This month, the Federal Circuit heard oral argument Marmen Inc. v. United States, a trade case that attracted an amicus brief. In this case, the Federal Circuit is reviewing a judgment of the Court of International Trade, which sustained a final antidumping duty determination that assigned a dumping margin on Marmen, a Canadian wind tower producer. Judges Prost, Taranto, and Chen heard oral argument. This is our argument recap.
Jay Campbell argued for Marmen. Early in his argument, he maintained that “Commerce’s refusal to accept a minor correction to the cost reconciliation is unsupported by substantial evidence.” He explained that Marmen is a Canadian company, “its financial statements are expressed in Canadian dollars,” and all “adjustments in between must also be expressed in Canadian dollars.” According to Campbell, however, Marmen “inadvertently . . . failed to ensure that one value on one line was entirely expressed in Canadian dollars.” He went on to explain that Marmen “submitted the corrected cost reconciliation,” but the Department of Commerce “unreasonably rejected the correction.” He claimed Commerce’s reasoning that the correction Marmen sought to make was “already accounted for in the cost of goods sold” was “illogical.”
A judge asked a series of questions about Commerce’s argument that Marmen’s “one year average for the exchange rate was incorrect” and whether that would “not be enough to reject this point.” In response, Campbell argued that the exchange used in the correction was for the entire year 2018 and explained that “Commerce could look on its website and corroborate that it’s an accurate exchange rate.” Another judge quickly asked if he was suggesting that “Commerce, instead of using the exchange rate you proposed, should have unilaterally gone to its website and figured out what the correct exchange rate should have been.” Campbell indicated that was not what he was suggesting, but still argued it is “unreasonable to reject an exchange rate on the grounds that it’s unsupported” when “Commerce could easily check” that “the exchange rate we used and reported is accurate.”
Campbell argued that “Commerce’s excuse or justification for not ensuring that the assumptions underlying the Cohen’s d coefficient are satisfied is wrong.” (The Cohen’s d coefficient measures the extent to which prices applied to a particular purchaser, region, or time period differ significantly compared to prices of all other sales of comparable merchandise.) Campbell explained that, if any of the assumptions are violated–“normal distribution, equivalent variances, equal size”–then the “coefficient is essentially meaningless.” A judge, however, suggested that “maybe you don’t need the assumptions to be perfectly satisfied.” In response, Campbell argued that, although there is some indication “that there could be a modest departure,” “the overall point is the same, which is that the assumptions do matter.”
Another judge remarked that Marmen hasn’t showed why its “numbers would produce a different analysis” and asked whether there is “prejudice from the erroneous assumption [and] reliance on Cohen’s d.” In response, Campbell argued it is “Commerce’s responsibility . . . to find a pattern of significant price difference.” And, he continued, “the way Commerce’s uses the Cohen’s d coefficient without adherence to any of the assumptions that make it meaningful is completely impermissible.”
Robert Kiepura argued for the United States. He suggested that, “in the multiple cases that have addressed this issue, the Cohen’s d coefficient is only one part of Commerce’s differential pricing.” In response, a judge remarked that Commerce “has not even come close to establishing that, if this first step . . . is unsupported, the rest makes up for it.” Kiepura responded by arguing that “the appellants have tried to say that the Cohen’s d coefficient is what Commerce is using to establish the pattern that the statutory language calls for,” but “it’s only one piece of that.”
A judge asked Kiepura to identify “the standard for determining whether the difference discovered is meaningful.” Kiepura replied that, “in that final step of the differential pricing analysis,” to find “the meaningful difference,” Commerce will run two tests “on the sales.” Kiepura argued, after some back and forth, that all Cohen’s d “is attempting to do is tell us mathematically, are these two groups different” and, in particular, whether “the difference between these two groups practically significant.”
Kiepura later argued Commerce is “not using Cohen’s d to conduct a comparison of sample groups,” but rather “to compare entire populations of sales data which they have.” A judge asked what evidence “remotely suggests that, only when it comes to sampling . . . do you need the assumptions to be in effect.” Kiepura argued that the reason to go “through all of these discussions about the assumptions and the sampling” is “to ensure that” the “samples being compared are reliable representations of the populations that they’re intended to represent.” The judge questioned him multiple times for a direct citation to “any other literature that says these assumptions are needed because we’re sampling.” Though Kiepura was unable to provide an exact citation, he argued generally that the assumptions are used “to get to a place where . . . samples . . . are reliable representations of their population.”
Maureen Thorson argued for the Wind Tower Trade Coalition. She argued that Commerce reasonably decided to “figure out some relatively predictable way of figuring out whether the [relevant] conditions are met”. Thorson admitted she doesn’t “doubt Congress is not using Cohen’s d in the sense that Cohen developed it for” because “Commerce is not a social scientist trying to design experiments.” But, she claimed, “Cohen’s d is just the same kind of test that Commerce already had.”
Another judge asked about the currency exchange. Thorson responded by arguing that Marmen hasn’t “documented the origin of this exchange rate” and “it, in fact, differs from exchange rates used elsewhere.” Furthermore, she refuted Marmen’s argument that Commerce must “take independent steps to verity its accuracy using [its] own information.” She labeled this argument “both waived and unexhausted.” She also argued that the relevant calculation “is contradicted by other evidence in the record as to what would have been potentially exchange rates applicable to the correct period.”
In his rebuttal, Campbell indicated he would rest on the briefs for the issue of “averaging Marmen’s reported product-specific plate costs.” With “respect to the exchange rate,” he argued it is “outrageous to reject a correction on the grounds that Marmen failed to support an exchange rate” when Commerce “measures and records exchange rates on its website daily for purposes of antidumping margin calculations.” Regarding Cohen’s d, he argued it “only has meaning in relation to percentages of non-overlap” and that it should not be used “if the three assumptions are not satisfied.”
We will continue monitoring this case and report on developments.