US, AD Petitioner Defend 'Smoothing' of Wind Tower's Steel Plate Costs at CAFC

The U.S. and antidumping duty petitioner Wind Tower Trade Coalition defended the Commerce Department's decision to weight average, or "smooth," respondent Marmen's steel plate costs in the AD investigation on utility scale wind towers from Canada (Marmen v. United States, Fed. Cir. # 23-1877).

Filing their reply briefs to the U.S. Court of Appeals for the Federal Circuit on Jan. 8, the government argued that the record backs Commerce's finding that "significant differences in Marmen's steel plate costs among" control numbers (CONNUMs) were mainly due to factors unrelated to the wind tower products' physical differences linked with the CONNUMs. As a result, the agency "reasonably determined" that the "significant factor" leading the steel plate cost differences between finished towers was the timing of the steel plate purchases, the brief said, adding that smoothing the costs was the right response.

Both the U.S. and the petitioner addressed the smoothing point, along with Marmen's claim that Commerce illegally rejected an additional adjustment included in Marmen's revised cost reconciliation. Both parties said the new reconciliation item "adjusts for amounts already reflected in the costs Marmen reported to Commerce" and that the adjustment "was otherwise unreliable." The briefs also defended the agency's use of the Cohen's d test to root out "masked" dumping in the investigation.

The Court of International Trade sustained the agency's decision to smooth the steel plate costs in a 2021 decision (see 2110220069). While Marmen argued that Commerce should have used its reported costs since differences in its reported costs were related to differences in physical characteristics, the court said Marmen's questionnaire responses and supplier agreements showed that the costs didn't vary for plates of different thickness, length, width and weight.

At the Federal Circuit, Marmen claimed that Commerce ditched its standard cost-smoothing practice by failing to look at whether Marmen's reported plate costs significantly differed in "nearly identical or very similar products." The government said this "misunderstand[s]" the agency's practice, since the practice is not limited to cases where the products are identical or very similar. The "key factor is whether a respondent's reported cost differences properly reflect production differences associated with physical characteristics, as distinguished from other unrelated factors."

The U.S. and the petitioner also said evidence supports the finding that the steel plate cost differences don't flow from differences in physical characteristics since Marmen's suppliers "did not charge prices that correlated to plates of different grade, thickness, width, or length." The only exception was for certain high-thickness plates greater than 50.8 mm, which included a surcharge.

The petitioner added that a "reasonable mind could easily accept Commerce's conclusion that the cost differences here were related to 'timing of production,'" since the majority of individual towers corresponding to the CONNUMs with the highest per-unit plate costs were made during one part of the investigation period, while a majority of the towers linked with the lowest price CONNUMs were made during another part of this period. As a result, "there is a clear correlation between the time of production and the per-unit plate cost differences," the brief said.

The government and petitioner also disputed Marmen's claim that its reconciling item should be affirmed. The respondent said the items are needed to convert the cost of the company's wind tower purchases from its affiliate Marmen Energie from U.S. dollars to Canadian dollars, which also offsets adjustments Marmen's auditor made to the company's financial statements.

The record shows that Marmen "definitively stated that its reported costs account for exchange gains and losses based on fixed rates during the year," and "that its auditors made an initial adjustment to convert the costs to reflect actual exchange rates for purchases originally made in USD," thus showing that there was "no basis for Marmen to make a further adjustment."