Federal Circuit Sends Back Commerce's Use of Statistical Test to Detect 'Masked' Dumping

The Commerce Department must reconsider its decision to use a simple average to calculate the pooled standard deviation when using the Cohen's d test in its differential pricing analysis to target "masked dumping," the U.S. Court of Appeals for the Federal Circuit said in an April 21 opinion. Ruling that Commerce strayed from the statistical literature without a proper explanation, Judges Pauline Newman, Alan Lourie and Richard Taranto said the agency should reconsider whether a weighted average for calculating the Cohen's d denominator is more appropriate.

"We believe that Judge Taranto’s decision was the final nail in the coffin of [Commerce's] calculation of the denominator of the 'significantly different' section of the Cohens d test based on a simple average of groups with unequal quantities," Ned Marshak, counsel for defendant-appellants led by PT Enterprise, said in an email. "Hopefully [Commerce] -- after losing this issue three times at the CAFC -- will now agree, thereby avoiding the need for yet another appeal."

In antidumping duty investigations, such as the one contested by appellant PT Enterprise into steel nails from Taiwan (see 2107270042), Commerce may identify goods that are dumped into the U.S. market through "masked" dumping. Since Commerce typically conducts its investigations by comparing the average home market price of the good in question with its U.S. price, exporters may work around this by dumping the goods in certain areas and selling them at a higher price in another or at another time. To combat this, Commerce may compare the weighted average of sales in the home country to individual sales prices.

Before conducting this analysis though, Commerce must first detect the masked dumping using a differential pricing analysis. The agency breaks down the U.S. sales data into sets based on comparable product groups. Once in the product group, Commerce then breaks that data into various subsets, including the region where the U.S. sales took place, the purchasers involved in the sales and even the time periods of the sales. Commerce will then pick one subset as the "test group" while aggregating the remaining subset into the "comparison group." The agency then employs the Cohen's d test to find whether the test group is significantly different from the comparison group.

The test comes out with a coefficient, or a ratio, wherein the numerator is the difference between the mean prices of the two groups and the denominator is a benchmark from which to judge the significance of this difference. It was the benchmark for the coefficient that served as the point of contention for PT Enterprises' appeal. In fact, this case was before the Federal Circuit on this exact issue, where the court remanded Commerce's position. Then the court again remanded Commerce's position, finding that the agency's decision to use a simple average of the pooled standard deviation for the Cohen's d denominator instead of a weighted average was unlawful.

The statistical literature on the d test says that the ideal way to come up with the denominator is to use the standard deviation for the whole population, which can be approximated using a pooled estimate. In the present case, Commerce did not use the entire population's standard deviation, despite acknowledging that it had the whole population's data, opting instead for the simple average.

In defense of this move, Commerce said that the data in each group represented equally rational and genuine pricing choices, warranting equal weighting for each group. While the court said it had no basis for questioning the premise of equal rationality of the pricing behavior, it said that the agency has not offered an "adequate explanation" for why that premise supports its decision on how to form the denominator. "The fact that the seller is acting rationally and genuinely in its pricing choices in both the test and comparison groups provides no apparent reason for assigning equal weight to each group’s standard deviation when computing the pooled standard deviation," the opinion said.

While Commerce is not "duty-bound" to follow the statistical literature in antidumping matters, the court said that the agency surely had to be more partial to it since it relied on the literature in every other step of its reliance on the Cohen's d test. "In this situation, Commerce needs a reasonable justification for departing from what the acknowledged literature teaches about Cohen’s d," the opinion said. "It has departed from those teachings about how to calculate the denominator of Cohen’s d, specifically in deciding to use simple averaging when the groups differ in size. And its explanations for doing so fail to meet the reasonableness threshold (a deferential one, in recognition of expertise) for the reasons we have set forth."

(Mid Continent Steel & Wire v. United States, Fed. Cir. #21-1747, dated 04/21/22, Judges Pauline Newman, Alan Lourie and Richard Taranto. Attorneys: Adam Gordon of The Bristol Group for plaintiff-appellee Mid Continent; Mikki Cottet for defendant U.S. government; Ned Marshak of Grunfeld Desiderio for plaintiff-appellants)