Commerce Maintains 16.17% Rate for Chinese Wood Flooring Exporter Upon Second Remand

After two remands by Court of International Trade Judge Jennifer Choe-Groves, the Commerce Department continued to sustain its use of Brazilian and Malaysian surrogate data in the final results of its 2019-2020 administrative review of the antidumping duty order on multilayered wood flooring from China, again assigning a plaintiff exporter a 16.17% AD margin (Jiangsu Senmao Bamboo and Wood Industry Co. v. U.S., CIT # 22-00190).

In her April 19 remand order, Choe-Groves again said that Commerce hadn’t adequately proved that the surrogate value data for both Brazil and Malaysia were publicly available and representative of the review’s subject merchandise (see 2404190032). She rebuked the department in the order, saying it hadn’t fulfilled her original order’s same requirement (see 2308250059) because it only cited its own memos as evidence (see 2310260032). She also remanded the department’s decision to remove Spanish import data from the Brazilian surrogate value dataset.

In its new results upon remand released June 18, Commerce continued to both use Brazilian and Malaysian surrogate values and again refused to use the Spanish import data.

The department chose to use Brazilian data generally to construct a value for exporter Jiangsu Senmao Bamboo and Wood Industry’s wood flooring. However, it used Malaysian data to value Senmao’s “European oak,” “red oak” and non-oak logs because Brazil’s data only listed its logs under a general basket category, while Malaysia’s differentiated between oak and non-oak. It said it did so because it couldn’t tell whether Brazil’s log data included any oak logs at all.

After the first remand order asked why Brazil’s log data wasn’t enough on its own for the review, the department said in response that it had discovered that Brazil did have import data for oak logs prior to the period of review, between 2015 and 2017, but none contemporaneous to it. As a result, Commerce said, “a sufficient basis in the record exits [sic] to show that the Brazilian oak log data for use in valuing Senmao’s oak log inputs … are unavailable for the POR.”

In response to the second remand order, the department said that it “provided the Court with a more detailed discussion of the evidence” to demonstrate that the Brazilian and Malaysian surrogate values met Commerce’s criteria of being “publicly available, contemporaneous with the POR, representative of a broad market average, tax duty-exclusive, and specific to the inputs being valued."

Both the Malaysian and Brazilian data, submitted by the review’s petitioner and respondent respectively, “include sources such as the Global Trade Atlas (GTA), the World Bank’s Doing Business reports, and Brazilian and Malaysian government and agency websites,” Commerce said. These sources were publicly available, it said.

It also noted that the data represented broad market averages; the GTA data, it said, averaged the prices of inputs imported into the two countries. And the GTA data has also “previously been found to be tax and duty-exclusive, and no parties have argued otherwise,” it said.

Finally, it said that both countries’ data was specific to the inputs because both “include raw GTA data for Brazilian and Malaysian [Harmonized System] codes covering the vast majority of inputs used in the production of subject merchandise.”

Regarding the removal of Spanish import data from the Brazilian plywood dataset, Commerce said the decision had been necessary because Spanish imports were erroneously expressed as kg/m³.

In her remand order, Choe-Groves explained that “Commerce determined that the Spanish import data for 2020 were incorrect because the data reported the same quantity of plywood in cubic meters ('m³') as it did in kilograms ('kg'). Because the m³ unit measures volume and the kg unit measures weight, Commerce concluded that it was ‘illogical for the Spanish import data to report the same quantity in these two different units of measure.’”

But the department relied on an exhibit that it had never placed on the record to make that determination, and it didn’t “provide adequate explanations” for its overall decision, she said.

In response, Commerce said in its results upon remand that “if both m³ and kg are equal values, it would result in a density of one, which is erroneous” because plywood density “ranges in the hundreds.”

But that flaw only occurred in one line of Spanish data representing one month, the department said. It argued that removing that one line didn’t have much impact on the overall quality of the dataset. The data still “represents a broad market average comprised of imports from numerous countries: Argentina, Belgium, Hong Kong, Latvia, Netherlands, Paraguay, Russia, Spain, and the United States,” it pointed out.

“The Court has upheld Commerce’s approach of eliminating aberrational values from data sets on which it relies,” it said.