Risen Energy Co., Ltd. v. United StatesRisen Energy Co., Ltd. v. United States
ALEXANDRA H. SALZMAN, DeKieffer & Horgan, PLLC, Washington, DC, argued for plaintiff-appellant. Also represented by JAMES KEVIN HORGAN, GREGORY S. MENEGAZ.
ASHLEY AKERS, Commercial Litigation Branch, Civil Division, United States Department of Justice, Washington, DC, argued for defendant-appellee. Also represented by REGINALD THOMAS BLADES, JR.,
JONATHAN STOEL, Hogan Lovells US LLP, for plaintiffs-appellees. Also represented by LINDSAY BROWN, CRAIG A. LEWIS, NICHOLAS SPARKS.
Before DYK, STOLL, and STARK, Circuit Judges.
Opinion for the court filed by Circuit Judge DYK.
Opinion concurring-in-part and dissenting-in-part filed by Circuit Judge STARK.
DYK, Circuit Judge.
This appeal concerns the Sixth Administrative Review of an antidumping order concerning crystalline silicon photovoltaic cells (commonly referred to as “solar cells“) from the People‘s Republic of China.
Appellant Risen Energy Co., Ltd. (“Risen“) is a Chinese exporter of solar cells, whose products are subject to the antidumping order imposed by the Department of Commerce (“Commerce“). Risen was selected as a mandatory respondent for such review. Since China is a nonmarket economy to calculate a dumping margin, Commerce used surrogate values from Malaysia for computing normal values (home market price) for the Sixth Administrative Review. The Court of International Trade (“Trade Court“) sustained Commerce‘s surrogate value calculations for Risen‘s physical inputs and its surrogate financial ratio calculations. See Risen Energy Co. v. United States, 569 F. Supp. 3d 1315, 1326 (Ct. Int‘l Trade 2022) (Risen I); Risen Energy Co. v. United States, 611 F. Supp. 3d 1384, 1389-94 (Ct. Int‘l Trade 2022) (Risen II). Risen appeals, challenging Commerce‘s surrogate value calculations for its backsheet and ethyl vinyl acetate (“EVA“), and Commerce‘s overhead ratio calculation. Because Commerce‘s selections of surrogate values for Risen‘s backsheet and EVA inputs were supported by substantial evidence, but Commerce‘s surrogate overhead ratio calculation was not, we affirm in part, vacate in part, and remand.
BACKGROUND
I
The government imposes antidumping duties on foreign merchandise sold “in the United States at less than its fair value.” Changzhou Trina Solar Energy Co. v. United States, 975 F.3d 1318, 1321 (Fed. Cir. 2020) (quoting
“Normal value” generally will be “the price at which the foreign like product is first sold . . . for consumption in the exporting country, in the usual commercial quantities and in the ordinary course of trade and, to the extent practicable, at the same level of trade as the export price or constructed export price[.]”
The factors of production that Commerce must value include, but are not limited to, “hours of labor required, quantities of raw materials employed, amounts of energy and other utilities consumed, and representative capital cost, including depreciation.”
By identifying a surrogate country and surrogate values for the factors of production, Commerce approximates “what a non-market economy manufacturer might pay in a market economy setting.” Shakeproof Assembly Components Div. of Ill. Tool Works v. United States, 268 F.3d 1376, 1382 (Fed. Cir. 2001).
II
In March 2019, Commerce initiated this Sixth Administrative Review of an earlier antidumping order covering solar cells from China for a period of review from December 1, 2017, through November 30, 2018. See Initiation of Antidumping Duty and Countervailing Duty Administrative Reviews, 84 Fed. Reg. 9300 (Dep‘t of Com. Mar. 14, 2019). Risen was selected as a mandatory respondent. See Crystalline Silicon Photovoltaic Cells, Whether or Not Assembled Into Modules, From the People‘s Republic of China: Preliminary Results, 85 Fed. Reg. 7532 (Dep‘t of Com. Feb. 10, 2020); see also
China is a nonmarket economy, so Commerce was required to select a primary surrogate country and individual surrogate values for Risen‘s various inputs. See
In October 2020, Commerce published the final results of its administrative review. See Crystalline Silicon Photovoltaic Cells, Whether or Not Assembled Into Modules, From the People‘s Republic of China: Final Results, 85 Fed. Reg. 62,275 (Dep‘t of Com. Oct. 2, 2020). Commerce selected Malaysia as the primary surrogate country, and it used import data from certain of the Malaysia Harmonized Tariff Schedule (“HTS“) categories applicable to “plates and sheets” to value Risen‘s backsheet and EVA inputs,2 rejecting Risen‘s position that Commerce should use the import data related to the HTS categories that apply to “film” instead. Commerce additionally used the 2018 financial statement from Malaysian solar cell producer Hanwha Q Cells Malaysia to calculate surrogate financial ratios, including overhead. Risen filed suit in the Trade Court, arguing that Commerce‘s determinations
were not supported by substantial evidence. See Risen I, 569 F. Supp. 3d at 1320.
The Trade Court initially agreed with Risen that Commerce‘s valuations of its backsheet and EVA inputs under the HTS
However, the Trade Court sustained Commerce‘s surrogate financial ratio calculation for overhead, despite its reservations about Commerce‘s rationale, finding “Commerce‘s reasoning could be clearer.” Id. at 1332. Nonetheless, the court could “reasonably discern from Commerce‘s citation” to two notes within the 2018 Hanwha financial statement that Commerce‘s allocation methodology was compliant with a standard in the International Financial Reporting Standards (“IFRS“) applicable to inventories. Id. at 1333-34.
On remand, Commerce reopened the record to further substantiate its choice of HTS categories for Risen‘s backsheet and EVA inputs. See Risen II, 611 F. Supp. 3d at 1391-93. Commerce placed on the record abstracts from two standards of the American Society for Testing and Materials (“ASTM“), ASTM D4801 and ASTM D6988, relating to film and sheet. See id. at 1392-93. Commerce continued to value Risen‘s backsheet and EVA with the HTS categories applicable to “sheet,” as opposed to the “film” categories Risen wanted, because in Commerce‘s view, the ASTM standards define “sheet” as materials with a thickness greater than 0.25 mm, and both of Risen‘s inputs meet that definition. The Trade Court sustained Commerce‘s remand determinations, concluding that they were reasonable in light of the new definition provided by the ASTM standards. See id. at 1392-94.
Risen appeals. We have jurisdiction pursuant to
DISCUSSION
We review Commerce‘s determinations using the same standard as the Trade Court—that is, whether those determinations are “unsupported by substantial evidence on the record, or otherwise not in accordance with law[.]”
I
We first address the issue of Commerce‘s classification of Risen‘s backsheet and EVA inputs under the Malaysia HTS categories applicable to “sheet” instead of “film.”
In assessing factors of production, Commerce is obligated to use the “best available information” from the surrogate country to identify an exporter‘s inputs and assign surrogate values to them. See Shakeproof Assembly Components, 268 F.3d at 1381 (quoting
Risen agrees that valuation using import data from HTS categories is appropriate but objects to Commerce‘s choice of HTS “sheet” categories for its backsheet and EVA inputs, arguing that Commerce could not reasonably rely upon the ASTM standards because those standards do not bear a “reasonable relationship to the inputs in question or defin[e] film and sheet.” Appellant Br. 16. Instead, Risen argues, Commerce should have used the product specifications and marketing materials Risen placed on the record to select HTS categories. We disagree.
We have recognized that, under the statute, industry standards are highly relevant to determining what merchandise is subject to an antidumping duty order. For example, in ArcelorMittal Stainless Belgium N.V. v. United States, 694 F.3d 82 (Fed. Cir. 2012), we stated that “[b]ecause the primary purpose of an antidumping order is to place foreign exporters on notice of what merchandise is subject to duties, the terms of an order should be consistent, to the extent possible, with trade usage.” Id. at 88. More recently, in Saha Thai Steel Pipe Public Co. v. United States, 101 F.4th 1310 (Fed. Cir. 2024), we affirmed Commerce‘s scope ruling in part because the fact that the exporter‘s merchandise was certified “in compliance with ASTM specifications” supported Commerce‘s reading of the plain language of the order to include the exporter‘s products. Id. at 1327. Moreover, Commerce‘s regulations specifically direct the agency to consult industry standards where the scope of an antidumping duty order is ambiguous. See
The first standard relied upon by Commerce, ASTM D6988, is titled “Standard Guide for Determination of Thickness of Plastic Film Test Specimens,” and “covers the determination of the thickness of plastic films where the thickness is used directly in determining the results of tests for various properties.” J.A. 7417-18. The standard defines “film” as an “optional term for sheeting having a nominal thickness no greater than 0.25 mm[.]” J.A. 7418. The second standard, ASTM D4801, is titled “Standard Specification for Polyethylene Sheeting in Thickness of 0.25 mm (0.010 in.) and Greater,” and “covers the requirements for extruded (cast or blown) and compression-molded sheeting made from low-, medium-, and high-density polyethylenes and copolymers[.]” J.A. 7421-22. Taken together, Commerce understood these standards as setting forth a distinction between polyethylene “sheet” and “film,” which turned on the thickness of the material. The fact that the thickness of Risen‘s backsheet and EVA inputs fell within the “sheet” definition (and not the “film” definition) led Commerce to conclude that those inputs were more appropriately categorized under the HTS categories for “sheet” rather than “film.”
Risen‘s arguments to the contrary are not persuasive. First, we reject Risen‘s argument that the ASTM standards should be disregarded because they do not mention Risen‘s inputs or the solar industry more generally. The standards cited by Commerce are not limited to particular products and appear to cover a broad array of plastic materials. Risen offers no alternative industry standards, and we conclude that it was reasonable for Commerce to determine that the industry standards it introduced constituted the best available information to distinguish between “sheet” and “film.”
As the Trade Court indicated, the product specifications and marketing materials submitted by Risen do not provide better information than the definitions found in the ASTM standards. See Risen II, 611 F. Supp. 3d at 1392-93. Beyond identifying backsheet and EVA as “film,” the materials submitted by Risen do not demonstrate why those inputs are considered “film” or shed light on the distinction between “sheet” and “film.” Further, at least as to EVA, these materials use the terms “sheet” and “film” interchangeably, such that the use of the term “film” in those materials is stripped of any definitional quality that it might otherwise hold.3
Nor are we persuaded that Commerce erred by selecting new HTS categories for Risen‘s inputs in this review. “Commerce may change its conclusions from one review to the next based on new information and arguments, as long as it does not act arbitrarily and it articulates a reasonable basis for the change.” Qingdao Sea-Line, 766 F.3d at 1387. Here, Commerce provided
sufficient explanation for why the HTS categories for “sheet” were preferable to the HTS categories for “film” in this Sixth Administrative Review.
Commerce explained that it chose the HTS category applicable to “sheet” for Risen‘s backsheet, but did not do so in prior reviews, because the ASTM standards were placed on the record here, whereas Commerce did not have the benefit of those standards in previous reviews. That is sufficient to sustain Commerce‘s determination, as Risen has not shown that “Commerce consistently followed a contrary practice in similar circumstances and provided no reasonable explanation for the change in practice.” Consol. Bearings Co. v. United States, 348 F.3d 997, 1007 (Fed. Cir. 2003).
We further agree with the Trade Court that Commerce‘s characterization of EVA as “sheet” as opposed to “film” was not inconsistent with Commerce‘s selection of HTS categories in prior administrative reviews where Thailand was selected as the surrogate country. Risen II, 611 F. Supp. 3d at 1393-94. In those proceedings, Commerce used an “other” Thai HTS category that covered “plates, sheets, film, foil and strips of polymers of ethylene.” J.A. 19. Commerce explained that Thailand‘s tariff schedule did not distinguish between “sheet” and “film” of polyethylene, but instead grouped them together in a single category. Malaysia, by contrast, had separate categories for polyethylene “sheet” and “film.” Using Malaysia‘s more precise HTS categories, Commerce determined
We conclude that Commerce‘s classifications of Risen‘s backsheet and EVA inputs under the Malaysia HTS categories applicable to “sheet” were reasonable and supported by substantial evidence. We affirm the Trade Court on this point. We further reject Risen‘s alternative argument that Commerce should have calculated surrogate values by averaging the values applicable to the HTS categories for “sheet” and “film” for each of its inputs because we do not find the record ambiguous on this issue.
II
The second issue is whether Commerce properly characterized certain unidentified costs in the 2018 Hanwha financial statement as overhead.
Separate from physical inputs, Commerce must calculate surrogate financial ratios for manufacturing overhead, selling, general, and administrative expenses, and profit. See
Although Commerce enjoys discretion in how it calculates surrogate financial ratios, see Fujitsu Gen. v. United States, 88 F.3d 1034, 1045 (Fed. Cir. 1996), including in how it “valu[es] the factors of production on which factory overhead is based[,]” Magnesium Corp. of Am. v. United States, 166 F.3d 1364, 1372 (Fed. Cir. 1999), “[a]n overriding purpose of Commerce‘s administration of the antidumping laws is to calculate dumping margins as accurately as possible.” Yangzhou Bestpak Gifts & Crafts Co. v. United States, 716 F.3d 1370, 1379 (Fed. Cir. 2013) (citing Rhone Poulenc, Inc. v. United States, 899 F.2d 1185, 1191 (Fed. Cir. 1990)). Commerce is required to demonstrate that its calculations are supported by “substantial evidence” and otherwise “in accordance with law[.]”
Risen primarily argues that Commerce‘s surrogate overhead ratio calculation is not supported by substantial evidence because Commerce‘s calculation is not supported by the financial statement on which Commerce relies. The Trade Court observed that Commerce‘s explanation for why it determined that unidentified costs were allocable to overhead “could be clearer,” but nonetheless sustained Commerce‘s determination. Risen I, 569 F. Supp. 3d at 1332. We think Commerce‘s
Commerce began calculating its overhead ratio by selecting the 2018 Hanwha financial statement as the best available information from which to derive its surrogate ratios. Risen does not object to Commerce‘s selection of this financial statement as the best available information. Using that statement, Commerce calculated a final overhead ratio of 21.70 percent for Risen by dividing what it deemed “overhead costs” by the costs for materials, labor, and energy (“MLE“).5
Commerce began its analysis with Hanwha‘s costs of goods sold, which was 2,003,400 Malaysian ringgits.6 From that total, Commerce sought to identify what proportion of the costs of goods sold represented MLE. Note 17 to the financial statement explained that, of the total costs of goods sold, 1,648,000 ringgits are attributable to “inventories.” Commerce considered these inventory costs to be roughly synonymous with Hanwha‘s total MLE costs, based on Note 2.12 of the financial statement. That Note provided as follows:
Costs incurred in bringing the Inventories to their present location and condition are accounted for as follows: . . .
Finished goods and work-in-progress: costs of direct materials and labour and a proportion of manufacturing overheads based on normal operating capacity.
J.A. 6666. Commerce concluded that Note 2.12‘s reference to a “proportion of manufacturing overheads based on normal operating capacity” was “a reference largely to energy costs” and not production overhead. J.A. 7165-66. Based on this assumption, after minor adjustments to account for the change in goods in inventory, Commerce determined that Hanwha‘s overhead expenses totaled essentially the difference between its costs of goods sold and inventory costs.
Commerce concluded that the difference had to be overhead (other than energy costs) because Commerce had already identified MLE costs (as included in inventories) and “the income statement explicitly identifies the sales, general, administrative, and interest costs . . . as separate line items.” J.A. 7132. Stated differently, Commerce found that because Hanwha‘s financial statement “specifies that MLE costs are included in the ‘Inventories’ portion of the ‘[c]ost of sales,’ . . . the remaining, unidentified costs of sales were not MLE but rather overhead.” Appellee Br. 37.
Nonetheless, Commerce argues that its calculation is supported by substantial evidence because the Hanwha financial statement was composed in compliance with the IFRS standard applicable to inventories.7 The standard does not support Commerce‘s approach. The standard explains that inventories should include “a systematic allocation of fixed and variable production overheads that are incurred in converting materials into finished goods.” Int‘l Fin. Rept. Standards Found., IAS 2 Inventories ¶ 12 (Mar. 2024) (“IAS 2“).
Given the similarity in language, we understand the standard‘s use of the term “production overheads” to be synonymous with the term “manufacturing overheads” in Note 2.12 of the Hanwha financial statement. As we have explained, and as Note 2.12 confirms, manufacturing overheads already were included in Hanwha‘s inventory costs. J.A. 6666.
We are not persuaded by Commerce‘s argument that it was reasonable to understand the phrase “a proportion of manufacturing overheads” in Note 2.12 to be “a reference largely to energy costs” included in inventories. J.A. 7165-66; Oral Arg. at 31:34-45. At no point has Commerce provided an adequate explanation or, indeed, any explanation for why it drew this conclusion. Further,
nothing in the Hanwha financial statement or the IFRS standard can be read to suggest that proportional “production overheads” are coterminous with, or even largely made up by, a company‘s energy costs, as Commerce asks us to conclude.
Finally, Commerce suggests that under the IFRS standard, inventory costs do not include fixed overhead costs and that those fixed expenses would be allocable to overhead separately from inventories in the costs of goods sold. Appellee Br. 35. This is an incorrect interpretation of the standard, which requires “fixed and variable production overheads that are incurred in converting materials into finished goods” to be allocated on a systematic basis to the cost of inventories. IAS 2 ¶ 12 (emphasis added). To be sure, the standard states that “administrative overheads that do not contribute to bringing inventories to their present location and condition[,] and selling costs” are to be excluded from the inventories total. IAS 2 ¶ 16. But the fact that administrative overheads and selling costs are excluded from the inventories total does not mean that those costs are additional overhead included in the costs of goods sold, as Commerce suggests. Indeed, the Hanwha financial statement specifically identifies the company‘s “[s]elling and administrative expenses” separately from cost of goods sold, consistent with the standard. J.A. 6648.
On the present record, Commerce‘s allocation of the remaining 257,063 ringgits in
CONCLUSION
We sustain the Trade Court‘s affirmance of Commerce‘s categorization of Risen‘s backsheet and EVA inputs under the Malaysia HTS categories applicable to sheet. We vacate the Trade Court‘s decision sustaining Commerce‘s surrogate financial ratio calculation for overhead. We direct the Trade Court to remand the matter regarding the overhead issue to Commerce for further proceedings consistent with this opinion.
AFFIRMED IN PART, VACATED AND REMANDED IN PART
COSTS
No costs.
STARK, Circuit Judge, concurring-in-part and dissenting-in-part.
I agree with the majority that Commerce‘s surrogate values for Risen‘s backsheet and EVA inputs are supported by substantial evidence. Accordingly, I join in that portion of the majority opinion. See Maj. at 7-12. However, I believe that Commerce‘s surrogate financial ratio calculations are also supported by substantial evidence. Therefore, I would affirm the judgment of the Court of International Trade (“Trade Court“), which reached this same conclusion, in its entirety.
I
As an initial matter, I choose not to join the majority on the surrogate financial ratios issue because the majority faults Commerce on grounds that the appellant, Risen, has not raised. And it provides relief, a remand, that Risen never requested.
A
The majority is vacating and remanding because “Commerce‘s approach is so unclear that it is insufficient.” Maj. at 13; see also id. at 16 n.7 (“Commerce‘s explanation is inadequate, regardless of when it was first raised.“). Yet at no point, at either the Trade Court or here, has Risen, the party that brings this case to us, argued that Commerce‘s approach is unclear.
Risen understands what Commerce did in calculating the surrogate financial ratios and why it did so. It just disagrees with Commerce – and, before us, insists that Commerce‘s determination is not supported by substantial evidence. As Risen accurately summarizes, “Commerce explained that it understood that labor and energy were already included in the [materials labor and energy (“MLE“)] denominator because the ‘Inventories’ line item [in the Hanwha Q Cells Malaysia (“Hanwha“) financial statement] included these expenses.” Open. Br. at 22 (emphasis added). The sole issue Risen presses has nothing
The majority claims that it is “not going beyond the parties’ arguments” and is “simply holding that the explanation given was insufficient because it lacked substantial evidence.” Maj. at 13 n.4 (emphasis added). But the majority points to nowhere that Risen actually argued either that Commerce‘s explanation was insufficient or that the purported lack of substantial evidence has anything to do with the clarity of Commerce‘s explanation. Risen‘s briefing makes clear it did not raise either of these points. See, e.g., Open. Br. at 6 (“Commerce calculated the surrogate financial ratios in a manner unsupported by the record. . . . The resulting calculation significantly overstated the overhead costs [and was] [c]ontrary to accounting principles, Commerce‘s usual understanding of ratio calculations, and the information in the financial statement itself.“); id. at 20 (“Commerce‘s calculation of the financial ratios are inaccurate [and] does not reasonably interpret the record information in the financial statement . . . .“); id. at 21-22 (“[Commerce‘s allocation of remaining costs to overhead, instead of MLE,] is contrary to Commerce‘s practice and contrary to the notes of the [Hanwha] statement.“); Reply Br. at 7 (“[Commerce‘s] allocation is contrary to the notes of the statement and accounting principles.“); id. at 11 (“Commerce has calculated inaccurate ratios not based on substantial evidence . . . .“).
While “Commerce is required to demonstrate that its calculations are supported by ‘substantial evidence,‘” Maj. at 12 (quoting
“In our adversary system . . . we follow the principle of party presentation,” which instructs us to “rely on the parties to frame the issues for decision and assign to courts the role of neutral arbiter of matters the parties present.” Greenlaw v. United States, 554 U.S. 237, 243 (2008). Hence, the Supreme Court has been clear that, with rare exceptions, “in both civil and criminal cases, in the first instance and on appeal,” id., we should “decide only questions presented by the parties,” United States v. Sineneng-Smith, 590 U.S. 371, 376 (2020). We have ourselves on multiple occasions recognized this constraint on our review, including very recently in Astellas Pharma, Inc. v. Sandoz Inc., 117 F.4th 1371, 1377 (Fed. Cir. 2024), where we vacated a judgment of patent invalidity because “the district court disregarded the longstanding principle of party presentation and, in doing so, abused its discretion.”
The majority does not attempt to show the presence of circumstances that could make it “appropriate” for us “to take a ‘modest initiating role’ in the shape of the
B
The majority says that Risen “explicitly sought a remand.” Maj. Op. at 13 n.4. I disagree.
The only reference Risen has made to a possible remand is an aside in the middle of its reply brief, which is untimely. See In re Cygnus Telecomms. Tech., LLC Patent Litig., 536 F.3d 1343, 1356 (Fed. Cir. 2008) (holding appellant forfeited argument for reversal of summary judgment by failing to raise request in opening brief); SmithKline Beecham Corp. v. Apotex Corp., 439 F.3d 1312, 1320 (Fed. Cir. 2006) (“Our law is well established that arguments not raised in the opening brief are waived.“). In any event, even the tangential, belated reference to a remand cannot fairly be read as an actual request. Instead, in the course of complaining that the IFRS standards only first came up at the Trade Court, Risen observes that “[t]his has made briefing at the Court more cumbersome” and then adds: “This alone should require remand to Commerce to consider as it is an agency role to do in the first instance on all issues.” Reply Br. at 7. But Risen never asks for such a remand – and most certainly not on the grounds of a purportedly unclear explanation by Commerce.
Rather, Risen is consistent and explicit about its sole goal on appeal, which is reversal of the Trade Court‘s entry of judgment for the government. In both its Opening and Reply Briefs, under sections headed “Conclusion and Statement of Relief Sought,” it writes a single, identical sentence:
In light of the foregoing, Plaintiff-Appellant requests that this Court enter judgment in its favor and find Commerce‘s determination of the best available information for backsheet and EVA and Commerce‘s financial ratio calculation are not supported by substantial evidence.
Open. Br. at 27; Reply Br. at 12 (same).
Therefore, I would limit our review to deciding whether to reverse or affirm the Trade Court‘s judgment, which I will turn to now.
II
The question actually presented in this appeal is far easier to state than it is to answer. As context, it is undisputed that because Risen is a Chinese company, and China has a nonmarket economy, Commerce had to calculate a dumping margin by using surrogate values – here, from Malaysia – to estimate the “normal value” at which Risen would sell its products in its home market (China). There is no challenge before us to Commerce‘s decision to use the financial statement of a Malaysian manufacturer of solar cells, Hanwha, as the best available information from which to calculate the necessary surrogate financial ratios, such as overhead ratio. There is also no disagreement between the parties that what Commerce did, as pertinent to this appeal, was to start with Hanwha‘s reported “cost of sales,” RM2,003,400 (Malaysian ringgits), subtract certain costs appearing on other lines in Hanwha‘s statement – including inventories, which note 17 of the Hanwha financial statement reports as RM1,648,000 for 2018, adjusted to RM1,646,244 for the change in finished
materials, labor, and energy – which would put it in the denominator of the ratios.
In my view, Commerce had substantial evidence for its decision to allocate the unidentified costs to overhead. That becomes clear when looking more closely at what Commerce did and why.
The issue of how to allocate the unidentified costs was first addressed by Commerce in its Preliminary Results issued in January 2020. See J.A. 6838-39 (calculating 6.29% overhead ratio). In those Preliminary Results, Commerce used a “constructed MLE” for Hanwha. J.A. 7131. Both the petitioner in the administrative review (SolarWorld Americas Inc.) and Risen sought adjustments to the Preliminary Results and, consequently, Commerce, having not done so before, considered two notes in the Hanwha financial statement: note 2.12, describing what Hanwha included in its reported “Inventories,” and note 17, reporting figures for 2017 and 2018 “Inventories.” J.A. 6666, 6688. Commerce found that these notes, which I describe in more detail below, “specifically identified direct product costs,” making the reported Inventories figure “a more appropriate reflection of MLE” than the constructed figure on which Commerce had earlier relied. J.A. 7131. Based on this new insight, in its Final Results, issued in October 2020, Commerce decided to “treat[] the difference between the total manufacturing costs and MLE” – that is, the unidentified costs – “as overhead costs.” J.A. 7132; see also J.A. 7133 (“[B]ased on the information contained in the [Hanwha] financial statements, we have concluded that the remaining unidentified costs are overhead costs.“); id. (calculating 21.70% overhead ratio).2 Risen disagreed with Commerce,
insisting “it is Commerce practice to classify unidentified costs in financial statements as [MLE] costs” and not overhead. J.A. 7133. Commerce responded that it was “unaware of any such practice.” Id.
Thereafter, in a November 2, 2020 memo responding to allegations of ministerial errors in the Final Results, Commerce stated that it had calculated the surrogate financial ratios based “solely on the financial statements of Hanwha.” J.A. 7165. Commerce explained:
In doing so, we determined that labor and energy, as well as material costs, were included in the category identified as “inventories recognized as an expense in cost of sales.” We made this conclusion based in part on the statement from the financial statements that “inventories” include “costs of direct materials and [labor] and a proportion of manufacturing overheads based on normal operating capacity.” We believed the “proportion of manufacturing overheads based on normal operating capacity” to be a reference largely to energy costs.
J.A. 7165.
In other words, Commerce read Hanwha‘s financial statement as disclosing that
Id.; see also id. (Commerce stating it “made a methodological decision based on record information“).3
Commerce‘s reading of the Hanwha financial statement – that Hanwha‘s reported inventories included all MLE, so the unidentified costs cannot also be MLE but instead should be allocated to overhead – was reasonable and supported by substantial evidence. Commerce based its conclusion primarily on note 2.12 of Hanwha‘s statement. Note 2.12, entitled “Inventories,” begins by referencing “Costs incurred in bringing the inventories to their present location and condition,” J.A. 6666, which Commerce reasonably understood to be a category of costs that includes the cost of energy to make and move the items found in inventory. The note then continues:
[These costs] are accounted for as follows:
Raw materials: purchase costs are derived by using the weighted average cost method.
Finished goods and work-in-progress: costs of direct materials and labour and a proportion of manufacturing overheads based on normal operating capacity.4 These costs are assigned by using the weighted average cost method.
J.A. 6666 (emphasis added). In sum, then, as can be seen, the note expressly references materials and labor, and implicitly describes energy.
Thus, I agree with the Trade Court that it was “reasonable” for Commerce to have “relied in part” on note 2.12 “as evidence that labor and energy costs are included in the valuation of” Hanwha‘s inventories. J.A. 53-54. I further agree with the Trade Court that this understanding of the Hanwha statement gains further support from the fact that the statement was prepared in accordance with the IFRS. J.A. 54; see also J.A. 6653 (“The financial statements of the [Hanwha] Group and of the Company have been prepared in accordance with . . . [the IFRS].“). IFRS Standard IAS2, which “provid[es] guidance for determining the cost of inventories and the subsequent
I agree with this analysis of the Trade Court, which (along with what I have set out here) describes the substantial evidence basis for Commerce‘s decision. While, of course, Commerce could have done a better job explaining itself, it does not follow that, as my colleagues conclude, Commerce‘s decision was “based on nothing more than guesswork or speculation.” Maj. at 17. Commerce‘s finding was grounded in the record evidence and its explanation of its reasoning was adequate to enable appellate review. The Trade Court was right to affirm.
III
Commerce confronted a complicated, case-specific fact question, calling on its expertise and experience with financial statements and accounting standards. It was a question that the government candidly acknowledges (as do I) has no “black-and-white answer.” Oral Arg. at 26:11-22. Reasonable minds could well differ as to whether the unidentified costs in the Hanwha financial statement should be allocated to MLE or to overhead. But “the possibility of drawing two inconsistent conclusions from the evidence does not prevent an administrative agency‘s finding from being supported by substantial evidence.” Consolo v. Fed. Mar. Comm‘n, 383 U.S. 607, 620 (1966). At best for Risen, that is the situation presented by this appeal.
Accordingly, we should affirm the Trade Court‘s affirmance of Commerce‘s determination of the surrogate financial ratios. Thus, I respectfully dissent from the majority‘s decision to remand this issue for further proceedings.