CS Wind Malaysia Sdn. Bhd. v. United States

Slip Op. 26-126 UNITED STATES COURT OF INTERNATIONAL TRADE Court No. 24-00079 CS WIND MALAYSIA SDN. BHD. and CS WIND CORPORATION, Plaintiffs, v. UNITED STATES, Defendant, and WIND TOWER TRADE COALITION, Defendant-Intervenor. Before: M. Miller Baker, Judge OPINION [Sustaining the Department of Commerce’s redetermi- nation.] Dated: October 14, 2026 Jarrod M. Goldfeder, Kenneth N. Hammer, and Sezi Erdin, Trade Pacific PLLC, Washington, DC, on the comments for Plaintiffs. Isabelle Aubrun, Trial Attorney, Commercial Litiga- tion Branch, Civil Division, U.S. Department of Jus- tice, Washington, DC, on the comments for Defendant. Ct. No. 24-00079 Page 2 Of counsel for Defendant was Brien Stonebreaker, At- torney, Office of the Chief Counsel for Trade Enforce- ment & Compliance, U.S. Department of Commerce, Washington, DC. Alan H. Price, Robert E. DeFrancesco III, Laura El-Sabaawi, Derick G. Holt, and Kimberly A. Reyn- olds, Wiley Rein LLP, Washington, DC, on the com- ments for Defendant-Intervenor. Baker, Judge: This case involving an administra- tive review of a countervailing duty order on utility- scale wind towers from Malaysia returns from a court- ordered remand. See CS Wind Malaysia Sdn. Bhd. v. United States, Slip Op. 25-149, 2025 WL 3496986 (CIT 2025) (Remand Op.). 1 A Malaysian producer and its South Korean parent (collectively CS Wind or the com- pany) challenged certain aspects of the Department of Commerce’s determination. The court remanded two issues for reconsideration. The first involved the rental rates the Department used to assess the adequacy of remuneration for CS Wind Malaysia’s property lease from that country’s government. The agency averaged information from two sources, the Ellis report and governmental data from the Malaysia Investment Development Author- ity. Remand Op. at 11–12, 2025 WL 3496986, at *4. In doing so, it did not address the producer’s arguments that differences in price and parcel size rendered use 1 The court presumes the reader’s familiarity with its pre- vious decision. Ct. No. 24-00079 Page 3 of the former source unreasonable. Id. at 12–13, 2025 WL 3496986, at *5. The second was CS Wind’s request for an entered- value adjustment. Commerce did not articulate what methodology it used for part of its calculations. It was also unclear whether the Department employed U.S. sales data or worldwide sales data. As the court ex- plained, if Commerce used worldwide pricing, it needed to explain the relevance of the criterion on which it relied—whether “the U.S. invoice establishes the customs value to which the CVD duties are ap- plied.” Id. at 26 & n.6, 2025 WL 3496986, at *10 & n.6 (quoting the agency’s decision memo, Appx1026). On remand, as to the land issue the Department stood by its previous decision to use the Ellis data. It observed that while the size of the lot from that source was smaller than the producer’s parcel, the alternative government-sourced information the company advo- cated did not specify lot size at all. Appx10201–10202. The agency found that while CS Wind emphasized price differences between two different Malaysian states, those distinctions did not render either area’s land pricing objectively better or worse. Appx10202. And the data the company favored came from a region known for government subsidies, such that one would expect prices there to be lower. Appx10202–10203. Thus, Commerce used the average of the two imperfect datasets because of its regulatory preference for using multiple sources and the fact that the Malaysian gov- ernment data reflected transactions from the area in which CS Wind operated. Appx10203. Ct. No. 24-00079 Page 4 For the entered-value adjustment, the Department acknowledged that it previously used companies’ worldwide sales in making its calculation. Appx10207. It explained that it discarded that methodology in fa- vor of using U.S. sales. Appx10207–10208. It also ex- plained that it treats the reported U.S. invoice value as a proxy for entered value, but only after it first ex- amines customs documentation to verify that the in- voices match the entered value. Appx10210. Here, CS Wind did not produce any customs docu- mentation for two of the four transactions reviewed, and for the other two sales the invoices disagreed with the customs forms. The result, the Department ex- plained, was that the information on the record was insufficient to allow it to determine the customs value or that the U.S. invoices were valid proxies for that value, and it therefore continued to deny the request for an entered-value adjustment. Appx10211. For these reasons, Commerce made no change to CS Wind’s subsidy rate. Appx10222. The company now challenges both aspects of the remand results. I Rather than exporting wind towers directly to the United States, the Malaysian producer instead sells them to its South Korean parent, which in turn sells them in the U.S. at a marked-up price. Remand Op. at 23–24, 2025 WL 3496986, at *9. That was the com- pany’s basis for claiming an entered-value adjustment. CS Wind now characterizes the Department’s de- nial of the adjustment as “unlawful.” ECF 72, at 7. Its Ct. No. 24-00079 Page 5 arguments, however, fail to address why the court re- manded. The company asserts, for example, that “[t]he Court directed Commerce to explain what justifies its deter- mination that the U.S. invoice issued by CS Wind Ko- rea to the unaffiliated U.S. customer did not establish the customs value to which the CVD duties are ap- plied.” ECF 72, at 13 (purporting to cite Remand Op. at 26, 2025 WL 3496986, at *10). But that is, at best, a misinterpretation. The court instructed the agency “to articulate what methodology it uses to calculate the denominator” and, “insofar as it looks to . . . some other metric rather than the affili- ate’s marked-up U.S. sales, Commerce must explain the relevance of the third criterion on which it relied to deny CS Wind’s requested adjustment.” Remand Op. at 26, 2025 WL 3496986, at *10 (emphasis in orig- inal and footnote omitted). The government, in turn, accurately summarizes the court’s request as “seeking clarity” about the relevance of U.S. sales data. ECF 78, at 12–13. Thus, what matters here is the Department’s expla- nation of its approach to granting an adjustment, not the merits of CS Wind’s data reporting. Commerce stated that it looks to U.S. sales and U.S. entered value because doing so focuses on “the core issue of po- tential overcollection of duties at the U.S. border.” Appx10208 (quoting Preliminary I&D Memo at 9 ac- companying 88 Fed. Reg. 28,496). That issue, in turn, corresponds to ascertaining the countervailing duty rate “to be applied to the entered value of the respon- Ct. No. 24-00079 Page 6 dent’s imports into the United States.” Appx10209 (em- phasis added). And the agency explained that because the record does not always contain “the entered value or invoice information for all sales” during the relevant period, it selects sample transactions “to ensure that the entered value is reflective of the claimed markup.” Appx10210. If the values match, it adjusts the duty rate to account for the marked-up entered value. Appx10209. That explanation answers the court’s question. Per- haps recognizing that fact, CS Wind does not meaning- fully address any of the foregoing. Instead, it attacks the agency’s analysis of the record. Commerce essentially made two key findings. First, it observed that for two of the four examined transac- tions, the company provided neither the Customs Form 7501 nor any alternative customs documenta- tion to establish the entered value. Appx10210–10211. Second, it emphasized that for the other two sales it reviewed, the values listed on the invoices did not match the entered values reported on the Customs 7501s, and the discrepancies “were not uniform in terms of value or direction.” Appx10211. The agency explained that those deficiencies meant it was unable to determine the customs value for those sales or to verify that CS Wind’s invoices accurately establish the customs values. Id. The company does not dispute that its invoices did not match the Customs forms. Instead, it disparages Commerce’s requirement that the numbers match, calling it unreasonable and insisting that the values Ct. No. 24-00079 Page 7 on the invoices and 7501s were “virtually identical,” ECF 72, at 6, and that the discrepancies were “insig- nificant,” id. at 8, or “minuscule,” id. at 13, 20. It in- sists that the Department “unlawfully ignored” that its documentation (purportedly) showed that the Ko- rean parent’s invoice price “reflects the imported value . . . as reported on the Forms 7501,” and it demands that the court not only remand, but order Commerce to grant the entered-value adjustment. Id. at 12. But a flaw runs through all those arguments: The numbers were not identical. The company’s character- ization of the discrepancies as “tiny” or “insignificant” overlooks that the court does not weigh the evidence. See, e.g., Celgene Corp. v. Peter, 931 F.3d 1342, 1352 (Fed. Cir. 2019). And even if the court could do so, CS Wind also cites no authority offering guidance on what magnitude of discrepancy is so small as to be “insignificant.” Cf. OCP S.A. v. United States, Slip Op. 26-75, at 22, 2026 WL 2269141, at *8 (CIT 2026) (“That’s presumably be- cause ‘significance’ is necessarily a fact-bound deter- mination left to the agency’s discretion.”). 2 And while 2 Similarly, the company’s argument that Commerce did not address “the much larger discrepancy between CS Wind Malaysia’s invoice price to CS Wind Korea and the Forms 7501 on the record,” ECF 72, at 17–18, is a strawman at best. Of course there was a discrepancy be- tween the Malaysian company’s invoice price to its parent and the value the latter declared to U.S. Customs. That’s what a “markup” is. The company’s argument amounts to, “The documents show there was a markup, so we are (footnote continues on next page) Ct. No. 24-00079 Page 8 the company insists that it is not “necessary” for the invoice and the 7501 to match, Commerce has the dis- cretion to decide what information it needs. 3 See, e.g., Shelter Forest Int’l Acquisition, Inc. v. United States, 831 F. Supp. 3d 1304, 1340 (CIT 2026). CS Wind’s only other argument is that the Depart- ment had a statutory obligation to ask for an explana- tion if it “required additional information to under- stand why invoice prices could differ slightly from the entered values reflected in the Forms 7501.” ECF 72, at 26 (citing 19 U.S.C. § 1677m(d)). The company (yet again) incorrectly states the issue. The agency wasn’t concerned about why the values didn’t match. It made clear that what mattered was that they did not match and that there was no consistency to the discrepancies in either amount or direction. 4 See, e.g., Appx10210– entitled to an adjustment.” The problem, as the Depart- ment explained, is that the record does not reliably estab- lish how much of a markup there actually was. 3 For this reason, the court need not address CS Wind’s ar- guments that are simply variations on the same theme— that the numbers were close enough to require Commerce to grant the adjustment. All those contentions rely on the same flawed premise—that CS Wind proved that the U.S. invoices in fact show the entered value despite its admis- sion that they do not reflect the same pricing reported on the Customs forms. 4 Put differently, the court can reasonably discern that if the discrepancy had been consistent, Commerce could have asked the company to explain it with a goal of applying a consistent correction to all sales. There was no consistency, so the court understands the agency’s point to be that no explanation was possible. See Motor Vehicle Mfrs. Ass’n of (footnote continues on next page) Ct. No. 24-00079 Page 9 10211 (noting both of those points); Appx10219 (em- phasizing, twice, that the values did not match). It is not apparent that § 1677m(d) applies at all in the context of an entered-value adjustment because, read in full, that provision addresses when Commerce may disregard all or part of a respondent’s submission and what prerequisites it must satisfy before doing so. See Daikin Am., Inc. v. United States, Slip Op. 24-32, at 9 n.3, 2024 WL 1171736, at *3 n.3 (CIT 2024) (ob- serving that where a respondent fails to show eligibil- ity for a constructed export price offset, the agency need not disregard submitted information and instead must simply deny the offset). The logic of Daikin ap- plies equally here. The Department did not “disregard” any part of CS Wind’s entered-value submissions. It reviewed the materials and concluded that they did not justify granting the adjustment, so it denied the request. But even assuming that § 1677m(d) does apply in this context, the government persuasively argues that the statute’s essential words are “to the extent practi- cable.” ECF 78, at 27. That is, when the Department “determines that a response to a request for infor- mation . . . does not comply with the request,” it is to inform the submitter of the problem “and shall, to the extent practicable,” provide an opportunity to cure the U.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983) (stating that courts will “uphold a decision of less than ideal clarity if the agency’s path may reasonably be discerned”). Ct. No. 24-00079 Page 10 deficiency or to explain it. 19 U.S.C. § 1677m(d) (em- phasis added). CS Wind admitted it could not provide Customs Form 7501 or bills of lading for two of the requested transactions because it did not “maintain the re- quested documentation.” Appx5458–5459 (question- naire responses stating, in part, “CS Wind is unable to provide the bill of lading or the CBP Forms 7501 in connection with this project”); ECF 78, at 26–27 (gov- ernment’s comments citing that statement). When a respondent admits that documents do not exist, it would be pointless for the court to remand for the agency to request them. Cf. Hung Vuong Corp. v. United States, 483 F. Supp. 3d 1321, 1354 (CIT 2020) (finding that where respondent admitted it had dis- posed of documents, “Commerce had no obligation un- der 19 U.S.C. § 1677m(d) to provide [the company] an opportunity to cure these deficiencies[, which] would have been futile because the documents no longer ex- isted”). That is the situation here. 5 In sum, Commerce’s decision to deny the entered- value adjustment was both lawful and supported by substantial evidence. The court sustains it. 5 Similarly, the Federal Circuit has recognized the fallacy of allowing respondents to knowingly submit incomplete or inadequate data only to turn around and “demand an op- portunity to remedy their intentionally deficient data.” Pa- pierfabrik August Koehler SE v. United States, 843 F.3d 1373, 1384 (Fed. Cir. 2016). Ct. No. 24-00079 Page 11 II The other issue is the Department’s averaging of land price data. 6 CS Wind contends that the agency failed to address record evidence that land prices in the Malaysian states of Penang (reflected in the Ellis data) and Pahang (where the company leases land from the Malaysian government) are not comparable because the former are roughly triple the latter. It asks the court to remand with instructions that Com- merce consider “the disparity in real estate prices.” ECF 72, at 32. The agency, however, acknowledged the price dif- ferences between the two states. Appx10202. But it ex- plained that “the existence of differences in land prices in and of themselves does not render the Penang prices objectively worse or the Pahang prices objectively bet- ter.” Id. Excluding “non-aberrational benchmarks due solely to price differences would not necessarily yield a more accurate benchmark.” Id. Commerce observed that CS Wind admits the Ma- laysian government subsidizes businesses investing in the region that includes Pahang. Appx10202 (citing Appx5595, CS Wind’s pre-remand agency brief). The 6 As noted above, in remanding the court also instructed the Department to address the producer’s contention that the agency ignored the size difference between CS Wind’s larger lot in Pahang and the smaller Penang parcel cited in the Ellis data. Remand Op. at 13–14, 2025 WL 3496986, at *5. Commerce discussed that issue in the remand results. See Appx10201–10202. The company fails to respond in its post-remand comments. Any objection is therefore waived. Ct. No. 24-00079 Page 12 Ellis data the Department placed on the record said the same. Appx10202–10203 (citing Appx10083). Thus, the agency concluded, it was reasonable to ex- pect land prices in Pahang to be lower, though it ad- mitted it was unclear to what extent government sub- sidization was to blame. Appx10203. And it noted that subsidization in Pahang was why it had originally been skeptical of the government data CS Wind favors. Id. The company’s claim that the Department “ig- nored” the difference in pricing is therefore baseless. But CS Wind also asserts that the agency’s regulation “prohibits including actual land transactions from a region that record evidence demonstrates is not com- parable in price with those” where the respondent is located, and it claims that “Commerce ignores the baseline requirement that the benchmark include only comparable prices stemming from actual transac- tions.” ECF 72, at 40 (emphasis in original). The com- pany cites no authority in support of either proposi- tion. As the government correctly notes, the regulation simply requires the Department “to consider compara- bility factors; it does not make any one factor alone dis- positive.” ECF 78, at 39 (citing 19 C.F.R. § 351.511(a)(2)(i)). Ultimately, Commerce found problems with both the Ellis dataset it originally placed on the record and the Malaysian government information CS Wind fa- vored. Appx10203. So it used both, “consistent with its preference for using multiple benchmark sources and in light of the fact that the [latter] data reflected trans- actions from the same region as CS Wind Malaysia.” Ct. No. 24-00079 Page 13 Id. The Department summarized that the Ellis data were favorable because the transaction reported was between two private parties, while the government in- formation benefited from location similarity and inclu- sion of multiple sales. Appx10203–10204. In sum, the agency’s obligation on remand was to consider CS Wind’s argument that the land prices in Penang and Pahang were not comparable. It did that and found that while there is indeed a difference, it is not disqualifying. The court did not purport to instruct the agency on what the outcome should be, nor did it direct the Department to exclude the Penang prices, reject them, or do anything else. Finally, the producer complains that Commerce’s “decision to average the [government] data with the single transaction contained in the CB Richard Ellis data is contrary to law.” ECF 72, at 37–38. But the court sustained the lawfulness of averaging the two datasets (provided the agency explained why it deemed the transactions sufficiently comparable). See Remand Op. at 14–15, 2025 WL 3496986, at **5–6. The Department reasonably explained why it deemed both sources acceptable despite their flaws. * * * The court sustains the agency’s redetermination. A separate judgment will issue. See USCIT R. 58(a). Dated: October 14, 2026 /s/ M. Miller Baker New York, NY Judge