.F.R. § 351.213. Commerce then initiated an administrative review, covering the period of review (“POR”) of October 1, 2014, through September 30, 2015. Initiation of Antidumping and Countervailing Duty Administrative Reviews, 80 Fed. Reg. 75,657, 75,658 (Dec. 3, 2015). Commerce selected Deacero S.A.P.I. de C.V. and one other company as mandatory respondents. Id.; see J.A. 695. 4 Having petitioned for the review, Nucor partici- pated as an interested party. J.A. 1029; see J.A. 1029–42 (Final I&D Mem.). In December 2015, Commerce issued Deacero its initial questionnaire, instructing Deacero to report its cost of pro- duction and constructed value figures based on the “actual costs incurred by [Deacero] during the [POR], as recorded under [Deacero’s] normal accounting system.” J.A. 89; see J.A. 85–90 (AD Duty Questionnaire). Commerce explained that if Deacero’s cost accounting system was based upon “standard or budgeted costs” or planned production, Deac- ero should provide cost variance information. J.A. 90. In January and February 2016, Deacero submitted re- sponses to all five sections. J.A. 91 (Excerpt from Deacero’s Section A Response), 93–114 (Excerpts from Deacero’s Sec- tion B and C Responses), 115–56 (Excerpts from Deacero’s 4 Commerce subsequently determined that this sec- ond respondent did not have any shipments into the United States of subject merchandise during the relevant POR and, therefore, did not calculate its separate rate. J.A. 695–96. Case: 20-1918 Document: 62-2 Page: 10 Filed: 04/13/2021 10 DEACERO S.A.P.I. DE C.V. v. US Section D and E Responses). In its Section D response, Deacero stated that it had “reported [its] cost of manufac- ture . . . based on the actual costs incurred during the POR as recorded in its accounting system, with certain revi- sions . . . described fully” in its response, “to meet [Com- merce’s] cost reporting requirements.” J.A. 117. Deacero indicated that it had not submitted any cost variance infor- mation because “Deacero’s cost accounting system [was] based on actual costs, not standard or budgeted costs.” J.A. 148. Deacero confirmed that its “reported costs reflect the actual costs incurred and recorded by Deacero in its normal cost accounting system.” J.A. 149; see J.A. 150–51 (explaining that Deacero’s calculated cost for steel scrap in- put was “calculated based on Deacero’s actual processing yield on a diameter-specific basis during the POR”). 5 In June 2016, Commerce issued Deacero a supple- mental questionnaire, requesting additional information about Deacero’s production process, accounting system, and reporting of its home market and U.S. sales. J.A. 157; J.A. 157–63 (Supplemental Questionnaire). Deacero re- sponded to the supplemental questionnaire. J.A. 164; see J.A. 164–665 (Deacero’s First Supplemental Response). However, in addition to responding to Commerce’s ques- tions, Deacero also submitted, unsolicited, a revised Sec- tion D cost of production database, J.A. 164–68. Deacero summarily explained that it had made “minor corrections 5 Deacero reported its cost for production of billet (the alloyed, but unshaped steel from which wire rod is sub- sequently produced), based on its “consumption costs” for “steel scrap,” J.A. 127, and “[a]dditions, such as ferroalloys and carbon, . . . used to produce the desired grade of steel,” J.A. 128; see J.A. 128 (explaining that “Deacero converts billet into wire rod in its rolling mills” and that “[a]ddi- tional direct materials are not consumed at [the rolling] stage [of production]”). Case: 20-1918 Document: 62-2 Page: 11 Filed: 04/13/2021 DEACERO S.A.P.I. DE C.V. v. US 11 to [its] sales and cost databases,” including “correct[ing] the assignment of steel scrap costs to each grade of billet produced during the POR.” J.A. 167. On September 29, 2016, Nucor submitted to Commerce deficiency comments concerning Deacero’s revised cost of production database. J.A. 671; see J.A. 671–78 (Deficiency Comments). Nucor asserted that the changes Deacero made to its revised cost of production database resulted in “significant changes” to Deacero’s reported costs “without any explanation.” J.A. 672. Nucor argued that, because the changes concerned the product identification and con- trol number (“CONNUM”) that “represented [the vast ma- jority] of Deacero’s U.S. sales during the POR,” and because those changes indicated that billet costs for that CONNUM substantially decreased Deacero’s original cost of production database, these changes would result in a sig- nificant reduction in Deacero’s total cost of manufacturing and, therefore, “Deacero’s overall dumping margin.” J.A. 672–73. Nucor requested that, “given the proximity of [Commerce’s] preliminary results,” Commerce “immedi- ately issue another supplemental questionnaire to clarify [the] issues [presented]” by Deacero’s revised cost of pro- duction database. J.A. 672. On November 7, 2016, Commerce issued Deacero a post-preliminary supplemental questionnaire. J.A. 718; see J.A. 718–20 (Post-Preliminary Supplemental Question- naire). Commerce requested that Deacero “provide a de- tailed explanation of billet cost changes during the POR and how such changes [were] reflected in [Deacero’s re- vised] questionnaire response,” and, more specifically, that Deacero “[e]xplain fully” the “decrease[] in cost in [its] cost database” for its main U.S. sales CONNUM and “provide a revised . . . cost buildup” for that CONNUM. J.A. 720. On November 16, 2016, Commerce published its preliminary results, calculating a 17.02 percent AD margin for Deacero, based on Deacero’s revised Section D database. Deacero I, 353 F. Supp. 3d at 1306; J.A. 707, 709–17; see Carbon and Case: 20-1918 Document: 62-2 Page: 12 Filed: 04/13/2021 12 DEACERO S.A.P.I. DE C.V. v. US Certain Alloy Steel Wire Rod From Mexico: Preliminary Results of Antidumping Duty Administrative Review; 2014–2015, 81 Fed. Reg. 80,638, 80,639 (Nov. 16, 2016); J.A. 709 (Preliminary I&D Mem., dated Nov. 3, 2016). 6 On November 25, 2016, Deacero proffered its response to Commerce. J.A. 912; see J.A. 912–26 (Deacero’s Second Supplemental Response). Deacero alleged that “[t]he main reason for the decrease in the cost of production [for Deac- ero’s primary U.S. export CONNUM] [wa]s that Deacero [had] corrected the allocation of costs for billet production.” J.A. 914. Deacero explained that, “[i]n [its] original Section D response, Deacero [had] allocated billet costs based on planned production” and had “failed to take billet reclassi- fications into account.” J.A. 914; see J.A. 914 (explaining that where Deacero “finds that the output of a particular production run fails to meet the specifications of the billet product Deacero planned to produce, [Deacero] reclassifies the billet to a product code for which the specifications are met”). In its revised cost of production database, Deacero had “allocated billet costs to the billet products actually produced.” J.A. 914. Deacero further alleged that, “[u]nder [its] original allocation . . . steel scrap costs assigned to some billet product codes were overstated, while” others were “understated,” and that Deacero’s revised cost of pro- duction database corrected these errors because its “steel scrap costs are now based on actual—not planned—produc- tion.” J.A. 915; see J.A. 915 (explaining that “Deacero cor- rected the allocation of steel scrap costs to take billet reclassifications into account”). In May 2017, Commerce issued its Final Results. 82 Fed. Reg. at 23,190; see J.A. 1029–42 (Final I&D Mem.). 6 Deacero asserts that this rate should have been 3.65 percent, “based on Deacero’s calculations,” with the correction of “several minor clerical errors” made by Com- merce in the Preliminary Results. Appellant’s Br. 9, 9 n.2. Case: 20-1918 Document: 62-2 Page: 13 Filed: 04/13/2021 DEACERO S.A.P.I. DE C.V. v. US 13 Commerce applied AFA to Deacero “and assigned . . . the highest margin alleged in the [P]etition, i.e., 40.52 percent, as Deacero’s AFA rate.” Final Results, 82 Fed. Reg. at 23,190. Commerce found that Deacero had “stated in its initial questionnaire response that it [had] based its [S]ec- tion D database on actual costs,” and that “in its supple- mental questionnaire response, Deacero [had] submitted an unsolicited and revised [S]ection D database stating that it [had] made ‘minor corrections’ to account for revi- sions to the allocation of billet costs.” J.A. 1032; see J.A. 1031 (noting that “[o]ther than stating that [Deacero] had ‘corrected the assignment of steel scrap costs to each grade of billet produced during the POR,’ Deacero did not explain these significant changes in the first supplemental questionnaire response” (quoting J.A. 167)). Commerce further found that Deacero did not, until its “post-prelimi- nary questionnaire response,” state that “its reporting of the costs in [its] initial [S]ection D database reflected ‘planned production’ while the reporting of steel scrap costs in the revised [S]ection D database reflected ‘actual produc- tion.’” J.A. 1032 (quoting J.A. 915). Commerce determined that, contrary to Deacero’s statements, the revised Sec- tion D database “significantly and disproportionately changed the billet costs associated with the CONNUM comprising the vast majority of Deacero’s U.S. sales of sub- ject merchandise during the POR.” J.A. 1032. Commerce concluded that recourse to facts otherwise available “w[as] appropriate under [19 U.S.C. §§ 1677e(a)(2)(A), (B), (C)].” J.A. 1033. Commerce ex- plained that this was appropriate because Deacero had “mischaracterized the nature of its [revised Section D da- tabase],” “withheld critical information from [Commerce]” when it submitted the revised database by representing that the changes were “minor,” and, despite further “oppor- tunity to explain the revisions,” “Deacero’s response” re- mained “not satisfactory.” J.A. 1033; see J.A. 1033 (noting that “in [its] post-preliminary attempt to explain the Case: 20-1918 Document: 62-2 Page: 14 Filed: 04/13/2021 14 DEACERO S.A.P.I. DE C.V. v. US significant and unsolicited changes made to its first [Sec- tion D] database,” Deacero had “provided an explanation of its allocation methodology that” was both new and contrary to its prior statements). Commerce concluded that it was “unable to use the [revised] cost information . . . because [it was] unable to determine whether the reallocation of the billet cost information [wa]s reliable,” and further, that it “h[ad] no confidence in either the originally submitted cost information or the reallocation of the costs” and, therefore, could not “rely on Deacero’s reported cost information, and without reliable cost information, [was] unable to calculate a margin for Deacero.” J.A. 1034. Commerce further concluded that use of an adverse in- ference in selecting from facts otherwise available was ap- propriate under 19 U.S.C. § 1677e(b)(1). J.A. 1034. Commerce explained that this was appropriate because Deacero’s “significant failings” during the review estab- lished that “Deacero ha[d] failed to act to the best of its ability to comply with [Commerce’s] request for infor- mation.” J.A. 1034. Specifically, while Deacero was an ex- perienced company that had “participated as a mandatory respondent in prior administrative reviews,” and therefore “is knowledgeable of the process and understands what is required to be prepared to participate and provide complete and reliable responses in an antidumping duty administra- tive review,” J.A. 1032, it had failed to do so, J.A. 1034 (finding that Deacero’s submission of “an unsolicited sec- ond [Section D] database in a supplemental response prior to the Preliminary Results in which it made significant and unexplained revisions to reported billet costs,” “amount[ed] to withholding of information, failure to submit infor- mation by the deadline for submission, and significantly impeding the proceeding,” and that, despite opportunity to correct the situation, “Deacero [had] not provide[d] an ade- quate explanation of its [cost of production] revisions, sup- ported by record evidence”). Case: 20-1918 Document: 62-2 Page: 15 Filed: 04/13/2021 DEACERO S.A.P.I. DE C.V. v. US 15 Commerce selected the highest margin alleged in the Petition, 40.52 percent, as Deacero’s AFA rate. J.A. 1034; see J.A. 1034 (explaining that Commerce’s practice is to se- lect the higher of “(a) [the] highest margin alleged in the petition, or (b) the highest weighted-average calculated rate for any respondent in the investigation”). Commerce relied on its prior “pre-initiation [of investigation] analysis of the adequacy and accuracy of the information in the [P]etition” to corroborate the selected rate. J.A. 1035; see 19 U.S.C. § 1677e(c) (providing that, when Commerce “re- lies on secondary information rather than on information obtained in the course of an investigation or review,” it “shall, to the extent practicable, corroborate that infor- mation from independent sources that are reasonably at [its] disposal”); J.A. 1035 (explaining that, when deciding whether to initiate its investigation, Commerce had “exam- ined information from various independent sources pro- vided either in the petition” or in Petitioners’ responses to Commerce’s requests, “which corroborated key elements” of the relevant calculations). C. Proceedings Before the CIT Deacero filed suit against the Government in the CIT, challenging the Final Results as unsupported by substan- tial evidence and contrary to law. Deacero I, 353 F. Supp. 3d at 1307, 1310. The CIT sustained “Commerce’s deter- mination to apply total facts available with an adverse in- ference” against Deacero as “supported by substantial evidence,” id. at 1306–07, while remanding for further ex- planation or reconsideration on the issue of Commerce’s se- lection of 40.52 percent, from the 2001 petition, as Deacero’s AFA rate, id. at 1314. The CIT explained that, while Commerce was required to corroborate its selected AFA rate, it had “not place[d] any corroborating infor- mation on the record.” Id.; see id. (“Here, although Com- merce purports to rely upon information it obtained during the initiation of the investigation, namely a pre-initiation analysis memorandum and documentation supporting the Case: 20-1918 Document: 62-2 Page: 16 Filed: 04/13/2021 16 DEACERO S.A.P.I. DE C.V. v. US calculations in the petition, that information has not been placed on the record of this proceeding.”). The CIT ex- plained that “[t]o the extent practicable, at a bare mini- mum,” Commerce must “produce the documents it relied upon to analyze why the chosen rate is probative.” Id.; see 19 U.S.C. § 1677e(c)(1) (requiring that Commerce corrobo- rate “to the extent practicable”). “In accordance with [Deacero I], Commerce . . . pro- vided additional information and explanation to corrobo- rate the total [AFA] margin applied to Deacero.” J.A. 1644. Specifically, Commerce supplemented the administrative record with a copy of the Initiation Notice and “the public version of the accompanying . . . Initiation Checklist[.]” J.A. 1649. Commerce explained that the Initiation Notice and checklist “evince our pre-initiation examination of the independent information provided in the Petition, includ- ing the information used to calculate [normal value] and [export price] in the [P]etition” and its “determination that such information had probative value and was a reasonable basis for initiating an investigation,” such that the docu- ments “corroborate[d] the 40.52 percent AFA rate applied to Deacero[.]” J.A. 1657. The CIT remanded again, concluding that Commerce’s First Remand Results “d[id] not comply with the [CIT’s] re- mand order in Deacero I and [Commerce’s] decision to ap- ply the 40.52 [percent] AFA-rate to Deacero continue[d] to be unsupported by substantial evidence.” Deacero II, 393 F. Supp. 3d at 1281. The CIT noted that, while Commerce had supplemented the record, it “[ha]d not rel[ied] upon” the documents presented “to corroborate Deacero’s rate,” as the documents only summarized Commerce’s previous conclusions. Id. at 1285; see id. at 1286 (explaining that “[Commerce’s] pre-initiation analysis itself is not an inde- pendent source”). The CIT explained that “[i]f the obliga- tion to demonstrate the probative value of a rate is to have any meaning, Commerce must do more than refer to Case: 20-1918 Document: 62-2 Page: 17 Filed: 04/13/2021 DEACERO S.A.P.I. DE C.V. v. US 17 conclusions of calculations it carried out previously.” Id. at 1286. 7 “Based on the CIT’s holding in Deacero II,” Commerce “placed on the record . . . the information . . . that was used to corroborate the petition margin that” served as the basis for Deacero’s AFA rate. J.A. 4964; see J.A. 1692–94 (Mem. re Placement of Factual Information on Record of Remand) (adding the Petition, various supporting documents, and Deacero’s records from the 2013–2014 review, with oppor- tunity for “interested parties to place rebuttal factual in- formation” on the record). Specifically, Commerce placed on the record “the business proprietary and public versions of the Petition and Petition Supplements, and various memoranda on which Commerce relied to corroborate [Deacero’s] 40.52 [percent] margin.” J.A. 4961; see J.A. 1695–1759 (Supplement 1.A), 1760–67 (Supplement 1.B) 1768–1855 (Supplement 1.C), 1856–1919 (Supplement 1.D), 1920–21 (Supplement 1.E), 1922–23 (Supplement 1.F). Commerce explained that these documents “consti- tute the independent sources of information . . . that Com- merce relied upon to derive the [AFA] margin” assigned to Deacero. J.A. 4965. Commerce also placed on the record “the margin calculations for Deacero” in connection with the prior, 2013–2014 review and “a summary spreadsheet that identifies individual transactions from that review with margins” for individual transactions “in the range of or exceed[ing] the 40.52 [percent] petition margin” as- signed to Deacero in the Final Results, “to further demon- strate that the petition margin has probative value [for] Deacero in the 2014–2015 review.” J.A. 4965–66. The CIT sustained Commerce’s selection of a 40.52 percent margin, based on this additional corroboration, as “reasonable on 7 The CIT also denied Deacero’s request to provide express “instructions to Commerce to abandon its chosen 40.52 [percent] AFA-rate.” Id. at 1287. Case: 20-1918 Document: 62-2 Page: 18 Filed: 04/13/2021 18 DEACERO S.A.P.I. DE C.V. v. US th[e] record” and “comport[ing] with the [CIT’s] order.” Deacero III, 456 F. Supp. 3d at 1272. DISCUSSION Deacero argues, first, that “Commerce’s decision to ap- ply AFA to Deacero is unsupported by substantial evi- dence,” Appellants’ Br. 30 (capitalization normalized); and, second, that “Commerce’s selection of a 40.52 [percent] AFA rate is unsupported by substantial evidence,” id. at 41 (capitalization normalized). We address each argument in turn. I. Standard of Review We “apply . . . the same standard” of review as the CIT, Downhole Pipe & Equip., L.P. v. United States, 776 F.3d 1369, 1373 (Fed. Cir. 2015) (internal quotation marks and citation omitted), upholding Commerce’s determinations if they are supported “by substantial evidence on the record” and otherwise “in accordance with law,” 19 U.S.C. § 1516a(b)(1)(B)(i). “Although we review the decisions of the CIT de novo, we give great weight to the informed opin- ion of the CIT and it is nearly always the starting point of our analysis.” Nan Ya Plastics, 810 F.3d at 1341 (internal quotation marks, alterations, and citation omitted). We review Commerce’s “statutory interpretations as to the appropriate methodology” under Chevron U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837 (1984). Pesquera Mares Australes Ltda. v. United States, 266 F.3d 1372, 1379 (Fed. Cir. 2001). If “Congress has di- rectly spoken to the precise question at issue,” then “that is the end of the matter; for the court, as well as the agency, must give effect to the unambiguously expressed intent of Congress.” Chevron, 467 U.S. at 842–43. “In order to de- termine whether a statute clearly shows the intent of Con- gress[,] . . . we employ traditional tools of statutory construction and examine the statute’s text, structure, and legislative history, and apply the relevant canons of Case: 20-1918 Document: 62-2 Page: 19 Filed: 04/13/2021 DEACERO S.A.P.I. DE C.V. v. US 19 interpretation.” Kyocera Solar, Inc. v. U.S. Int’l Trade Comm’n, 844 F.3d 1334, 1338 (Fed. Cir. 2016) (internal quotation marks and citations omitted). If, however, “the statute is silent or ambiguous with respect to the specific issue,” we evaluate whether Commerce’s interpretation is reasonable. Chevron, 467 U.S. at 843. The agency’s con- struction need not be the only reasonable interpretation or even the most reasonable interpretation. See Zenith Radio Corp. v. United States, 437 U.S. 443, 450 (1978). We review Commerce’s findings of fact for substantial evidence. See SolarWorld Ams., Inc. v. United States, 910 F.3d 1216, 1222 (Fed. Cir. 2018). Substantial evidence is “more than a mere scintilla”; it is such “evidence that a rea- sonable mind might accept as adequate to support a con- clusion.” Downhole Pipe, 776 F.3d at 1374 (internal quotation marks and citations omitted). “We look to the record as a whole, including evidence that supports as well as evidence that fairly detracts from the substantiality of the evidence.” SolarWorld, 910 F.3d at 1222 (internal quo- tation marks and citation omitted). II. Adverse Facts Available A. Legal Standard In the course of an investigation or review, “the burden of creating an adequate record lies with interested parties.” QVD Food Co. v. United States, 658 F.3d 1318, 1324 (Fed. Cir. 2011) (internal quotation marks, alterations, and cita- tion omitted). If Commerce determines that “necessary in- formation is not available on the record” or “an interested party or any other person . . . withholds information that has been requested by [Commerce],” “fails to provide such information by the deadlines . . . or in the form and manner requested,” “significantly impedes a proceeding,” or “pro- vides such information but the information cannot be veri- fied,” then Commerce is permitted to use “facts otherwise available” in making its determinations. 19 U.S.C. § 1677e(a); see 19 C.F.R. § 351.308 (providing for Case: 20-1918 Document: 62-2 Page: 20 Filed: 04/13/2021 20 DEACERO S.A.P.I. DE C.V. v. US “[d]eterminations on the basis of facts available”). Com- merce uses “facts otherwise available” to “fill in . . . gaps” in the administrative record. Nippon Steel Corp. v. United States, 337 F.3d 1373, 1381 (Fed. Cir. 2003). Further, if an interested party “fail[s] to cooperate by not acting to the best of its ability to comply with a request for information,” then Commerce “may use an inference that is adverse to the interests of that party in selecting from among the facts otherwise available,” commonly re- ferred to as AFA. 19 U.S.C. § 1677e(b); see 19 C.F.R. § 351.308 (similar). Because Commerce “has no subpoena power,” Nan Ya Plastics, 810 F.3d at 1338, AFA is “an es- sential investigative tool” in AD proceedings, Statement of Administrative Action Accompanying the Uruguay Round Agreements Act (“SAA”), H.R. REP. NO. 103–316, at 868 (1994). 8 To avoid AFA, “interested parties” must “cooper- ate . . . to the best of [their] ability,” 19 U.S.C. § 1677e(b), meaning they must “do the maximum [they are] able to do,” Nippon Steel, 337 F.3d at 1382. This standard “does not require perfection and recognizes that mistakes sometimes occur[.]” Id. However, “it does not condone inattentive- ness, carelessness, or inadequate record keeping.” Id.; see id. at 1383 (explaining that, “[w]hile intentional conduct, such as deliberate concealment or inaccurate reporting,” may show “a failure to cooperate, the statute does not con- tain an intent element”). 8 The SAA is “an authoritative expression by the United States concerning the interpretation and applica- tion of the Uruguay Round Agreements and [enacting leg- islation] in any judicial proceeding in which a question arises concerning such interpretation or application.” 19 U.S.C. § 3512(d). Case: 20-1918 Document: 62-2 Page: 21 Filed: 04/13/2021 DEACERO S.A.P.I. DE C.V. v. US 21 B. Commerce’s Decision to Apply AFA is Supported by Substantial Evidence and Otherwise in Accordance with Law In its Final Results, Commerce concluded that, in set- ting Deacero’s antidumping duty margin, recourse to facts otherwise available “w[as] appropriate under [19 U.S.C. §§ 1677e(a)(2)(A), (B), (C)],” and further, that use of an ad- verse inference in selecting from facts otherwise available was appropriate under 19 U.S.C. § 1677e(b)(1), given Deac- ero’s “significant failings” in responding to Commerce’s re- quests for information during the administrative review. J.A. 1033–34. The CIT sustained “Commerce’s determina- tion to apply total facts available with an adverse infer- ence” against Deacero as “supported by substantial evidence.” Deacero I, 353 F. Supp. 3d at 1306–07. On ap- peal, Deacero argues that “Commerce’s decision to apply AFA to Deacero is unsupported by substantial evidence[.]” Appellants’ Br. 30. Deacero asserts that recourse to AFA was unwarranted because “Deacero provided complete, ac- curate, and timely information.” Id. (capitalization nor- malized). We disagree with Deacero. First, Commerce’s recourse to facts otherwise available under 19 U.S.C. § 1677e(a) is supported by substantial ev- idence. Specifically, Commerce found that Deacero had “mischaracterized the nature of its [revised Section D da- tabase],” and, further, that Deacero “withheld critical in- formation from [Commerce],” when it submitted the revised database by representing that the changes were “minor.” J.A. 1033; see J.A. 166. Commerce explained that the changes were not “minor,” but rather “significant.” J.A. 1032; see J.A. 1032 (explaining that the revised Sec- tion D database “significantly and disproportionately changed the billet costs associated with the CONNUM comprising the vast majority of Deacero’s U.S. sales of sub- ject merchandise during the POR”). Commerce further found that, despite “opportunity to explain the revisions,” “Deacero’s response” remained “[un]satisfactory.” Case: 20-1918 Document: 62-2 Page: 22 Filed: 04/13/2021 22 DEACERO S.A.P.I. DE C.V. v. US J.A. 1033; see J.A. 117 (Deacero representing that it had “reported [its] cost of manufacture . . . based on the actual costs”), 166 (Deacero representing that it had made “minor corrections to [its] sales and cost databases”), 914 (Deacero representing that “[i]n [its] original Section D response, Deacero [had] allocated billet costs based on planned pro- duction” not “actual costs”). Commerce concluded that it could not “rely on Deacero’s reported cost information,” be- cause it “h[ad] no confidence in either the originally sub- mitted cost information or the reallocation of the costs.” J.A. 1034; see J.A. 1033 (noting that Deacero had never oth- erwise raised the “planned” versus “actual cost” accounting system in any of its prior submissions in this or any other segment under the AD Order), 1033–34 (noting that Deac- ero had yet to produce any records to support or clarify its “planned” versus “actual cost” explanation). Deacero, therefore, “with[held] [requested] information,” “fail[ed] to provide [necessary] information by the deadlines for sub- mission of the information,” and, thereby, “significantly im- pede[d]” the administrative review, such that Commerce properly “use[d] facts otherwise available.” 19 U.S.C. §§ 1677e(a)(2)(A), (B), (C); see Ad Hoc Shrimp, 802 F.3d at 1355 (concluding that Commerce’s use of facts otherwise available was supported by substantial evidence where the respondent’s “withholding of information and its repeated misrepresentations rendered the company’s submissions unreliable”). Accordingly, Commerce’s recourse to facts otherwise available is supported by substantial evidence. Second, Commerce’s determination to use an adverse inference under 19 U.S.C. § 1677e(b) in selecting from facts otherwise available is supported by substantial evidence. Specifically, Commerce used an adverse inference because Deacero’s “significant failings” during the review estab- lished that “Deacero ha[d] failed to act to the best of its ability to comply with [Commerce’s] request for infor- mation[.]” J.A. 1034. Commerce explained that, while Deacero had “participated as a mandatory respondent in Case: 20-1918 Document: 62-2 Page: 23 Filed: 04/13/2021 DEACERO S.A.P.I. DE C.V. v. US 23 prior administrative reviews,” and therefore “is knowledge- able of the process and understands what is required . . . in an [AD] administrative review,” J.A. 1032, it had failed to do so, J.A. 1033–34. Rather, Deacero had “mischaracter- ized the nature” of its revised Section D database, “misrep- resented the magnitude of the changes made to [that] database,” and “withheld critical information from Com- merce,” by representing that its revised Section D database presented only “minor corrections” to its original Section D database. J.A. 1032–33. Compare J.A. 167 (Deacero rep- resenting that its revised Section D database made “minor corrections to [its] sales and cost databases”), with J.A. 1032 (Commerce finding that Deacero’s revised Sec- tion D database “significantly and disproportionately changed the billet costs associated with the CONNUM comprising the vast majority of Deacero’s U.S. sales of sub- ject merchandise during the POR”). Further, despite being given an opportunity to explain and correct the situation, J.A. 718–20, “Deacero [did] not provide an adequate expla- nation of its [cost of production] revisions, supported by rec- ord evidence,” J.A. 1034; see J.A. 1033–34 (noting that Deacero had failed to produce any records to support or clarify its “planned” versus “actual cost” explanation). Whether through active misrepresentation or “inadequate record keeping,” Deacero failed to “act to ‘the best of its ability’”—it did not “do the maximum it [wa]s able to do”— in responding to Commerce’s request for information, and was therefore uncooperative. Nippon Steel, 337 F.3d at 1382–83 (quoting 19 U.S.C. § 1677e(b)(1)); see SAA, H.R. REP. NO. 103–316, at 870 (“A party is uncooperative if it has not acted to the best of its ability to comply with re- quests for necessary information.”). Accordingly, Com- merce’s decision to apply AFA to Deacero, based on Deacero’s, at best, inconsistent representations, and fail- ure to timely explain and meaningfully support those rep- resentations, is supported by substantial evidence. Case: 20-1918 Document: 62-2 Page: 24 Filed: 04/13/2021 24 DEACERO S.A.P.I. DE C.V. v. US Deacero’s counterarguments are unpersuasive. First, Deacero asserts that “Commerce’s determination to apply AFA” relies on “a mischaracterization of Deacero’s submis- sions and responses.” Appellants’ Br. 22. Deacero argues that “[t]he record demonstrates [it] acted exactly as a coop- erative respondent should” and, therefore, Commerce’s de- termination that Deacero failed to cooperate to the best of its ability “is unsupported by substantial evidence on the record.” Id.; see id. at 34 (arguing that “[t]he fact that Deac- ero did not provide information [to explain its revised Sec- tion D database] prior to being asked by Commerce does not evince a flaw in the data or lack of cooperation”). This argument is without merit. Even if Deacero’s alternative interpretation of the record was within the scope of our re- view, see SolarWorld, 910 F.3d at 1222 (“Commerce’s find- ing may still be supported by substantial evidence even if two inconsistent conclusions can be drawn from the evi- dence.” (internal quotation marks and citation omitted)), the record belies Deacero’s argument, J.A. 1032–34. In- deed, Deacero’s conduct “points to why the use of an ad- verse inference is a useful tool in antidumping determinations.” Mukand, Ltd. v. United States, 767 F.3d 1300, 1307 (Fed. Cir. 2014). An adverse inference is appro- priate not only when an interested party fails to respond, but also where “it is reasonable for Commerce to expect that more forthcoming responses should have been made.” Nippon Steel, 337 F.3d at 1383. Cost of production data “is a fundamental element in the [anti]dumping analysis, and it is standard procedure for Commerce to request” such data. Mukand, 767 F.3d at 1307; see 19 U.S.C. § 1677b(b)(2)(A)(i) (providing that, in an AD investigation or review, Commerce “shall request information necessary to calculate the constructed value and cost of production” to determine whether “sales of the foreign like product have been made at prices that represent less than the cost of production of the product”). “It was thus reasonable for Commerce to expect . . . more accurate,” transparent, “and responsive answers” from Deacero, and to conclude that Case: 20-1918 Document: 62-2 Page: 25 Filed: 04/13/2021 DEACERO S.A.P.I. DE C.V. v. US 25 Deacero was an uncooperative respondent when it failed to provide such answers. Mukand, 767 F.3d at 1307. Second, Deacero asserts that “Commerce abused its discretion and acted contrary to law when it refused to con- sider Deacero’s corrected cost database.” Appellants’ Br. 32. Deacero argues that “when a respondent seeks to correct an error prior to the final results,” Commerce must “analyze the new information,” id. at 31 (citing Timken U.S. Corp. v. United States, 434 F.3d 1345, 1353 (Fed. Cir. 2006)), and that, “if there were any deficiencies with [Deacero’s] submission, Commerce failed to follow its stat- utory obligation to promptly notify Deacero of the ‘nature of the deficiency,’” id. at 33 (quoting 19 U.S.C. § 1677m(d)). This argument is without merit. It is premised on a mis- reading of Commerce’s determination. Commerce did not refuse to consider Deacero’s revised Section D database; it declined to rely on that database because it found it unre- liable. J.A. 1033–34. While “Commerce is free to correct any type of importer error—clerical, methodology, substan- tive, or one in judgment—in the context of making an anti- dumping duty determination,” it is incumbent on the importer to “seek[] correction before Commerce issues its final results and adequately prove[] the need for the re- quested corrections.” Timken, 434 F.3d at 1353. Deacero did not adequately explain the need for its requested cor- rections. J.A. 1033. Contrary to its arguments here, Deacero’s “significant failing” was not that it submitted corrected information; ra- ther, it was that Deacero failed to timely notify Commerce of the nature and import of those corrections, and failed to adequately explain the corrections when given the oppor- tunity to do so. J.A. 1033–34. After receiving Deacero’s unsolicited, revised Section D database, Commerce “promptly inform[ed]” Deacero “of the nature of [its] defi- cienc[ies]” and provided Deacero “with an opportunity to remedy or explain th[ose] deficienc[ies],” through its post- preliminary supplemental questionnaire. 19 U.S.C. Case: 20-1918 Document: 62-2 Page: 26 Filed: 04/13/2021 26 DEACERO S.A.P.I. DE C.V. v. US § 1677m(d); see J.A. 718–20. Commerce found that “Deac- ero’s response to [that] questionnaire” was “not satisfac- tory,” J.A. 1033; see J.A. 912–26, and that Deacero had not acted to the best of its ability in providing its response, J.A. 1033–34. Commerce, therefore, acted within its dis- cretion when it declined to use Deacero’s revised Section D database. See 19 U.S.C. § 1677m(d)(1), (e)(4) (providing that, where Commerce “finds that [a] response [to a defi- ciency notice] is not satisfactory,” Commerce may “disre- gard all or part of the original and subsequent responses” unless, inter alia, “the interested party has demonstrated that it acted to the best of its ability in providing the infor- mation”). Accordingly, Commerce’s decision to apply AFA in the selection of Deacero’s AD duty rate is supported by substantial evidence and otherwise in accordance with law. III. Corroboration A. Legal Standard In selecting from adverse facts available, Commerce may use “information derived from” the following: “the pe- tition,” the “final determination in the investigation,” “any previous [administrative] review,” or “any other infor- mation placed on the record.” 19 U.S.C. § 1677e(b)(2); ac- cord SAA, H.R. REP. NO. 103–316, at 870; see Gallant Ocean (Thail.) Co. v. United States, 602 F.3d 1319, 1323 (Fed. Cir. 2010) (providing that, “in the case of uncoopera- tive respondents,” Commerce has discretion to “select from a list of secondary sources as a basis for its adverse infer- ences”). When Commerce “relies on secondary information rather than on information obtained in the course of an in- vestigation or review,” it “shall, to the extent practicable, corroborate that information from independent sources that are reasonably at [its] disposal.” 19 U.S.C. § 1677e(c); see Nan Ya Plastics, 810 F.3d at 1338 (“Secondary infor- mation does not include information obtained from the sub- ject [review], which is known as ‘primary information.’”). “Corroborate means that [Commerce] will examine Case: 20-1918 Document: 62-2 Page: 27 Filed: 04/13/2021 DEACERO S.A.P.I. DE C.V. v. US 27 whether the secondary information to be used has proba- tive value.” 19 C.F.R. § 351.308(d); see SAA, H.R. REP. NO. 103–316, at 870 (providing that, before Commerce may use “secondary information” in calculating a duty rate, it must establish that the information “has probative value”). “Independent sources may include, but are not limited to, published price lists, official import statistics and customs data, and information obtained from interested parties during the instant investigation or review.” 19 C.F.R. § 351.308(d). A source’s “independence” is a question of “the nature of the information, not . . . whether the source of the information was referenced in or included with the petition” or an interested party’s submission. KYD, Inc. v. United States, 607 F.3d 760, 765 (Fed. Cir. 2010). B. Commerce’s Selection and Corroboration of Deacero’s AFA Rate Is Supported by Substantial Evidence and Otherwise in Accordance with Law In its Final Results, Commerce “selected the highest margin alleged in the [P]etition,” 40.52 percent, “as Deac- ero’s AFA rate[.]” J.A. 1034. Commerce explained that it “relied on [its] pre-initiation analysis of the adequacy and accuracy of the information in the petition” to corroborate the rate. J.A. 1035. Following two remands, see Deacero I, 353 F. Supp. 3d at 1314; Deacero II, 393 F. Supp. 3d at 1281, Commerce placed on the record the Petition Sup- plements as “the independent sources of information . . . Commerce [had] relied upon to derive the margin” assigned to Deacero, J.A. 4965; see J.A. 4961. Commence also placed on the record transaction-specific “margin calculations for Deacero” from the 2013–2014 review, that were in “the range of or exceed[ed] . . . 40.52 [percent].” J.A. 4965–66. The CIT sustained Commerce’s selection of a 40.52 percent margin as “reasonable on th[e] record[.]” Deacero III, 456 F. Supp. 3d at 1272. Deacero asserts that “Commerce’s se- lection of a 40.52 [percent] AFA rate is unsupported by sub- stantial evidence[.]” Appellants’ Br. 41 (capitalization normalized). Deacero argues that “Commerce failed to Case: 20-1918 Document: 62-2 Page: 28 Filed: 04/13/2021 28 DEACERO S.A.P.I. DE C.V. v. US properly corroborate [its selected] AFA rate,” id. (capitali- zation normalized), because “[t]he petition-related evi- dence” it placed on the record is not “reliable and relevant in the context of [the] 2014–2015 administrative review,” id. at 43. We disagree with Deacero. Substantial evidence supports Commerce’s corrobora- tion of Deacero’s selected AFA rate. Commerce relied on secondary information—the highest Petition rate—in set- ting Deacero’s AFA rate. J.A. 1034; see 19 U.S.C. § 1677e(b)(2)(A) (listing “the petition” as a “[p]otential source[] of information for adverse inferences”). Commerce found that the highest Petition rate “ha[d] probative value,” because it was “both reliable and relevant” to Deac- ero. Ad Hoc Shrimp, 802 F.3d at 1354 (internal quotation marks and citation omitted); see J.A 4963–67. Commerce demonstrated that its selected AFA rate was reliable be- cause it was derived from independent sources of infor- mation submitted with the Petition and Petition Supplements, including “market research information on the terms of sale and price of wire rod sold by Mexican pro- ducers in Mexico and the United States.” J.A. 4965; see J.A. 1695–1759 (Supplement 1.A), 1760–67 (Supplement 1.B) 1768–1855 (Supplement 1.C), 1856–1919 (Supplement 1.D), 1920–21 (Supplement 1.E), 1922–23 (Supplement 1.F); see also 19 C.F.R. § 351.308(d) (“Independent sources may include, but are not limited to . . . information ob- tained from interested parties”); KYD, 607 F.3d at 765 (ex- plaining that a source’s independence is a question of “the nature of the information, not . . . whether [it] was refer- enced in or included with the petition”). Commerce further demonstrated that its selected AFA rate was relevant be- cause it aligned with transaction-specific “margin calcula- tions for Deacero” from the 2013–2014 review, which were in “the range of or exceed[ed] . . . 40.52 [percent].” J.A. 4965–66; see Ta Chen Stainless Steel Pipe, Inc. v. United States, 298 F.3d 1330, 1340 (Fed. Cir. 2002) (ex- plaining that where “Commerce select[s] a dumping Case: 20-1918 Document: 62-2 Page: 29 Filed: 04/13/2021 DEACERO S.A.P.I. DE C.V. v. US 29 margin within the range of [a respondent’s] actual sales data, we cannot conclude that Commerce overreached real- ity” (internal quotation marks omitted)). Commerce thus corroborated the highest Petition rate using “independent sources . . . reasonably at [its] disposal. 19 U.S.C. § 1677e(c)(1). Accordingly, Commerce’s corroboration of Deacero’s AFA rate is supported by substantial evidence. Deacero’s counterarguments are unpersuasive. First, Deacero asserts that “Commerce’s chosen rate is impermis- sibly punitive.” Appellants’ Br. 45 (capitalization normal- ized). This argument is without merit. “[A]s long as a rate is properly corroborated according to the statute, Com- merce has acted within its discretion and the rate is not punitive.” Papierfabrik Aug. Koehler SE v. United States, 843 F.3d 1373, 1382 (Fed. Cir. 2016) (citing KYD, 607 F.3d at 768); see Ad Hoc Shrimp, 802 F.3d at 1354 (explaining that “corroborat[ing]” an AFA rate means “demonstrating the rate is both reliable and relevant” (internal quotation marks and citations omitted)). As Commerce has properly corroborated Deacero’s AFA rate, Commerce acted within its discretion in its selection of that AFA rate. Second, Deacero argues that Commerce failed to “con- sider the overall facts and circumstances of each case, in- cluding the level of culpability of the non-cooperating party in an AFA analysis.” Appellants’ Br. 46 (quoting BMW of N. Am. LLC v. United States, 926 F.3d 1291, 1301 (Fed. Cir. 2019)); see id. at 47–48 (arguing that Commerce’s “re- fusal to consider the totality of the circumstances, as well as the punitive nature of its proposed AFA rate,” render Commerce’s selected rate unreasonable). This misreads Commerce’s determination. “We agree, and common sense dictates, that Commerce should consider the overall facts and circumstances of each case, including the level of cul- pability of the non-cooperating party in an AFA analysis.” BMW, 926 F.3d at 1301. It is well established both that “the purpose of” AFA is to incentivize cooperation, “not to impose punitive, aberrational, or uncorroborated margins,” Case: 20-1918 Document: 62-2 Page: 30 Filed: 04/13/2021 30 DEACERO S.A.P.I. DE C.V. v. US F.lli De Cecco Di Filippo Fara S. Martino S.p.A. v. United States, 216 F.3d 1027, 1032 (Fed. Cir. 2000), and that Com- merce’s AFA determinations must be reasonable on the record, see Yangzhou Bestpak Gifts & Crafts Co. v. United States, 716 F.3d 1370, 1378 (Fed. Cir. 2013) (providing that we “review whether substantial evidence supports Com- merce’s [AFA] determination”). It did so here. Commerce found that Deacero had “mischaracterized,” “misrepre- sented,” and “withheld critical information from Com- merce,” J.A. 1032–33, and, further, that Deacero failed to produce any records to support or clarify its representa- tions when given the opportunity to do so, J.A. 1034. That is, Commerce considered the “unique factual circum- stances” of Deacero’s situation and its “level of culpability,” BMW, 926 F.3d at 1302, and concluded that the highest Petition rate was appropriate, J.A. 1043. Third, Deacero argues that “Commerce acted unrea- sonably when it ignored evidence on the record of the ac- tual weighted-average calculations from prior administrative reviews.” Appellants’ Br. 48; see id. (argu- ing that, “[d]espite the fact that Commerce had more re- cent, relevant information on the record that it did not have to corroborate . . . Commerce erroneously insisted on rely- ing on the [P]etition rate”). This argument is without legal basis. “Commerce has wide discretion to assign the ‘high- est calculated rate’ to uncooperative parties.” BMW, 926 F.3d at 1300 (quoting KYD, 607 F.3d at 766). Further, in applying AFA, Commerce is “not required” “to estimate what the . . . dumping margin would have been if the inter- ested party . . . had cooperated,” id. § 1677e(d)(3)(A), or “to demonstrate that the . . . dumping margin used by the ad- ministering authority reflects an alleged commercial real- ity of the interested party,” id. § 1677e(d)(3)(B). In short, Commerce was not required to select Deacero’s AFA rate to reflect Deacero’s alleged commercial reality and, therefore, did not err in failing to do so. See Nan Ya Plastics, 810 F.3d at 1347 (“The statute simply does not require Commerce to Case: 20-1918 Document: 62-2 Page: 31 Filed: 04/13/2021 DEACERO S.A.P.I. DE C.V. v. US 31 select facts that reflect a certain amount of sales, yield a particular margin, fall within a continuum according to the application of particular statistical methods, or align with standards articulated in other statutes and regulations.”). We decline to “impose conditions not present in or sug- gested by the statute’s text.” Id. Deacero also asserts that “the TPEA established a pre- sumption of validity for AFA rates that [are] based on weighted average margins from prior reviews,” because 19 U.S.C. § 1677e(c)(2) “remov[es] the corroboration require- ment” for use of such rates. Appellants’ Br. 48; see 19 U.S.C. § 1677e(c)(2) (providing that Commerce is not “re- quired to corroborate any dumping margin or countervail- ing duty applied in a separate segment of the same proceeding”). However, as the CIT noted “[s]imply because Congress established a presumptive validity for AFA rates based on margins from previous reviews does not preclude the use of other corroborated rates. Had Congress intended to limit Commerce’s consideration to margins from previ- ous reviews it could have done so.” Deacero III, 456 F. Supp. 3d at 1272. In effect, Deacero “invite[s] [us] to re- weigh” the evidence it presented to Commerce concerning what it considered a more appropriate AFA rate. Downhole Pipe, 776 F.3d at 1376. We decline to do so. See Matsushita Elec. Indus. Co. v. United States, 750 F.2d 927, 936 (Fed. Cir. 1984) (“That [an appellant] can point to evidence of record which detracts from the evidence which supports the [agency’s] decision and can hypothesize a reasonable basis for a contrary determination is neither surprising nor persuasive.”). Accordingly, Commerce’s selection and cor- roboration of Deacero’s AFA rate is supported by substan- tial evidence and otherwise in accordance with law. CONCLUSION We have considered Deacero’s remaining arguments and find them unpersuasive. For the foregoing reasons, the Judgment of the U.S. Court of International Trade is Case: 20-1918 Document: 62-2 Page: 32 Filed: 04/13/2021 32 DEACERO S.A.P.I. DE C.V. v. US AFFIRMED