Mukand, Ltd. v. United States

Mukand, Ltd. (“Mukand”) appeals a decision of the Court of International Trade affirming the Department of Commerce’s application of adverse facts available in its calculation of an antidumping duty on Mu-kand’s imports of stainless steel bar from India. The Department of Commerce applied adverse facts available (“AFA”) after Mukand failed to provide production cost data broken down by product size as requested on five separate occasions. For the reasons set forth below, we affirm.

I

Upon the receipt of a proper request, the Department of Commerce (“Commerce”) is required to review and reassess its antidumping duty orders at least once each year. 19 U.S.C. § 1675(a). On March 30, 2010, at the request of domestic interested parties, Commerce initiated the current administrative review on an outstanding antidumping duty order on stainless steel bar from India for the period of February 1, 2009, through January 31, 2010. As part of this review, Commerce *1303 issued to Mukand a series of questionnaires designed to obtain information necessary to calculate Mukand’s dumping margin. These questionnaires asked Mu-kand to provide, among other things, its costs of producing different sizes of stainless steel bar. Product size is one of six product characteristics determined by Commerce to be significant in differentiating between steel bar products, the other five being general type of finish, grade, remelting, type of final finish, and shape. Commerce thus sought product-specific cost information to ensure that it compared similar products in its priee-to-price comparisons, calculated a correct difference-in-merchandise adjustment, and arrived at an accurate constructed normal value for Mukand’s merchandise.

Upon receiving Mukand’s response to its initial questionnaire, Commerce discovered that Mukand assigned the same production costs across all product sizes. Mukand did not explain its rationale for this approach despite the questionnaire’s request to “quantify and explain” any belief that size, or any other physical characteristic, is an insignificant cost factor. Commerce informed Mukand that it did not consider this approach to be reasonable and asked that Mukand produce size-specific cost information, regardless of whether it tracked such information in its normal accounting records. Alternatively, Commerce again asked Mukand to “quantify and explain” any reasons for believing that size-based cost differentials are insignificant. Mu-kand responded with a brief statement that where product grade and type of finishing operation are the same, direct material costs do not vary with size. In a second supplemental questionnaire, Commerce reiterated its need for either size-specific cost estimates or a more thorough narrative quantifying and explaining Mu-kand’s belief that size is not a cost factor. Again, Mukand asserted without detailed support that size does not affect costs when all other physical characteristics remain the same. In a third supplemental questionnaire, Commerce again reiterated its need for size-specific cost information, noting:

[I]t is not necessary for Mukand to calculate [control number (“CONNUM”)] specific costs in its normal books and records in order to differentiate cost differences between CONNUMs that have different physical characteristics when reporting to the Department as long as the cost differences reported to the Department are based on reasonable and verifiable methods.

J.A.2063 (emphasis added). Mukand responded with a short statement that it does not keep track of size-specific costs and reasserted its belief that size-based costs are insignificant “as smaller sizes can be processed at higher speed than to [sic] larger size.” J.A.2064.

Unsatisfied with Mukand’s response, Commerce sought this information one last time. In a fourth supplemental questionnaire, Commerce noted that it sought cost data with respect to two factors— rolling time and weight—and asked a series of specific questions designed to obtain the elicited information. These questions included a sample chart for Mukand to complete regarding size, weight, and rolling time. Commerce instructed Mukand to contact it if its request was unclear, if Mukand was unable to supply the information, or if Commerce was otherwise mischaracterizing Mukand’s production process. Commerce also warned that “[flailure to provide the requested information may result in the Department deciding to rely on facts available, as required by section 776(a) of the Tariff Act of 1930, as amended, in our preliminary results.” J.A.2075. In its response, Mu-kand again restated its reasons for not *1304 reporting size-specific costs, concluding that there “is no reasonable and verifiable way to do what is requested.” J.A.2074 (emphasis altered). Mukand never contacted Commerce directly to ask for clarification or assistance of any kind.

Commerce determined that Mukand’s responses were deficient and resorted to facts otherwise available. Pursuant to statute, Commerce may resort to facts otherwise available to complete the record when an interested party fails, for whatever reason, to provide requested information. 1 Before resorting to facts otherwise available, Commerce must notify the respondent of the nature of the deficiency and, to the extent practicable, provide an opportunity for the respondent to remedy or explain the deficiency. 19 U.S.C. § 1677m(d). If the respondent’s explanation is unsatisfactory or untimely, Commerce may “disregard all or part of the original and subsequent responses.” Id. Commerce may not, however, refuse to consider necessary information that satisfies the five criteria outlined in section 1677m(e).

Commerce may further rely on an adverse inference against a respondent when selecting among the facts otherwise available if it concludes that the respondent failed to cooperate to the best of its ability. 19 U.S.C. § 1677e(b). The “best of its ability” standard requires the respondent to put forth its maximum effort to investigate and obtain full and complete answers to Commerce’s inquiries. Nippon Steel, 337 F.3d at 1382. While this standard does not require perfection on the respondent’s part, it does not allow for “inattentiveness, carelessness, or inadequate record keeping.” Id.

In its preliminary results, Commerce applied an adverse inference against Mukand after concluding that Mukand (i) repeatedly failed to provide product-specific cost data by size; (ii) failed to provide a meaningful explanation of why it could not provide such data; and (iii) failed to provide factual information supporting its claim that product size did not significantly affect production cost. 2 Commerce noted that requesting product-specific cost data is standard procedure, and that a respondent has a duty to provide a “full explanation and suggested alternative forms” if it is unable to provide requested information. Id. Accordingly, Commerce concluded that applying AFA against Mukand was justified.

Mukand responded to the preliminary results and claimed that it materially complied with Commerce’s requests. Mukand argued that it could not report size-specific production costs because any information it would generate would not be subject to reasonable verification. At the same time, Mukand offered to submit the same information it previously declared was not reasonably available, and that it could do so “immediately on request.” Commerce refused to consider this new information because it lacked time to review and solicit comments on the data within the statutory deadlines. Commerce also noted that Mu-kand had numerous opportunities during the questionnaire process to provide this data. Commerce further found that Mu-kand’s failure to provide size-specific cost information rendered its response “so incomplete that it could not serve as a reliable basis for reaching a final determination” and could not be used without undue difficulty. Commerce thus continued to *1305 apply AFA to all of Mukand’s sales under review and assigned Mukand an AFA rate of 21.02 percent ad valorem. 3

Mukand appealed to the Court of International Trade (“Trade Court”), and the court affirmed Commerce’s application of AFA. 4 The Trade Court noted that Commerce asked for size-specific cost information on five separate occasions, and Commerce explained on four of those occasions that it was unsatisfied with Mukand’s response and reiterated both the type of information it needed and why it was important. The Trade Court also rejected Mukand’s assertion that it complied with Commerce’s questionnaires when it explained that it had no reasonable and verifiable way to report size-specific costs. The Trade Court agreed with Commerce that Mukand’s responses consisted of vague, unsupported assertions that the requested information was not reasonably available and that size was not a significant cost factor. The Trade Court also noted that, despite these repeated assertions, Mukand was suddenly willing and able to provide the requested information after Commerce issued its preliminary results. The Trade Court thus affirmed Commerce’s conclusion that Mukand failed to cooperate to the best of its ability.

The Trade Court also rejected Mukand’s argument that Commerce should have applied “partial” AFA because Mukand’s deficiency did not infect the entire record. The Trade Court noted that the use of partial AFA may be appropriate to fill gaps in a record that otherwise contains usable data and is incomplete with respect to only a discrete category of information. 5 In contrast, Commerce applies total AFA when none of the reported data is reliable or usable because, for example, the data contains pervasive and persistent deficiencies that cut across the entire record. In such situations, Commerce applies an adverse inference to all of the respondent’s sales covered by the relevant antidumping duty order. 6

The Trade Court held that Commerce did not err in applying total AFA against Mukand because Mukand’s failure to provide size-specific cost information rendered its responses so incomplete that they could not be used without undue difficulty. Without size-specific cost information, Commerce could not conduct an adequate sales-below-cost test, accurately calculate a difference-in-merchandise adjustment for size, or arrive at an accurate constructed value for any of Mukand’s sales. As the Trade Court noted, the absence of information so vital to the antidumping determination rendered Mukand’s responses too incomplete for Commerce to calculate a reliable margin. The Trade Court thus affirmed Commerce’s application of total AFA.

Mukand appealed the Trade Court’s decision to this court. We have jurisdiction pursuant to 28 U.S.C. § 1295(a)(5).

II

We review decisions of the Trade Court de novo and apply anew the same *1306 standard used by the Trade Court. 7 Commerce’s antidumping determinations are reviewed for substantial evidence. 19 U.S.C. § 1516a(b)(l)(B)(i). Substantial evidence is defined as “more than a mere scintilla,” as well as evidence that a “reasonable mind might accept as adequate to support a conclusion.” 8 Our review is limited to the record before Commerce in the particular review proceeding at issue and includes all evidence that supports or detracts from Commerce’s conclusion. 9 An agency finding may still be supported by substantial evidence even if two inconsistent conclusions can be drawn from the evidence. 10

A

Mukand argues that Commerce erred in applying AFA because it “fully answered Commerce’s specific questions on this issue and so acted to the best of its ability.” Appellant Br. 4. According to Mukand, the wording of Commerce’s questionnaires suggested that it should submit size-specific cost information only if it could be obtained in a reasonable and verifiable way. We disagree.

Commerce’s decision to resort to facts otherwise available and to apply an adverse inference against Mukand is supported by substantial evidence. Mukand’s narrow focus on the exact wording of Commerce’s questionnaires ignores the main import of Commerce’s repeated attempts to obtain size-specific cost information. Commerce requested this information from Mukand on five separate occasions— in the initial questionnaire and in four supplemental questionnaires. In each of the supplemental questionnaires, Commerce explained why it was unsatisfied with Mukand’s response and reiterated both the type of information it needed and why it was important. Commerce even went so far as to provide Mukand with a sample chart to complete and encouraged Mukand to reach out if it needed assistance or additional clarification. Commerce further warned Mukand that its continued failure to provide the requested information may force Commerce to resort to facts otherwise available. Commerce was thus justified in resorting to facts otherwise available based on Mukand’s repeated failures to provide the requested size-specific cost data.

Commerce’s decision to adopt an adverse inference against Mukand is also supported by substantial evidence. Commerce reasonably concluded that Mukand failed to cooperate to the best of its ability when responding to Commerce’s requests for information. To avoid the risk of an adverse inference, respondents must take reasonable steps to maintain full and complete records and put forth maximum effort to investigate and obtain all requested information. Nippon Steel, 337 F.3d at 1382. Mukand thus had a duty to account for size-specific cost differences in its responses using reasonably available information or explain why such information was not available. As Commerce highlighted in its Issues and Decision Memo *1307 randum, Mukand failed to provide any of the requested information:

We provided Mukand with several opportunities to submit factual information to support its claim that cost differences between sizes were insignificant. We requested the weight to length conversion factors, rolling times, and separate conversion cost fields for the rolling and other finishing stages of production in an attempt to analyze the potential significance or insignificance of cost differences due to size. Mukand provided none of the requested data.

J.A. 1815. Product-specific information is a fundamental element in the dumping analysis, and it is standard procedure for Commerce to request product-specific data in antidumping investigations. It was thus reasonable for Commerce to expect from Mukand more accurate and responsive answers to the questionnaire. Relevant here is that Mukand evaded providing a direct response to Commerce’s specific questions, and it was not until Mukand responded to the third supplemental questionnaire that it informed Commerce it did not maintain cost accounting records on the basis of product size. Indeed, Mukand was suddenly able to provide the requested information after Commerce published its preliminary results and applied an adverse inference despite repeated claims that the data was not reasonably available.

We agree with Commerce that Mukand’s change in position further demonstrated its failure to cooperate to the best of its ability. This circumstance points to why the use of an adverse inference is a useful tool in antidumping determinations. The statement of administrative action on the Uruguay Round Agreements Act (“URAA”) provides that the purpose of the adverse inference provision is to encourage future cooperation and ensure that a respondent does not obtain a more favorable antidumping rate by failing to cooperate. 11 Absent the threat of an adverse inference, respondents could sit out the preliminary phase of the investigation and submit requested data only when the resulting preliminary antidumping rates are higher than the rate that would have been established with the withheld data. Hence, we hold that Commerce’s decision to apply AFA against Mukand is supported by substantial evidence.

B

Mukand argues in the alternative that Commerce erred in applying total AFA. According to Mukand, Commerce should have applied partial AFA because it complied with the majority of Commerce’s requests for information on U.S. and home market sales and costs. Again, we do not agree.

Commerce’s decision to apply total AFA is supported by substantial evidence. Commerce noted that the “requirement to report product-specific sales and cost data is one of the most basic and significant requirements in performing the dumping analysis and margin calculation.” J.A. 1808. Hence, Mukand’s refusal to break down its cost information by product size prevented Commerce from conducting an adequate sales-below-cost test, accurately calculating a difference-in-merchandise adjustment for size, or arriving at an accurate constructed value for any of Mukand’s sales. Commerce thus reasonably concluded that Mukand’s submissions were so *1308 incomplete that they could not be used without undue difficulty.

Contrary to Mukand’s argument, the deficiencies in its responses were not limited to a discrete category of information. As Commerce noted, Mukand assigned the “same amount of conversion costs per kilogram of bar produced, irrespective of the final size of the product produced.” J.A. 1604. Mukand thus premised all of its production cost data on the assumption that product size is not a significant cost factor—an assumption it failed to support. In general, use of partial facts available is not appropriate when the missing information is core to the anti-dumping analysis and leaves little room for the substitution of partial facts without undue difficulty. 12 Without cost data broken down by product size, Commerce was unable to differentiate between different types of steel bar products and could not calculate an accurate constructed value for any of Mukand’s products. We therefore hold that Commerce’s reliance on total AFA is supported by substantial evidence.

Ill

For the reasons set forth above, we affirm the decision of the Trade Court.

AFFIRMED