Siemens Gamesa Renewable Energy v. United States

Slip Op. No. 23-ŗŚş UNITED STATES COURT OF INTERNATIONAL TRADE SIEMENS GAMESA RENEWABLE ENERGY, Plaintiff, v. UNITED STATES, Before: Timothy C. Stanceu, Judge Defendant, Court No. 21-00449 and WIND TOWER TRADE COALITION, Defendant-Intervenor.

OPINION AND ORDER [Ordering a second remand in litigation contesting an agency determination concluding an antidumping duty investigation of wind towers from Spain] Dated: October 11, 2023 Daniel J. Cannistra, Crowell & Moring LLP, of Washington, D.C., for plaintiff.

With him on the briefs were Pierce Lee and Simeon Yerokun.

Sara E. Kramer, Trial Attorney, Commercial Litigation Branch, Civil Division, U.S. Department of Justice, of Washington, D.C., for defendant. With her on the brief were Brian M. Boynton, Principal Deputy Assistant Attorney General, Patricia M. McCarthy, Director, and Reginald T. Blades, Jr., Assistant Director. Of counsel on the brief was Shelby M. Anderson, Senior Attorney, Office of the Chief Counsel for Trade Enforcement & Compliance, U.S. Department of Commerce.

Alan H. Price, Wiley Rein LLP, of Washington, D.C., for defendant-intervenor.

With him on the brief were Robert E. DeFrancesco, III and Laura El-Sabaawi.

Court No. 21-00449 Page 2

Stanceu, Judge: In this litigation, plaintiff contested a “less-than-fair-value” (“LTFV”) determination by the International Trade Administration, U.S. Department of Commerce (“Commerce” or the “Department”) concluding an antidumping duty (“AD”) investigation of certain wind towers from Spain. The court previously ordered Commerce to reconsider its final LTFV determination. Siemens Gamesa Renewable Energy v. United States, 47 CIT __, 621 F. Supp. 3d 1337 (2023) (“Siemens Gamesa I”).

Before the court is a decision (the “First Remand Redetermination”), which Commerce issued in response to the court’s opinion and order in Siemens Gamesa I.

Final Results of Redetermination Pursuant to Court Remand (Int’l Trade Admin.

June 16, 2023), ECF No. 53 (“First Remand Redetermination”). Concluding that the First Remand Redetermination does not comply with the court’s order in Siemens Gamesa I and is contrary to law, the court directs Commerce to issue a new decision in conformity with the instructions set forth herein.

I. BACKGROUND Background for this case is presented in the court’s prior opinion and is supplemented herein. Siemens Gamesa I, 47 CIT at __, 621 F. Supp. 3d at 1339–40.

A. The Parties Plaintiff Siemens Gamesa Renewable Energy (“Siemens Gamesa” or “SGRE”) is a Spanish exporter of utility scale wind towers (the “subject merchandise”). Defendant is the United States. Defendant-intervenor Wind Tower Trade Coalition is an association Court No. 21-00449 Page 3 of U.S. producers of utility scale wind towers that was the petitioner in the underlying antidumping duty investigation.1 B. The Department’s Final Less-Than-Fair-Value Determination The agency decision contested in this litigation (the “Final LTFV Determination”) was published as Utility Scale Wind Towers From Spain: Final Determination of Sales at Less Than Fair Value, 86 Fed. Reg. 33,656 (Int’l Trade Admin. June 25, 2021) (“Final LTFV Determination”). The period of investigation (“POI”) was July 1, 2019, through June 30, 2020. Id. The Final LTFV Determination incorporated by reference an explanatory “Issues and Decision Memorandum.” Issues and Decision Memorandum for the Final Affirmative Determination in the Less-Than-Fair-Value Investigation of Utility Scale Wind Towers from Spain (Int’l Trade Admin. June 14, 2021), P.R. 149 (“Final I&D Mem.”).2 The Final LTFV Determination concluded the Department’s antidumping duty investigation of utility scale wind towers from Spain. In the course of its investigation, Commerce sent “Quantity and Value” (“Q&V”) questionnaires to nineteen known

“The members of the Wind Tower Trade Coalition are Arcosa Wind Towers Inc. and Broadwind Towers, Inc.” Utility Scale Wind Towers From Spain: Preliminary Affirmative Determination of Sales at Less Than Fair Value, 86 Fed. Reg. 17,354, 17355 n.6. (Int’l Trade Admin. Apr. 2, 2021) (“Prelim. Determination”).

Documents in the Joint Appendix (May 26, 2022), ECF Nos. 41 (public), (conf.) are cited as “P.R. __” (for public documents). Documents from the first remand proceeding, Remand Joint Appendix (Sept. 8, 2023), ECF Nos. 65 (public), 66 (conf.), are cited as “P.R.R. __” (for public documents). All information disclosed in this Opinion and Order is information for which there is no claim for confidential treatment.

Court No. 21-00449 Page 4 exporters and producers of the subject merchandise, thirteen of which filed responses.

Decision Memorandum for the Preliminary Determination in the Less-Than-Fair-Value Investigation of Utility Scale Wind Towers from Spain at 2 (Int’l Trade Admin. Mar. 29, 2021), P.R. 134. From among those thirteen companies, Commerce decided that it would “examine individually only one respondent (i.e., a ‘mandatory respondent’),” for which Commerce selected the company with the largest export volume, Vestas Eolica S.A.U. (“Vestas Eolica”). Siemens Gamesa I, 47 CIT at __, 621 F. Supp. 3d at 1341 (citing Less-Than-Fair-Value Investigation of Utility Scale Wind Towers from Spain: Respondent Selection at 6 (Int’l Trade Admin. Dec. 23, 2020), P.R. 106 (“Respondent Selection Mem.”) (“Based on our analysis of the Q&V questionnaire data submitted by exporters and producers, the exporter/producer with the largest value of entries of subject merchandise is Vestas Eolica.”)).

When Vestas Eolica notified Commerce that it would not participate in the investigation, Utility Scale Wind Towers from Spain: Notice of Decision to Not Participate in the Investigation at 1 (Jan. 28, 2021), P.R. 124, Siemens Gamesa filed a request with Commerce to be investigated individually, along with its affiliated supplier Windar Renovables (“Windar”), as a mandatory respondent. Less-Than-Fair-Value Investigation of Utility Scale Wind Towers from Spain: Request for Mandatory Respondent Selection at 1 (Feb. 17, 2021), P.R. 128 (“SGRE Request for Mandatory Respondent Selection”). Despite the absence of any mandatory respondents other than Vestas Eolica, Commerce rejected Court No. 21-00449 Page 5 this request. Siemens Gamesa I, 47 CIT at __, 621 F. Supp. 3d at 1342 (citing Utility Scale Wind Towers from Spain: Request to Select Replacement Mandatory Respondent (Int’l Trade Admin. Mar. 5, 2021), P.R. 132).

Commerce concluded that Vestas had “failed to cooperate by not acting to the best of its ability when it did not respond to the Department’s antidumping duty questionnaire.” Siemens Gamesa I, 47 CIT at __, 621 F. Supp. 3d at 1342 (citing Utility Scale Wind Towers From Spain: Preliminary Affirmative Determination of Sales at Less Than Fair Value, 86 Fed. Reg. 17,354, 17,355 (Int’l Trade Admin. Apr. 2, 2021)). Relying on “facts otherwise available” under 19 U.S.C. § 1677e(a) and an “adverse inference” under U.S.C. § 1677e(b) (collectively, “adverse facts available” or “AFA”), Commerce assigned Vestas Eolica a preliminary dumping margin of 73.00 percent ad valorem, a rate drawn from the petition. For the six companies, including Windar,3 that failed to respond to the Department’s initial Q&V questionnaire, Commerce also assigned a preliminary antidumping duty margin of 73.00 percent based on “total AFA.” Siemens Gamesa I, 47 CIT at __, 621 F. Supp. 3d at 1343. Commerce, further, preliminarily assigned the 73.00 percent rate as an “all-others” rate to the exporters and producers of

These six companies were Acciona Windpower S.A., Gamesa Energy Transmission, Haizea Wind Group, Kuzar Systems S.L., Proyectos Integrales y Logisticos S.A.A. (“Proinlosa”), and Windar Renovables. Prelim. Determination, 86 Fed. Reg. at 17,355 n.5.

Court No. 21-00449 Page 6 the subject merchandise that Commerce did not individually examine, including Siemens Gamesa. Id. Commerce did not alter its analysis in issuing the Final LTFV Determination, which applied “total AFA” to assign the 73.00 percent dumping margin to the sole mandatory respondent, Vestas Eolica, and assigned that same rate to five of the six companies the Department preliminarily had determined not to have cooperated with the investigation by failing to respond to the Q&V questionnaires. Id. (citing Final LTFV Determination, 86 Fed. Reg. at 33,657 (explaining that one of those six companies attempted to cooperate with the investigation, and so “we no longer find that application of total AFA is appropriate with respect to Proinlosa.”)). Commerce also made no change to its preliminary determination of an “all-others” rate of 73.00 percent.

Id. Thus, Commerce assigned the 73.00 percent rate to every respondent in the investigation.

After receiving notice of an affirmative final determination of material injury by the U.S. International Trade Commission (“ITC”), Utility Scale Wind Towers From Spain; Determination (Int’l Trade Comm’n Aug. 13, 2021), 86 Fed. Reg. 44,748, Commerce published the antidumping duty order (the “Order”), Utility Scale Wind Towers From Spain: Antidumping Duty Order, 86 Fed. Reg. 45,707 (Int’l Trade Admin. Aug. 16, 2021).

In the Order, Commerce directed U.S. Customs and Border Protection to collect 73.00 percent cash deposits on all imports of subject merchandise, “effective on the date of Court No. 21-00449 Page 7 publication in the Federal Register of the ITC’s final affirmative injury determination.”

Id., 86 Fed. Reg. at 45,708.

C. Submission of the First Remand Redetermination and Comments In response to the court’s opinion and order in Siemens Gamesa I, Commerce submitted the First Remand Redetermination to the court on June 16, 2023. Plaintiff and defendant-intervenor filed comments on July 17, 2023. Plaintiff Siemens Gamesa Renewable Energy’s Comments on Draft Remand Redetermination, ECF Nos. 55 (conf.), (public) (“SGRE’s Comments”); Defendant-Intervenor Wind Tower Trade Coalition’s Comments on Remand Redetermination, ECF Nos. 58 (conf.), 59 (public) (“Def.-Int.’s Comments”). The government responded to those comments on July 31, 2023.

Defendant’s Response to Plaintiff’s and Defendant-Intervenor’s Comments on Commerce’s Remand Redetermination, ECF No. 61 (“Def.’s Resp.”).

II. DISCUSSION A. Jurisdiction and Standard of Review The court exercises jurisdiction under section 201 of the Customs Courts Act of 1980, 28 U.S.C. § 1581(c),4 pursuant to which the court reviews actions commenced under section 516A of the Tariff Act of 1930, as amended (“Tariff Act”), 19 U.S.C. § 1516a,

All citations to the United States Code herein are to the 2018 edition. All citations to the Code of Federal Regulations are to the 2023 edition.

Court No. 21-00449 Page 8 including an action contesting a final determination that Commerce issues to conclude an antidumping duty investigation.

In reviewing an agency determination, including one made upon remand to the agency, the court “shall hold unlawful any determination, finding, or conclusion found . . . to be unsupported by substantial evidence on the record, or otherwise not in accordance with law.” 19 U.S.C. § 1516a(b)(1)(B)(i). Substantial evidence refers to “such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.” SKF USA, Inc. v. United States, 537 F.3d 1373, 1378 (Fed. Cir. 2008) (quoting Consol. Edison Co. v. NLRB, 305 U.S. 197, 229 (1938)).

B. The Court’s Prior Opinion and Order In Siemens Gamesa I, the court held that “[t]he assignment of the 73.00 percent rate to Siemens Gamesa was unlawful because it resulted from an unlawful respondent selection method, Commerce having limited its individual examination to a single respondent.” 47 CIT at __, 621 F. Supp. 3d at 1348. The court ruled that the statute, in U.S.C. § 1677f-1(c)(2), as interpreted by the Court of Appeals for the Federal Circuit (“Court of Appeals”) in YC Rubber Co. (North America) LLC v. United States, No. 21-1489, 2022 WL 3711377 (Fed. Cir. Aug. 29, 2022) (“YC Rubber”), requires Commerce to “determine the weighted average dumping margins for a reasonable number of exporters or producers,” where “a ‘reasonable number’ is generally more than one.” Siemens Gamesa I, 47 CIT at __, 621 F. Supp. 3d at 1345 (quoting YC Rubber, 2022 WL 3711377, Court No. 21-00449 Page 9 at *4). Noting that “Commerce announced its decision to examine individually only one respondent in the Respondent Selection Memorandum and never departed from that decision throughout the conduct of the entire investigation,” the court held that “[t]he Department’s assigning the 73.00 percent rate to Siemens Gamesa was a result of that unlawful decision, which, when viewed according to the holding YC Rubber, was not based on a permissible interpretation of 19 U.S.C. § 1677f-1(c)(2).” Id. The court ruled, additionally, that the Department’s assignment of the 73.00 percent rate to Siemens Gamesa as an “all others” rate did not satisfy the “reasonable method” requirement of the Tariff Act. Id., 47 CIT at __, 621 F. Supp. 3d at 1345–47.

Commerce earlier had determined that it would select its mandatory respondent based on “largest export volume under 19 U.S.C. § 1677f-1(c)(2)(B),” id., 47 CIT at __, 621 F. Supp. 3d at 1348 (citing Respondent Selection Mem. at 6), a decision not challenged in this litigation and therefore final. Accordingly, the court held that the Department’s unlawful decision “not to examine Siemens Gamesa individually as the largest remaining exporter . . . must be remedied by an individual investigation of Siemens Gamesa during the remand proceeding the court is ordering.” Id., 47 CIT at __, 621 F. Supp. 3d at 1349.

C. The First Remand Redetermination In the First Remand Redetermination, Commerce reported to the court that it “has now individually investigated SGRE.” First Remand Redetermination at 1.

Court No. 21-00449 Page 10

Commerce further informed the court that it conducted a “collapsing” analysis under 19 C.F.R. § 351.401(f), under which it decided to treat Siemens Gamesa and six other companies as a single entity for purposes of the remand proceeding. Id. at 1–2, 6 (citing Remand for the Less-Than-Fair-Value Investigation of Utility Scale Wind Towers: Preliminary Affiliation and Collapsing Memorandum for Siemens Gamesa Renewable Energy S.A. and Windar Renovables S.A. (Int’l Trade Admin. Apr. 25, 2023), P.R.R. 69 (“Collapsing Mem.”)). The six companies were Windar and five other companies Commerce described as wholly-owned subsidiaries of Windar: Tadarsa Eolica SL, Windar Offshore SL, Windar Wind Services SL, Aemsa Santana SA, and Apoyos Metalicos SA. Id. at 2 n.4. Commerce found that the seven companies were “affiliated” within the meaning of section 771(33) of the Tariff Act, 19 U.S.C. § 1677(33), and that “it is appropriate to treat SGRE and Windar (and certain Windar subsidiaries) as a single entity, because their operations with respect to the sale and production of subject merchandise are intertwined.” Id. at 1–2 (citing Collapsing Mem.).

In the First Remand Redetermination, Commerce assigned the seven-company entity an estimated weighted average dumping margin of 73.00 percent. Commerce gave as its rationale: (1) that, following its finding that Windar failed to respond to the Q&V questionnaire in the underlying investigation, it had determined in the Final LTFV Determination to assign to one of the seven companies in the collapsed entity, Windar, a 73.00 percent rate as an adverse inference, id. at 22; (2) that Windar’s 73.00 percent rate Court No. 21-00449 Page 11 was “final” because Windar did not challenge it in the Final LTFV Determination, id.; (3) that in Siemens Gamesa I “the Court did not require (or even permit) the agency to revisit its final/unchallenged decision applying AFA to Windar,” id. at 26; and (4) that the assignment of the 73.00 rate to the entire seven-company entity “is consistent with extensive agency practice,” id. at 22.

Because the decision Commerce reached in the First Remand Redetermination was unlawful, the court sets it aside and orders Commerce to conduct additional proceedings. The court concludes that the First Remand Redetermination was contrary to law in three major respects.

First, Commerce relied on a conclusion that the 73.00 percent adverse inference rate Commerce assigned Windar in the original investigation is final and controlling with respect to its decision. This was incorrect. Any decision the court might sustain that involves collapsing of the seven companies necessarily would render null and void the original assignment of the 73.00 percent rate to Windar and would supplant it on remand with a newly-determined rate for the combined entity.

Second, in using an inference adverse to Siemens Gamesa, the First Remand Redetermination does not comply with section 776 of the Tariff Act, 19 U.S.C. § 1677e.

The record evidence does not support, and instead refutes, a finding that Commerce could resort to “facts otherwise available” under section 776(a) of the Tariff Act, U.S.C. § 1677e(a). Substantial evidence on the record fails to support the Court No. 21-00449 Page 12

Department’s finding that the absence of a response by Windar to the Q&V questionnaire in the underlying investigation impaired the Department’s ability to investigate Siemens Gamesa individually and assign it an estimated dumping margin in the remand proceeding, as the court directed it to do. To the contrary, Commerce reopened the record during that remand proceeding, conducted a questionnaire process, and did not find that Siemens Gamesa failed to provide the information it requested or otherwise failed to cooperate.

Third, Commerce repeated the error it made in the Final LTFV Determination with respect to an all-others rate. Noting that “we have assigned the SGRE/Windar entity a single dumping margin, i.e., 73.00 percent,” Commerce decided that “[b]ecause there are no other rates on the record of this proceeding from which to select a different ‘all-others rate,’ the ‘all others’ rate remains unchanged.” Id. at 37–38.

1. Effect of the “Finality” that Attached to the Rate Commerce Assigned to Windar in the Final LTFV Determination Commerce noted that Windar did not contest its assignment in the Final LTFV Determination of the 73.00 percent rate. First Remand Redetermination at 8. Defendant and defendant-intervenor also point out that Windar’s 73.00 percent adverse inference rate, as assigned in the Final LTFV Determination, was final and unchallenged.

Def.-Int.’s Comments 8; Def.’s Resp. 7. These conclusions are correct, as Windar is not a plaintiff in this case and did not otherwise contest the Final LTFV Determination before this Court.

Court No. 21-00449 Page 13

Commerce misinterpreted the consequence of the finality that attached to Windar’s adverse inference rate as a result of the Final LTFV Determination. The consequence of that finality is that any exports of subject merchandise by Windar that are occurring or may occur in the future will be subject to a 73.00 percent deposit rate, but only for so long as that rate is in effect.5 The First Remand Redetermination would collapse Windar with six other companies and assign the combined entity a newly determined rate (which the First Remand Redetermination would set at 73.00 percent).

Were the court to sustain a future determination by Commerce upon remand that assigns a rate to a collapsed entity that includes Windar, the court’s sustaining of that remand redetermination necessarily would vacate the existing 73.00 percent rate for Windar, as determined by Commerce in the Final LTFV Determination, and supplant it with a newly determined rate. In its reasoning, Commerce overlooked that subject merchandise exports by Windar (were any to occur) could not be subject to the old rate and the newly determined rate at the same time. The First Remand Redetermination reasoned that the “finality” of Windar’s rate supports the decision it reached in the First

5 Plaintiff states that, according to all record evidence, Windar provided wind tower components to Siemens Gamesa but did not itself export subject merchandise to the United States during the period of investigation. See Plaintiff Siemens Gamesa Renewable Energy’s Comments on Draft Remand Redetermination 2, ECF Nos. 55 (conf.), 56 (public). As discussed later in this Opinion and Order, record evidence supports plaintiff’s statement.

Court No. 21-00449 Page 14

Remand Redetermination to subject Siemens Gamesa to a 73.00 percent rate. See First Remand Redetermination at 26. It does not.

Siemens Gamesa I, while directing Commerce to investigate plaintiff Siemens Gamesa individually, did not address the issue of whether on remand Commerce could collapse Siemens Gamesa with any other companies. The court neither requires nor prohibits Commerce, in going forward, from using a collapsing analysis. In the Second Remand Redetermination the court is ordering, Commerce has a choice between two options. Commerce may submit a new determination that would apply to Siemens Gamesa alone and allow to stand as “final” the uncontested, 73.00 percent rate it assigned to Windar in the Final LTFV Determination. This option necessarily would foreclose any collapsing of Siemens Gamesa with Windar. The other option open to Commerce is to substitute for Windar’s existing rate a new rate that it would apply to a collapsed entity. But Commerce cannot repeat the internally inconsistent approach it took in the First Remand Redetermination, which in effect attempted to do both.

Defendant points out that plaintiff did not object to the decision in the First Remand Redetermination to collapse it with Windar and the Windar subsidiaries.

Def.’s Resp. 4–7 (quoting SGRE’s Letter Regarding Collapsing (May 1, 2023), P.R.R. 74 (“SGRE concurs with the Department’s preliminary determination regarding collapsing in the instant investigation.”)). That is true, but Siemens Gamesa did object to the outcome of the Department’s collapsing analysis, which was to assign the 73.00 percent Court No. 21-00449 Page 15 rate to all seven companies in the collapsed entity. SGRE’s Comments 2. Because the determination to assign that rate was unlawful, the court must set it aside (along with its reasoning), and as a result it is up to Commerce to decide whether or not it will use a collapsing methodology in the second remand proceeding.

Defendant argues that “[n]either SGRE nor Windar challenged Windar’s rate before Commerce during the investigation, and thus failed to exhaust their administrative remedies.” Def.’s Resp. 7. Defendant-intervenor argues, similarly, that Siemens Gamesa “failed to exhaust its administrative remedies regarding Windar’s AFA rate,” Def.-Int.’s Comments 8, on the premise that Siemens Gamesa “did not raise the issue of Windar’s rate until Commerce’s remand proceeding,” id. at 9. These arguments are meritless. Commerce decided to apply a 73.00 percent adverse inference rate to Siemens Gamesa (as a member of the collapsed entity) for the first time on remand, not in the Final LTFV Determination, and plaintiff has a right to contest that decision upon judicial review. The Final LTFV Determination had applied a 73.00 percent rate to Siemens Gamesa as an all-others rate (which Siemens Gamesa successfully contested before the court), not as an adverse inference rate. In this way, the First Remand Redetermination would adopt an approach different than the one Commerce took in the Final LTFV Determination. In its comments on the Department’s draft version of the First Remand Redetermination, Siemens Gamesa exhausted its administrative remedies when it contested the assignment of Windar’s adverse Court No. 21-00449 Page 16 inference rate to the collapsed entity. See Antidumping Duty Investigation of Utility Scale Wind Towers from Spain: Comments on Draft Remand Determination at 4–5, 7 (May 24, 2023), P.R.R. 80.

2. The Department’s Use of an Adverse Inference Rate in the First Remand Redetermination Section 776(b)(1) of the Tariff Act authorizes Commerce to use an inference “adverse to the interests” of an “interested party” that “has failed to cooperate by not acting to the best of its ability to comply with a request for information.” 19 U.S.C. § 1677e(b)(1). Although they are affiliated, and despite a “collapsing” determination by Commerce, Siemens Gamesa, Windar, and Windar’s subsidiaries remained separate entities. Commerce itself acknowledged this point. First Remand Redetermination at 31 (“[B]oth SGRE and Windar remain separate legal entities, even though collapsed for AD purposes.”).

Courts have recognized limited situations in which an interested party’s failure to cooperate can have an adverse collateral effect on a fully cooperative party. This case does not present one of them. The collapsing procedure is a creation of the Department’s regulation, 19 C.F.R. § 351.401(f), that is not contained in any provision of the Tariff Act. From the perspective of 19 U.S.C. § 1677e(b)(1), which contains no exception broadening its scope in the situation presented by this case, Siemens Gamesa, Windar, and Windar’s subsidiaries remained separate “interested parties.”

Nevertheless, Commerce applied an adverse inference to the prejudice of Siemens Court No. 21-00449 Page 17

Gamesa (the sole plaintiff in this case) in the remand proceeding, based entirely on Windar’s failure to submit a response to the Q&V questionnaire in the original investigation. Rather than allow Commerce to act punitively, § 1677e is intended to induce cooperation on the part of an interested party to a proceeding. According to the Department’s own findings, Siemens Gamesa, the party Commerce was charged with investigating individually on remand, did not fail to cooperate, either in the investigation culminating in the Final LTFV Determination or in the remand proceeding. When an agency’s regulation (in this case, 19 C.F.R. § 351.401(f)) conflicts with the intent and purpose of a statute (here, 19 U.S.C. § 1677e), the statute must prevail.

Applying an adverse inference rate to Siemens Gamesa in the remand proceeding also was unlawful because it was unsupported by valid factual findings as required by 19 U.S.C. § 1677e(a). Referring to subsection (a), subsection (b) of section of the Tariff Act provides for the use of an adverse inference “in selecting from among the facts otherwise available.” 19 U.S.C. § 1677e(b). With respect to the “facts otherwise available,” the First Remand Redetermination relied on its findings that, due to Windar’s failure to provide a response to the Q&V questionnaire, “relevant information remains missing from the record,” that “the time to supply that information has long passed,” and that “incomplete” information provided by SGRE in the remand proceeding did “not overcome Windar’s failure to provide a timely Q&V Court No. 21-00449 Page 18 questionnaire response.” First Remand Redetermination at 34. These findings are contrary to the record evidence, which: (1) refutes the finding that the lack of a response to the Q&V questionnaire deprived Commerce of information it needed to examine Siemens Gamesa individually in the remand proceeding (as the court ordered it to do); and (2) shows that Commerce reopened the record during the remand proceeding to collect additional information, obtained the information it sought, and did not find that Siemens Gamesa withheld any information or was untimely in responding to the Department’s requests. The Department’s use of an adverse inference against Siemens Gamesa was not supported by valid findings for the use of facts otherwise available, as the court explains in further detail below.

Addressing the lack of a response from Windar to the Q&V questionnaire, Commerce found as follows in the First Remand Redetermination: Given Windar’s failure to provide a response to the Q&V questionnaire, Windar effectively prevented itself from consideration as an individually examined respondent in the LTFV investigation. Nothing on the record remedies this deficiency. SGRE’s section A response, in turn, only confirms that Windar was the source of SGRE’s exports; it neither constitutes a timely Q&V response from Windar nor necessarily identifies the full extent of Windar’s exports to the United States (through SGRE, or otherwise).

First Remand Redetermination at 33–34 (footnote omitted). The record evidence does not support, and instead refutes, the findings that “nothing on the record remedies this deficiency,” that SGRE’s Section A response does no more than “confirm that Windar was the source of SGRE’s exports,” and that this response did not identify “the full Court No. 21-00449 Page 19 extent of Windar’s exports to the United States (through SGRE, or otherwise).”

Commerce overlooked the point that questionnaire responses Siemens Gamesa submitted during the remand proceeding remedied any deficiency that could have arisen from Windar’s earlier failure to provide the Q&V response. They also provided information in addition to the fact that Windar was the source of SGRE’s exports, including the information that Windar had no exports to the United States during the POI, either directly or through a third country.

The Q&V questionnaire Commerce used in the underlying antidumping duty investigation contained the following instruction: Please include only sales exported by your company directly to the United States. However, if your company made sales to third countries for which you have knowledge that the merchandise was ultimately destined for the United States, please separately identify these sales quantities and the location (i.e., countries) to which you made the sales.

Antidumping Duty Investigation of Utility Scale Wind Towers from Spain: Issuance of Quantity and Value Questionnaires at Attachment I (Int’l Trade Admin. Nov. 25, 2020), P.R. 49 (“Q&V Questionnaire”). In its Section A questionnaire response, SGRE identified the “distribution channels” through which Windar sold subject merchandise.

Antidumping Duty Investigation of Utility Scale Wind Towers from Spain: Siemens Gamesa Renewable Energy Section A Questionnaire Response at 16–17 (Mar. 10, 2023), P.R.R. 8–40 (“Section A Response”). These channels were home market sales to its affiliate SGRE, sales to an unaffiliated home market customer, and sales to an unaffiliated company in Court No. 21-00449 Page 20 a third country, for which the ultimate destinations of the subject merchandise did not include the United States. Id.; see also id. at 2–5 (explaining that “Windar has two methods of selling wind towers and wind tower sections to entities other than SGRE”: sales to unaffiliated “non-SGRE customers in the Spanish market” and sales to one unaffiliated “third party customer” in a third country for final delivery in countries other than the United States). According to SGRE’s Section A questionnaire response, Windar made no sales “exported . . . directly to the United States,” nor did Windar make any “sales to third countries . . . ultimately destined for the United States.” Q&V Questionnaire at Attachment I. The Q&V instructions and SGRE’s Section A questionnaire response, read together, informed Commerce that no export sales of Windar were reportable in response to the Q&V questionnaire.

Despite the record evidence, Commerce reached the unsupported finding that the Section A response did not identify “the full extent of Windar’s exports to the United States (through SGRE, or otherwise).” First Remand Redetermination at 33–34.

Commerce further erred in finding that “where Windar was the first company in the chain of distribution with knowledge that the wind towers were destined for the United States, Windar should have reported these transactions in its Q&V response, consistent with Commerce’s practice.” Id. at 34 n.144 (citation omitted). This finding is invalidated by the Department’s instructions for the Q&V questionnaire, which expressly told respondents to report only direct sales to the United States except for Court No. 21-00449 Page 21 sales in third countries for which the respondent has knowledge that the merchandise was ultimately destined for the United States. Q&V Questionnaire at Attachment I.

In summary, the Department’s use in the First Remand Redetermination of facts otherwise available and an adverse inference under 19 U.S.C. § 1677e did not comply with the purpose and intent of that statutory provision and lacked an evidentiary basis.

The various findings Commerce put forth in support of its use of facts otherwise available were refuted by the record information Commerce obtained during the remand proceeding.

3. The Department’s Finding that the Record Did Not Allow It to Perform a Margin Calculation In the context of discussing why “corroboration” for its adverse inference rate is not necessary or feasible, Commerce stated as follows in the First Remand Redetermination: While the record now contains additional information with respect to SGRE’s U.S. prices and Windar’s costs of production, none of this information is useable as the basis for a margin calculation. Of note, SGRE’s U.S. prices are transfer prices from Windar (i.e., an affiliated party), which are generally not used under section 772 of the Act [19 U.S.C. § 1677a] as the basis for a calculated dumping margin; Windar’s costs consist of a single, aggregate figure, not differentiated by product or broken into its component elements. Further, Commerce did not analyze the reported prices or costs for accuracy or attempt to identify any deficiencies in them that needed correction. Finally, Commerce did not collect pricing information related to home market or third country sales (although Windar had viable third country markets), and Commerce did not establish a deadline for the petitioner to allege that the multinational corporation provision applied to those foreign market sales (although the petitioner requested that Commerce permit such an allegation). Thus, the Court No. 21-00449 Page 22 information on the record with respect to SGRE/Windar’s U.S. prices and normal values are potentially inaccurate, unusable, and/or incomplete.

First Remand Redetermination at 35–36. In this quoted passage from the First Remand Redetermination, Commerce indicates that it cannot determine an estimated weighted average dumping margin that would apply to Siemens Gamesa (either individually or as part of collapsed entity). This rationale is unsatisfactory, in three respects.

First, the state of the record resulted in part from the inadequate “investigation” Commerce performed prior to the issuance of the Final LTFV Determination, during which Commerce, by its own choice, did not examine individually the export sales of any respondent and thereby failed to conduct what could be described as a valid antidumping duty investigation. Instead, Commerce assigned to every respondent in the investigation a 73.00 percent rate, which was derived from an adverse inference rate based solely on information in the petition. As the court concluded previously: Congress entrusted Commerce with the responsibility to conduct an antidumping duty investigation, and to assign individual and, if necessary, all-others rates, according to detailed requirements set forth in the Tariff Act. Here, it was not lawful for Commerce to evade that investigative responsibility by outsourcing the critical determination to the petitioner.

Siemens Gamesa I, 47 CIT at __, 621 F. Supp. 3d at 1347. In the original investigation, the absence of record information from which to calculate an individual estimated dumping margin for Siemens Gamesa (whether or not as part of a collapsed entity) also resulted from the Department’s own rejection of the request of Siemens Gamesa, on Court No. 21-00449 Page 23 behalf of itself and Windar, to be a mandatory respondent. After Vestas Eolica declined to participate in the investigation, Siemens Gamesa sent a letter to Commerce “[o]n behalf of Siemens Gamesa Renewable Energy (SGRE), an exporter of subject merchandise . . . and SGRE’s affiliated supplier, Windar Renovables” requesting to be selected as “a mandatory respondent for individual investigation.” SGRE Request for Mandatory Respondent Selection at 1. The reasons upon which Commerce unlawfully rejected that request had nothing to do with Windar’s failure to respond to the Quantity and Value questionnaire. Siemens Gamesa I, 47 CIT at __, 621 F. Supp. 3d at 1342. That unlawful rejection left the court with no alternative but to remand the Final LTFV Determination to Commerce for correction of the Department’s investigative error through an individual investigation of Siemens Gamesa—the only plaintiff in this litigation—that will satisfy the agency’s statutory obligation.

Second, the state of the record also is the result of the information collection process Commerce employed in the remand proceeding. It appears that Commerce is now informing the court that it considers the record inadequate to allow it to determine an estimated weighted average dumping margin for Siemens Gamesa. If such is the case, the problem is one of the Department’s own making. In the remand proceeding, Commerce reopened the record, designed a questionnaire procedure for the purpose of obtaining the information it needed, and acknowledged that it had obtained the information it requested.

Court No. 21-00449 Page 24

Third, Commerce included in its rationale a finding that is unsupported by the record evidence. Further to the Department’s conclusion that record evidence did not allow it to calculate a dumping margin was the finding that “SGRE’s U.S. prices are transfer prices from Windar (i.e., an affiliated party), which are generally not used under section 772 of the Act [19 U.S.C. § 1677a] as the basis for a calculated dumping margin.” First Remand Redetermination at 35. The record does not establish that SGRE’s U.S. prices are transfer prices from Windar. See Section A Response at 16. Section 772 of the Tariff Act requires Commerce to determine U.S. price by one of various methods and to do so reasonably. See 19 U.S.C. § 1677a; see also 19 C.F.R. § 351.402. SGRE provided information pertinent to a determination of U.S. price. Section A Response at 16–17; Antidumping Duty Investigation of Utility Scale Wind Towers from Spain: Siemens Gamesa Renewable Energy Section C Questionnaire Response at 1–4 (Apr. 14, 2023), P.R.R. 58–66. It was impermissible for Commerce to ignore and misstate record evidence in an attempt to justify its use of AFA.

With respect to the current state of the record, Commerce rejected as “unsolicited,” and thus excluded from the record, the “data sourced from Windar.”

First Remand Redetermination at 24–25. Commerce stated that it sent SGRE its standard antidumping duty questionnaire, that SGRE responded and also indicated that it would file a consolidated response on behalf of itself, Windar, and Windar’s subsidiaries, and that Commerce requested additional information on March 28, 2023, which SGRE Court No. 21-00449 Page 25 provided on March 30 and April 3, 2023. Id. at 4. Commerce added that “[o]n March 30, 2023, Commerce directed SGRE not to provide information sourced from Windar, and instructed SGRE to limit its reporting to the company’s own information.”

Id. (citing Utility Scale Wind Towers from Spain: Submission of Questionnaire Response (Int’l Trade Admin. Mar. 30, 2023), P.R.R. 45 (“Questionnaire Resp. Submission Mem.”)).

“SGRE, nonetheless, filed a joint response to the remaining sections of Commerce’s questionnaire.” Id. (footnote omitted). Commerce stated, further, that “[b]ecause SGRE submitted an unsolicited questionnaire response containing extensive data sourced from Windar, we rejected SGRE’s response and afforded SGRE an opportunity to resubmit it in the form and manner requested in our March 30, 2023, instruction.” Id. (footnote omitted). Commerce added that “SGRE resubmitted this information on April 14, 2023.” Id. (footnote omitted).

The First Remand Redetermination reasoned that “[g]iven that SGRE’s responses to these questionnaires revealed that SGRE was functioning as a single entity with Windar—a company that received a margin based on AFA in the LTFV investigation— Commerce immediately instructed SGRE not to provide data sourced from Windar.”

Id. at 24. Siemens Gamesa objected to the exclusion of the Windar data, arguing that it was not “unsolicited” because Commerce actually did request it. SGRE’s Comments 3 (quoting Antidumping Duty Investigation Initial Questionnaire at G-10 (Int’l Trade Admin.

Feb. 17, 2023), P.R.R. 1 (directing Siemens Gamesa to “[p]repare only a single response Court No. 21-00449 Page 26 for you and your affiliates involved with the production or sale of the products under investigation . . . .”)). The First Remand Redetermination acknowledged that Commerce initially requested the Windar-related information but insisted that “SGRE/Windar misconstrues Commerce’s practice, as well as the purpose behind what is, at best, a generic instruction given to all questionnaire respondents” that is “intended to cover routine, non-controversial situations.” First Remand Redetermination at 25. The First Remand Redetermination further explained that “Commerce (as is its prerogative) instructed SGRE to exclude data sourced from Windar—an affiliate that received an AFA rate for failing to cooperate in the underlying LTFV investigation.” Id. From the statements in the First Remand Redetermination, it appears that the underlying reason for the rejection of the Windar-related data, which SGRE provided, stemmed from the Department’s erroneous conclusion that “finality” attaching to the 73.00 percent rate Commerce assigned to Windar in the Final LTFV Determination would affect the rate to be assigned to the collapsed entity comprised of the seven companies. But any “finality” attaching to Windar’s 73.00 percent rate, as assigned in the Final LTFV Determination, fails as a justification to support the rejection of the Windar-related data. As the court explained earlier in this Opinion and Order, the 73.00 percent Windar rate is “final” for only so long as it remains in effect, and if replaced by a rate determined for a collapsed entity, it no longer can be in effect. Further, it is reasonable to presume that an actual examination of the collapsed entity would require Court No. 21-00449 Page 27 consideration of data of the members of that entity, including Windar. Accordingly, it is not apparent how the Department’s decision to remove the Windar data from the record can be reconciled with an objective of calculating an estimated dumping margin for the entity Commerce identified when it collapsed Siemens Gamesa with Windar and the Windar subsidiaries.

The record does not give the court confidence that Commerce had the objective of investigating and giving Siemens Gamesa an estimated dumping margin, as the court directed it to do. The decision Commerce made on March 30, 2023, to instruct SGRE to exclude the Windar data from subsequent questionnaire responses, Questionnaire Resp. Submission Mem. at 1, in conjunction with the decision on April 25, 2023, to collapse Windar and its subsidiaries with SGRE, Collapsing Mem. at 1, reasonably indicates to the court that Commerce already had reached a tentative decision not to calculate an actual dumping margin for the collapsed entity, a decision Commerce never reversed during the remand proceeding, and to assign it the 73.00 percent rate instead. Nevertheless, defendant requested an extension of time for Commerce to submit the First Remand Redetermination. Motion For an Extension of Time for Department of Commerce to File its Remand Redetermination (May 11, 2023), ECF No. 48. Defendant explained in its request that: Good cause exists for granting this extension. Commerce has made progress in preparing its remand redetermination. Commerce issued its Section A questionnaire to Siemens Gamesa Renewable Energy (SGRE) the day after receiving the Court’s order. Commerce received multiple Court No. 21-00449 Page 28 requests for extensions from SGRE to file its questionnaire responses, observing in one request the “extremely time consuming” process an investigation requires. Commerce granted the requested extensions.

Id. at 1–2. Reasonably presuming Commerce was conducting the individual investigation of Siemens Gamesa that it had ordered Commerce to perform, the court granted defendant’s request, enlarging the time for submission of a remand redetermination from 90 to 120 days, despite plaintiff’s objection that the extension would unduly delay the proceeding. See Order 2–3 (May 17, 2023), ECF No. 52. The Department’s decision to assign the collapsed entity the 73.00 percent adverse inference rate, based essentially on nothing more than its collapsing decision and what it described as its “practice,” unfortunately has delayed this litigation even further.

Having itself designed its questionnaires and having chosen to pursue a collapsing analysis, it was implausible for Commerce to maintain that the lack of Windar’s response to Commerce’s Q&V questionnaire in the original investigation prevented it from determining an actual estimated dumping margin that would apply to Siemens Gamesa, which the court’s order required it to do. If the Department’s exercising its “prerogative” to reject the Windar-related information contributed to the Department’s perceived inability to calculate an estimated dumping margin for SGRE, Commerce will have the opportunity to address this problem by taking steps to supplement the record during the next remand proceeding.

Court No. 21-00449 Page 29

4. The Reliance on a “Practice” of Assigning an Adverse Inference Rate of a Single Company to an Entire Collapsed Entity The First Remand Redetermination puts forth its own administrative practice as a justification for subjecting Siemens Gamesa to a 73.00 percent adverse inference margin. First Remand Redetermination at 6, 31 (“Commerce’s practice, when collapsing two companies, one of which has an existing AFA rate, into a single entity, is to assign the existing rate to the collapsed entity.”) (footnote omitted). This justification is unavailing.

Whatever it may consider its practice to be, Commerce is not permitted to apply it contrary to a statute. With respect to 19 U.S.C. § 1677e, it has sought to do just that in the circumstance presented here. Citing Zhaoqing New Zhongya Aluminum Co. v. United States, 36 CIT 1390, 1399, 887 F. Supp. 2d 1301, 1310 (2012) (“Zhaoqing New Zhongya Aluminum”), Commerce also asserted that “CIT precedent” supports this practice. Id. at 31. This case is inapposite, for two reasons. First, Commerce determined in that case that “each of the three companies that makes up the collapsed entity failed to cooperate.” Zhaoqing New Zhongya Aluminum, 36 CIT at 1394, 887 F. Supp. 2d at 1306.

Therefore, the case did not present the issue of whether Commerce could apply an adverse inference rate to a fully cooperative interested party. Second, although the opinion in Zhaoqing New Zhongya Aluminum mentions the Department’s practice “to apply AFA to the entire entity when one producer within it fails to cooperate,” it did Court No. 21-00449 Page 30 not hold that the practice was lawful, having also stated that the plaintiffs in the case did not challenge that practice. Id., 36 CIT at 1399, 887 F. Supp. 2d at 1310–11.

In this litigation, defendant argues that Siemens Gamesa “does not challenge Commerce’s practice of assigning the adverse facts available rate of one company in an entity to the entity as a whole,” Def.’s Resp. 10, but acknowledges that Siemens Gamesa “disagrees with Commerce’s application of its long-standing practice that resulted in Commerce [sic] applying Windar’s adverse facts available rate to the entire collapsed entity,” id. at 5. It is sufficient that plaintiff opposed the Department’s applying its practice so as to assign it a 73.00 percent rate. The court need not, and does not, hold that Commerce will never face a circumstance allowing it to apply a company’s adverse inference rate to an entire collapsed entity but holds that doing so was unlawful in the circumstance of this remand proceeding.

Defendant-intervenor argues that it was proper for Commerce to apply Windar’s adverse inference rate to SGRE, “which forecloses Windar’s ability to obtain a more favorable dumping rate by shipping towers to the United States through SGRE.”

Def.-Int.’s Comments 8. This argument is unconvincing. The record refutes any inference that Windar’s failure to submit a response to the Q&V questionnaire was an attempt to obtain a more favorable dumping rate. To the contrary, the record shows that Windar joined Siemens Gamesa in the mandatory respondent request that Commerce unlawfully rejected in the original investigation. SGRE Request for Court No. 21-00449 Page 31

Mandatory Respondent Selection at 1. The request is record evidence refuting any finding that Windar declined to participate in the investigation in order to obtain an advantage for itself or for any affiliate.

5. The First Remand Redetermination Unlawfully Determined an “All Others” Rate of 73.00 Percent Siemens Gamesa I held that Commerce erred in selecting as an all-others rate the rate of 73.00 percent. One of the reasons the court gave was that assignment of the 73.00 percent “all others” rate did not satisfy the “reasonable method” requirement of the Tariff Act. Siemens Gamesa I, 47 CIT at __, 621 F. Supp. 3d at 1345–47 (citing 19 U.S.C. § 1673d(c)(5)(B)). Regardless, the First Remand Redetermination concluded that “[b]ecause there are no other rates on the record of this proceeding from which to select a different ‘all-others rate,’ the ‘all-others’ rate remains unchanged.” First Remand Redetermination at 38. The court rejected this same rationale in Siemens Gamesa I. 47 CIT at __, 621 F. Supp. 3d at 1347. The court recognizes that Siemens Gamesa, although assigned an all-others rate in the Final LTFV Determination, was not assigned one in the First Remand Redetermination and therefore, at this stage of the proceedings, no longer has standing to object to the all-others rate. At the same time, the court also recognizes that the First Remand Redetermination, which unreasonably would adopt the 73.00 percent rate as an all-others rate, does not comply with the holding in Siemens Gamesa I.

Court No. 21-00449 Page 32

III. CONCLUSION AND ORDER The assignment of the 73.00 percent adverse inference rate to plaintiff Siemens Gamesa in the First Remand Redetermination was unlawful for the multiple reasons the court discussed above and has caused an unwarranted and unnecessary delay in the conduct of this judicial proceeding. Whether or not Commerce chooses to employ a collapsing analysis going forward, Commerce must prepare, as expeditiously as possible, a Second Remand Redetermination in accordance with this Opinion and Order.

Therefore, upon consideration of the First Remand Redetermination, the comments submitted thereon, and all other papers and proceedings had herein, and upon due diligence, it is hereby ORDERED that Commerce shall submit a redetermination in accordance with this Opinion and Order (a “Second Remand Redetermination”) within 90 days of the date of issuance of this Opinion and Order; it is further ORDERED that plaintiff and defendant-intervenor may submit comments on the Second Remand Redetermination within 30 days of the date of submission of the Remand Redetermination to the court; and it is further ORDERED that defendant may submit a response to the comments of plaintiff and defendant-intervenor within 15 days of the date of the last comment submission.

/s/ Timothy C. Stanceu Timothy C. Stanceu Judge Dated: October 11, 2023 New York, New York