Earlier this month, the Federal Circuit heard oral argument in Stupp Corp. v. United States, an international trade case we have been following because it attracted an amicus brief. In this case, the Federal Circuit is reviewing a judgment of the Court of International Trade, which sustained a decision of the Department of Commerce to use a particular test in a differential pricing analysis used to calculate antidumping margins. Judges Lourie, Bryson, and Stark heard the oral argument. This is our argument recap.
Jeffrey Winton argued for SeAH Steel Corporation. Winton opened by discussing how this appeal is the second time the case is before the Federal Circuit. He noted how “in the first go-around” the Federal Circuit “held that when Commerce applies a mathematical tool like Cohen’s d it has to apply it in a manner consistent with the tool’s assumptions.” He explained the court remanded the case for the Department of Commerce to explain either “why the assumptions underlying Cohen’s d were met” or “why Commerce could ignore those assumptions.” He said the court previously concluded “that Commerce’s use of Cohen’s d is not consistent with normal statistical practice.”
One judge asked “how much discretion does Commerce have after the remand?” In response, Winton argued that the Department of Commerce is not doing “what Professor Cohen said” and that this is not reasonable. According to Winton, “when the conditions that are specified by Professor Cohen are not satisfied, the Cohen’s d test basically gives you a random result.” He argued it would not be a problem if the Department of Commerce “were using a statistical test in the proper way” and “if the conditions for using Cohen’s d” were met, but, he said, the Department of Commerce didn’t “use Cohen’s d properly.”
Robert Kiepura argued for the United States. Kiepura explained that, following the Federal Circuit’s partial remand in its prior decision, the Department of Commerce “addressed the points regarding comparison groups that were not normally distributed, having small numbers of data points or having disparate variances.” Kiepura suggested that, though the Department of Commerce recognizes the court’s concerns, “as Commerce demonstrated in its remand determination the Cohen’s d equation is a robust mathematical tool used for the purpose of telling the difference between two groups.” Kiepura highlighted that the “Cohen’s d itself is one part of the differential pricing analysis” and “the Cohen’s d coefficient does not tell you whether or not there’s a pattern.” Rather, he explained, “it’s just the first step in that process; before Commerce will determine that there’s a pattern, they look to the ratio test.”
A judge asked whether, if the court “were to say that Cohen’s d is simply inapplicable under the circumstances, such as in this case,” then “do you think that the judgment in this case from Commerce could be sustained without any regard for Cohen’s d?” Kiepura responded by arguing that the Department of Commerce “has an instruction from Congress . . . to look for these patterns” and suggested it would “shift to an alternative comparison methodology.” As far as the rest of the differential pricing analysis, Kiepura explained, “it’s possible to continue using . . . the meaningful difference tests, the ratio tests,” and other tests that have been affirmed by the court. Ultimately, he contended, “if the court takes issue with the Cohen’s d itself, it’s possible for Commerce to determine another way to show this difference.”
A judge suggested the court might say that Cohen’s d in certain circumstances “is simply not reliable enough to be given any credence, and therefore you’re down to the two remaining tests.” If the court said this, he continued, would the other two tests “independently establish everything we need to find in order to find dumping in this case,” or “would you have to go back and say, well, we’ll approach this in another way, or we’ll use different a different test?” In response, Kiepura argued the Department of Commerce “would have to come up with an explanation” for how it was “showing the difference” because that’s all “Cohen’s d is doing.”
Toward the end of his argument, Kiperua rejected the “idea that somehow Cohen’s d is producing false results, or false positives, or something to that effect.” He argued that “a lot of those comments come from places in the academic literature where they’re . . . talking about these statistical results.” But, he said, the Department of Commerce has said “we don’t need to worry about” whether a sample is reflective of its population because “we have the full population.”
Jeffrey Gerrish argued for Welspun Tubular. Gerrish argued that “the core issue here is whether Commerce’s methodology is reasonable–whether it provides a reasonable means of effectuating the statutory text and purpose of determining whether there are significant price differences among customers, regions, or time periods in order to unmask targeted dumping.” He argued “the methodology that Commerce uses here is an adaptation of the Cohen’s d test,” and he suggested that it is “not using the Cohen’s d test as Professor Cohen and others have used it for purposes of conducting behavioral research experiments.” He argued that, “in applying the test in this way, Commerce is not beholden to conform to what the statistical literature says about the correct application for conducting behavioral science experiments.”
A judge suggested that “Commerce decided to call” the test “a version of Cohen’s d,” and then “defends its use of it based on Cohen.” As a result, the judge indicated, “it seems only fair that we would assess what Commerce is doing” by comparing what it is doing “to what the statistical literature would say you should do with Cohen’s d.” In response, Gerrish admitted the Department of Commerce “has created confusion” by using “the Cohen’s d name.” But, he said, “it’s a test that’s like the the tests” the Department of Commerce has “used in the past; it uses standard deviations” and average price differences and is “akin to rules of thumb they have used in other contexts.”
In his rebuttal, Winton argued that, after looking “at the effect of exchange rates” on the differential pricing analysis–“and exchange rates are not under our control”–it turns out that “just the variations in exchange rates would lead to a finding” of no dumping. Regardless, he argued, “the academic literature shows [that] when the requirements set forth by Professor Cohen are not satisfied, the Cohen’s d test gives you a number that’s random.” And, he continued, “that’s what makes [its use] unreasonable.” Ultimately, he argued, the Department of “Commerce hasn’t shown that the results” of using Cohen’s d “when the conditions set forth by Professor Cohen are not satisfied” provides meaningful results.
We will continue monitoring this case and report on developments.