CIT Says Commerce Erroneously Used AFA as Factual Basis to Find AD/CVD Circumvention
The Court of International Trade remanded multiple aspects of the Commerce Department's use of adverse facts available in its inquiry into whether certain Vietnamese goods circumvented the antidumping duty and countervailing duty orders on hardwood plywood products from China.
Sign up for a free preview to unlock the rest of this article
Timely, relevant coverage of court proceedings and agency rulings involving tariffs, classification, valuation, origin and antidumping and countervailing duties. Each day, Trade Law Daily subscribers receive a daily headline email, in-depth PDF edition and access to all relevant documents via our trade law source document library and website.
In a decision made public on Feb. 13, Judge Mark Barnett found that Commerce failed to give certain exporters adequate notice that their responses were insufficient, as required, to warrant the use of AFA; didn't properly support its finding that two exporters failed to cooperate to the best of their ability; and, perhaps more crucially, "failed to support with substantial evidence the factual bases for its adverse inferences" that the goods at issue in the inquiry were "completed or assembled from relevant Chinese components."
Commerce initially opened the circumvention inquiry following a scope ruling on exporter Vietnam Finewood's products that said the scope of the AD/CVD orders "unambiguously covers hardwood plywood and certain veneered panels that ... 'consist[] of two or more layers or plies of wood veneers and a core.'" The inquiry was requested by the Coalition for Fair Trade in Hardwood Plywood, which focused its petition on two production scenarios in which the main parts of the plywood were shipped from China to Vietnam for assembly and exportation.
However, the agency asked the petitioner to consider adding three additional production scenarios, and Commerce ultimately included these additional scenarios in the circumvention inquiry.
Commerce then issued quantity and value questionnaires to 57 exporters and producers. Fifty-one companies responded, including two unsolicited responses, meaning eight companies didn't respond to the initial questionnaires. Supplemental questionnaires were issued to the 51 respondents, of which 46 timely responded. These 46 companies were then issued a second supplemental questionnaire, to which 45 timely responded.
The agency then issued 11 company-specific questionnaires. However, since there wasn't a single respondent that said it made inquiry merchandise, Commerce didn't select any mandatory respondents.
Commerce preliminarily issued an affirmative circumvention finding, deciding to use AFA for non-responding companies and the companies that responded to all the questionnaires but reported "significant discrepancies, inconsistencies, or misleading information."
The agency also relied on AFA in part to determine that the "process of assembly and completion" in Vietnam "was minor or insignificant." In addition to the use of AFA, Commerce cited record evidence from the petitioner, namely a declaration and data regarding patterns of trade, and the Finewood scope proceeding.
Commerce then imposed a country-wide circumvention finding for Vietnam and implemented a certification program under which companies can certify on an entry-specific basis that their imports don't contain inquiry merchandise. However, the agency said the 36 companies subject to AFA weren't eligible to take part in the certification program, though Commerce said it would "reconsider the eligibility" of those companies if they show in a future segment of the AD/CVD proceedings that their goods weren't made under any of the production scenarios subject to the inquiries.
Barnett rejected the agency's use of AFA in multiple different areas as unsupported by substantial evidence. Regarding Commerce's use of AFA to find that merchandise of the "same class or kind" as merchandise subject to the AD/CVD orders is "completed or assembled" in Vietnam using certain Chinese components, the U.S. alleged that the use of AFA in this way is "moot" with respect to the companies who are allowed to take part in the certification regime.
The judge said that argument "fails," since Commerce's fundamental circumvention finding regarding the statutory factor on the completion of goods in Vietnam "rested primarily, if not completely, on AFA." While the agency said its use of AFA is "bolstered" by other data on the record, Barnett said "Commerce approached its analysis backwards."
Instead of using an adverse inference when selecting from the facts available and explaining the link "between the source ... and the result," Commerce "drew an adverse conclusion (completion of inquiry merchandise) in order to fulfill that statutory requirement and identified evidence the agency claims 'bolstered' the finding Commerce already made." The court said it's "unclear" whether Commerce considered the petitioner's evidence "sufficient on its own to make an affirmative finding regarding the completion or assembly of inquiry merchandise by each respondent subject to AFA."
The judge emphasized that the petitioner's evidence only relates to two production scenarios and not to the three others added to the inquiry by Commerce itself. While a finding of circumvention "with respect to two scenarios may have the same impact as one based on five scenarios," it's up to Commerce first to state this based on the evidence, the court held.
Addressing prudential concerns raised by the government, Barnett said it's well aware that the respondents are the ones in possession of the relevant information and that addressing AD/CVD circumvention is akin to "an exercise of 'Whack-A-Mole." However, the "adherence to the requirements of the statute ensures that legitimate market-driven supply shifts not amounting to circumvention or evasion are not mistaken for a 'mole,'" the judge said.
Greg Menegaz, counsel for various exporters and importers led by American Woodmark, said he hopes a remand proceeding will end favorably for the respondents given that many of them were able to certify in the fifth review of the AD order that their products weren't of Chinese origin. He said he hopes Commerce will take all of the "extensive efforts" the respondents undertook in the AD review "into account and bring this to a close."
Barnett also rejected Commerce's use of AFA for many of the respondents due to a lack of notice the agency gave to some of the respondents of deficiencies in their responses. While the government argued that the supplemental questionnaires gave the requisite notice of deficiencies, Barnett rejected this claim for the generic supplemental questionnaires issued to many of the respondents.
Missing from these questionnaires are words like "clarify," "deficient," "discrepancy," or "inconsistency." Commerce's attempt to equate the questionnaires' attempt to seek "additional details" with "clarification" isn't persuasive, the court held. While Commerce doesn't have to use "magic words" to notify a respondent of a deficiency, "these generic and widely-issued supplemental questionnaires are the type of '[b]roadly drawn' questionnaires that do not provide notice of any particular deficiency," the court said.
The court then went through the company-specific questionnaires and found that all but the questionnaires to respondent Win Faith failed to notify the companies of deficiencies in their responses.
Menegaz said he was "very pleased that the judge recognized the all-essential guarantee in the statute that the Commerce Department has to notify respondents of deficiencies and provide them an opportunity to address deficiencies."
Barnett then admonished Commerce for improperly using AFA for two companies, Win Faith and Vietnam Zhongjia Wood, for their alleged failure to cooperate to the best of their ability. However, the court sustained the use of AFA against Cam Lam, TL Trung, Bao Yen and Thang Long for their failure to cooperate to the best of their ability.
Prior to his discussion of the use of AFA, Barnett rejected various procedural challenges brought against the circumvention proceeding. For instance, the judge held that Commerce didn't err in its data collection practice or use of secondary sources, including the Finewood scope ruling and petitioner's declaration and data. Barnett added that Commerce didn't err in not selecting mandatory respondents, since nothing in the circumvention statute "suggests, let alone requires, the selection of mandatory respondents," and he said Commerce didn't violate the law by "taking roughly 1,120 days to complete its investigation, rather than the 300 days 'contemplated' by the statute."
The court lastly rejected many of the respondents' claims that Commerce unlawfully expanded the scope of the AD/CVD orders and implemented an unlawful certification regime. Goods subject to a circumvention inquiry will lawfully be considered in-scope of the relevant orders, and the certification regime, and related cash deposit requirement, isn't an "improper duty assessment," the judge said.
Barnett also granted the government's voluntary remand request so that Commerce can tweak its instructions to CBP regarding exporters An An Plywood and Greatwood Hung Yen, since the agency declined to apply AFA to the companies.
(Shelter Forest International Acquisition v. United States, Slip Op. 26-5, CIT Consol. # 23-00144, dated 01/28/26; Judge: Mark Barnett; Attorneys: William Chandler of Pillsbury for plaintiff Shelter Forest International Acquisition; William Marshall of Sandler Travis for consolidated plaintiffs led by USPLY; Bryan Cenko of Mowry & Grimson for consolidated plaintiffs led by Concannon Lumber Co.; Gregory Menegaz of The Inter-Global Trade Law Group for consolidated plaintiffs led by American Woodmark Corp.; Stephen Brophy of Husch Blackwell for consolidated plaintiff Greatriver Wood Co.; Gregory McCue of Steptoe for consolidated plaintiff Tumac Lumber Co.; Sosun Bae for defendant U.S. government; Stephanie Bell of Wiley Rein for defendant-intervenor Coalition for Fair Trade in Hardwood Plywood)