Primesource Building Products, Inc. v. United States

111 F.4th 1320

Opinion

United States Court of Appeals for the Federal Circuit

______________________

PRIMESOURCE BUILDING PRODUCTS, INC., CHENG CH INTERNATIONAL CO., LTD., CHINA STAPLE ENTERPRISE CORP., DE FASTENERS INC., HOYI PLUS CO., LTD., LIANG CHYUAN INDUSTRIAL CO., LTD., TRIM INTERNATIONAL INC., UJL INDUSTRIES CO., LTD., YU CHI HARDWARE CO., LTD., ZON MON CO., LTD., Plaintiffs-Appellants

v.

UNITED STATES, MID CONTINENT STEEL & WIRE, INC.,

Defendants-Appellees

______________________

2022-2128, 2022-2129

______________________

Appeals from the United States Court of International Trade in Nos. 1:20-cv-03911-MAB, 1:20-cv-03934-MAB, Chief Judge Mark A. Barnett.

______________________

Decided: August 7, 2024

______________________

BRYAN PATRICK CENKO, Mowry & Grimson, PLLC, Washington, DC, argued for plaintiff-appellant PrimeSource Building Products, Inc. Also represented by JILL CRAMER, JEFFREY S. GRIMSON, YIXIN LI, KRISTIN HEIM MOWRY, SARAH WYSS.

2 PRIMESOURCE BUILDING PRODUCTS, INC. v. US

KELLY ALICE SLATER, Appleton Luff Pte. Ltd., Washington, DC, argued for plaintiffs-appellants Cheng Ch International Co., Ltd., China Staple Enterprise Corp., De Fasteners Inc., Hoyi Plus Co., Ltd., Liang Chyuan Industrial Co., Ltd., Trim International Inc., UJL Industries Co., Ltd., Yu Chi Hardware Co., Ltd., Zon Mon Co., Ltd.

SOSUN BAE, Commercial Litigation Branch, Civil Division, United States Department of Justice, Washington, DC, argued for defendant-appellee United States. Also represented by BRIAN M. BOYNTON, PATRICIA M. MCCARTHY; VANIA WANG, Office of the Chief Counsel for Trade Enforcement and Compliance, United States Department of Commerce, Washington, DC.

ADAM H. GORDON, The Bristol Group PLLC, Washington, DC, argued for defendant-appellee Mid Continent Steel & Wire, Inc. Also represented by BENJAMIN JACOB BAY, JENNIFER MICHELE SMITH-VELUZ.

______________________

Before LOURIE, DYK, and STOLL, Circuit Judges.

Opinion for the court filed by Circuit Judge LOURIE.

Opinion concurring in part and dissenting in part filed by Circuit Judge DYK.

LOURIE, Circuit Judge.

PrimeSource Building Products, Inc. (“PrimeSource”) and Cheng Ch International Co., Ltd., et al. 1 (“Cheng Ch”

1 China Staple Enterprise Corporation, De Fasteners Inc., Hoyi Plus Co., Ltd., Liang Chyuan Industrial Co., Ltd. (“Liang Chyuan”), Trim International

PRIMESOURCE BUILDING PRODUCTS, INC. v. US 3

or the “non-selected respondents”) separately appeal from the final judgment of the United States Court of International Trade (the “Trade Court”) on the fourth administrative review of an antidumping duty order on certain steel nails from Taiwan. PrimeSource Bldg. Prod., Inc. v. United States, 581 F. Supp. 3d 1331 (Ct. Int’l Trade 2022) (“Decision”). The Trade Court sustained the United States Department of Commerce’s (“Commerce”) use of the expected method to calculate an all-others rate for the non- selected respondents equal to the adverse facts available (“AFA”) rate that was applied to all the mandatory respondents in Commerce’s review. Id.

PrimeSource is an importer of steel nails from Taiwan and appeals Commerce’s calculation and application of the all-others rate solely with respect to Liang Chyuan, one of the non-selected respondents. Cheng Ch appeals the rate with respect to the non-selected respondents, generally. Because Commerce’s calculation and application of the all- others rate is supported by substantial evidence and otherwise in accordance with law, we affirm.

BACKGROUND

I

We begin with a brief overview of the statutory framework for determining antidumping duty rates.

Commerce is authorized by statute to impose antidumping duties on goods sold in the United States below fair market value. See 19 U.S.C. § 1673. Those duties are equal to the amount by which the normal value of the merchandise exceeds the export price, i.e., the dumping margin. Id. at §§ 1673e(a)(1), 1677(35); see

Inc., UJL Industries Co., Ltd., Yu Chi Hardware Co., Ltd., and Zon Mon Co., Ltd.

4 PRIMESOURCE BUILDING PRODUCTS, INC. v. US

Albemarle Corp. & Subsidiaries v. United States, 821 F.3d 1345, 1348 (Fed. Cir. 2016).

Commerce is generally charged with calculating individual dumping margins for each exporter of the subject merchandise during an administrative review of a countervailing duty order. 19 U.S.C. §§ 1677f–1(c)(1), 1675(a); Albemarle, 821 F.3d at 1348. However, when it is “not practicable” to calculate a rate for each exporter because of the large number, Commerce may limit its examination to “a reasonable number” of exporters constituting a statistically representative sample of all known exporters or accounting for the largest volume of the subject merchandise from the exporting country. 19 U.S.C. § 1677f–1(c)(2). The exporters selected for individual examination are referred to as mandatory respondents. See Yangzhou Bestpak Gifts & Crafts Co. v. United States, 716 F.3d 1370, 1372 (Fed. Cir. 2013) (“Bestpak”). Commerce calculates dumping margins for the mandatory respondents based on data provided by the respondents through their responses to an antidumping questionnaire. Id. at 1372. However, if a respondent fails to act to the best of its ability to respond to the questionnaire, Commerce may assign it a dumping margin based on an adverse inference from the facts available in the petition or elsewhere, i.e., based on AFA. 19 U.S.C. § 1677e(b).

When Commerce limits the number of individually examined exporters under § 1677f–1(c)(2), it calculates an “all-others” rate for the non-selected respondents by weight-averaging the dumping margins assigned to the mandatory respondents, excluding any margins that are zero, de minimis, or determined entirely based on AFA. 19 U.S.C. § 1673d(c)(1)(B), (c)(5)(A). The statute provides an exception when the dumping margins for all mandatory respondents are zero, de minimis, or determined entirely based on AFA; it instructs Commerce to “use any reasonable method” to calculate the all-others rate, including averaging the dumping margins for the

PRIMESOURCE BUILDING PRODUCTS, INC. v. US 5

mandatory respondents. 19 U.S.C. § 1673d(c)(5)(B). The method for calculating the all-others rate when that exception applies is the primary issue on appeal.

Additional guidance on that provision is provided in the Statement of Administrative Action (“SAA”), which is legislative history that Congress has mandated “as an authoritative expression concerning the interpretation and application of the Tariff Act.” Bestpak, 716 F.3d at 1373; 19 U.S.C. § 3512(d). For the § 1673d(c)(5)(B) exception, the SAA provides that:

In such situations, Commerce may use any reasonable method to calculate the all others rate. The expected method in such cases will be to weight-average the zero and de minimis margins and margins determined pursuant to the facts available, provided that volume data is available. However, if this method is not feasible, or if it results in an average that would not be reasonably reflective of potential dumping margins for non- investigated exporters or producers, Commerce may use other reasonable methods.

SAA, accompanying the Uruguay Round Agreements Act, H.R. Doc. No. 103–316, at 873 (1994), reprinted in 1994 U.S.C.C.A.N. 4040, 4201.

The statute also provides non-selected respondents the opportunity to complete the antidumping questionnaire to request individual examination as a voluntary respondent. 19 U.S.C. § 1677m(a); Bestpak, 716 F.3d at 1373. The voluntary respondent may submit its responses to the questionnaire by the date specified for the respondents “that were initially selected for examination.” 19 U.S.C. § 1677m(a). However, Commerce may still decline to individually examine the voluntary respondent if it determines that it would be unduly burdensome to do so. Id.

6 PRIMESOURCE BUILDING PRODUCTS, INC. v. US

II

We now turn to Commerce’s fourth administrative review of the antidumping duty order covering steel nails from Taiwan with a July 1, 2018, to June 30, 2019 period of review. See Certain Steel Nails From Taiwan, 85 Fed. Reg. 76,014 (Dep’t Commerce Nov. 27, 2020) (“Final Results”); see also Certain Steel Nails From the Republic of Korea, Malaysia, the Sultanate of Oman, Taiwan, and the Socialist Republic of Vietnam, 80 Fed. Reg. 39,994 (Dep’t Commerce July 13, 2015). For the fourth administrative review, Commerce determined that it was necessary to limit the number of individually examined exporters in accordance with 19 U.S.C. § 1677f–1(c)(2). It chose the two largest exporters of the subject merchandise by volume as mandatory respondents: Bonuts Hardware Logistics Co., LLC, (“Bonuts”) and Create Trade Co., Ltd., (“Create”). 2 Commerce issued antidumping duty questionnaires to Bonuts and Create on October 23, 2019, and October 28, 2019, respectively. Bonuts did not respond to the questionnaire, and Create submitted a letter stating that it had no reviewable sales because of Commerce’s reseller policy. Commerce accepted the representations made by Create and did not require it to respond to the questionnaire. Commerce then selected Pro-Team to replace Create as a mandatory respondent because it was the next highest exporter by volume. On January 31, 2020,

2 The Customs and Border Protection data relied on by Commerce in the selection process indicated that Bonuts accounted for 73.67% of the entry volume of the subject merchandise for the relevant period of review. J.A. 142. Create, the second largest exporter by volume, accounted for 5.68%, and Pro-Team Coil Nail Enterprise Inc. (“Pro-Team”), the third largest exporter by volume, accounted for 4.32%. Id.

PRIMESOURCE BUILDING PRODUCTS, INC. v. US 7

Pro-Team submitted a letter indicating that it would not respond to Commerce’s questionnaire.

On February 3, 2020, after learning that neither mandatory respondent would respond to its antidumping duty questionnaire, Liang Chuyan submitted a letter requesting that Commerce calculate a dumping rate for it based on its own data. It did not, however, submit questionnaire responses or any data for the period of review to Commerce. If Commerce would not give it a rate calculated based on its own data, it alternatively requested that Commerce pull forward and apply its calculated rate from the previous administrative review or pull forward and apply the all-others rate from the original investigation.

Two months later, on April 6, 2020, Commerce published its preliminary results assigning both mandatory respondents an AFA rate of 78.17%, the highest margin applied in prior segments of the proceeding, due to their failure to respond to the questionnaires. Certain Steel Nails From Taiwan, 85 Fed. Reg. 19,138 (Dep’t of Commerce Apr. 6, 2020) (preliminary results); see Decision Memorandum for Preliminary Results of Antidumping Duty Administrative Review: Certain Steel Nails from Taiwan; 2018–2019, J.A. 543. Commerce then used the expected method to calculate the all-others rate for the non-selected respondents. J.A. 543–44. Because both mandatory respondents received an AFA rate of 78.17%, the weighted average of those two rates—the expected method—produced an all-others rate of 78.17%, which was equal to the AFA rate. See id. After considering the interested parties’ letters and case briefs, Commerce issued its Final Results and the accompanying issues and decision memorandum on November 27, 2020. See Final Results, 85 Fed. Reg. at 76,014; Issues and Decision Memorandum for the Final Results of the Antidumping Duty Administrative Review: Certain Steel Nails from Taiwan; 2018–2019, J.A. 612–33. It continued to assign the 8 PRIMESOURCE BUILDING PRODUCTS, INC. v. US

antidumping duty rate of 78.17% to the mandatory respondents, Bonuts and Pro-Team, and therefore to the non-selected respondents as well, including Liang Chyuan. J.A. 621–22 (citing Albemarle, 821 F.3d at 1352).

Commerce determined that the application of the expected method to the non-selected respondents was reasonable, despite all mandatory respondents receiving an AFA rate, because it examined the largest exporters by volume, which also accounted for the “vast majority” of total volume during the period of review. J.A. 623–24. Commerce examined the history of assigned rates in the proceedings on certain steel nails from Taiwan and found that those rates did not undermine the representativeness of the mandatory respondents. Id. Finally, Commerce determined that it could not deviate from the expected method to pull forward previous review-specific rates as requested by the non-selected respondents absent substantial evidence in the record to justify doing so. J.A. 628–30. It determined that such evidence did not exist and thus it continued to apply the 78.17% all-others rate produced by the expected method to the non-selected respondents. See J.A. 630.

Commerce further determined that Liang Chyuan was not entitled to an individual rate. J.A. 630–31. In particular, Commerce found that Liang Chyuan did not meet the requirements to be selected as a mandatory respondent because it was not the next largest producer and that Liang Chyuan had not satisfied the statutory requirements to be considered a voluntary respondent. Id. It explained that Liang Chyuan’s letter expressing willingness to submit questionnaire responses was not an acceptable substitute for the statutory requirements to be a voluntary respondent. J.A. 631. Commerce also rejected Liang Chyuan’s alternative request to pull a rate forward from an earlier review, finding that there was no record evidence supporting Liang Chyuan’s claim that it would have received the same result in the fourth review as it did

PRIMESOURCE BUILDING PRODUCTS, INC. v. US 9

in the previous review, had it been selected for individual examination. Id.

III

PrimeSource and Cheng Ch separately filed suit in the Trade Court to challenge the rate received by the non- selected respondents in Commerce’s Final Results. Decision, 581 F. Supp. 3d at 1334. The court sustained the Final Results, finding that Commerce’s reliance on the expected method was supported by substantial evidence and in accordance with the law. Id. at 1334–35. The court also rejected PrimeSource’s attempt to distinguish Liang Chyuan from the other non-selected respondents. Id. at 1343–44.

Specifically, the court held that Commerce’s use of the expected method was lawful. Examining our case law and the statutory framework, it determined that “the expected method is the default method and that the burden of proof lies with the party seeking to depart from the expected method.” Id. at 1338. It explained that, because the largest exporters are assumed to be representative of the non-selected respondents, Commerce is expected to use the mandatory respondents’ rates to determine the rate to be assigned to non-selected respondents. Id. at 1340.

The court then turned to whether or not the respondents had provided sufficient evidence to rebut the presumption of representativeness and thereby justify a departure from the expected method. Id. at 1341. The court noted that the respondents identified no evidence from the current period of review to support their assertion that the expected method was not reasonable. Id. It therefore reviewed Commerce’s analysis of the history of dumping margins assigned in the previous reviews. Id. at 1342. It determined that the prior dumping margins “support Commerce’s conclusion that the rates fluctuated significantly from review to review and, thus, that looking to past reviews for evidence of current dumping lacks a 10 PRIMESOURCE BUILDING PRODUCTS, INC. v. US

logical foundation.” Id. at 1343. It found that Commerce’s determination “that there was insufficient evidence on the record to rebut the presumed representativeness of the mandatory respondents’ rates” was supported by substantial evidence and it therefore sustained Commerce’s use of the expected method. Id.

Finally, the Trade Court determined that that Liang Chyuan was not entitled to a different rate than the other non-examined respondents. Id. at 1343. It noted that non- selected respondents are not generally entitled to individually determined rates, but that they may qualify as a voluntary respondent if they submit the necessary information in a timely fashion. Id. at 1344. It found that Liang Chyuan had not submitted the necessary information and “[had] not express[ed] its willingness to participate until February 3, 2020, roughly two months after the deadlines.” Id. The court then explained that, simply because Liang Chyuan received a calculated rate as a mandatory respondent in an earlier review, that did not entitle it to retain that rate for the present review. Id. The Trade Court therefore sustained Commerce’s Final Results.

PrimeSource and Cheng Ch timely appeal. We have jurisdiction pursuant to 28 U.S.C. § 1295(a)(5).

DISCUSSION

We review decisions of the Trade Court concerning Commerce’s antidumping determinations by applying the same standard of review as the Trade Court. Bestpak, 716 F.3d at 1377. At the same time, “‘we give great weight to the informed opinion’ of that court, which has expertise in international trade matters.” Chemtall, Inc. v. United States, 878 F.3d 1012, 1018 (Fed. Cir. 2017) (quoting Schlumberger Tech. Corp. v. United States, 845 F.3d 1158, 1162 (Fed. Cir. 2017)). Commerce’s determination will be sustained unless it is “unsupported by substantial evidence on the record, or otherwise not in accordance with law.”

PRIMESOURCE BUILDING PRODUCTS, INC. v. US 11

19 U.S.C. § 1516a(b)(1)(B)(i). A finding is supported by substantial evidence if a reasonable mind might accept the evidence to support the finding. Consol. Edison Co. v. NLRB, 305 U.S. 197, 229 (1938). “An agency finding may still be supported by substantial evidence even if two inconsistent conclusions can be drawn from the evidence.” Ad Hoc Shrimp Trade Action Comm. v. United States, 802 F.3d 1339, 1348 (Fed. Cir. 2015) (quoting Consolo v. Fed. Mar. Comm’n, 383 U.S. 607, 619–20 (1966)).

PrimeSource challenges Commerce’s application of the all-others rate solely with respect to Liang Chyuan, while Cheng Ch challenges it with respect to all non-selected respondents. Nevertheless, they raise two similar challenges on appeal. They first argue that, even though Commerce used the “expected method” to determine the dumping margin for the non-selected respondents, Commerce was required to demonstrate that the calculated all-others rate was reasonably reflective of the non-selected respondents’ potential dumping margin. They also argue that Commerce’s application of the all-others rate, which was equal to the AFA rate, to the non-selected respondents was unreasonable and not supported by substantial evidence. PrimeSource additionally argues that Commerce erred by not assigning an individual rate to Liang Chyuan.

I

We first address Commerce’s use of the expected method to calculate the all-others rate for the non-selected respondents. PrimeSource’s arguments against Commerce’s use of the expected method in the fourth administrative review focus on two issues: (1) that the statutory language places an affirmative burden on Commerce to show that its chosen method produces results reasonably reflective of the non-selected respondents’ potential dumping margin, and (2) that the presumption of representativeness of the mandatory respondents was rebutted by substantial evidence on the record, thus 12 PRIMESOURCE BUILDING PRODUCTS, INC. v. US

rendering the expected method unreasonable. See PrimeSource Br. at 13. Cheng Ch similarly argues that (1) the statute requires Commerce to select a method that produces results reasonably reflective of the non-selected respondents dumping margin, and (2) selecting a method based solely on the AFA rate was unreasonable for the cooperative non-selected respondents. See Cheng Ch Br. at 8–14.

A

Regarding Commerce’s burden with respect to the expected method, both appellants point to the language of 19 U.S.C. § 1673d(c)(5)(B) instructing Commerce to select “any reasonable method” to determine the all-others rate for exporters and producers not individually investigated when the rates calculated for the mandatory respondents are either zero, de minimis, or based entirely on AFA. PrimeSource argues that that language is “unequivocal” and that the term “reasonable” imposes a duty on Commerce to show that its chosen method, even when it is the expected method, is reasonable as applied to the facts of the case. PrimeSource Br. at 20. It continues that any reading of the SAA that does not place a burden on “Commerce to find that expected method results [are] a rate that reasonably reflects the potential dumping margins” of the non-selected respondents ignores that statutory mandate. Id. at 22–23. Similarly, Cheng Ch argues that it is unreasonable for Commerce to select a methodology based entirely on AFA rates. Cheng Ch Br. at 8. It argues that Commerce must undertake an examination of the non-selected respondents to show that its calculated rate reasonably reflects the non-selected respondents’ actual dumping margin. Id. at 9–10. We disagree.

Reading the SAA as prescribing the weight average as the default or expected “reasonable method” does not contradict the statute’s requirement that Commerce use

PRIMESOURCE BUILDING PRODUCTS, INC. v. US 13

“any reasonable method.” The SAA is the congressionally mandated “authoritative expression” of the Tariff Act in judicial proceedings. 19 U.S.C. § 3512(d); Bestpak, 716 F.3d at 1373. The SAA directly addresses § 1673d(c)(5)(B) and provides that, when all individually examined respondents receive rates of zero, de minimis, or based entirely on AFA:

Commerce may use any reasonable method to calculate the all others rate. The expected method in such cases will be to weight-average the zero and de minimis margins and margins determined pursuant to the facts available, provided that volume data is available.

SAA at 4201.

The SAA echoes the language of the statute highlighted by the appellants—that Commerce “may use any reasonable method” to calculate the all-others for the non-selected respondents. Id. (emphasis added); 19 U.S.C. § 1673d(c)(5)(B). However, the SAA goes on to explain that the reasonable method Commerce is “expected” to use to calculate the all-others rate is “to weight-average the zero and de minimis margins and margins determined pursuant to the facts available, provided that volume data is available.” SAA at 4201; Albemarle, 821 F.3d at 1352. There is no contradiction between the statute and the SAA. The statute requires selecting “any reasonable method,” 19 U.S.C. § 1673d(c)(5)(B), and the SAA merely prescribes what the default methodology is expected to be when “volume data is available,” SAA at 4201.

That is equally true when all mandatory respondents receive an AFA rate. Neither the statute nor the SAA distinguishes scenarios where the examined respondents all received a zero, de minimis, or AFA rate, or some combination of the three. See 19 U.S.C. § 1673d(c)(5)(B); SAA at 4201. As such, the expected method is just that—expected—even when all mandatory respondents 14 PRIMESOURCE BUILDING PRODUCTS, INC. v. US

receive an AFA rate. See Albemarle, 821 F.3d at 1352 (explaining that the SAA makes it clear that Commerce is to apply the expected method even “when all individually examined respondents are assigned de minimis margins”); see also Bestpak, 716 F.3d at 1379 (explaining that “§ 1673d(c)(5)(B) and the SAA explicitly allow Commerce to factor both de minimis and AFA rates into the calculation methodology”).

The SAA then goes on to provide additional guidance as to when Commerce may deviate from the prescribed methodology. Following the description of the expected method above, the SAA provides:

However, if this method is not feasible, or if it results in an average that would not be reasonably reflective of potential dumping margins for non- investigated exporters or producers, Commerce may use other reasonable methods.

SAA at 4201.

As we explained in Albemarle, “Commerce may use ‘other reasonable methods,’ but only if Commerce reasonably concludes that the expected method is ‘not feasible’ or ‘would not be reasonably reflective of potential dumping margins.’” 821 F.3d at 1352 (quoting SAA at 4201) (emphasis added); see also id. at 1348 n.3 (noting that “[t]he [Trade Preferences Extension Act of 2015, Pub. L. No. 114–27, 129 Stat. 362 (2015)] makes a number of changes to the antidumping duty laws, none of which is relevant to this case”). In other words, to deviate from the expected method, Commerce must affirmatively determine, based on substantial evidence, that the expected method is not feasible or would not be reasonably reflective of the potential dumping margin of the non- selected respondents. Changzhou Hawd Flooring Co. v. United States, 848 F.3d 1006, 1012 (Fed. Cir. 2017) (“Commerce could not deviate from the expected method unless it found, based on substantial evidence, that the

PRIMESOURCE BUILDING PRODUCTS, INC. v. US 15

separate-rate firms’ dumping is different from that of the mandatory respondents.”)

The burden is on Commerce to justify a departure from the expected method, not to justify its use. Id.; Albemarle, 821 F.3d at 1353. The converse of that conclusion is also true; when Commerce applies the expected method, the party that desires Commerce to deviate from the expected method bears the burden to identify and present substantial evidence on the record that either the expected method was “not feasible” or produced results not “reasonably reflective of potential dumping margins for non-investigated exporters or producers.” SAA at 4201; see Decision at 1338. The Trade Court therefore correctly held that “the expected method is the default method and that the burden of proof lies with the party seeking to depart from the expected method.” Decision at 1338.

That conclusion is further bolstered by the statute’s recognition of Commerce as an organization of finite resources. See 19 U.S.C. § 1677f–1(c)(2) (allowing Commerce to limit its investigation to a subset of the exporters or producers when the large number of exporters or producers means it is “not practicable” to determine individual dumping margins for each); 19 U.S.C. § 1677m(a) (allowing Commerce to decline to investigate a voluntary respondent when it would be “unduly burdensome” to do so). Placing an affirmative burden on Commerce to investigate the non-selected respondents and determine that the expected method produced results reasonably reflective of their dumping margin, as suggested by appellants, would contravene the purpose of those statutory provisions that allow Commerce to limit the number of parties individually investigated under certain circumstances. See Decision at 1341 (“Such an interpretation would defeat the purpose of the respondent selection process.”).

16 PRIMESOURCE BUILDING PRODUCTS, INC. v. US

PrimeSource argues that our decisions in Bestpak and Bosun Tools Co. v. United States, No. 2021-1929, 2022 WL 94172 (Fed. Cir. Jan. 10, 2022), nevertheless support placing a burden on Commerce to justify the use of the expected method. PrimeSource Br. at 23–26. Those cases, however, are distinguishable.

In Bestpak, Commerce did not employ the expected method, and instead used “a simple average rather than a weighted average.” 716 F.3d at 1378. In deviating from the expected method, Commerce was required to use “other reasonable methods.” Id. This court determined that the administrative record lacked substantial evidence supporting a conclusion that the chosen method was reasonable as applied to the facts of that case. Id. Bestpak does not address Commerce’s burden when it calculates the all-others rate using the expected method.

Bosun is a nonprecedential opinion which therefore does not control here. Regardless, it also does not support PrimeSource’s position. Bosun presents a scenario similar to the one here in that Commerce used the expected method 3 to calculate an all-others rate by averaging a zero and an AFA rate. Bosun, 2022 WL 94172, at *3. Bosun argued that the resulting rate was not reasonably reflective of its potential dumping margin. Id. at *3. Commerce reviewed the history of the rates in the proceeding and came to the opposite conclusion. Id. at *5– 6. We affirmed Commerce’s decision as supported by

3 Commerce stated that it used the expected method to calculate the all-others rate despite having used a simple average rather than a weighted average. See Bosun Tools Co. v. United States, 493 F. Supp. 3d 1351, 1354 (Ct. Int’l Trade 2021). That was not challenged, and Commerce was therefore treated as if it had used the expected method. See generally Bosun, No. 2021-1929, ECF No. 18 (Appellant’s opening brief on appeal.).

PRIMESOURCE BUILDING PRODUCTS, INC. v. US 17

substantial evidence. Id. at *6. Bosun thus provides only an example of a party challenging Commerce’s use of the expected method and failing to meet its burden to show that the results of the expected method were not reasonably reflective of its potential dumping margin.

We therefore agree with the Trade Court’s analysis and conclusion that “[n]othing in the statute, SAA, or jurisprudence suggests” that Commerce has an affirmative burden to justify its use of the expected method. Decision at 1341. The Trade Court thus correctly concluded that “the non-selected respondents bear the burden of providing evidence that the results of the expected method would not reasonably reflect the potential dumping margins of the non-selected respondents.” Id.

There is no dispute that Commerce employed the expected method. There is also no contention that the expected method is not feasible. The remaining question is therefore whether or not substantial evidence supports Commerce’s determination that the non-selected respondents failed to demonstrate that the expected method produced results not reasonably reflective of their potential dumping margins.

B

We next turn to Appellants’ arguments that the record provides substantial evidence to rebut the presumption of representativeness and therefore renders the expected method unreasonable. The mandatory respondents are presumed representative of the non-selected respondents. Changzhou Hawd Flooring, 848 F.3d at 1012 (“The very fact that the statute contemplates using data from the largest volume exporters suggests an assumption that those data can be viewed as representative of all exporters.” (quoting Albemarle, 821 F.3d at 1353)). That presumption is essential to the justification for calculating the “all-others” rate based on the weighted average of the mandatory respondents. Id. However, the presumption of 18 PRIMESOURCE BUILDING PRODUCTS, INC. v. US

representativeness is rebuttable, and a party wishing to depart from the expected method can satisfy its burden by showing that the non-examined respondents’ dumping is different from the mandatory respondents. Id. Commerce found that “the expected method is reasonable here because the record evidence does not rebut the presumption that the mandatory respondents are representative.” J.A. 620.

Appellants first point to the fact that both mandatory respondents received AFA rates rather than calculated rates in the fourth administrative review. PrimeSource argues that the presumption of representativeness has been rebutted because the mandatory respondents’ rates were based entirely on AFA rates. Specifically, it argues that the use of an AFA rate creates a weak presumption of representativeness because it is “divorced from a respondents’ [sic] contemporaneous data on the record,” PrimeSource Br. at 32, and that an AFA rate based on data from the petition cannot constitute substantial evidence to support the presumption of representativeness in the fourth administrative review, id. at 35. Cheng Ch similarly takes issue with the application of the AFA rate to the non- selected respondents, asserting that the AFA rate has no relationship to the administrative record in the fourth administrative review and that it is merely a punitive measure for non-cooperative respondents. Cheng Ch. Br. at 11–12.

Those arguments are unpersuasive. Appellants attempt to shift the burden to Commerce to show that the mandatory respondents’ AFA rate was similar to the non- selected respondents’ potential dumping margin during the fourth administrative review. But as earlier stated, it is their burden to justify a departure from the expected method. Here, they fail to point to any information from the fourth administrative review to rebut the presumption of representativeness of the mandatory respondents. Instead, they place too much weight on the fact that the

PRIMESOURCE BUILDING PRODUCTS, INC. v. US 19

mandatory respondents received AFA rates based on data from the original petition.

The mere fact that all mandatory respondents received an AFA rate cannot, in and of itself, undermine the presumption of representativeness. First, the respondent selection process under 19 U.S.C. § 1677f–1(c)(2) supports the presumption of representativeness based on exporters’ volume, not the results of Commerce’s dumping margin analysis. Albemarle, 821 F.3d at 1353; Changzhou Hawd Flooring, 848 F.3d at 1012. Additionally, as discussed above, the statute and SAA expressly require Commerce to factor in AFA rates when calculating the all-others rate using the expected method. See 19 U.S.C. § 1673d(c)(5)(B); SAA at 4201. Finally, under 19 U.S.C. § 1677e(b)(2), when determining an AFA rate, Commerce may rely on information from prior proceedings, including the original petition. In fact, Commerce must rely on that earlier information because receipt of an AFA rate means the respondent failed to provide adequate information to calculate a dumping margin in the current proceeding. See 19 U.S.C. § 1677e(b)(1). An AFA rate will therefore always be at least partially based on data that are not contemporaneous to the current proceeding. Simply pointing out the realities of the statutory framework when a respondent receives an AFA rate does nothing to undermine the presumption of representativeness of the mandatory respondents.

Furthermore, Appellants’ assertion that the mandatory respondents’ AFA rate is somehow punitive or divorced from any contemporaneous evidence is incorrect. Receiving an AFA rate is not a punitive measure, “[r]ather, it reflects a common sense inference that the highest prior margin is the most probative evidence of current margins because, if it were not so, the importer, knowing of the rule, would have produced current information showing the margin to be less.” Rhone Poulenc, Inc. v. United States, 899 F.2d 1185, 1190 (Fed. Cir. 1990). Here, the 20 PRIMESOURCE BUILDING PRODUCTS, INC. v. US

respondents were aware of the 78.17% AFA rate because that rate had been applied in previous proceedings. Decision at 1341. The mandatory respondents’ decision to not provide questionnaire responses is therefore at least circumstantial, contemporaneous evidence that their dumping margin was equal to or higher than the AFA rate during the relevant period of review. That inference is particularly strong with a respondent such as Pro-Team because it had complied in prior proceedings and previously received lower calculated rates between 0% and 6.72%, but then chose to not provide questionnaire responses in the fourth administrative review, essentially guaranteeing that it would receive the 78.17% AFA rate. See Decision at 1342; J.A. 523. We therefore reject Appellants’ arguments that the mandatory respondents’ receipt of the AFA rate provides substantial evidence to undermine the presumption of representativeness.

Appellants next assert that the all-others rate applied to the non-selected respondents was unreasonable and not supported by substantial evidence because the history of calculated rates in the proceeding on certain steel nails from Taiwan were significantly lower than the AFA rate assigned to the mandatory respondents in the fourth administrative review. PrimeSource Br. at 42; Cheng Ch Br. at 13. Specifically, PrimeSource points to the rates assigned to individually investigated respondents from the investigation and first through third periods of review (“POR”), summarized in the chart below.

PRIMESOURCE BUILDING PRODUCTS, INC. v. US 21

PrimeSource Br. at 39.

PrimeSource argues that rather than a trend of non-

cooperation, the history of rates in the proceeding demonstrates that respondents received low, calculated rates far more often than AFA rates. Id. at 38–40. And it argues that the history of calculated rates shows numerous zero, de minimis, or other low calculated rates making the application of the AFA rate to all non-selected respondents in the fourth administrative review unreasonable. Id. With respect to Liang Chyuan, PrimeSource argues that its offer to submit questionnaire responses in the fourth administrative review and its calculated rate of 2.54% in the third administrative review establish that its dumping margin was likely lower than the AFA rate, and therefore that Commerce’s decision to apply the all-others rate to Liang Chyuan was not supported by substantial evidence. PrimeSource Br. at 40–42.

Here, Commerce considered and rejected Appellants’

suggestion of a pattern of lower rates, instead finding a history of AFA rate usage in previous reviews. See J.A. 624–26. Commerce “reviewed the information proffered by [respondents], which [was] a listing of the calculated rates throughout this proceeding, including the investigation” ranging from 0% to 27.69%, which respondents characterized as low margins. Id. at 624–25. Commerce determined that it was improper to look at only the calculated rates and ignore the AFA rates of 78.17% previously assigned in the proceeding because it “assigned AFA in three out of five segments,” which was “more than half of the reviews.” Id. at 625. It found that there was a pattern of non-cooperation and receipt of AFA rates, including from the mandatory respondents in the fourth administrative review. Id. Pro-Team, “a mandatory respondent in every segment, including the investigation, has been assigned margins ranging from zero to 78.17.” Id. Additionally, it found that “Bonuts[] has a history of 22 PRIMESOURCE BUILDING PRODUCTS, INC. v. US

uncooperative behavior, having been assigned AFA in [the first and second administrative review] and this review.” Id. Relying on that evidence, Commerce reasonably determined that examining only the calculated rates would not provide a full picture of the historical rates in the proceeding.

Commerce went on to examine the calculated rates of another frequent respondent, Unicatch. Id. at 626. It determined that the 27.69% calculated rate in the third administrative review was not, in fact, “low,” as asserted by respondents. Id. Rather, it demonstrated that “the percentage increase in Unicatch’s margin from [the second to third administrative review] is 350 percent.” Id.

Commerce also noted the lack of evidence of review-

specific rates for the vast majority of non-examined companies, finding that “73 of 75 of the non-examined companies have never been examined in any segment of the proceeding,” and that “there is no evidence on this record or any other record that the 78.17 percent rate does not reflect their commercial reality.” J.A. 626. Weighing those findings together, Commerce determined that, from its “analysis of all the assigned rates, segment to segment, it is apparent that there is no pattern of ‘low’ margins in this proceeding, as claimed” by the respondents. Id. It therefore determined that “the record does not show that the assumed representativeness (as recognized in Albemarle) for mandatory respondents should not apply” and that the facts did not present a situation where the use of the expected method was unreasonable. Id.

Commerce engaged with the evidence of record and came to the reasonable conclusion that the facts of the case did not support a departure from the expected method. Commerce’s decision to apply the expected method, resulting in a 78.17% all-others rate, to the non-selected respondents was therefore supported by substantial evidence and in accordance with the law. See Albemarle,

PRIMESOURCE BUILDING PRODUCTS, INC. v. US 23

821 F.3d at 1352 (“Commerce may use ‘other reasonable methods,’ but only if Commerce reasonably concludes that the expected method is ‘not feasible’ or ‘would not be reasonably reflective of potential dumping margins.’” (quoting SAA at 4201)); Changzhou Hawd Flooring, 848 F.3d at 1012.

PrimeSource’s attempt to distinguish Liang Chyuan from the other non-selected respondents is also unpersuasive. See PrimeSource Br. at 41–46. Although it is true that Liang Chyuan offered to submit questionnaire responses, it did not do so until it was aware that both mandatory respondents would likely receive AFA rates, J.A. 528, and it never followed up to submit any questionnaire responses, J.A. 631. Liang Chyuan is thus no different from the other non-selected respondents in that the record contains no contemporaneous data regarding its potential dumping margin. Additionally, its rate moving from 2.74% in the third administrative review to 78.17% after the fourth administrative review is not unreasonable. As Commerce identified, Pro-Team, a mandatory respondent in the initial investigation and every administrative review received calculated rates as low as 0% but then received the AFA rate of 78.17% in the fourth administrative review. J.A. 625. In fact, similar to Liang Chyuan, Pro-Team’s rate moved from a single digit rate, 6.72%, in the third review to the AFA rate in the fourth administrative review. See Decision at 1342. Without more information, Liang Chyuan’s calculated margin of 2.74% in the third administrative review does not demonstrate that the 78.17% all-others rate in the fourth administrative review is unreasonable as “[t]here is no basis to simply assume that the underlying facts or calculated dumping margins remain the same from period to period.” Albemarle, 821 F.3d at 1356. Commerce therefore reasonably determined that “there is no record evidence substantiating [Liang Chyuan]’s claim that it would have received the same result in this review as it did 24 PRIMESOURCE BUILDING PRODUCTS, INC. v. US

in a previous review, had it been selected for individual examination.” J.A. 631.

Appellants additionally argue that pulling forward rates from earlier administrative reviews would have been a more reasonable method for calculating a dumping margin for the non-selected respondents. PrimeSource Br. at 46; Cheng Ch Br. at 13. We need not address that argument. When all mandatory respondents receive a rate that is zero, de minimis, or based entirely on AFA rates, Commerce’s statutory obligation is to select “any reasonable method,” not the most reasonable method. 19 U.S.C. § 1673d(c)(5)(B) (emphasis added). The SAA dictates that that method must be the expected method unless it is not feasible or not reasonable—only then may Commerce select “other reasonable methods.” See SAA at 4201. Here, Commerce’s decision not to depart from the expected method was in accordance with the law and supported by substantial evidence. There is thus no need to evaluate Appellants’ other suggested method.

II

Finally, we turn to PrimeSource’s argument that Liang Chyuan was entitled to an individual rate in the fourth administrative review. PrimeSource argues that Liang Chyuan was not required to “submit a full questionnaire response to be considered a voluntary respondent.” PrimeSource Br. at 48. It argues that because Liang Chyuan gave Commerce notice that it was willing to submit a response, Commerce should have used its authority to solicit information from Liang Chyuan. Id. at 49–50.

We disagree, as that argument is expressly foreclosed by the text of the statute.

The statute provides the requirements to be considered as a voluntary respondent. Relevant here is that the respondent “submits to the administering authority the

PRIMESOURCE BUILDING PRODUCTS, INC. v. US 25

information requested from exporters or producers selected for examination . . . by the date specified . . . for exporters and producers that were initially selected for examination.” 19 U.S.C. § 1677m(a)(1). Liang Chyuan failed to satisfy that requirement. It did not submit responses to the antidumping questionnaire provided to the mandatory respondents at all, and its letter expressing willingness to submit response did not arrive until “well after the deadlines” for the initially selected mandatory respondents. J.A. 631. Furthermore, there is no requirement for Commerce to solicit information from a potential voluntary respondent. Commerce therefore correctly determined that Liang Chyuan’s “‘letter of willingness’ to be a respondent was an unacceptable substitute for the requirements established under [§ 1677m(a)] of the Act.” Id.

Notably, Liang Chyuan’s letter did not arrive until after it was aware that all mandatory respondents were to receive an AFA rate based on their non-cooperation. See J.A. 527 (“[T]he dumping rates for both mandatory respondents in this proceeding could be calculated on the basis of total AFA, as they are non-cooperative. However, that does not mean that the dumping rates calculated for unsampled respondents such as [Liang Chyuan] should be based on total AFA.”). The statutory directive for voluntary respondents to submit the request information “by the date specified . . . for exporters and producers that were initially selected for examination” indicates that that type of wait-and-see approach is not sanctioned. 19 U.S.C. § 1677m(a)(1). Commerce’s decision to not grant Liang Chyuan an individual rate was therefore in accordance with the law.

CONCLUSION

We have considered Appellants’ remaining arguments, and do not find them persuasive. For the above reasons, we conclude that Commerce’s Final Results decision is 26 PRIMESOURCE BUILDING PRODUCTS, INC. v. US

supported by substantial evidence and in accordance with the law. Accordingly, the Trade Court’s decision sustaining Commerce’s Final Results is affirmed.

AFFIRMED

United States Court of Appeals for the Federal Circuit

______________________

PRIMESOURCE BUILDING PRODUCTS, INC., CHENG CH INTERNATIONAL CO., LTD., CHINA STAPLE ENTERPRISE CORP., DE FASTENERS INC., HOYI PLUS CO., LTD., LIANG CHYUAN INDUSTRIAL CO., LTD., TRIM INTERNATIONAL INC., UJL INDUSTRIES CO., LTD., YU CHI HARDWARE CO., LTD., ZON MON CO., LTD., Plaintiffs-Appellants

v.

UNITED STATES, MID CONTINENT STEEL & WIRE, INC.,

Defendants-Appellees

______________________

2022-2128, 2022-2129

______________________

Appeals from the United States Court of International Trade in Nos. 1:20-cv-03911-MAB, 1:20-cv-03934-MAB, Chief Judge Mark A. Barnett.

______________________

DYK, Circuit Judge, concurring in part and dissenting in part. 1

1 I agree with the majority that, based on this record, there is no requirement that Liang Chyuan be individually 2 PRIMESOURCE BUILDING PRODUCTS, INC. v. US

This appeal presents the question of whether Commerce can treat an adverse facts available (“AFA”) rate as presumptively reasonable when calculating the all others rate for non-selected respondents, particularly where there is evidence that the AFA rate is not a reasonable approximation of the all others rate. I respectfully dissent from the majority’s conclusion that an AFA rate is presumptively reasonable even though Congress in 2015 determined that an AFA rate need not be reasonable, and from its conclusion that the record fails to show that the all others rate is unreasonable.

I

Dumping margins for all importers are generally determined based on the calculation of dumping margins for the individually examined respondents (the largest importers). Those margins are generally presumed to be representative of the “all others” rates. Albemarle Corp. & Subsidiaries v. United States, 821 F.3d 1345, 1353 (Fed. Cir. 2016). However, a problem arises if the mandatory respondents refuse to cooperate, and Commerce cannot calculate a rate for the mandatory respondents and assigns them a so- called AFA rate. The Statement of Administrative Action (“SAA”) accompanying the Uruguay Round Agreements Act provides that the “expected method in such cases will be to weight-average the zero and de minimis margins and margins determined pursuant to the facts available.” H. DOC. NO. 103-316, at 873 (1994), reprinted in 1994 U.S.C.C.A.N. 4040, 4201. Here there are no de minimis rates, and Commerce used an average of the two AFA rates as the all others rate. The statute gives Commerce wide discretion in calculating an AFA rate, and there is no contention in this case that the AFA rate for the mandatory respondents was not accurate. But the AFA rate calculated

examined, and therefore join Part II of the majority opinion .

PRIMESOURCE BUILDING PRODUCTS, INC. v. US 3

for mandatory respondents can only be used as an all others rate if the rate is “reasonable” under the circumstances. 19 U.S.C. § 1673d(c)(5); see also SAA, 1994 U.S.C.C.A.N. at 4201.

Before 2015, a presumption that the overall AFA rate was reasonable was supported by the prior version of the statute because the AFA rate itself had to be commercially reasonable. See Gallant Ocean (Thailand) Co. v. United States, 602 F.3d 1319, 1323 (Fed. Cir. 2010). But, as appellants point out, Cheng Ch Br. at 12, in 2015 Congress amended the statute to eliminate any such requirement. See Trade Preferences Extension Act of 2015, Pub. L. No. 114-27, § 502, 129 Stat. 362, 384. The purpose of the 2015 amendments was to create additional incentives to cooperate with Commerce in investigations. The 2015 amendments removed the requirement of commercial reasonableness for AFA rates in order to address situations in which “a foreign party fails to cooperate with the agency’s request for information in a proceeding.” S. REP. NO. 114-45, at 37 (2015). Under the 2015 amendment, when AFA rates are applied to respondents that “failed to cooperate,” Commerce no longer has an obligation to demonstrate that the rate “reflects an alleged commercial reality of the interested party.” 19 U.S.C. §§ 1677e(b)(1), 1677e(d)(3)(B).

In such cases after 2015, where, as here, an AFA rate is applied to the non-examined parties, there is no basis for assuming that the AFA rate is reasonable as to the non- examined parties. The 2015 amendments did not change Commerce’s obligation to use a “reasonable method” to set the all others rate, which applies to cooperative non-selected parties. See 19 U.S.C. § 1673d(c)(5). Rather, the reasonable method requirement was left in place. Thus, while it may have been safe to assume that an AFA rate was presumptively reasonable as applied to the non-examined parties before the 2015 amendments, the 2015 4 PRIMESOURCE BUILDING PRODUCTS, INC. v. US

amendments explicitly removed any reasonableness requirement for calculating an AFA rate.

In this case, appellants contend that the burden is on Commerce to establish reasonableness. The majority rejects that contention on the ground that the “expected method” provided in the statute is to use AFA rates if there are no other rates available. But an “expected” method is not necessarily a reasonable method. “Expected” and “reasonable ” are different words carrying different requirements , and the mere fact that the method resulting in an AFA all others rate is “expected” does not render it “reasonable .” In particular, the problem with the majority’s reasoning is that, after the 2015 amendments, there is no basis for assuming that the expected method rate is a reasonable rate, nor any legislative history suggesting that it is presumptively reasonable for the non-examined parties. Because an AFA rate no longer must reflect commercial reality , there can be no presumption that an AFA rate resulting from the so-called “expected method” is necessarily “reasonable” for the cooperating respondents.

II

As appellants argue, there was no finding that the AFA rate of 78.17% was reasonable as applied to the non-examined parties (apart from the fact that it was the rate determined for the largest importers), and no evidence to support a finding that here the AFA rate was reasonable as applied to the non-examined parties. The AFA rate was the rate applied in each of the previous reviews to non-cooperative individually examined parties, not the rate that was applied to non-examined parties. The non-examined parties never received the 78.17% rate. In other words, the 78.17% AFA rate was itself not considered to be reasonably reflective of the all others parties’ dumping margins in the prior periods. Rather, these parties received at the highest a 35.30% rate reflecting the average of the AFA rate and a calculated zero percent rate. First Review of Certain Steel

PRIMESOURCE BUILDING PRODUCTS, INC. v. US 5

Nails from Taiwan, 87 Fed. Reg. 45,758, 45,759 (July 29, 2022). There is also no subsequent data to suggest that the 78.17% rate is reflective of the actual dumping margins of the all others parties. While Commerce found it pertinent that the dumping margins had increased since the first review , none of the calculated rates in prior proceedings remotely approach the 78.17% rate applied to the non- examined parties.

There is other significant evidence that the 78.17% rate was not reasonable as applied to the non-examined parties. In the initial investigation, Commerce calculated a margin of 2.16% for the only mandatory respondent found to be dumping, which “resulted in a revised rate of 2.16 percent for all other producers and exporters.” Investigation of Certain Steel Nails from Taiwan, 82 Fed. Reg. 55,090 (Nov. 20, 2017). In the first review period, which applied the AFA rate to a non-cooperative mandatory respondent for the first time, Commerce calculated a margin of zero for one of the largest importers and a “rate of 35.30 percent for the non-examined companies.” 87 Fed. Reg. at 45,759. Similarly, in the second and third reviews Commerce calculated margins of zero percent, 2.54%, 6.16%, 6.72%, and 27.69%, and assigned a rate of 12.90% to non-examined companies. See Second Review of Certain Steel Nails from Taiwan, 84 Fed. Reg. 11,506, 11,507 (Mar. 27, 2019); Third Review of Certain Steel Nails from Taiwan, 85 Fed. Reg. 14,635, 14,636 (Mar. 13, 2020). Here, there is simply no basis for assuming that the AFA rate is reasonable for the non-examined parties.

As we said in Albemarle, outside of the AFA context “accuracy and fairness must be Commerce’s primary objectives .” 821 F.3d at 1354; see also Yangzhou Bestpak Gifts & Crafts Co. v. United States, 716 F.3d 1370, 1379 (Fed. Cir. 2013) (“An overriding purpose of Commerce’s administration of antidumping laws is to calculate dumping margins as accurately as possible.”). Commerce’s approach here is neither accurate nor fair.

6 PRIMESOURCE BUILDING PRODUCTS, INC. v. US

I would vacate and remand with instructions for Commerce to reconsider the all others rate.