CAFC Again Rejects Use of 'd' Test in 2nd Decision on the Test in As Many Days
The U.S. Court of Appeals for the Federal Circuit on April 23 again rejected the Commerce Department's use of the Cohen's d test as part of its analysis to detect and address "masked" dumping. A day after the court resoundingly struck down the agency's use of the test in a separate case (see Ref:2504220030]), Judges Alan Lourie, William Bryson and Leonard Stark said they were bound by the court's day-old ruling.
Sign up for a free preview to unlock the rest of this article
Timely, relevant coverage of court proceedings and agency rulings involving tariffs, classification, valuation, origin and antidumping and countervailing duties. Each day, Trade Law Daily subscribers receive a daily headline email, in-depth PDF edition and access to all relevant documents via our trade law source document library and website.
Writing individually, Stark said that were it not for the court's ruling in Marmen v. U.S., he might have come to a different conclusion on whether Commerce's use of the d test, "as one step in its three-step differential pricing analysis, is reasonable.”
The agency frequently used the test in antidumping proceedings to test if a respondent's U.S. sales differed significantly across purchasers, regions and time periods. The appellate court first called Commerce's use of the test into question in 2021, since the agency didn't adhere to three main statistical assumptions: normal distribution of data, roughly equal variances and equal groups.
In Marmen, the issue returned to the Federal Circuit, which struck down Commerce's use of the test. Judges Sharon Prost, Richard Taranto and Raymond Chen rejected the government's claim that it didn't need to adhere to the statistical assumptions, since Commerce used the entire population of data instead of just samples. The court also struck down the government's defense that it didn't need to adhere to the assumptions, since the test isn't used to determine dumping margins but only as one part of the broader differential pricing analysis.
In the second decision on the issue in as many days, the court said it confronts the same issue and is "compelled to reach the same result."
Stark said in additional remarks that if Commerce "had labeled what it is doing at step one as 'Cohen's d,' and had" not tried to "borrow credibility for its test by reference to Dr. Cohen's extensive work and other literature endorsing Cohen’s d," he might agree the test could be used as a "rough, initial 'measur[e] [of] the practical significance of price difference[s].'" He said any alternative approach to the test by the court is now thwarted due to Marmen, additionally noting that "this case does not call on us to assess the broader question of whether Commerce can ever reasonably rely on rules of thumb that are not statistically grounded."
(Stupp Corp. v. United States, Fed. Cir. # 23-1663, dated 04/23/25; Judges: Alan Lourie, William Bryson, Leonard Stark; Attorneys: Jeffrey Gerrish of Schagrin Associates for plaintiff-appellee Welspun Tubular; Robert Kiepura for defendant-appellee U.S. government; and Jeffrey Winton of Winton & Chapman for defendant-appellant SeAH Steel Corp.)