OCTG Exporter Challenges Commerce's Mid-Review Switch Off 'd' Test
The Commerce Department erred in backing off its use of the Cohen's d test to identify targeted dumping in the middle of an antidumping duty review and introducing a new "two-percent threshold," review respondent Tubos de Acero de Mexico (TAMSA) argued in a Nov. 6 complaint at the Court of International Trade. TAMSA said that while Commerce said it was backing off the d test due to the U.S. Court of Appeals for the Federal Circuit rejecting the agency's use of the test, the agency didn't have to make a change, since CAFC's decision wasn't "final and conclusive" (Tubos de Acero de Mexico v. United States, CIT # 25-00221).
Sign up for a free preview to unlock the rest of this article
Export Compliance Daily combines U.S. export control news, foreign border import regulation and policy developments into a single daily information service that reliably informs its trade professional readers about important current issues affecting their operations.
The respondent added that the agency's decision to "suddenly use an untested new calculation methodology that was not court-approved at a late stage in the proceeding," instead of using the d test or the "average-to-average method," was "arbitrary and punitive" to the respondent.
In its complaint, TAMSA challenged other elements of the 2021-22 review of the AD order on oil country tubular goods from Mexico, including Commerce's decision to use a recoded field in the respondent's sales and cost databases when identifying the foreign like product. TAMSA also contested Commerce's rejection of certain information in the company's case brief as new factual information and the agency's denial of constructed export price offset in the review. The respondent also said the review unlawfully exceeded the statutory maximum 18-month period for conducting reviews.
Initially in the review, Commerce used the Cohen's d test to identify if TAMSA was engaging in targeted, or "masked," dumping. Using the test, the agency preliminarily decided to use the "average-to-transaction" comparison method only for the respondent's U.S. sales that failed the test, assigning the respondent a 30.38% AD rate.
However, during the review, the Federal Circuit in Marmen v. U.S. struck down Commerce's use of the test "when that test is applied to data that do not satisfy the statistical criteria of normal distribution, equal variances, and sufficiently numerous data." After this holding, in the present review, Commerce "introduced a new two-percent threshold in place of the Cohen's d test," and discontinued its use of the "mixed method" as a "potential alternative comparison methodology and reduced the threshold for applying the A-T method from 66 percent to 33 percent," the complaint said.
As a result of this change, Commerce compared an average of TAMSA's home market sales to an individual U.S. transaction for all of TAMSA's U.S. sales, leading to a 42.65% AD rate.
At the trade court, the respondent said Commerce was too quick to abandon the d test and that the new two-percent threshold used in all cases isn't consistent with the Statement of Administrative Action's mandate that the agency use a "case-by-case" approach to targeted dumping. TAMSA said "Commerce failed to address the case-specific arguments" the company raised "regarding why the use of two-percent threshold is not appropriate based on the record of the First Review."
The respondent also said Commerce's discontinuance of the "mixed method" as a potential alternative comparison methodology is contrary to CAFC's decisions, "which instruct that on remand Commerce must discontinue the use of the Cohen’s d test under certain circumstances, and provide for the continued use of the mixed or 'hybrid' method."
Regarding the agency's model match methodology, TAMSA said it used the exact methodology and product characteristic coding Commerce verified and accepted in the original AD investigation, which included two codes for "field TYPE added by TAMSA in the investigation." The agency later required TAMSA to eliminate one of the two codes added to the TYPE field even though that code "distinguished certain products" that are "commercially different from casing and tubing," and instructed the respondent to replace the code with two codes from the questionnaire covering tubing or casing.
While TAMSA complied with this request, the respondent challenged Commerce's use of the recoded field and failure to use the "SUBTYPE field" reported by the respondent "to differentiate the products." The respondent said Commerce's decision to use the recoded TYPE field also "yielded an unreasonable comparison, distorting both product matching and the resulting margins."