No Adverse Inference Without Finding AD Respondent Failed to Cooperate, CIT Rules

The Commerce Department must reconsider its use of an adverse inference in an antidumping review on Italian pasta since it failed to find out whether a respondent did not to cooperate to the best of its ability, the Court of International Trade said in a Nov. 30 opinion. However, the court upheld the remaining elements of the decision, including Commerce's use of facts available and the agency's rejection of the respondent's post-verification arguments for different classification systems for the pasta's protein content and shape.

The case was over the 22nd administrative review of the antidumping duty order on certain pasta from Italy, in which affiliated plaintiffs Ghigi 1870 and Pasta Zara served as a mandatory respondent. Following the questionnaire period of the review, Commerce undertook a verification of Ghigi's U.S. constructed export price sales, when Ghigi told the agency that the U.S. payment dates were errant. A programming error had altered the dates. Ghigi subsequently requested that Commerce revert to using its original database to get a more accurate list of the U.S. sales dates.

The agency declined to do this, declaring that its practice is to use the most recently submitted data. Settling on using the wrong data, Commerce then found it couldn't verify the dates, and thus relied on facts available with an adverse inference. In the court's opinion on the matter, Judge Richard Eaton upheld Commerce's use of facts available but not the adverse inference. Eaton said that Commerce's practice of relying on the most recently submitted data as the basis for verification is reasonable, finding that "Commerce cannot be said to err when it uses a respondent’s own responses to the Department’s last-in-time requests."

The judge also said that the plaintiffs did not make their argument against the revised database until post-verification, precluding Commerce from being able to verify the original database. So, it was not unreasonable to decline to use the unverified original database. What was unreasonable was the use of an adverse inference, the judge said. An adverse inference can only be used when a respondent has been found to not cooperate to the best of its ability in the investigation. Commerce did not establish that here, so it cannot rightfully claim AFA, the judge said. Eaton gave Commerce a chance to either drop the adverse inference or further explain how Ghigi and Zara did not cooperate to the best of their ability.

Also, during the post-verification stage of the review, Ghigi and Zara argued against Commerce's classification of their pasta's protein content and shape. Commerce requires that a pasta's protein content be reported as it appears on the label of the product. But Ghigi instead converted the protein percentage on the label to reflect the amount of protein under Italian protocols and then coded the protein content based on this conversion, the opinion said. This conversion changed the classification of Ghigi and Zara's pasta from premium to standard pasta. Commerce discovered the mishap, then instructed them to recode their protein content.

During the post-verification stage, Ghigi and Zara argued that there was an ambiguity in the instructions since the protein measurement protocols were different between the U.S. and Italy. Commerce rejected this argument. Eaton said the agency was right to reject this rationale since there is no ambiguity in Commerce's instructions and the respondent did not tell Commerce that it was dealing with any sort of ambiguity.

A similar issue unfolded over the proper pasta shape classification for the Ghigi and Zara noodles. Commerce requests that pasta be classified into eight categories, with fusilli and cavatappi falling into specialty short cuts. Ghigi instead reported them as simply short cuts. The agency rejected this classification, applying its preferred shape codes, and rejecting the respondent's argument that the pasta could be reclassified based on company-specific practices.

"The problem for Ghigi/Zara is that (1) it decided to modify the model-match method with respect to protein content and shape on its own without alerting Commerce to its activities, and (2) it did not raise the argument in favor of its preferred methods for calculating protein content and for classifying shapes during the questionnaire phase, when Commerce could have considered its claims, sought evidence of industry-wide changes, and, importantly, made findings on the proposed modifications to the methods based on substantial record evidence," the judge said.

Eaton relied on a recent U.S. Court of Appeals for the Federal Circuit decision in this case, Goodluck India Limited v. United States, which found that "[v]erification represents a point of no return" (see 2108310040). Verification is not meant as a chance to submit new factual information and is rather the "culmination of an orderly process of information-gathering." No good reason was given to have Commerce accept new information at verification in this review, and so the plaintiff's arguments fall short, the judge ruled.

(Ghigi 1870 S.P.A., et al. v. United States, Slip Op. 21-159, CIT #20-00023, dated 11/30/21, Judge Richard Eaton. Attorneys: David Simon of Law Offices of David L. Simon for plaintiffs Ghigi and Pasta Zara; Sosun Bae for defendant U.S. government)