SeAH Steel Corp. v. United States , 764 F. Supp. 2d 1322 ( 2011 )

764 F.Supp.2d 1322 (2011)

SEAH STEEL CORPORATION, Plaintiff,
v.
UNITED STATES, Defendant, and
Bristol Metals, Defendant-Intervenor.

Slip Op. 11-33. Court No. 09-00248.

United States Court of International Trade.

March 29, 2011.

Troutman Sanders LLP (Donald B. Cameron; Julie C. Mendoza; R. Will Planert; Brady W. Mills; Mary S. Hodgins), Washington, DC, for Plaintiff.

Tony West, Assistant Attorney General; Jeanne E. Davidson, Director; Patricia M. McCarthy; Assistant Director, Commercial Litigation Branch, Civil Division, United States Department of Justice (Claudia Burke); Scott D. McBride, Office of the Chief Counsel for Import Administration, United States Department of Commerce, Washington, DC; for Defendant.

Schagrin Associates (Roger B. Schagrin; Michael J. Brown), Washington, DC, for Defendant-Intervenor.

OPINION & ORDER

CARMAN, Judge:

This case disputes the results of a Court-ordered remand of the Final Results of the 2006-2007 administrative review of the antidumping duty order covering certain steel products manufactured by SeAH, and now reviews Commerce's Final Results of Redetermination Pursuant to Court Remand ("Remand Redetermination"). The results of the initial administrative review were published as Certain Welded Stainless Steel Pipes from the Republic of Korea: Final Results of Antidumping Duty Administrative Review, 74 Fed.Reg. 31,242 (June 30, 2009) ("Final Results"), and incorporated by reference an accompanying Issues and Decision Memorandum (PR^[1] 74, "IDM"), setting forth the Department's analysis in detail. The Court remanded in part and affirmed in part the Final Results in a Slip Opinion on Plaintiff's USCIT R. 56.2 Motion challenging the Final Results. SeAH Steel Corp. v. United States, 34 CIT ___, 704 F.Supp.2d 1353 (2010). In that opinion, the Court granted Commerce's request for a voluntary remand so that the agency could reconsider the use of steel specification as well as grade when conducting its major input analysis for SeAH Steel Corporation ("Plaintiff" or "SeAH"). The Court also remanded to Commerce for more complete data and explanation regarding the methodology by which Commerce conducted the cost recovery test set out in section 773(b)(2) of the Tariff Act of 1930, as amended, 19 U.S.C. § 1677b(b)(2).^[2] The Court directed Commerce to provide additional data regarding the effect of its cost recovery methodology on the computation of Plaintiff's normal value, to explain its methodology more clearly, and to reconsider whether the methodology employed was consistent with the cost recovery section of the antidumping statute.

In the Remand Redetermination, Commerce determined, based on additional information provided by Plaintiff, to consider steel specification as well as steel grade in applying the major input analysis, and as a result decreased the dumping margin calculated for SeAH from 9.05 percent to 7.92 percent. (Id. at 1.) As for the cost recovery methodology, Commerce complied with the Court's requirement to provide a clear analysis of its methodology and additional data showing how it implemented that methodology. Commerce, upon reconsideration, determined that its methodology was consistent with the cost recovery statute and therefore continued to employ it.

The Court affirms the portion of the Remand Redetermination in which Commerce decided to consider both steel grade and steel specification in its major input analysis. However, the Court holds that the cost recovery methodology employed by Commerce violates requirements imposed by the statute, and therefore exceeds Commerce's authority. The case is remanded for Commerce to employ a cost recovery methodology consistent with the statute, as discussed in detail below.

I. Major Input Analysis

A. Background

The issue in this case regarding major input analysis stems from SeAH's purchase of stainless steel hot-rolled coils, an input in the production of stainless steel pipe, from affiliate POSCO. (Remand Redetermination at 25.) The amount paid to an affiliate for purchases of a major input, the market price of those inputs, and the affiliate's cost of producing major inputs are all relevant to the statutory and regulatory analysis conducted by Commerce. (Id. at 26.) SeAH thus reported these values to Commerce on the basis of both steel grade and, within each grade, steel specification. The specifications included "STS" (for steel coil meeting the Japanese Industrial Standards and Korean Industrial Standards, which are effectively identical) and "ASTM" (for steel meeting the American Society of Testing Materials standards). (Id.) In the Final Results, Commerce "considered differences among the steel grades reported by SeAH," but "aggregated price and cost data for the ASTM and STS specifications." (Id. at 27.) SeAH argued in the brief accompanying its USCIT R. 56.2 motion that disregarding specification was unlawful and improper, and Commerce requested a voluntary remand to gather further information and reexamine the question. (Id.) The Court granted the voluntary remand request. 704 F.Supp.2d at 1378-79.

B. Remand Results

On remand, Commerce gathered additional data from SeAH regarding its steel specifications and accepted argument from Defendant-Intervenor and SeAH as to whether specification distinctions should affect Commerce's major input analysis. (Remand Redetermination at 27.)

The new record data demonstrated "physical and chemical differences" between the ASTM and STS specifications within each grade of steel coil purchased by SeAH from POSCO. (Id. at 28.) Steel within a single grade was more expensive when produced to STS standards than when produced to ASTM standards due to these differences, especially the requirement that STS steel contain one percent more nickel, "a very expensive raw material." (Id.) Noting that production of STS-specified steel was more costly than production of ASTM-specified steel of the same grade, and that record evidence showed that sales of steel in each specification were correspondingly priced, the Department revised its major input calculations to take specification into account and recalculated SeAH's dumping margin. (Id. at 29.)

Defendant-Intervenor did not comment on the Department's draft of the Remand Redetermination (id. at 29-30), but has opposed the Department's consideration of specification in comments to this Court (Comments of Defendant Intervenor Bristol Metals to the Commerce Final Results of Redetermination Pursuant to Court Remand ("Bristol Comments") 8-15.) SeAH supports the Department's revised major inputs analysis and urges the Court to affirm that part of the Remand Redetermination. (Pl. SeAH Steel Corp.'s Comments on the U.S. Dep't of Comm's Sept. 17, 2010 Final Results of Redetermination Pursuant to Court Remand ("SeAH Comments") at 1.) In its response to the comments of the parties, Commerce argues that the Court should disregard Defendant-Intervenor's objections to the major inputs analysis because Defendant-Intervenor failed to comment on the draft Remand Redetermination and thus failed to exhaust its administrative remedies. (Def.'s Response to Comments Regarding the Remand Determination ("Commerce Response") at 2, 11-12.)

C. Analysis

Plaintiff and Defendant agree that the Remand Redetermination should be upheld on the issue of the major input analysis. The only party that contests this issue is Defendant-Intervenor Bristol Metals ("Bristol"). However, as Commerce points out, this Court takes a strict view of the administrative exhaustion doctrine. See Jiaxing Brother Fastener Co., Ltd. v. United States, 751 F.Supp.2d 1345, 1355-1357 (2010). This is not from caprice or blind adherence to custom, but rather due to statutory mandate. See 28 U.S.C. § 2637(d) (requiring that the Court of International Trade, "where appropriate, require the exhaustion of administrative remedies" in trade cases).

Strict construal of this statute conforms with guidance from the Court of Appeals for the Federal Circuit. Jiaxing, 751 F.Supp.2d at 1355-56 (quoting Corus Staal BV v. United States, 502 F.3d 1370, 1379 (Fed.Cir.2007)). Certain exceptions to the exhaustion requirement apply: where exhaustion would be "a useless formality," intervening legal authority "might have materially affected the agency's actions," the issue involves "a pure question of law not requiring further factual development," where "clearly applicable precedent" should have bound the agency, or where the party "had no opportunity" to raise the issue before the agency. Jiaxing, 751 F.Supp.2d at 1355-56 (citing cases). None of these exceptions applies here, however, and the Court therefore will not consider Bristol's opposition to the major input analysis in the Remand Redetermination.

There being no cognizable objections on this issue, and Commerce's approach being supported on its face by substantial evidence upon the record, the Remand Redetermination is affirmed as to the major input analysis.

II. Commerce's Implementation of the Cost Recovery Test

The central issue in dispute here is the method by which Commerce conducted the cost recovery test, a process set forth in the antidumping statute which describes the particular sales that Commerce is to include when it calculates the "Normal Value" ("NV") of the subject merchandise in the home market. The Court's role here is to "hold unlawful any determination, finding, or conclusion found . . . to be unsupported by substantial evidence on the record, or otherwise not in accordance with law." 19 U.S.C. § 1516a(b)(1). Because the Court determines that the Department's methodology violated the cost recovery analysis required by the antidumping statute, the Court remands this case for further action by Commerce, as described below, to correct the defect.

A. Statutory Scheme

The antidumping statute provides that home market sales may be disregarded when calculating NV if certain conditions are met.

If the administering authority determines that sales made at less than the cost of production—

(A) have been made within an extended period of time in substantial quantities, and

(B) were not at prices which permit recovery of all costs within a reasonable period of time,

such sales may be disregarded in the determination of normal value.

19 U.S.C. § 1677b(b)(1). Although the statute's language is permissive regarding Commerce's ability to disregard sales below cost of production—"such sales may be disregarded" (emphasis added)—the statute speaks in mandatory terms when describing whether sales were "at prices which permit recovery of all costs within a reasonable period of time":

If prices which are below the per unit cost of production at the time of sale are above the weighted average per unit cost of production for the period of investigation or review, such prices shall be considered to provide for recovery of costs within a reasonable period of time.

19 U.S.C.A. § 1677b(b)(2)(D) (emphasis added).^[3]

The dispute between the parties hinges on the second prong of the sales below cost test—whether Commerce improperly excluded a large number of SeAH's home market sales from the calculation of NV in determining that those sales were not at prices which "provide for recovery of costs within a reasonable period of time." 19 U.S.C. § 1677b(b)(2)(D).

B. Final Results and Court Remand Order

In the Final Results, which were before the Court when SeAH brought its USCIT 56.2 Motion, Commerce "adopted an alternative methodology that we determine complies with the statute's weighted average costs requirements while taking into account the distortive effect of significant cost changes." (IDM at 19.)

To account for these distortions, Commerce applied a quarterly indexing methodology within the cost recovery test. The quarterly indexing consisted of restating the costs of the various quarters into a constant cost figure based on costs at the end of the period of review, extending each quarter's restated costs by that quarter's production quantities, and calculating a restated annual average direct material cost by weight averaging the restated constant cost levels over the period of review. This figure was then restated back to each quarter's calculated cost level using the quarterly indices. (See IDM at 18-19, Def.'s 56.2 Mem. at 33-34.)

Commerce took the position that this adjustment to its ordinary methodology still resulted in a "weighted average per unit cost required by the statute." (Def.'s 56.2 Mem. at 33.) Commerce conceded that it would usually be proper to use an unadjusted weighted average for the cost recovery test; however, in the Department's view, the fluctuations in SeAH's production costs were great enough that they required quarterly indexing in order to neutralize distortion. (Id. at 33-34.) In Commerce's view, the quarterly indexing adjustment did not violate the statute because the Department continued to apply a period-wide weighted average per unit cost, although it was adjusted by incorporation of the quarterly indexing methodology. (Id. at 35.)

In resolving the 56.2 Motion, the Court reviewed the Department's methodology to observe whether it conformed with the 19 U.S.C. § 1677b(b)(2)(D) cost recovery test and found the record insufficient to allow a conclusion. The Court characterized the core question in this way: "did Commerce's quarterly indexing adjustments produce a ``weighted average per unit cost of production for the period of . . . review' as required by 19 U.S.C. § 1677b(b)(2)(D)?" SeAH, 704 F.Supp.2d at 1367. The Court concluded that it could not answer this question for two reasons. First, Commerce failed to clearly explain how its complex analysis neutralized the perceived distortion caused by significant input cost changes, and merely asserted without adequate explanation that its quarterly indexing adjustment did not transform "the weighted average per unit cost of production for the period of . . . review" required by the statute into something different. Id. at 1367-68 (quoting 19 U.S.C. § 1677b(b)(2)(D)). Second, Commerce provided exemplars of its methodology without providing sufficient data to allow the Court to evaluate how the Department's methodology affected the consideration of the actual sales under consideration. Id. at 1368.

As a result, the Court remanded the cost recovery issue to Commerce, with the following directions:

Commerce shall calculate the normal value of Plaintiff's home market sales using both the quarterly-indexed cost recovery test employed in the Final Results and using the ordinary weighted average per unit cost of production for the period of review. Second, Commerce shall include in the record the specific figures used in and resulting from these calculations. Third, in its remand redetermination, Commerce shall identify all those sales that are recoverable using the ordinary weighted average per unit cost of production for the period of review, but subject to exclusion under the quarterly indexed version of the cost recovery test. Fourth, Commerce shall explain which of the two methodologies it adopts to conduct the cost recovery test, stating in clear terms why the particular steps of that methodology are appropriate in the context of the requirements of 19 U.S.C. § 1677b(b)(2)(D).

Id. at 1370.

C. Remand Redetermination

The Court thanks Commerce for responding to the Court's remand order with a lucid and thorough explanation of its quarterly indexed cost recovery test, a detailed comparison of that methodology to an ordinary, unindexed cost recovery test, and a clear argument for the legitimacy of its methodology in terms of the statutory framework.

1. Explanation and Justification of Commerced Methodology

Commerce reasoned that an indexing methodology was necessary to account for significant changes in SeAH's "metal purchasing power" throughout the period of review due to significant spikes in the price of nickel. (Remand Redetermination at 6.) The Department, to mitigate the distortions it believed were caused by these price spikes, adapted a methodology it has used in relation to "economies experiencing high inflation where purchasing power levels differ remarkably during periods of high inflation" and also looked to the International Accounting Standards ("IAS"). (Id. at 8.) Specifically, IAS 29 explains that, in the case of rapid devaluation of currency in a hyperinflationary economy, financial reports must be restated to end-of-term values in order to be meaningful. Financial Reporting in Hyperinflationary Economies (IAS 29), in HENNIE VAN GREUNING & MARIUS KOEN, INTERNATIONAL ACCOUNTING STANDARDS: A PRACTICAL GUIDE 110 (2000). The Department, asserting that significant cost spikes during the period of review caused a "distortive impact" similar to high inflation, concluded that "by adjusting its cost recovery methodology in a manner similar to that applied in the high-inflation scenario. . . it could appropriately address the different purchasing power levels . . . in each quarter of the POR during periods of rapidly changing costs." (Remand Redetermination at 10.)

These concerns led Commerce to conclude that "without indexing, the agency would be required to weight-average costs from disparate periods which do not represent the same level of purchasing power," resulting in a "flawed normal value calculation" inconsistent with "generally accepted international accounting standards," the agency's treatment of high inflation contexts, and its goal of calculating as accurate a margin as possible. (Id. at 11.)

2. Conformance of Methodology With Statutory Framework

Responding to the Court's concerns as to whether the quarterly indexing methodology was consistent with the cost recovery statute, Commerce argued that its approach was legal because it satisfied what Commerce saw as the "three elements" of the 19 U.S.C. § 1677b(b)(2)(D) cost recovery test:

1) Does the Department's methodology cover the "period of review?"

2) Does the Department's methodology use a "per-unit cost of production?"

3) Does the Department's methodology calculate a "weighted average" of the total cost of production?

(Id. at 12.)

a. Coverage of the Period of Review

In 74 words, Commerce maintained that its methodology "uses all of the historic costs reported by SeAH" and therefore "covers the period of review" in satisfaction of "the first element" of § 1677b(b)(2)(D). (Id.)

b. Usage of a "Per-Unit Cost of Production"

Commerce next reasoned that its analysis used a "per-unit cost of production," and asserted that "[t]he fact that these costs were restated using a price index to account for known distortions in no way undermines this fact." (Id. at 13.) Commerce then explained the development of its quarterly price indexing as a six-step process.

First, "the Department developed numerical indices to measure the relative changes between quarters of the POR in the cost of hot-rolled stainless steel coil" by measuring SeAH's "consumption cost changes" of this input, "in terms of a percentage," for each CONNUM and quarter. (Id.)

Second, SeAH's quarterly direct material costs, extended by quantity, were then restated to end-of-POR terms by multiplying each quarter's extended direct material costs by a "factor," consisting of "a ratio of the index for the last quarter of the POR to the index of each respective quarter." (Id. at 14.) Commerce, referring to this as a "critical step," explained its effect as converting each quarter's costs to "a consistent end-of-POR purchasing power level" and equating each quarter's costs "to the same level of metal purchasing power." (Id. at 15.)

Third, Commerce summed the restated extended direct material costs for all four quarters, and also summed the quarterly production quantities. (Id. at 16.) This set the stage for "Step Four," in which Commerce divided the total extended end-of-POR direct material cost by the total production quantity within each CONNUM, resulting in weighted average per-unit costs. (Id. at 16-17.)

Fifth, the quarterly indices were applied, again, to the results of the previous calculations "in order to reflect SeAH's actual purchasing power for each quarter during the POR." (Id. at 17.) Thus "the single weighted-average cost calculated in step four was . . . restated into four values, depending on the price level of a given quarter," and these values were compared to contemporaneous sales in the given quarter. (Id.)

Sixth, the Department added to the restated POR weighted average per-unit material costs from step five and added SeAH's POR weighted average direct labor, variable overhead, fixed overhead expenses, general & administrative expenses, and financial expenses,^[4] and produced the cost of production of the subject merchandise. (Id. at 18.) The Department asserted that results of its calculations were "consistently applied on a per-unit basis" in satisfaction of what it characterized as the third element of the statutory cost recovery test. (Id.)

Commerce rejected interpreting the cost recovery statute to mean that only a single non-indexed amount for the entire POR could be used in the cost recovery test, noting that the statutory text nowhere contained a prohibition on adjustments and that the Department is permitted to adjust costs for significant variations caused by, for example, the performance of major facility maintenance during the POR. (Id. at 19 (citing Uruguay Round Agreements Act, Statement of Administrative Action, H.R.Rep. No. 103-316, at 832 (1994), reprinted in 1994 U.S.C.C.A.N. 4040, 4169 ("SAA").))

c. Calculation of a Weighted Average Total Cost of Production

Finally, the Department viewed its methodology as satisfying the cost recovery statute because it resulted in a weighted average cost of production, as shown by the fact that "the average took into consideration the proportional relevance of production quantities in each quarter." (Id. at 19-20.)

d. Consistency With Text of Cost Recovery Statute

Commerce agreed with the Court that "the statute does not give the Department discretion to compare prices to a weighted-average per-unit cost [of production] for a time span other than the period of investigation or review." (Id. at 21 (citing SeAH, 704 F.Supp.2d at 1368 n. 18).) Commerce explained that while its use of a price index caused the cost of production to rise or descend "throughout the POR depending on SeAH's metal purchasing power at a given point in time," giving the initial appearance of four separate costs, the methodology actually collected the four separate values into a weighted average per-unit cost of production for the period of review, as required by the statute, through the use of the "well-known, commonly-applied accounting tool" of a price index. (Id.)

3. Comparison of Indexed and Unindexed Methodologies

Commerce supplied, as requested, data underlying the remand analysis. That data is contained in numerous attachments to a memorandum regarding the Department's calculations. (See Adjustments to the Cost of Production ("COP") and Constructed Value Information Pursuant to Court Remand (Aug. 6, 2010) ("Remand Calculation Memo"), Def.'s Confid. App'x to Response to Comments Regarding Remand Determination, Ex. B.)

Examining the data, Commerce stated that it supported the use of quarterly indexing because it showed that the unindexed cost recovery test resulted in dramatically increased recovery of sales during the third quarter, the quarter during which nickel prices spiked most dramatically. (Remand Redetermination at 22-23.) Commerce also pointed out that its calculation of the total extended cost of all sales in the home market during the POR, when compared to its calculation of the total extended cost of those same sales, showed that SeAH's costs exceeded the value of its sales in the POR and that it did not, in fact, recover its costs during the POR. (Id. at 23.) Commerce found this important because

we believe the cost recovery test allows for the ``recovery' (i.e., the inclusion of sales in normal value) of sales made at a loss when, in the end, respondent's overall sales are found to have been made above the average cost of production for the review

(Id. at 23-24.)

Addressing a concern raised by the Court in its remand opinion, Commerce disagreed that the indexing methodology "results in the erection of an artificially raised floor vis-á-vis the unadjusted POR weighted average per-unit COP." In this regard, the Department cited to the fact that for certain sales, the indexed cost was actually lower than the unindexed cost as proof that "the Department's methodology created no artificial ``floor' or ``ceiling' in its calculation." (Id. at 24.)

D. Comments of Parties and Response of Commerce

SeAH argues that the quarterly indexing methodology is not in accordance with law for three reasons. First, SeAH contends that Commerce's methodology actually creates four separate quarterly average costs and excludes from the NV calculation any sales priced below the quarterly average cost without consideration of whether they are priced above the per unit weighted average COP for the POR. (SeAH Comments at 3.) SeAH attacks Commerce's three-element restatement of the statute (quoted supra at 1328), calling it a "transparent attempt to rewrite the statute" by "deconstruct[ing]" its plain language. (Id. at 4) Focusing on the first question ("Does the Department's methodology cover the ``period of review?'"), SeAH argues that the question "deviates from the plain language of the statute and thus invalidates the entire analysis." (Id.) SeAH emphasizes that the statute requires that the cost recovery test compare sales prices against a "weighted-average per unit cost of production for the period of . . . review." (Id. (citing 19 U.S.C. § 1677b(b)(2)(D) (emphasis added by SeAH).)) SeAH insists that even if the Department's methodology covers the period of review, that does not mean that it satisfies the statute, because an average covering the period of review is different from an average being "for" the period of review. (Id. at 4-5.)

Second, SeAH contends that the Department's quarterly indexing is contrary to law because it fails to recover home market sales that were made at prices above the unadjusted weighted average per-unit cost of production for the POR, but below the adjusted COP resulting from Commerce's methodology. (Id. at 6-7.) SeAH points out that the record identifies many hundreds of these sales, while showing that a comparatively trivial number of home market sales are recovered under the indexed methodology that would have been excluded by the unadjusted methodology. (Id. at 7.)

Third, SeAH challenges Commerce's stated reasoning that it must adopt an indexed methodology in order to avoid illogical or distortive "absurd" results. (Id. at 8-11.) SeAH states that Commerce's methodology must yield to the clear intent of Congress, which has spoken in unambiguous terms through the language of the statute as to when sales must be considered to be at prices providing for the recovery of costs within a reasonable period. (Id. at 10-11.)

SeAH also objects to Commerce's comparison of the spike in nickel prices to hyperinflationary contexts, pointing out that hyperinflation involves currency devaluation across all costs and prices in an economy simultaneously, making it impossible to distinguish real from illusory price changes in that currency without an index. (Id. at 12.) SeAH maintains that this case, in contrast, involved a spike in the real price of a single commodity within an economy that was, by Commerce's admission, not hyperinflationary. (Id.) As SeAH sees it, the cost recovery statute definitively clarifies how quickly SeAH must adjust home market selling prices to account for its increased COP: the statute states that even sales below COP at the time they are made "will nevertheless be considered to provide for recovery of costs within a reasonable period of time if those prices are above the weighted-average per unit cost for the POR." (Id. at 13 (citing 19 U.S.C. § 1677b(b)(2)(D).))

Bristol's comments essentially support the positions expressed on this issue in the Remand Redetermination, and the response of Defendant reiterates its position in the Remand Redetermination.

E. Analysis

After careful consideration of the Remand Redetermination, the cost recovery statute, and the arguments of all parties, the Court finds that the Department's quarterly indexing methodology violates the plain language of the cost recovery statute. The Court is therefore compelled to remand this case to Commerce a second time with the instruction that Commerce must conduct the cost recovery analysis using the unindexed weighted average per-unit COP for the POR that it calculated for the present remand results.

A. The Language of the Cost Recovery Statute

The Court finds that Commerce's methodology is contrary to the language of the cost recovery statute. It is worth once more quoting in full the section of the antidumping statute on cost recovery:

If prices which are below the per unit cost of production at the time of sale are above the weighted average per unit cost of production for the period of investigation or review, such prices shall be considered to provide for recovery of costs within a reasonable period of time.

19 U.S.C. § 1677b(b)(2)(D).

The Court has already declared that this statutory language "does not give Commerce discretion to compare prices to a weighted average per unit cost for a different time span" than the POR, adding that "Congress intended the cost recovery statute to limit Commerce's ability to exclude certain home market sales from normal value calculations." SeAH, 704 F.Supp.2d at 1368 n. 18. The Court also described the effect of this language as creating "a price floor, above which home market sales ``shall be' recovered and considered in establishing normal value," and stated that the price floor must be established "by calculating the ``weighted average per unit cost of production for the period of . . . review[.]'" Id. at 1369 (quoting 19 U.S.C. § 1677b(b)(2)(D)). The Court declared that it "must invalidate as contrary to law any cost recovery test that excludes home market sales prices that are above that price floor." Id.

Commerce and Bristol both emphasize that § 1677b(b)(2)(D) does not declare a prohibition against the Department making adjustments to the weighted average per unit COP for the POR. This focuses, however, on what the statute does not say, rather than on what the statute does say and does require. Although Commerce may have discretion to develop its own interpretations and methodologies where the statute is ambiguous or silent, the Department does not have the discretion to ignore the plain meaning of the statute.

The statute comes first, and any methodology employed by the Department must come after, with implementation of the statute as its goal. Commerce appears to have adopted its quarterly indexing methodology due to perceived distortions, without first considering whether those perceived distortions were actually addressable under the statute. Commerce appears now to seek, post hoc, a way of justifying its preferred result by bending the plain meaning of the statute to conform to its methodology. Doing so impermissibly places the statute in service of the methodology, rather than the methodology in service of the statute.

A prime example of Commerce's errant approach is the three-element restatement of the statute that Commerce uses to argue that its methodology complies with the law. (See supra at 1328.) The statute quite clearly states that the cost recovery test must compare sales below costs against the single weighted average per unit cost of production for the period of review. By using the definite article ("the weighted average per unit cost of production for the period of . . . review"), the statute demands comparison to the single benchmark number—what the Court has also referred to as the "price floor"—and not to an ascending and descending landscape that varies with input commodity prices.

When Commerce fractured the statute's syntax into a series of disjunct questions about whether the methodology "covers" the POR, uses "a" per unit cost of production, and calculates "a" weighted average, Commerce shifted (perhaps unintentionally) to the indefinite article, as if the statute demanded comparison not with one single benchmark number, but with one of a set of numbers to be picked by Commerce. This interpretation is wrong because it reads the statute as providing Commerce with options, rather than as constraining Commerce to one single cost recovery test to be conducted in the specified manner.

B. The Difference in Results Between the Two Methodologies

The Court rules that the language of the cost recovery statute unambiguously requires Commerce to use one single benchmark value—the weighted average per unit COP for the POR—in the cost recovery analysis. Commerce violated that statutory requirement by calculating instead a weighted average per unit COP for the POR that was adjusted to restate SeAH's costs in quarterly indexed terms, and then recovered only those sales which fell below these varying quarterly values.

The data provided by Commerce reveals that its chosen methodology resulted in the exclusion of many hundreds of sales above the weighted average per unit COP for the POR. (See Remand Calculation Memo, especially Attach. 1 and Attach. 5.) The number of SeAH's total home market sales that is excluded using Commerce's methodology, but which would be included under the ordinary unindexed methodology, comprises more than 10% of SeAH's total home market sales. The two methodologies include and exclude from NV different sets of sales; but while Commerce's methodology includes some sales that the unindexed methodology would exclude, Commerce's indexing excludes a far greater number of sales that the unindexed methodology would include. The net result is that Commerce's decision to use an indexing methodology excluded a great number of sales that would otherwise be used in calculating NV.

To reiterate this point using the metaphor employed by the Court in its prior Slip Opinion, the statutory benchmark establishes a floor and mandates that prices above that floor be included in NV calculation. Commerce's quarterly indexing methodology raised a stage above the floor in certain quarters and added steps descending below the floor in other quarters. Commerce then included in the NV calculation all sales above the stage, floor, or steps its indexing had crafted. This violated the statute where a sale occurred above the floor, but below a stage erected over that floor by Commerce with its methodology. The data shows that many hundreds of SeAH's home market sales fell into the space between the statutory floor and the stage erected by Commerce's methodology and that Commerce, contrary to law, ignored those sales in its calculation of SeAH's NV.

Additionally, the difference in result between the statutory methodology and Commerce's quarterly indexed methodology, in and of itself, indicates that the quarterly indexed methodology and the unindexed methodology are distinct entities with sharply differing outcomes. Commerce's ability to highlight the difference reinforces the Court's conclusion that the quarterly indexed methodology is different from the single benchmark calculation required by the plain language of the cost recovery statute.

C. Commerce's Analogy to Hyperinflationary Contexts

SeAH's point that it experienced real spikes in the price of nickel during the third quarter of the POR, as opposed to experiencing an illusory change in purchasing power due to devaluation, is well-taken. Commerce's analogy to the hyperinflationary context may have some appeal, but the analogy is faulty precisely because there was no hyperinflation in Korea during the POR. SeAH actually incurred significantly increased costs during the spike in nickel prices in real terms, not merely due to a general failure of the Korean won to maintain meaningful value as a currency.

In defending its ability to make adjustments for extraordinary events such as hyperinflation, Commerce briefly mentions the SAA. The relevant language is:

The determination of cost recovery is based on an analysis of actual weighted-average prices and costs during the period of . . . review, except that, before testing for cost recovery, such costs incurred during the period of . . . review may be adjusted as appropriate to take account of variations in unit costs caused by periodic temporary disruptions to production that occur on a less frequent than annual basis. For example, major maintenance may be scheduled every three years. While this maintenance is performed, output is suspended or reduced. This results in unit costs being artificially increased in years when the maintenance is performed and depressed in other years. To account for this, Commerce will spread out the effect of such disruptions over the appropriate period of time so that a proportional effect is recognized. The party claiming the adjustment must demonstrate that the disruptions have recurred at regular and predictable intervals.

SAA at 832, 1994 U.S.C.C.A.N. at 4170. This language indicates that Congress intended to permit adjustments to account for disruptions "on a less frequent than annual basis . . . at regular and predictable intervals." The Court cannot see how such language could apply to market-driven input material price spikes—or even, for that matter, to a hyperinflationary currency emergency.

SeAH goes to the core of the problem when it comments that "[t]he entire question, for purposes of the cost recovery test, is how rapidly SeAH should be required to adjust its home market selling prices to reflect [a] change in its real COP." (SeAH Comments at 13.) Congress has set up a mechanism for dealing with price spikes like those in nickel prices here; where, as here, there are no extraordinary pressures except those ordinarily present in a market, Commerce need not draw from methodologies such as IAS 29 for hyperinflationary economies. The analogy is not only inapt, but unnecessary.

D. Commerce's "Absurd Results" Argument

Commerce takes the position, in part, that its use of quarterly indexing is justified by the absurd results that would occur absent indexing. Commerce states, in essence, that the price spike in nickel reduced SeAH's purchasing power and, without indexing, distorted the cost recovery test by making it too easy for low-priced sales during the spike to appear to recover their costs and thus be included in the NV. (Remand Redetermination at 5-7.) But whenever a commodity price spikes, even significantly, the antidumping statute does not require a manufacturer to pass all of those increased costs along to its customers within the quarter on pain of having its sales excluded from the calculation of NV.

Two arguments raised by Commerce deserve to be addressed here. First, Commerce calculated upon remand the total extended costs of SeAH's home market sales and the total extended sales price of SeAH's home market sales. (Id. at 23.) Finding that total extended costs for the POR exceeded total extended sales for the POR, Commerce maintained that "overall, SeAH incurred a loss on these sales (i.e. that SeAH did not recover its costs during the POR)." Id. Commerce found this significant given its view of the cost recovery test:

[W]e believe the cost recovery test allows for the "recovery" (i.e. the inclusion of sales in normal value) of sales made at a loss when, in the end, respondent's overall sales are found to have been made above the average cost of production for the review. . . . It simply makes no sense for the Department to "recover" sales in its calculation when the respondent had an overall loss on those sales during the period of review. . . . [Such an analysis] would introduce, rather than mitigate, distortion into the Department's calculations, and we do not believe such an analysis is required under the statute.

(Id. at 23-24.)

The first problem with this argument is that, by Commerce's logic, no sales should be subject to recovery in a situation where the total extended sales for the POR are less than the total extended costs for the POR. While Congress could certainly have established such an approach, it did not. Instead, Congress very specifically crafted a statute that compares the price of sales against the cost of production at the time of the sale, and permitted Commerce to disregard such sales if "made within an extended period of time," "in substantial quantities," and "not at prices which permit recovery of all costs within a reasonable period of time." 19 U.S.C. § 1677b(b)(1). Congress then established a comparison of sales prices below cost against the "weighted average per unit cost of production for the period of . . . review," mandating that sales made above that very specific weighted average COP be included in NV. 19 U.S.C. § 1677b(b)(2)(D). It is no wonder Commerce finds the result of this Congressionally-mandated test absurd if it truly believes that sales should be ignored wherever the company failed to turn an overall profit in the home market during the POR. That is simply not the test Congress established.

The second problem is that the statute does not require costs to be recovered within the POR, but rather within a reasonable period of time. See 19 U.S.C. § 1677b(b)(1), (b)(2)(D). The statute is focused on excluding sales that are not made pursuant to normal market forces, but shows an awareness that not every market enterprise will always be able to recover all of its costs within a single month, quarter, or even period of review. Therefore, the statute only permits exclusion of those sales below cost that are "were not at prices which permit recovery of all costs within a reasonable period of time." 19 U.S.C. § 1677b(b)(1)(B).

The POR becomes important to cost recovery only in that it provides part of the mechanism for determining when sales do recover costs within a reasonable period of time. Sales are to be compared to the weighted average per unit cost of production for the POR, but this is done not to see whether costs are recovered within the POR, but rather to determine which sales "shall be considered to provide for recovery of costs within a reasonable period of time." In other words, the POR is not the relevant period for cost recovery; "a reasonable period of time" is the relevant period, and the interpretation of what is "reasonable" is constrained by statute.

The mechanism provided by the cost recovery statute is consistent with a market-based approach on passing costs along to the consumer, and does not at all indicate the sort of absurdity feared by Commerce. Instead, it strikes a careful balance between excluding sales from NV when made at prices that are simply too low to reflect market forces, and, on the other hand, requiring consideration of those prices that, while low, are still driven by market factors.

III. Conclusion

In conclusion, the Court (1) affirms Commerce's major input analysis; (2) determines that the quarterly indexing methodology employed by the Department was contrary to the explicit mandate of the cost recovery statute at 19 U.S.C. § 1677b(b)(2)(D); and (3) determines that this case must be remanded to Commerce to conduct the cost recovery test as mandated by the statute.

It is thus hereby

ORDERED that the Remand Redetermination is affirmed as to the major input analysis; and it is further

ORDERED that this case be, and hereby is, remanded to Commerce; and it is further

ORDERED that, on remand, Commerce shall employ the cost recovery test using the unadjusted weighted average per unit cost of production for the period of review which Commerce employed for comparative purposes on the last remand; and it is further

ORDERED that Commerce shall file with the Court its second redetermination upon remand by April 26, 2011; and it is further

ORDERED that Plaintiff shall file any comments on the second redetermination upon remand, in a length not to exceed 10 pages, by May 10, 2011; and it is further

ORDERED that Defendant and Defendant-Intervenor shall file any comments or responses to comments regarding the second redetermination upon remand, in a length not to exceed 10 pages, by May 24, 2011.

NOTES

[1] "PR" indicates a reference to the Public Record of the administrative proceedings before Commerce.

[2] All citations to the United States Code are to the 2006 edition.

[3] Department of Commerce regulations appear to provide no agency interpretation of § 1677b(b)(2)(D). See generally 19 C.F.R. Part 351.

[4] None of these other costs were adjusted by quarterly indexing, as they did not experience price spikes during the POR.