Commerce Properly Set All-Others Rate by Averaging AFA Rates, CIT Says

The judge said importer CME Acquisitions failed to show that the agency's use of the expected method here wasn't "feasible" or resulted in a rate that wouldn't be "reasonably reflective" of potential AD margins for non-investigated exporters. Katzmann added that CME "had ample notice and opportunity to provide evidence to the contrary."

In the 2021-22 administrative review of the AD order on stainless steel sheet and strip in coils from Taiwan, the two mandatory respondents failed to complete the mandatory respondent questionnaire in time, leading to identical 21.1% AFA rates. In the preliminary results, Commerce assigned the non-selected companies a 4.3% rate, though the agency ultimately assigned the non-selected parties an average of the AFA rates, leading to a 21.1% rate.

Katzmann opened his discussion of the agency's actions by noting that the U.S. Court of Appeals for the Federal Circuit recently upheld Commerce's calculation of an all-others rate by averaging two identical AFA rates, in PrimeSource Building Products v. U.S. (see 2408070020). In PrimeSource, the Federal Circuit upheld Commerce's decision to weight average the two mandatory respondents' AFA rates to set the all-others rate, finding that the agency doesn't have an "affirmative burden to justify the use of the expected method and that the non-selected companies have the burden of showing that Commerce should buck the expected method.

The PrimeSource decision also said the "mandatory respondents are presumed representative of the non-selected respondent" and that this presumption is "rebuttable by the party wishing to depart from the expected method." Katzmann said the facts here largely mirror those found in PrimeSource, noting the only dispute is on whether Commerce used the expected method here, since it said it used a simple average rather than a weighted average.

The court first said Commerce didn't deviate from the expected method here, since the results of a simple average and a weighted average are the same. “Where the rate resulting from a simple average is identical to the rate that would result from a weighted average, that rate is also presumptively reasonable," the judge said. "It serves no purpose to impose an affirmative burden on Commerce to justify its use of one method versus another where the result is the same."

CME argued that the Federal Circuit's decision in Yangzhou Bestpak Gifts & Crafts v. U.S. changes the rate's reasonableness. In Bestpak, Commerce departed from the expected method and thus had an "affirmative obligation" to show that the expected method wasn't feasible or wouldn't reasonably reflect the non-selected parties' dumping margin. Katzmann said Bestpak doesn't change the result here, since it doesn't "change a rate’s presumptive reasonableness where it is calculated using the expected method."

And while CME argued that Commerce strayed from its past practice of pulling forward rates from past reviews where the mandatory respondents get AFA rates, Katzmann held that "Commerce does not have an affirmative burden to justify its use of the expected method." Since pulling rates from past reviews is a "departure from the expected method," CME had the burden of justifying this departure.

Katzmann then turned to the question of whether CME actually justified a departure from the expected method, finding that it didn't. First, the importer said the expected method couldn't be used, since "volume data was not available." The judge said volume data is only needed "to calculate the weighting factor," which is "irrelevant when calculating a weighted average of two identical values."

CME next said Commerce's respondent selection process undermined the presumption that the mandatory respondents are representative of the non-selected companies, noting that Commerce only showed data for three companies in its respondent selection memo. Katzmann held that while the memo only included three companies, "Commerce reviewed data covering all 61 companies" in its respondent selection process.

The importer also argued that there wasn't substantial evidence that the AFA rate has a reasonable relationship to the economic reality of the non-selected companies. Katzmann said the mere fact that all mandatory respondents got AFA rates can't, "in and of itself, undermine the presumption of representativeness." CME didn't show the non-examined respondents' dumping "was different from the mandatory respondents'," and thus can't rebut the presumption of representativeness, the court said.

Katzmann, finally, held that CME had sufficient notice and opportunity to submit evidence regarding the 21.1% rate. The importer said it had no reason to believe Commerce would switch off the 4.3% rate it used in its preliminary results. The judge rejected this claim, noting that the "expected method is just that: expected."

In addition, the "domestic interested parties demonstrated actual knowledge that Commerce’s calculation of the all-others rate was at issue here when they submitted pre-preliminary comments arguing that Commerce should calculate the all-others rate by 'averaging the estimated weighted-average dumping margins determined for the companies individually investigated,'" the court noted. CME, like the domestic parties, "had reason to know that Commerce could apply the expected method," Katzmann said.

(CME Acquisitions v. United States, Slip Op. 25-91, CIT # 24-00032, dated 07/18/25; Judge: Gary Katzmann; Attorneys: Ned Marshak of Grunfeld Desiderio for plaintiff CME Acquisitions; Collin Mathias for defendant U.S. government; Deanna Okun of Polsinelli for defendant-intervenor Outokumpu Stainless USA; William Jones of VanAntwerp Attorneys for defendant-intervenor North American Stainless)