2023-2245

Appeal from the United States Court of International Trade in No. 1:18-cv-00191-RKE, Senior Judge Richard K. Eaton.

Decided: July 28, 2025

BRITTNEY RENEE POWELL, Fox Rothschild LLP, Washington, DC, argued for plaintiff-appellee. Also represented by LIZBETH ROBIN LEVINSON.

BRENDAN DAVID JORDAN, Commercial Litigation Branch, Civil Division, United States Department of Justice, Washington, DC, argued for defendant-appellant. Also represented by BRIAN M. BOYNTON, TARA K. HOGAN, PATRICIA M. MCCARTHY; RACHEL BOGDAN, Office of the Chief Counsel, United States Department of Commerce, Washington, DC.

Page 2Before HUGHES, BRYSON, and STARK, Circuit Judges .

BRYSON, Circuit Judge .

The government appeals the decision of the Court of International Trade (“CIT”), which held that it was unlawful for the Department of Commerce (“Commerce”) to assign a non-market economy (“NME”) country-wide antidumping duty rate to Jilin Forest Industry Jinqiao Flooring Group Co. (“Jilin”). We reverse.

Jilin is an exporter of multilayered wood flooring in the People’s Republic of China (“China” or “PRC”). In November 2010, Commerce initiated an antidumping investigation into the sale of multilayered wood flooring from China. For purposes of the investigation, Commerce treated China as an NME country. Multilayered Wood Flooring From the People’s Republic of China: Final Determination of Sales at Less Than Fair Value, 76 Fed. Reg. 64,318 at 64,321 (Oct. 18, 2011) (“Final Determination”).

The Tariff Act defines an NME country as “any foreign country that the administering authority determines does not operate on market principles of cost or pricing structures, so that sales of merchandise in such country do not reflect the fair value of the merchandise.' 19 U.S.C. § 1677(18)(A). The Act further lists five nonexclusive factors for Commerce to consider in determining whether a country is an NME country. Id. § 1677(18)(B). Two of those factors are: “the extent of government ownership or control of the means of production” and “the extent of government control over the allocation of resources and over the price and output decisions of enterprises.” Id . § 1677(18)(B)(iv), (v). No party to the investigation challenged China’s designation as an NME country. Final Determination at 64,321.

Page 3Given China's NME status, Commerce applied a longstanding practice that it described as follows:

In proceedings involving NME countries, the Department holds a rebuttable presumption that all companies within the country are subject to government control and, thus, should be assessed a single antidumping duty rate. It is the Department’s policy to assign all exporters of the subject merchandise in an NME country this single rate unless an exporter can demonstrate that it is sufficiently independent so as to be entitled to a separate rate.

Id . In its Initiation Notice, Commerce notified the parties of the application process by which exporters may obtain separate rates in NME investigations. Multilayered Wood Flooring From the People’s Republic of China: Preliminary Determination of Sales at Less Than Fair Value, 76 Fed. Reg. 30,656 at 30,658 (May 26, 2011).

In determining the antidumping duty rate for the PRCwide entity, Commerce drew an adverse inference based on a finding that not all exporters responded to Commerce’s requests for information and thus 'the PRC-wide entity has failed to cooperate to the best of its ability.” Final Determination at 64,322. The final rate that Commerce calculated for the PRC-wide entity was 25.62 percent. See Baroque Timber Indus. (Zhongshan) Co. v. United States , 971 F. Supp. 2d 1333, 1339 (Ct. Int’l Trade 2014).

That PRC-wide rate was not assigned to the seventyfour exporters that timely filed separate rate applications. Commerce found that each of them demonstrated “both de jure and de facto absence of government control with respect to each company’s respective exports of the merchandise under investigation” and thus qualified for a separate rate instead of the PRC-wide rate. Final Determination at 64,322. Jilin was among the seventy-four exporters that received a separate rate (3.31 percent). Id . at 64,323.

Page 4In February 2017, Commerce initiated the fifth administrative review of the antidumping duty order. Due to the large number of exporters involved, Commerce selected the two largest PRC exporters by volume as mandatory respondents for individual examination, one of which was Jilin. See App. 83–84. Commerce once again explained the application process by which exporters could obtain separate rates. Initiation of Antidumping and Countervailing Duty Administrative Reviews, 82 Fed. Reg. 10,457 at 10,458 (Feb. 13, 2017).

Jilin cooperated with the review and asserted that it was not under the control of the Chinese government and should therefore receive a separate rate, as it had in prior reviews. However, this time Commerce found that Jilin failed to rebut the presumption of government control and was therefore regarded as a part of the PRC-wide entity. See App. 138-40.¹ Accordingly, Commerce assigned Jilin the PRC-wide antidumping duty rate of 25.62 percent, the rate calculated in the initial investigation.² Multilayered Wood Flooring From the People’s Republic of China: Final Results of Antidumping Duty Administrative Review, Final Determination of No Shipments, and Partial Rescission; 2015–2016, 83 Fed. Reg. 35,461 at 35,464 (July 26, 2018).

Jilin challenged the results of the fifth administrative review at the CIT. In Jilin I , the court ordered a remand, Page 5questioning the lawfulness of “applying a rebuttable presumption that all companies within a nonmarket economy country are controlled by the government of that country, i.e. , the ‘NME Policy.’” App. 20. The court ordered Commerce to calculate an individual antidumping duty rate for Jilin “or provide a reasonable explanation for why it need not.” App. 33.

On remand, Commerce identified two separate components to its practice: a presumption of government control over exporters in an NME country and the use of a single antidumping duty rate for the NME-wide entity. Commerce explained that both components were 'consistent with a general statutory recognition of a ‘close correlation between a nonmarket economy and government control of prices, output decisions, and the allocation of resources.’” App. 174 (quoting Sigma Corp. v. United States , 117 F.3d 1401, 1405–06 (Fed. Cir. 1997)). Commerce also cited 19 C.F.R. § 351.107(d) as supporting the use of a single rate for all exporters in an NME country that are unable to demonstrate independence from the government. App. 173.

In Jilin II , the CIT found “substantial evidence to support Commerce’s conclusion that Jilin has not rebutted the presumption of state control.” App. 45. Nevertheless, the court held that “Commerce has not shown that its NME Policy . . . is in accordance with law with respect to Jilin” and remanded the case again to Commerce to calculate an individual weighted average rate for Jilin. App. 37 (cleaned up). On the second remand, Commerce determined, under protest, an individual weighted average rate of zero percent for Jilin. The CIT entered that rate in its final judgment. App. 222.

The government timely filed this appeal.

Page 6The Tariff Act requires Commerce to “determine the estimated weighted average dumping margin for each exporter and producer individually investigated' once it makes a final determination that the subject merchandise is being, or is likely to be, sold in the United States at less than its fair value. 19 U.S.C. § 1673d(c)(1)(B)(i)(I). But “[i]f it is not practicable to make individual weighted average dumping margin determinations under [section 1673d(c)] because of the large number of exporters or producers involved in the investigation or review,” the Tariff Act authorizes Commerce to “determine the weighted average dumping margins for a reasonable number of exporters or producers,” known as mandatory respondents. Id . § 1677f-1(c)(2).

At the time of the fifth administrative review in this case, there was no statutory provision or regulation that expressly codified the presumption of government control over exporters in an NME country ('the NME presumption”), or the assignment of a single NME-wide rate on failing to rebut the presumption (“the NME policy').³ However, the version of 19 C.F.R. § 351.107(d) that was in effect at the time provided that “[i]n an antidumping proceeding involving imports from a nonmarket economy country, ‘rates’ may consist of a single dumping margin applicable to all exporters and producers.”

At issue in this case is whether Commerce lawfully assigned the PRC-wide antidumping duty rate to Jilin, a cooperative mandatory respondent that failed to rebut the NME presumption. Jilin does not appeal the CIT’s finding Page 7that Jilin failed to rebut the presumption of government control. As such, reversal is required so long as the NME policy, with the NME presumption at its core, is lawful.

The government argues that binding precedents since Sigma Corp. v. United States , 117 F.3d 1401 (Fed. Cir. 1997) require reversal. Specifically, the government asserts that this court has 'repeatedly upheld Commerce's NME policy” and recently confirmed that “Commerce may apply the NME-wide rate to a cooperative mandatory respondent who fails to rebut the presumption of government control even when the NME-rate was based on adverse facts available.” Appellant’s Br. at 17. According to the government, 'Commerce's decision here is no different' from China Manufacturers Alliance, LLC v. United States , 1 F.4th 1028 (Fed. Cir. 2021) (“ CMA ”), and “the same result-affirmance of Commerce's decision-is required.' Appellant’s Br. at 22.

In Sigma , we held that Commerce “has broad authority to interpret the antidumping statute and devise procedures to carry out the statutory mandate,' and that 'it was within Commerce’s authority to employ a presumption of state control for exporters in a nonmarket economy, and to place the burden on the exporters to demonstrate an absence of central government control.” 117 F.3d at 1405. We explained that the antidumping statute “recognizes a close correlation between a nonmarket economy and government control of prices, output decisions, and the allocation of resources,” citing 19 U.S.C. § 1677(18)(B)(iv) and (v). Id . at 1405-06. We further explained that 'because exporters have the best access to information pertinent to the ‘state control’ issue, Commerce is justified in placing on them the burden of showing a lack of state control.' Id . at 1406.

Since Sigma , we have consistently approved of Commerce’s practice of applying the NME presumption and assigning the NME-wide rate to exporters that fail to rebut the presumption. See, e.g. , Transcom, Inc. v. United States ,Page 8 294 F.3d 1371, 1373 (Fed. Cir. 2002); Changzhou Wujin Fine Chem. Factory Co. v. United States , 701 F.3d 1367, 1370 (Fed. Cir. 2012); Michaels Stores, Inc. v. United States , 766 F.3d 1388, 1390 (Fed. Cir. 2014); Dongtai Peak Honey Indus. Co. v. United States , 777 F.3d 1343, 1349–50 (Fed. Cir. 2015); Albemarle Corp. & Subsidiaries v. United States , 821 F.3d 1345, 1348 (Fed. Cir. 2016); Changzhou Hawd Flooring Co. v. United States , 848 F.3d 1006, 1009 (Fed. Cir. 2017); Diamond Sawblades Mfrs. Coal. v. United States , 866 F.3d 1304, 1311 (Fed. Cir. 2017); Zhejiang Mach. Imp. & Exp. Corp. v. United States , 65 F.4th 1364, 1372 (Fed. Cir. 2023); Pirelli Tyre Co. v. United States , 128 F.4th 1265, 1268 (Fed. Cir. 2025).

In CMA , we sustained Commerce's assignment of the PRC-wide rate to Double Coin, a cooperative mandatory respondent that failed to rebut the NME presumption in an administrative review. We held that “where a respondent in an NME country cooperates with an investigation or review but fails to rebut the presumption of government control, Commerce may permissibly apply the country-wide NME entity rate,” including a rate that was based in whole or in part on adverse facts available and carried forward from the initial investigation.⁴ 1 F.4th at 1039-40. We further held that “[t]his conclusion applies whether or not other members of the NME-wide entity are identified by name and subject to the administrative review at issue.” Id . at 1040.

Jilin, however, argues that “[t]here is simply no precedent regarding the precise legal question raised in this appeal that binds the Court’s hands.” Appellee’s Br. at 19.

Page 9According to Jilin, the question in this case is 'whether Commerce’s application of the NME Policy is lawful in the absence of a statute or regulation.' Id . at 15. Jilin asserts that “[n]one of this Court’s prior decisions have addressed why an uncodified policy should be given more deference than the statutory provision requiring Commerce to determine the ‘individual weighted average dumping margin for each known exporter and producer of the subject merchandise.’” Id . at 18–19 (quoting 19 U.S.C. § 1677f-1(c)(1)).

Specifically, Jilin argues that this court's decision in CMA 'rel[ied] on strict and unchallenged deference to Commerce, and it is that deference to an uncodified policy, with no statutory or regulatory basis, that is being challenged in this proceeding.' Id . at 17. Jilin contends that CMA 'relied on Sigma to conclude (and arguably 'bootstrap' without further analysis) that Commerce has ‘broad authority to interpret the antidumping statute and devise procedures to carry out the statutory mandate.’” Id . at 16-17 (quoting CMA , 1 F.4th at 1038).

Jilin's characterization of CMA reads out the court's analysis of the statutory and regulatory support for the NME policy to avoid its clear import in this case. In CMA , respondent Double Coin “recognize[d] that binding cases (too numerous to list in their entirety) have uniformly sustained Commerce’s recognition of an NME-wide entity as a single exporter for purposes of assigning an antidumping rate to the individual members of the entity.” 1 F.4th at 1036. Nevertheless, Double Coin “question[ed] . . . the authority for Commerce to so recognize such an NME-wide entity,” as Jilin does in this case. Id . at 1037.

The CMA court squarely addressed that question of Commerce's authority by first holding that 19 C.F.R. § 351.107 provides 'clear authority . . . for Commerce to fashion a single rate for all exporters and producers that qualify for the single rate.” Id . In light of that regulation, we found it 'clear that Commerce may, where the facts Page 10warrant, recognize a single NME-wide entity to include all exporters that fail to rebut the presumption of government control.” Id .

We further held that the 'country-wide NME entity rate may be an ‘individually investigated’ rate within the meaning of 19 U.S.C. § 1673d(c)(1)(B)(i)(I),” so that “Commerce may permissibly assign such a rate to the unitary group of exporters in an NME country that have failed to rebut the presumption of government control.' Id . at 1039. We reasoned that the “PRC-wide entity rate resulting from Commerce’s initial investigation constitute[d] an ‘individually investigated’ weighted average dumping margin . . . because Commerce treat[ed] the companies comprising the China-wide entity as a single entity and investigated them as such in the original investigation.” Id . at 1037 (cleaned up); see also Guizhou Tyre Co., Ltd. v. United States , 557 F. Supp. 3d 1302, 1323 (Ct. Int’l Trade 2022) (“[B]ased on the holding in China Mfrs. Alliance , the court concludes that the PRC-wide rate must be deemed to have been assigned to a known, 'individually-investigated' exporter and producer consisting of the PRC-wide entity.').⁵

The CMA court's reliance on the holding in Sigma that Commerce “has broad authority to interpret the antidumping statute and devise procedures to carry out the statutoryPage 11 mandate” was directed to a narrow issue. We cited that holding when discussing whether Commerce permissibly exercised discretion to take the simple average of two rates to determine the “proper PRC-wide entity rate.” 1 F.4th at 1038 (quoting Sigma , 117 F.3d at 1405, and holding that “Double Coin points to no precedent that precludes the discretion Commerce exercised in averaging the two rates in this case'). As explained above, we found statutory and regulatory support for the fundamental aspects of the NME policy at issue in CMA .

Accordingly, CMA and the long line of cases CMA relied on are binding precedent that Jilin may not avoid unless or until the court sits en banc . See Preminger v. Sec’y of Veterans Affs. , 517 F.3d 1299, 1309 (Fed. Cir. 2008) (“A prior precedential decision on a point of law by a panel of this court is binding precedent and cannot be overruled or avoided unless or until the court sits en banc .”). Our holding in CMA that Commerce may lawfully assign an NMEwide antidumping duty rate to a cooperative mandatory respondent that has failed to rebut the presumption of government control requires reversing the CIT in this case.

Even if CMA and the long line of cases sustaining the NME policy did not exist, Jilin’s argument that Commerce could not apply the NME policy absent notice-and-comment rulemaking would fail.

To be clear, the version of 19 C.F.R. § 351.107(d) applicable to this case expressly contemplated a single antidumping duty rate applicable to all exporters in an NME country. See 19 C.F.R. § 351.107(d) ('In an antidumping proceeding involving imports from a nonmarket economy country, ‘rates’ may consist of a single dumping margin applicable to all exporters and producers.”). If anything, the language of the regulation was broader than the NME policy in that it did not provide for the possibility of a separate rate upon rebutting the presumption of government Page 12control. To the extent that section 351.107(d) did not spell out the NME presumption, that silence does not render the NME presumption invalid.

The NME presumption is an evidentiary presumption, i.e. , a factual inference that arises from a finding that an exporting country has a non-market economy. From a finding that a country does not operate on market principles, Commerce infers that the country’s government exercises control over its exporters, unless proved otherwise in the case of individual exporters. “Presumptions normally arise when proof of one fact renders the existence of another fact ‘so probable that it is sensible and timesaving to assume the truth of [the inferred] fact . . . until the adversary disproves it.’” NLRB v. Curtin Matheson Sci., Inc. , 494 U.S. 775, 788-89 (1990) (alteration in original) (quoting E. Cleary, McCormick on Evidence § 343, p. 969 (3d ed. 1984)).

By its terms, the Administrative Procedure Act (“APA”) does not always require an agency to comply with the notice-and-comment procedure outlined in 5 U.S.C. § 553(b). The APA exempts “interpretive rules, general statements of policy, or rules of agency organization, procedure, or practice' from notice-and-comment requirements. Id . § 553(b)(A). Only “legislative rules” are subject to noticeand-comment requirements. See Perez v. Mortg. Bankers Ass’n , 575 U.S. 92, 101 (2015); Lincoln v. Vigil , 508 U.S. 182, 196 (1993).

Jilin argues that “[i]t is a fundamental principle of administrative law that an agency’s authority to promulgate legislative rules ( i.e. rules made pursuant to congressionally-delegated authority) is governed by the informal rulemaking procedures outlined in the [APA].” Appellee’s Br. at 13. But Jilin merely assumes that the NME presumption is a legislative rule. Jilin provides no authority or explanation as to why an evidentiary presumption, aPage 13 shorthand for adopting a factual inference, qualifies as a legislative rule.

The validity of an evidentiary presumption turns on its rationality. See Republic Aviation Corp. v. NLRB , 324 U.S. 793, 804–05 (1945) (“Like a statutory presumption or one established by regulation, the validity [of the Board’s presumption], perhaps in a varying degree, depends upon the rationality between what is proved and what is inferred.”); NLRB v. Baptist Hosp. , 442 U.S. 773, 787 (1979) (“It is, of course, settled law that a presumption adopted and applied by the Board must rest on a sound factual connection between the proved and inferred facts.”); Nat’l Min. Ass’n v. Babbitt , 172 F.3d 906, 912 (D.C. Cir. 1999) (“[A]n evidentiary presumption is only permissible if there is a sound and rational connection between the proved and inferred facts . . . .” (cleaned up)).

There is a sound and rational connection between a finding that a country is an NME country and the inference that exporters in that country are subject to government control. As noted, two of the factors that Commerce considers in determining whether a country is an NME country are “the extent of government ownership or control of the means of production' and 'the extent of government control over the allocation of resources and over the price and output decisions of enterprises.” 19 U.S.C. § 1677(18)(B)(iv), (v). Logically, greater government control over the means of production or the allocation of resources would lead to the conclusion that a country's economy does not operate on market principles. Thus, once a country has been found to be an NME country, there is nothing unreasonable about presuming that exporters in that country are subject to government control, unless proved otherwise in each individual case.

Accordingly, the NME presumption is a valid evidentiary presumption that Commerce was permitted to use.

Page 14For the foregoing reasons, the judgment of the Court of International Trade is reversed.

COSTS

No costs.