CIT Says Commerce Failed to Justify Quarterly Cost Methodology in Italian AD Review

The Court of International Trade on Sept. 17 sent back the Commerce Department's use of a quarterly cost methodology to analyze exporter Officine Tecnosider's sales during the 2020-21 review of the antidumping duty order on steel plate from Italy to address "shortcomings" in its analysis.

Judge Stephen Vaden said Commerce failed to explain why "focusing solely on Italian sales is a reliable indicator of linkage for U.S. sales," why it bucked its precedent by analyzing products jointly sold in both the U.S. and home markets, and how the agency analyzed the data to find there was "proper linkage between the cost of manufacturing and the sales price." The court said the agency can't use a "rote application of its test to a non-traditional dataset in the face of specific objections," though the court said the door is open for Commerce to stick with its present decision if sufficient justification is provided.

The agency decided to go with the alternative method for calculating costs, based on quarterly data, after conducting its two-step test on whether to buck the standard annual weighted-average cost methodology. The second step of the test looks at whether there were linkages between the cost of production and the sales price. Commerce asked the court for a voluntary remand to conduct this test in light of arguments from Tecnosider. The request was granted, and on remand, Commerce went with the alternative method and slashed the exporter's AD rate from 20.44% to zero percent (see 2309120010).

In conducting step two of the test, Commerce ordinarily examines two pools of the five largest sales volume control numbers (CONNUMs) in the home market and the five largest sales volume CONNUMs in the U.S. market to see if there was a reasonable correlation between changing costs and sales prices from quarter to quarter. Tecnosider's U.S. sales were made of only five CONNUMs, all of which occurred in the same quarter, making it impossible to perform the linkage analysis for the U.S. sales CONNUMs, the court said.

Based on this "limited data," Commerce said that as manufacturing costs went up or down, so did its sales prices in Italy. The agency didn't conduct a linkage analysis for the U.S. sales but said it could still find a reasonable linkage between changes in home market sales prices and changes in the cost of production.

Vaden first said Commerce's conclusion failed to respond to objections from petitioner Nucor Corp. related to the sole use of Italian data. The court said no party has cited a past case "where Commerce found a reasonable linkage based solely on home market sales data.” In addressing the lack of U.S. data, Commerce only said the lack of this data isn't "dispositive" due to other linkages in the home market data. This justification "fails to explain why the absence of data apparently analyzed by Commerce in every other application of its test does not undermine the results," the court said.

Commerce merely pointed to data and asked the court to trust it, Vaden said. "The agency creates a textbook example of failing to ‘articulate [a] rational connection between the facts found and the choice made,'" the opinion said.

The court said Commerce failed to justify why it departed from its past practice, established in the AD proceeding on ferrovanadium from South Korea. In that case, the respondent produced three total CONNUMs, two of which were sold exclusively in the home market and one that was sold in both the U.S. and the home markets. The agency in that instance only focused on the one shared CONNUM.

In the ferrovanadium case, Commerce was faced with a "non-standard data set" and "deviated from its test" to focus on the one CONNUM. "Here, Commerce has done the opposite" and rotely applied its standard analysis despite being faced "with another non-standard data set," the court said.

The agency lastly defended its practice by saying that if it used Nucor's preferred analysis, under which the agency would focus on the three CONNUMs sold in both the U.S. and the Italian market, the result would be the same. Commerce said that, under Nucor's approach, the petitioner's analysis is "questionable" since it could find sufficient evidence of linkage for two of the three overlapping CONNUMs, since they match to a large majority of U.S. sales.

Vaden said the mere identification of this argument isn't enough, adding that if the agency "believes there is more than one way to look at the data, it must state the alternatives, explain how it analyzes the available data, and offer a reasoned explanation."

On remand, if Commerce wants to still use the alternative quarterly cost methodology, it must "state what potential flaws caused it to abandon its 'normal' linkage analysis, explain how" the precedent proceeding on ferrovanadium from South Korea "informs its decision, and describe how any alternative analysis leads to trustworthy results," the court said.

(Officine Tecnosider v. United States, Slip Op. 24-102, CIT # 23-00001, dated 09/17/24; Judge: Stephen Vaden; Attorneys: Pierce Lee of Crowell & Moring for plaintiff Officine Tecnosider; Augustus Golden for defendant U.S. government; Stephanie Bell of Wiley Rein for defendant-intervenor Nucor Corp.)