Authored by: Jeremy J. Gustrowsky
Before the Commerce Department can launch an antidumping investigation, it has to confirm that the petition asking for one was actually filed “by or on behalf of the industry.” Under 19 U.S.C. § 1673a(c)(4)(A), that means the domestic producers and workers backing the petition must account for at least 25 percent of total domestic production of the like product, and more than 50 percent of the production of those who took a position either way. Commerce generally has just 20 days to sort this out. This appeal did not challenge the duties themselves. It challenged whether Commerce ever should have opened the door.
The case revolves around oil country tubular goods, or OCTG, the steel pipes used in oil and gas wells. In October 2021, four domestic producers and a labor union petitioned Commerce to investigate OCTG imports from Argentina, Mexico, and Russia. Because industry-wide production figures for 2020 were not yet available, the petitioners estimated total production using domestic shipment data adjusted by a historical production-to-shipment ratio. Tenaris Bay City, which describes itself as the largest domestic OCTG producer but also imports OCTG and is affiliated with foreign producers, opposed the petition and filed four sets of comments attacking the support calculations. Commerce sent the petitioners two questionnaires, tightened the numbers, and initiated the investigation just before the 20-day clock ran out. Antidumping duty orders followed in November 2022.
At the Court of International Trade, Tenaris got a partial win. The trade court noticed record evidence that some domestic companies both manufacture and finish OCTG, raising the possibility that the same length of pipe was counted once when formed and again when finished. It sent the issue back so Commerce could either explain or reconsider. On remand, Commerce stuck to its conclusion. It found nothing in the record showing that any producer literally counted the same tonnage twice, and it looked specifically at the two companies whose websites had triggered the remand. One turned out to be primarily a pipe producer that also happens to have finishing capability, and the other finished pipe imported from overseas rather than pipe made by another U.S. company. The trade court sustained that explanation.
The Federal Circuit affirmed, applying the deferential substantial evidence standard. Tenaris identified places where double counting was theoretically possible, the court said, but never pointed to evidence that it actually happened, in the numerator, the denominator, or anywhere else. The court also rejected Tenaris’s argument that Commerce had an affirmative duty to go find evidence ruling the problem out. Tenaris leaned on the statutory instruction that Commerce examine “the accuracy and adequacy of the evidence provided in the petition,” but the panel pointed out that this language sits in subdivision (i) of 19 U.S.C. § 1673a(c)(1)(A), which governs whether the petition adequately alleges dumping. The industry support determination lives in subdivision (ii), which contains no such language. Invoking the familiar rule that Congress is presumed to act deliberately when it includes words in one provision and omits them from a neighboring one, the court declined to import the requirement. Commerce still must examine the record and explain itself, and must account for evidence cutting against its conclusion, but the burden of building a record belongs to the interested parties. That principle carries extra weight on a 20-day timetable where, by statute, the industry support decision cannot be revisited once made.
Tenaris’s remaining theories fared no better. On appeal it argued that Commerce may have undercounted total production by leaving processors out of the denominator, and overcounted support by treating threading operations as production in the numerator. The Federal Circuit agreed with the trade court that these were new arguments. During the pre-initiation window, Tenaris had said only that the relationship between pipe forming and pipe finishing had “implications” for measuring the industry, and it had actually complained that too much finishing activity was being counted, the opposite of its undercounting theory. A passing reference like that does not give Commerce fair notice of specific objections, and since exhaustion in trade cases is a discretionary call, the court found no abuse of discretion in refusing to hear them. The panel likewise deferred to the trade court’s reading of its own remand order, which had been aimed squarely at double counting rather than a general audit of the data.
One small footnote is worth noting for anyone who reads agency math closely. Commerce had suggested that processing counted in both the numerator and denominator would simply “cancel out,” illustrating the point by striking the same term from top and bottom of a fraction. The panel observed that the algebra is not literally correct, since adding an equal amount to a numerator and denominator changes the ratio. But it credited the underlying idea: if double counting inflated both figures proportionally, the support percentage would not move. The larger takeaway for practitioners is a practical one. If you want to defeat a petition at the initiation stage, raise your specific objections in writing during the short comment period and back them with alternative data. Pointing out that something could have gone wrong will not carry the day.