CAFC Judges Ask Questions About Use of 'd' Test During Oral Argument in Lead Case on Issue
Judges at the U.S. Court of Appeals for the Federal Circuit last week questioned the Commerce Department's use of the Cohen's d test in identifying "masked" dumping in the lead case on the use of the test, which returned to the appellate court after its initial remand in 2023. Judges Alan Lourie, William Bryson and Leonard Stark asked counsel for exporter SeAH Steel Corp. if Commerce has a "lot of discretion" in how it uses the test, and they asked the government's attorney if the agency has discretion to use the test even if it's statistically unsound (Stupp Corp. v. United States, Fed. Cir. # 23-1663).
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.
CAFC previously remanded Commerce's use of the test since the agency didn't adhere to certain statistical assumptions, including the normal distribution of data when using the test. The government explained on remand that it didn't need to satisfy these assumptions since it was using the whole population of data instead of a sample. The Court of International Trade sustained that explanation, prompting the case's return to the Federal Circuit (see 2401110037).
Jeffrey Winton, counsel for SeAH, opened the oral argument by claiming that Commerce's use of the test isn't consistent with normal statistical practice. Winton invoked the work of Northwestern University professor Larry Hedges that, while not on the record, is purportedly influential in the field of statistics and decries Commerce's use of the test in this way.
Stark then asked Winton that it seems that "Commerce has a lot of discretion here," further asking how much discretion the agency has after the court's previous remand. Winton said that when the statistical assumptions imagined by the creator of the test, Dr. Jacob Cohen, aren't satisfied, use of the test isn't reasonable, since it "basically gives you a random result."
Stark suggested that maybe there's not a "blanket rule that Commerce has to use a statistical test in the way that the statisticians who developed it used it," but then questioned what the standard of reasonableness would be for the agency when it uses the test. Bryson asked Winton about the government's defense of its use of the test, which rested on its use of the entire population of data as opposed to just a sample.
Bryson then pressed DOJ attorney Robert Kiepura on the government's defense of the test, asking if the government could still identify masked dumping using its remaining tools if the court were to invalidate Commerce's use of the test. Kiepura said the agency could still identify masked dumping without the test.
Stark asked the DOJ attorney if Commerce still has the authority to use the test even if the court deems the agency's use of the test to be statistically unsound. Kiepura responded by saying there's nothing on the record "that says it's inappropriate to use Cohen's d in this manner." He added that there's "nothing that says that explicitly, that it's OK, but what we're saying is the academic literature is silent, and so Commerce's explanation as to how and why it's using this particular equation crosses that reasonableness threshold."
Kiepura added that the agency can use the test if it's statistically unsound, since it's being used for a "singular purpose" of identifying differences in groups of U.S. sales. He noted that every time the agency's use of the test has been pressed, it hasn't led to any errant results, making the test "robust enough to deal with these issues."