TENARIS BAY CITY INC IPSCO v. UNITED STATES PTC LLC AFL CIO CLC (2026)

In 2021, a coalition of domestic producers filed a petition urging the United States Department of Commerce (“Commerce”) to investigate whether steel pipes imported from Argentina were being “dumped” in the United States, i.e., sold at less than fair value. The ensuing investigation vindicated the petition, and Commerce imposed antidumping duties on the Argentine imports. This case isn't about that ultimate determination, but instead concerns a threshold question: whether the petition that kicked off the investigation was filed “by or on behalf of the industry.” 19 U.S.C. § 1671a(c)(1)(A)(ii). Because substantial evidence supports Commerce's determination that it was, we affirm.

Background

Oil country tubular goods (“OCTG”) are steel pipes used in oil and gas wells. On October 6, 2021, a group of four domestic OCTG producers and a union filed a petition alleging that OCTG were being imported from various countries, including Argentina, and sold in the United States for less than fair value. Their petition was brought under the antidumping provisions of the Tariff Act of 1930, 19 U.S.C. §§ 1673–1673h, which establish a process for Commerce to investigate allegations that foreign goods are being dumped in the United States.

After receiving such a petition, Commerce must determine whether “the petition has been filed by or on behalf of the industry.” Id. § 1673a(c)(1)(A)(ii). A petition is filed “by or on behalf of the industry” if the “domestic producers or workers who support the petition account for” (1) “at least 25 percent of the total production of the domestic like product” and (2) “more than 50 percent of the production of [those] expressing support for or opposition to the petition.” Id. § 1673a(c)(4)(A). In both calculations, the production attributable to supporters forms the numerator. What changes is the denominator—total domestic production for the 25% test, and the production of those expressing a position on the petition for the 50% test. Commerce must make the industry support determination quickly, generally within 20 days. Id. § 1673a(c)(1)(A).

Petitioners provided calculations indicating that their petition satisfied both the 25% and 50% thresholds. The numbers they used were from 2020, the most recent calendar year. For the numerator, petitioners provided their own production data, as well as data from several other domestic producers who wrote letters in support of the petition. Then, because information on total domestic OCTG production was not yet available for 2020, petitioners estimated industry-wide production based on domestic shipment numbers, adjusted using a historical ratio of production to shipments. They used that estimate of industry-wide production as the denominator for the 25% calculation, and for the 50% calculation, they used that number as a starting point, and then subtracted the production of two companies who stated that they took no position on the petition.

Commerce got right to work. The day after receiving the petition, the agency issued a questionnaire to petitioners asking for more detail about how they determined the universe of domestic OCTG producers. After petitioners provided the requested information, Commerce followed up with another questionnaire on October 19, asking about the production facilities served by the petitioner labor union. Petitioners provided this information as well.

Not everyone supported the petition, though. Notably, Tenaris Bay City, Inc., which claims to be the largest domestic producer of OCTG, opposed the petition. It turns out that even though Tenaris is a major domestic producer of OCTG, it also imports OCTG and is related to foreign producers who would be harmed by tariffs. The statute provides that interested parties “may submit comments or information on the issue of industry support” before Commerce determines whether to initiate an investigation. 19 U.S.C. § 1673a(c)(4)(E). So Tenaris availed itself of that opportunity and filed comments on October 8, 15, 20, and 22 challenging petitioners’ industry support calculations.

Tenaris made several different arguments, including that 2020 was an anomalous year and that shipment data was an unreliable proxy to determine domestic production. Tenaris also observed that “[t]he relationship of pipe formation and pipe finishing has implications for any assessment of a domestic OCTG industry.” J.A. 1712.

Tenaris did not, however, point to any other source of industry-wide production data that would be more accurate than petitioners’ source. Tenaris requested that Commerce either dismiss the petition or exercise its statutory authority to poll the industry to determine whether there was sufficient support. See 19 U.S.C. § 1673a(c)(4)(D)(i).

Commerce declined. Relying on the information provided by petitioners and its statutory authority, Commerce determined that even if all non-petitioning and non-neutral domestic OCTG producers opposed the petition, the statutory thresholds for industry support were met. Commerce rejected Tenaris's arguments to the contrary, saying that its concerns were unsubstantiated and it had not offered any alternative production data. On October 26, 2021, just before the 20-day deadline expired, Commerce initiated the requested antidumping investigation. See Oil Country Tubular Goods from Argentina, Mexico, and the Russian Federation: Initiation of Less-Than-Fair-Value Investigations, 86 Fed. Reg. 60,205 (Nov. 1, 2021) (initiation notice).

That investigation ultimately resulted in a finding of sales made at less-than-fair value and the imposition of antidumping duties on OCTG imports from Argentina. See Oil Country Tubular Goods from Argentina, Mexico, and the Russian Federation: Antidumping Duty Orders and Amended Final Affirmative Antidumping Duty Determination for the Russian Federation, 87 Fed. Reg. 70,785 (Nov. 21, 2022).

On December 16, 2022, Tenaris filed an appeal in the United States Court of International Trade (“CIT”), arguing that Commerce never should have initiated the investigation in the first place because the petition was not filed on behalf of the domestic industry. The CIT rejected most of Tenaris's arguments, but it agreed with Tenaris that there was a potential issue regarding “double counting.” See Tenaris Bay City, Inc. v. United States, 693 F. Supp. 3d 1314, 1326–28 (Ct. Int'l Trade 2024) (“Tenaris I”). As Tenaris pointed out, there was evidence in the record that “certain domestic companies both produce and finish OCTG.” Id. at 1326. It was thus possible that there was “pipe that was counted for the purposes of industry support when it was produced and again when it was finished.” Id. at 1327.

The CIT noted that “Commerce may have reasons to reject this inference,” but “it must acknowledge consideration of such evidence and explain why it nonetheless rejects the inference.” Id. So the CIT “remand[ed] determination on the double counting issue to Commerce for further explanation or reconsideration.” Id. at 1328.

On remand, Commerce maintained its position that the petition had sufficient industry support and the investigation was properly initiated. Commerce found that there was “no evidence to suggest that, to the extent the companies also engage in processing,² any of the U.S. producers who provided actual production data on the record literally counted each ton of pipe they produced twice—once upon the pipe formation and again upon heat treatment, threading, etc.” J.A. 33.

The agency also specifically addressed the two companies Tenaris had flagged as potential cause for concern, Borusan U.S. and PTC Liberty. Commerce concluded that the record did not show that PTC Liberty was solely a pipe finisher (which might imply that it was counting pipes already counted by another domestic producer that manufactured them). Rather, PTC Liberty was “first and foremost” a producer of OCTG, which also had finishing capabilities. J.A. 35–36. And Borusan's processing facility finished pipes imported from overseas rather than ones manufactured by another U.S. producer. Commerce therefore determined that “concerns about ‘double counting’ ” were “misplaced.” J.A. 32.

On December 2, 2024, the CIT sustained Commerce's remand results, finding that Commerce's industry support calculations were reasonable and supported by substantial evidence. See Tenaris Bay City, Inc. v. United States, 745 F. Supp. 3d 1336, 1345 (Ct. Int'l Trade 2024) (“Tenaris II”). The CIT also rejected two new arguments Tenaris advanced: that Commerce may have (1) undercounted total production in the denominator by failing to include processors, and (2) overcounted production in the numerator by including threading, a form of processing that does not count as OCTG production.³ The CIT concluded that Tenaris had failed to raise these arguments during the 20-day pre-initiation period, so they were unexhausted. Id. at 1342–43. While Tenaris had made vague reference to the “implications” of including both producers and processors in the industry support calculations, it could not use that as a hook “to fashion more specific arguments about potential undercounting or distinctions between processing that involves heat treatment as opposed to threading operations.” Id.

The CIT also rejected Tenaris's contention that Commerce had not complied with its remand order. The court had not asked Commerce to “assess the accuracy and completeness of the data more generally.” Id. at 1344. It had merely “ordered Commerce to ‘reconsider or further explain’ its determination that the record accurately reflected industry support, including whether finishing operations were counted twice”—and Commerce had done so. Id. at 1345 (quoting Tenaris I, 693 F. Supp. 3d at 1320).

Tenaris now appeals to this court, asking us to overturn Commerce's industry support calculation. See 28 U.S.C. § 1295(a)(5) (granting this court exclusive jurisdiction over appeals from final decisions of the CIT).

I

We review the CIT's decision de novo, “stepping into its shoes and applying the same standard of review.” JTEKT Corp. v. United States, 642 F.3d 1378, 1381 (Fed. Cir. 2011). Under that standard, our review of Commerce's industry support finding is deferential. We only overturn the agency's determination if it is “unsupported by substantial evidence on the record, or otherwise not in accordance with law.” 19 U.S.C. § 1516a(b)(1)(B)(i). Substantial evidence means “more than a mere scintilla” and “such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.” Suramerica de Aleaciones Laminadas, C.A. v. United States, 44 F.3d 978, 985 (Fed. Cir. 1994) (quoting Consolidated Edison Co. v. NLRB, 305 U.S. 197, 229, 59 S.Ct. 206, 83 L.Ed. 126 (1938)). Even if we “would have reached a different conclusion based on the same record,” Cleo Inc. v. United States, 501 F.3d 1291, 1296 (Fed. Cir. 2007), we must uphold Commerce's determination if it is “reasonable and supported by the record as a whole,” Altx, Inc. v. United States, 370 F.3d 1108, 1121 (Fed. Cir. 2004) (quotation omitted).

Applying that deferential standard of review, we affirm. As Commerce explained, Tenaris identified places where double counting was possible, but it did not identify any actual evidence of double counting. And Commerce was under no obligation to hunt for evidence dispelling Tenaris's theoretical concerns. As for Tenaris's arguments about potential undercounting and overcounting, we agree with the CIT that those arguments are unexhausted and decline to consider them.

A

We start with double counting. Everyone agrees that the domestic OCTG industry properly includes both manufacturing and finishing operations. So if a pipe is manufactured by a domestic producer, it should be counted. And if a pipe is finished by a domestic producer, it should be counted too. But a single pipe should not be counted twice.

There are two ways that could happen: (1) if a pipe was manufactured by one domestic producer but finished by another domestic producer (and counted by both), or (2) if a single domestic producer manufactured and then finished a pipe and counted it twice. Notably, double counting only helps Tenaris if it happens in the numerator, i.e., if it attributes too much production to the petition's supporters. That would mean that Commerce's calculation overstated support for the petition. Whereas if the double counting were in the denominator, that would mean domestic support was understated. And if double counting turned out to be equally a problem in the numerator and denominator, any double counting would cancel out.⁴

The problem for Tenaris is that there is no evidence in the record indicating that either form of double counting actually happened—not in the numerator, not in the denominator, not anywhere. The evidence prompting the CIT's initial remand was website information from two petitioners—PTC Liberty and Borusan—indicating that both had processing capabilities. See Tenaris I, 693 F. Supp. 3d at 1326. But as Commerce explained on remand, nothing in the record suggested that either company double counted its production. If anything, Commerce's further examination of those companies indicated that double counting was unlikely, because PTC Liberty was primarily a producer of OCTG with some finishing capabilities and Borusan's processing facility finished pipes imported from overseas rather than ones manufactured by another U.S. producer.

Tenaris contends that Commerce “erroneously considered the double-counting issue to be limited to two petitioning companies.” Opening Br. 16. But that misrepresents Commerce's analysis. The agency started by explaining that it found “no evidence” that “any of the U.S. producers who provided actual production data on the record literally counted each ton of pipe they produced twice.” J.A. 33 (emphasis added). So there was “no record evidence” of any double counting in the numerator. J.A. 34. It was only after Commerce conducted its general survey of the record that it went on to discuss the two specific companies whose websites had prompted the remand. Now on appeal, Tenaris identifies no evidence in the record that Commerce missed that would indicate double counting.

Because Tenaris has no record evidence of double counting, much of its argument boils down to its contention that the onus was on Commerce to “ensure” there was “no” double counting, after Tenaris raised it as a possibility. Opening Br. 44. Tenaris bases this argument on the statutory provision requiring Commerce to “examin[e] ․ the accuracy and adequacy of the evidence provided in the petition.” 19 U.S.C. § 1673a(c)(1)(A)(i).

But Tenaris misreads the statute. The full text of 19 U.S.C. § 1673a(c)(1)(A), with the portion Tenaris relies on in bold, is:

Except as provided in subparagraph (B), within 20 days after the date on which a petition is filed under subsection (b), the administering authority shall—

(i) after examining, on the basis of sources readily available to the administering authority, the accuracy and adequacy of the evidence provided in the petition, determine whether the petition alleges the elements necessary for the imposition of a duty under section 1673 of this title and contains information reasonably available to the petitioner supporting the allegations, and

(ii) determine if the petition has been filed by or on behalf of the industry.

Id. (emphases added).

So the part Tenaris relies on is in subdivision (i), which relates to the other determination Commerce must make during the 20-day pre-initiation period—not subdivision (ii), which provides for the industry support determination. If Congress had wanted the “accuracy and adequacy” provision to apply to both determinations, it could have easily accomplished that by putting it before the subdivisions, or by including it in both. Instead, Congress included that requirement in subdivision (i) but not subdivision (ii), a clear sign that it was meant to apply only to the first determination. See Chafin v. Off. of Pers. Mgmt., 177 F.4th 1183, 1186 (Fed. Cir. 2026) (“It is a fundamental canon of statutory construction that ‘where Congress includes particular language in one section of a statute but omits it in another section of the same Act, it is generally presumed that Congress acts intentionally and purposely in the disparate inclusion or exclusion.’ ” (quoting Russello v. United States, 464 U.S. 16, 23, 104 S.Ct. 296, 78 L.Ed.2d 17 (1983))).

This is not to say Commerce may accept industry support evidence at face value. Commerce's determination still must be supported by substantial evidence, meaning the agency must “examine the record and articulate a satisfactory explanation for its action.” Yangzhou Bestpak Gifts & Crafts Co. v. United States, 716 F.3d 1370, 1378 (Fed. Cir. 2013) (citation omitted). Commerce must also take account of evidence that “fairly detracts” from its conclusion. CS Wind Vietnam Co. v. United States, 832 F.3d 1367, 1373 (Fed. Cir. 2016) (quoting Gerald Metals, Inc. v. United States, 132 F.3d 716, 720 (Fed. Cir. 1997)). That is precisely why the CIT remanded when Commerce initially failed to address record evidence suggesting that some OCTG may have been double counted. Tenaris I, 693 F. Supp. 3d at 1326–28.

But we agree with the CIT that Commerce satisfied its obligations here. During the initial 20-day period, the agency sent petitioners multiple questionnaires, probing various aspects of petitioners’ industry support calculation and underlying data. This back and forth prompted petitioners to revise their calculations in certain respects, and Commerce ended up performing its own calculations that differed further from the calculations submitted by petitioners. Then on remand, Commerce specifically reconsidered the record evidence bearing on double counting and explained why it did not support Tenaris's inference.

What Commerce did not do was go out and affirmatively find evidence to supplement the record and disprove double counting (or the other hypothetical issues Tenaris raised). But “the burden of creating an adequate record lies with interested parties and not with Commerce.” QVD Food Co. v. United States, 658 F.3d 1318, 1324 (Fed. Cir. 2011) (citation modified). Tenaris itself had every opportunity to submit evidence corroborating its concerns or proposing alternative sources of data, but it did not do so. See Qingdao Sea-Line Trading Co. v. United States, 766 F.3d 1378, 1386 (Fed. Cir. 2014) (“Once Commerce selected [an] index as the best available adjustment method, [the objecting party] had the duty to submit verifiable evidence showing that the use of the index was not the best available method ․”). As the CIT explained in a prior antidumping case, Commerce “has no independent duty to make a diligent inquiry into any and all information highlighted by interested parties.” Coal. of Am. Flange Producers v. United States, 448 F. Supp. 3d 1340, 1357 (Ct. Int'l Trade 2020).

These principles have particular force in the pre-initiation context, where the statute requires Commerce to resolve industry support quickly and decisively. See 19 U.S.C. § 1673a(c)(1) (setting a presumptive 20-day deadline that can be extended to “a maximum of 40 days” “in exceptional circumstances”); id. § 1673a(c)(4)(E) (“After the administering authority makes a determination with respect to initiating an investigation, the determination regarding industry support shall not be reconsidered.”). Commerce has enough on its plate during that 20-day period without also having to seek out evidence to rebut unsubstantiated concerns like the double counting Tenaris theorizes here.

In a different case, the concerns raised by Tenaris might have required Commerce to do more—follow up with petitioners about their data, revise the calculation, etc. But here, the record did not bear those concerns out. After reexamining the evidence, Commerce found no indication that double counting had occurred. It was not required to reinvent the wheel to disprove a possibility that the record did not substantiate.

B

What about Tenaris's other concerns, regarding undercounting and overcounting?

The CIT found that those arguments were unexhausted, reasoning that Tenaris could not “rely on the word ‘implications’ to fashion more specific arguments about potential undercounting or [overcounting].” Tenaris II, 745 F. Supp. 3d at 1342–43. Tenaris had raised general objections to “the completeness of the Industry Source data,” but it had not raised these specific concerns within the 20-day comment period, so the arguments were “not exhausted, and therefore not reviewable by this Court.” Id.

As Tenaris points out, exhaustion of administrative remedies is not a jurisdictional requirement in this context. The CIT's statutory exhaustion provision provides that the court “shall, where appropriate, require the exhaustion of administrative remedies.” 28 U.S.C. § 2637(d). So the “application of exhaustion principles in trade cases is subject to the discretion of the judge of the [CIT].” Agro Dutch Indus. Ltd. v. United States, 508 F.3d 1024, 1029 (Fed. Cir. 2007) (internal quotation omitted). But “section 2637(d) ‘indicates a congressional intent that, absent a strong contrary reason, the [trade] court should insist that parties exhaust their remedies before the pertinent administrative agencies.’ ” Itochu Bldg. Prods. v. United States, 733 F.3d 1140, 1145 (Fed. Cir. 2013) (alteration in original) (citing Corus Staal BV v. United States, 502 F.3d 1370, 1379 (Fed. Cir. 2007)). And because it is a discretionary decision, we “review the CIT's failure to exhaust determination for an abuse of discretion.” Apex Frozen Foods Priv. Ltd. v. United States, 862 F.3d 1322, 1332 (Fed. Cir. 2017).

We find no abuse of discretion in the CIT's finding that Tenaris failed to exhaust its administrative remedies here. To start, the application of exhaustion principles makes good sense in the pre-initiation context. As noted above, the statute generally gives Commerce only 20 days to make its industry support determination, 19 U.S.C. § 1673a(c)(1)(A), and after it makes its decision, “the determination regarding industry support shall not be reconsidered,” id. § 1673a(c)(4)(E). A vague reference to the “implications” of including both producers and processors in the industry support calculation does not give Commerce enough notice to fairly expect it to address more specific objections like the ones Tenaris raises now, especially on such a compressed timetable. See Tenaris II, 745 F. Supp. 3d at 1342–43; Itochu Bldg. Prods., 733 F.3d at 1145 (explaining that exhaustion gives the agency a “full opportunity to correct errors and thereby narrow or even eliminate disputes needing judicial resolution”).

Take Tenaris's undercounting argument. The theory is that because the data Commerce used to calculate total industry production in the denominator was not “disaggregated” into manufacturing versus finishing, it might not have properly accounted for both. Opening Br. 27–28. Tenaris speculates, for example, that the denominator might not account for U.S. processor shipments, meaning that the industry support calculation overstates support for the petition. Tenaris similarly speculates that the ratio Commerce used to adjust the denominator might not account for finishing operations.

None of these arguments can be found anywhere in Tenaris's initial submissions. While Tenaris did make general objections to the reliability of the sources Commerce used to calculate the denominator,⁵ it did not argue that the source was flawed because it failed to include processor shipments. In fact, at the time, Tenaris's concern ran in the opposite direction: it argued that processing activity might be improperly included in the calculation and asked Commerce to exclude OCTG that petitioners “merely finish[ed] rather than produce[d].” J.A. 1450–51.

The same goes for Tenaris's overcounting argument, which focuses on the numerator. Tenaris points out that certain processing operations—such as threading alone—do not qualify as “production” of OCTG. If supporters of the petition counted those operations as production, that would inflate the numerator and overstate support for the petition. This argument, though, has even less basis in Tenaris's original submissions. During the pre-initiation period, Tenaris never argued that threading, as opposed to heat treatment or other processing, had been improperly counted as production. That distinction appeared only later on remand.

For these reasons, the CIT did not abuse its discretion in determining that Tenaris failed to exhaust these more specific arguments and we decline to consider them.

II

Tenaris also argues that Commerce failed to comply with the CIT's remand order by only addressing double counting. In its view, because the remand order directed Commerce to further explain or reconsider its “determination that the data relied upon accurately reflected industry support, including whether finishing operations were counted twice,” Tenaris I, 693 F. Supp. 3d at 1328 (emphasis added), double counting was just one issue Commerce should have reconsidered on remand, not the only issue.

Tenaris has an uphill battle here, because a court's interpretation of its own order “is entitled to deference unless the interpretation is unreasonable or is otherwise an abuse of discretion.” Amado v. Microsoft Corp., 517 F.3d 1353, 1358 (Fed. Cir. 2008) (citation omitted). And in Tenaris II, the CIT rejected Tenaris's interpretation of its prior order, finding that Commerce had complied with the remand. 745 F. Supp. 3d at 1345.

Tenaris's argument also requires ignoring the rest of the remand order. Besides double counting, the CIT's opinion rejects Tenaris's arguments. The court said that remand was necessary “[b]ecause Commerce did not adequately address Plaintiffs’ concerns and record evidence that finishing operations were not counted twice.” Tenaris I, 693 F. Supp. 3d at 1326. It then identified the particular information that troubled it—evidence that “certain domestic companies both produce and finish OCTG,” creating an inference that “some domestic pipe may have been double counted”—and directed Commerce to address that inference. Id. at 1326–27. And immediately before its conclusion, the CIT again stated that Commerce had failed to address “the possibility of double counting” and “[a]ccordingly” remanded “determination on the double counting issue to Commerce for further explanation or reconsideration.” Id. at 1328.

Because the CIT's interpretation of its order is reasonable and not an abuse of discretion, we reject Tenaris's argument that Commerce failed to comply with the remand order.

* * *

For these reasons, the CIT's decision in Tenaris II is

AFFIRMED

FOOTNOTES

2.   In this context, “processing” and “finishing” are used interchangeably.

3.   Technically, the 25% and 50% calculations use different denominators. But both are based on petitioners’ estimate of industry-wide production, so Tenaris's arguments—which are focused on that estimate—apply equally to both.

4.   Commerce itself made this point, albeit somewhat clumsily. It reasoned that processing counted in both the numerator and denominator would “effectively cancel[ ],” and illustrated the point by removing the same additive term from both sides of the fraction. J.A. 34 n.57. That algebra is not literally correct: adding the same amount to the numerator and denominator does not necessarily leave a ratio unchanged. But Commerce's underlying point was sound. If double counting inflated the numerator and denominator proportionally (e.g., increasing both by 50%), the industry support percentage would be unaffected.

5.   To the extent Tenaris reraises its general objection to the reliability of the sources Commerce used to calculate the denominator, we disagree. Petitioners submitted information from a well-known industry report that “maintains the largest market intelligence database on steel pipe and tube in the world and is the premier resource for pipe and tube statistics.” J.A. 1093. Commerce also reasonably determined that shipment data is an acceptable proxy for production in this case. Petitioners supported that approach with their own production-and-shipment data, and Commerce adjusted the shipment figures using historical production-to-shipment ratios drawn from ITC data. These aspects of Commerce's determination are supported by substantial evidence.

Subramanian, District Judge.