Catfish Farmers of Am. v. United States

2014 CIT 146

• Court: United States Court of International Trade
• Decided: December 18, 2014
• Docket: Consol. 12-00087
• Status: Published

Opinion

Slip Op. 14 - 146

UNITED STATES COURT OF INTERNATIONAL TRADE

: CATFISH FARMERS OF AMERICA, et al., : : Plaintiffs, : : v. : Before: R. Kenton Musgrave, Senior Judge : UNITED STATES, : Consol. Court No. 12-00087 : Defendant, : : and : : VINH HOAN CORP., QVD FOOD CO., LTD., : VIETNAM ASSOCIATION OF SEAFOOD : EXPORTERS AND PRODUCERS, ANVIFISH : JOINT STOCK CO., BIEN DONG SEAFOOD : CO., LTD., and VINH QUANG FISHERIES : CORP., : : Defendant-Intervenors. : :

OPINION AND ORDER

[Remanding seventh antidumping duty administrative review of frozen fish fillets from the Socialist Republic of Vietnam.]

Dated: December 18, 2014

Valerie A. Slater, Jarrod M. Goldfeder, and Nazak Nikakhtar, Akin, Gump, Strauss, Hauer & Feld, LLP, of Washington DC, for the plaintiffs.

Ryan M. Majerus, Attorney, Commercial Litigation Branch, Civil Division, U.S. Department of Justice, of Washington DC, argued for the defendant. On the brief were Stuart F. Delery, Assistant Attorney General, Jeanne E. Davidson, Director, and Franklin E. White, Jr., Assistant Director. Of Counsel was Elika Eftekhari, Attorney-International, Office of the Chief Counsel for Import Administration, U.S. Department of Commerce. Consol. Court No. 12-00087 Page 2

Matthew J. McConkey, Mayer Brown LLP, of Washington DC, for defendant-intervenors Vinh Hoan Corporation and QVD Food Company, Limited.

Mark E. Pardo, Andrew B. Schroth , and Andrew T. Schultz, Grunfeld Desiderio Lebowitz Silverman & Klestadt, LLP, of Washington DC, for the defendant-intervenor Vietnam Association of Seafood Exporters and Producers.

Robert G. Gosselink and Jonathan M. Freed, Trade Pacific, PLLC, of Washington DC, for defendant-intervenors Anvifish Joint Stock Company, Bien Dong Seafood Company Ltd., and Vinh Quang Fisheries Corporation.

Musgrave, Senior Judge: This opinion addresses consolidated lawsuits contesting

the administrative review portion of Certain Frozen Fish Fillets from the Socialist Republic of

Vietnam: Final Results of the Seventh Antidumping Duty Administrative Review and Sixth New

Shipper Review, 77 Fed. Reg. 15039 (Mar. 14, 2012), APDoc1 129 (“Final Results” or “Seventh

Review”). Compiled by the defendant, International Trade Administration, United States Department

of Commerce (“Commerce” or “Department”), the Seventh Review covers the period August 1, 2009

through July 31, 2010, and the administrative reasoning is in the issues and decision memorandum

(“IDM”) of record, A-PDoc 112 (“I&D Memo”).

Previously, further proceedings here were stayed pending the results of remand of

prior reviews, as those reviews implicated certain issues herein. See, e.g., Catfish Farmers of

America v. United States, 37 CIT ___, Slip Op. 13-63 (May 23, 2013). Redetermination of those

1 The antidumping duty order covers Pangasius hypophthalmus (also identified as Pangasius pangasius), Pangasius bocourti, and Pangasius micronemus. See Notice of Antidumping Duty Order: Certain Frozen Fish Fillets from the Socialist Republic of Vietnam, 68 Fed. Reg. 47909 (Aug. 12, 2003) (“Order”). The designation “A” herein preceding this court’s conventional citations to the public or confidential administrative record documents (PDoc or CDoc) are to those documents filed with the Import Administration’s Antidumping and Countervailing Duty Centralized Electronic Service System (“Access”). Consol. Court No. 12-00087 Page 3

reviews having been sustained in a separate slip opinion and judgments issued this date, the stay of

this matter is lifted hereby, sua sponte, and the merits addressed below.

As in those prior reviews, the plaintiffs’2 initial claims here concern Commerce’s

selection of Bangladesh as the primary surrogate country for the calculation of the respondents’

margins, which determination also resulted in surrogate valuation (“SV”) from Bangladesh of data

for the factors of production (“FOPs”) of the whole live fish input, farming inputs, labor, additives,

diesel fuel, and packing materials, as well as the use of financial statements for certain Bangladesh

shrimp processors to value foreign respondents’ overhead, SG&A, and profit. In addition, the

plaintiffs complain of the use of Indonesian import statistics rather than price quotes to value four

by-products: fish waste, fish oil, fresh broken fish fillets, and frozen broken fish fillets.

For their part, the defendant-intervenors collectively challenge Commerce’s use of

“zeroing” methodology in this administrative review, and the defendant-intervenor Vinh Hoan

Corporation challenges the determination to reject as untimely its request for company-specific

revocation as well as the determination to “cap” the SV for its fresh broken fillets at the level of the

SV for whole live fish.

Jurisdiction is pursuant to 28 U.S.C. §1581(c) and 19 U.S.C. §1516a(a)(2)(B)(iii).

The court is required to examine whether the Final Results are “unsupported by substantial evidence

on the record, or otherwise not in accordance with law.” 19 U.S.C. §1516a(b)(1)(B)(i). Certain

claims on this matter persuade that remand of the case is necessary.

2 The plaintiffs here again are industry petitioners Catfish Farmers of America, America’s Catch, Alabama Catfish Inc., d/b/a Harvest Select Catfish, Inc., Heartland Catfish Company, Magnolia Processing, Inc., d/b/a Pride of the Pond, and Simmons Farm Raised Catfish, Inc. Consol. Court No. 12-00087 Page 4

Discussion

I. Plaintiffs’ Motion for Judgment -- Selection of Primary Surrogate Country

A. Background

19 U.S.C. §1677b(c) mandates that the valuation of the factors of production

(“FOPs”) of a producer or exporter from a non-market economy (“NME”) “shall be based on the best

available information regarding the values of such factors in a market economy country or

countries”. Pursuant to its interpretation thereof, Commerce normally selects a “primary” surrogate

country based on a four-step sequence. See Import Administration Policy Bulletin 04.1: Non-Market

Economy Surrogate Country Selection Process (Mar. 1, 2004).3 Only the “Data Considerations” step

in that sequence is at issue here:

[D]ata quality is a critical consideration affecting surrogate country selection. After all, a country that perfectly meets the requirements of economic comparability and significant producer is not of much use as a primary surrogate if crucial factor price data from that country are inadequate or unavailable. . . .

In assessing data and data sources, it is the Department’s stated practice to use investigation or review period-wide price averages, prices specific to the input in question, prices that are net of taxes and import duties, prices that are contemporaneous with the period of investigation or review, and publicly available data.

Id.

For its preliminary determination, Commerce considered the Philippines, Indonesia,

and Bangladesh to be economically comparable to Vietnam and significant producers of comparable

merchandise, and it based the selection of the primary surrogate country on the record data for the

3 See, e.g., Jiaxing Brother Fastener Co., Ltd. v. United States, 38 CIT ___, ___, 961 F. Supp. 2d 1323, 1328 (2014). Consol. Court No. 12-00087 Page 5

main input for production of frozen fish fillets: whole live fish.4 See Certain Frozen Fish Fillets

from the Socialist Republic of Vietnam, 76 Fed. Reg. 55872 (Sep. 9, 2011) (“Preliminary Results”),

APDoc 7, at 55875-77. Commerce determined the Indonesian “FAO-FIGIS” data represented a

more reliable broad market average for purposes of valuing whole live fish, having also satisfied the

surrogate value criteria (according to its policy) of being publicly available, from an approved

surrogate country, sufficiently specific to the input in question, tax and duty exclusive, and

contemporaneous with the POR. Id. Commerce also determined to reject certain DAM-internal

worksheets, of the type used to calculate the whole live fish price in the final results of the prior

review, on the ground that questions remained unanswered as to the worksheets’ publicly

availability. Id. (citing DAM’s failure to respond to Commerce’s inquiries and an affidavit from a

Bangladeshi attorney who, as a member of the public, had tried unsuccessfully to obtain a copy of

the worksheets). See File Memorandum, “Placing Questions to Philippine and Bangladeshi

Governments on the Record” (June 27, 2011), APDoc 113. Additionally, Commerce valued the fish

waste, fish oil, fresh broken fish fillets, and frozen broken fish fillets by-products using SV import

data for Indonesia. See Preliminary Surrogate Value Memo, APDoc 5, at 6-7.

4 The three sources of data for these countries’ respective prices for whole live Pangasius fish were as follows: (1) Fisheries Statistics of the Philippines 2007-2009, an official government publication of the Philippine Government’s Bureau of Agricultural Statistics (“BAS”), Fisheries Statistics Division, containing 2009 Pangasius prices; (2) price and quantity data for 2009 for Pangasius pertaining to Indonesia from the U.N. Food and Agriculture Organization’s (“FAO”) Fisheries Global Information System (“FIGIS”); and (3) internal worksheets from the Department of Agriculture Marketing (“DAM”) of the Bangladesh Ministry of Agriculture pertaining to price data for “pangas” at the wholesale level of distribution (the “DAM worksheets”). Consol. Court No. 12-00087 Page 6

After the Preliminary Results, the parties placed further data on the record.5 For the

Final Results, Commerce changed its preliminary determination, and selected Bangladesh based on

the updated data for valuing whole live fish. I&D Memo at cmt. I.C. See Final Surrogate Value

Memo, APDoc 113, at 4-5. Commerce again found the data from all three countries for that input

to be “approved”, tax and duty exclusive, publicly available, specific, and contemporaneous with the

POR. But Commerce rejected the Philippine BAS data, after adopting its “observations and

concerns” from the previous review regarding them.6 Commerce also found that while the FAO-

FIGIS data for Indonesian Pangasius are “meant to capture all encompassing whole country data”,

representative of a broad-market average, and significant in volume in the amount of 109,685MT,

they are presented as a single average price for the whole country, include all four species of

Pangasius (of which Pangasius hypophthalmus is the farmed majority) and evince some uncertainty

as to total volume.

5 The additional data were as follows: (1) Fisheries Statistics of the Philippines 2008-2010 (“FSP08-10”), an updated Philippine Bureau of Agricultural Statistics (“BAS”) government publication, which reported farm-gate volume and value data for whole live Pangasius produced and sold in the Philippines during 2009 and 2010; (2) supplementation of the Indonesian FAO data; (3) printouts of weekly price data for whole “pangas” fish, covering a portion of Bangladesh’s districts from the DAM website (submitted at Commerce’s request, after the deadline for submitting surrogate data had already passed). See I&D Memo at 6-8; see also VASEP’s Surrogate Value Submission (Nov. 15, 2012) at Exhibit 5, APDocs 36-53; File Memorandum, “Phone Call to Counsel for VASEP” (Dec. 20, 2011), APDoc 69. 6 I&D Memo at 9. The plaintiffs’ challenges to certain of those findings and concerns were addressed in Slip Op. 13-63, supra. On this point, the plaintiffs argue the adoption of prior “observations” would include Commerce’s prior finding that the BAS data represented a broad market average. The defendant argues there was no specific broad market average finding on the Philippine BAS data in the instant review, only Commerce’s statement that “[a]ll other observations and concerns about the [BAS] data remain the same as in . . . prior segments”. The court, however, is inclined to agree with the plaintiffs on the import of that statement, which would include the prior administrative finding of the BAS data as constituting a broad market average. Consol. Court No. 12-00087 Page 7

Such observations apparently persuaded Commerce when it considered the BAS data

both as a whole and in comparison with the DAM data for Bangladesh. Considering the DAM data

further, Commerce continued to reject the DAM worksheets as lacking public availability. With

respect to certain DAM website data submitted after the preliminary results, see note 5, Commerce

rejected the “grower” prices that consisted of two data points as being too limited to constitute a

broad market average. However, as in the prior review, Commerce was not persuaded by the

domestic industry’s argument that the DAM data should be rejected on the ground that they represent

wholesale prices and not “farm gate” prices, reasoning that “it is uncertain the extent to which such

a distinction is relevant” in SV analysis, which “seeks to determine the price a respondent would pay

for an input if it were to produce subject merchandise in the surrogate country, not necessarily what

producers/sellers of the input in the surrogate country receive.” I&D Memo at 12.

Commerce then noted that the record indicated that the whole fish processed in

Vietnam range from 1-1.5 kilograms and that the DAM website data for “pangash, small” consisted

of fish that included that size range, i.e., up to 1.5 kilograms. Commerce also noted that the terms

“pangas” and “pangash” are used interchangeably, that “pangas” is the local name for Pangasius

hypophthalmus, and that Pangasius hypophthalmus is the only species listed under “pangas” in the

Fisheries Statistical Yearbook of Bangladesh 2009-2010 published by the Government of

Bangladesh. Commerce therefore found that the DAM website data was sufficiently species-and-

size specific to the whole live fish input utilized by the respondents.

With regard to the broad market average criterion, Commerce noted that the DAM

website data represented weekly prices for “only 31 of the 68 districts, [but] this still represents 767 Consol. Court No. 12-00087 Page 8

price observations from a considerable portion of the country, a significant number” that included

the largest producing district (Mymensing), “thereby indicating that the vast majority of production

was captured.” Commerce therefore found the source to be a broad market average.

Finally, Commerce compared the prices in the “hard copy” (i.e. the DAM worksheets)

to those from the DAM website and found the differences between the two “minute.” Thus,

Commerce found that:

1) the data are species-specific (unlike the FIGIS data); 2) this still represents 767 price observations; 3) the largest district, by far, is included; 4) the numbers tie to the hard copy; 5) the data exactly match the POR; and 6) the data are publicly available. In addition, the Fisheries Statistical Yearbook of Bangladesh 2009-2010, establishes that cultured species-specific Pangasius hypophthalmus[ ] production in Bangladesh was 124,760MT, greater than the volume from the FIGIS data (109,685MT). Although we do not question the reliability of the FIGIS data, we find the DAM data to be a more robust data source, given its breadth and focus, especially with respect to specificity and contemporaneity. We thus find that the DAM data represent the best option for valuing the whole fish input.

I&D Memo at cmt. I.C (footnote omitted). Commerce reinforced this finding by noting that

“Bangladesh also has multiple viable surrogate financial companies” for the purpose of calculating

SV financial ratios.

B. Analysis

The plaintiffs advance various arguments attempting to convince that Commerce’s

choice of surrogate for valuing the whole live fish input is unsupported by substantial evidence and

contrary to law. One is sufficient for remand.

First contended is that Commerce has imposed a new and unexplained test of

“robustness” without notice and comment or support in the record. The court disagrees. Consol. Court No. 12-00087 Page 9

“Robustness” appears inherent in the statutory term “best available information,” and is not a new

statistical concept. It originally referred to the sensitivity of the method of analysis, i.e., to the degree

of resistance to errors produced by deviations from the assumptions employed, see, e.g., Peter J.

Huber, Robust Statistics, ch. 1 (Wiley & Sons ed. 1981), but has since been utlized to cover the

quality of the data being analyzed, cf. id., p. v (“[a]musingly, . . . ‘robust’ has now become a magic

word, which is invoked in order to add respectability”). Commerce in this matter simply found the

qualities of breadth and focus, as further defined in terms of specificity and contemporaneity,

compelling. The plaintiffs chip away at a selective juxtaposition of Commerce’s findings of record

they deem relevant to those terms,7 but in the end the arguments are insufficient to render

7 For example, the plaintiffs argue: that Commerce has placed apparent undue emphasis on the sheer number of “data points” embodied by the DAM data; that it is unreasonable to conclude that the “higher-level” yearly average prices aggregated at the provincial levels and published in the Philippine FSP08-10 data and the yearly total quantity and value aggregated at the national level published in the FAO data for Indonesia are any less “robust” merely because they presented fewer numbers than the lower-level district-aggregated DAM website data; that it is misleading for Commerce to attempt to bolster its case with respect to the DAM website data by referring to the 124,760MT figure for Bangladesh Pangas production in 2009-2010 (as reported in the Fisheries Statistical Yearbook prepared by the Department of Fisheries, a different government agency than DAM); that the actual DAM website data cannot compare to the data of record for Indonesia -- 109,685MT total volume of Pangasius production for 2009 based on FIGIS data that were specifically associated with the reported FAO data value; and that the Indonesian FAO-FIGIS data were not “overbroad” since Commerce acknowledged that Pangasius hypophthalmus is the majority farm-raised species and the antidumping duty order covers three of the four species of Pangasius. Further on this latter point, the plaintiffs argue that given the fact that the antidumping duty order covers three of the four Pangasius species Commerce fails to explain why a data set that underincludes two of the Pangasius species subject to the AD order is more “robust” for purposes of the whole live fish input SV than a data set that overincludes a single Pangasius species not subject to the AD order that is not the set’s majority species. E.g., Pls’ Br. at 13 n.5, referencing Certain Frozen Fish Fillets from the Socialist Republic of Vietnam, 72 Fed. Reg. 13242 (Mar. 21, 2007) and accompanying IDM (Mar. 12, 2007) at cmt. 8 (using genus-level data to value whole live fish); Pls’ Reply at 8, referencing Certain Frozen Fish Fillets from the Socialist Republic of Vietnam, (continued...) Consol. Court No. 12-00087 Page 10

unreasonable the administrative determination of the DAM website data as displaying “more robust”

qualities than the Indonesian or Philippine data sets. Cf. Bowman Transp., Inc. v. Arkansas-Best

Freight System, Inc., 419 U.S. 281, 286 (1974) (“a decision of less than ideal clarity” must be upheld

“if the agency’s path may reasonably be discerned”) with Universal Camera Corp. v. NLRB, 340

U.S. 474, 488 (1951) (court may not displace Commerce’s choice between two fairly conflicting

views). Unstated in the summary paragraph in which Commerce’s ultimate finding appears, but

stated elsewhere, are Commerce’s observations regarding the DAM website data’s individuated

weekly coverage of the particular size Pangasius hypophthalmus utilized in the respondents’

production, which “lower level of aggregation” (about which the plaintiffs complain in other context)

is what enabled Commerce’s deeper consideration of the DAM data’s “breadth” and “focus”.8

Regarding contemporaneity, in light of Commerce’s finding that every source was

“sufficiently” contemporaneous, the plaintiffs’ argument that the DAM website data cannot validly

be determined temporally “more robust” overlooks the finding that the FAO-FIGIS data covered only

5 months of the POR. It is not inconsistent for the agency preliminarily to find competing data sets

sufficient for the purpose of further consideration, and then, upon such further consideration, to find

that the coverage of one is temporally fuller, even exact, and the other less so.

7 (...continued) 75 Fed. Reg. 56062, 56067 (Sep. 15, 2010) (prelim. results) (the record did not show that using genus-level rather than species-level sources “would necessarily generate a difference in price”). 8 Among the various definitions of “breadth”, only four are apt here: (1) spaciousness or extent (distance), (2) largeness or liberality, as of views or vision, (3) the quality of having details so massed as to produce an impression of largeness and unity, and (4) in logic, the meaning of extension or denotation. Webster’s New International Dictionary (Unabridged) (2d ed. 1954), p. 329. “Focus” in the context of this matter obviously means center, concentration or convergence. See id. p. 978. The parties’ arguments over the terms do not extend beyond such parameters. Consol. Court No. 12-00087 Page 11

Similarly regarding specificity, the fact that Commerce found the FAO-FIGIS data

“sufficiently” specific (to the genus level) for further consideration does not mean, contrary to the

plaintiffs’ argument thereon, that it was unreasonable for Commerce to further find the DAM

website data “more robust” in terms of its specificity or focus upon Pangasius hypophthalmus, i.e.,

the specific whole live fish input utilized by respondents, and in terms of size.

As between the Bangladeshi DAM website data and the BAS data, although

Commerce did not directly compare their volumes, the plaintiffs’ attempt to refute the defendant’s

point that the total BAS volume “pales” in comparison with Bangladesh Pangasius production

overall is insufficient to impugn the reasonableness of Commerce’s “more robust” finding with

respect to the DAM website data. And as between the DAM website data and the Indonesian data,

the plaintiffs do not adequately explain or persuade why it was unreasonable for Commerce to have

preferred a “view” of the data, in the form of weekly snapshots of district markets, as opposed to a

single average price summation of them and regardless of the comparative volumes involved.

Nonetheless, the plaintiffs make the point that it is difficult to understand how

Commerce could possibly have found the DAM data more “robust” when DAM did not respond to

two separate requests from Commerce for information to clarify aspects of the worksheet data and

the nature and soundness of DAM’s collection procedures -- in contrast, the plaintiffs maintain, to

the Philippine BAS, which they characterize as having provided a prompt and complete response to

Commerce’s inquiries. See Commerce’s File Memorandum, “Response to Questions for the

Philippine Bureau of Agriculture Statistics Regarding Price Data in the Fisheries Statistics of the Consol. Court No. 12-00087 Page 12

Philippines” (July 15, 2011), APDoc 8. Thus returning to methodology, the plaintiffs vigorously

contest Commerce’s “corroboration” of the DAM website data, arguing that Commerce ignored

significant discrepancies between those data and the unpublished internal worksheets that

undermined their reliability. VASEP had claimed that the DAM website data may be corroborated

by comparing the percentage of matching “pangas, small” data points among the unpublished

worksheets data, which Commerce did “by comparing the instances where a field for both sources

contained data.” I&D Memo at 12. Commerce “found that the numbers were identical except for

a few observations, and even then the differences between the two were minute.” Id. According to

the defendant, this “enabled the agency to ascertain whether the DAM website data w[ere] indeed

sourced from the internal worksheets”. Def’s Resp. at 24. The plaintiffs contend this is cherry-

picking and faulty logic, as their analysis shows that the percentage of DAM website data fields with

varying degrees of differences, as compared with the corresponding fields of the DAM worksheets

and either containing different numbers or missing data, was over 52 percent.9

9 In greater detail, the plaintiffs contend that the DAM wholesale price data were from “only” 31 of 68 districts in Bangladesh and provided 53 weekly price observations for the 12-month POR of August 1, 2009 through July 31, 2010. See VASEP’s Surrogate Value Submission (Dec. 22, 2011), at Ex. 2, APDoc 70. Assuming steady production over time, when the plaintiffs performed the same analysis for all 32 districts for which DAM included any data, the data reported by DAM were for only for 45.2% of all weeks, so that even for the 32 of 68 (i.e., 47%) of Bangladesh districts included in the DAM data, the plaintiffs point out that the data are only partial. See Pls’ Br. at Att. A. In particular, the plaintiffs point out that since the DAM-reported data for Mymensingh (which Commerce claims the DAM data “covered”) are for only 17 of 53 weeks of the POR, by their reckoning this means that even if the pricing in each week was representative of all production in that week, which is not known, the DAM data could, at most, account for only 32% of total POR production in Mymensingh. This last point arguably applies with equal force to the method used to produce the BAS data, and even if all the foregoing is true, it is insufficient on its own to diminish the “substantiality” of the DAM website data’s reliability determination on the reviewing standard before the court. But see infra. Consol. Court No. 12-00087 Page 13

Generally speaking, the court is not free to disagree with Commerce’s choice of

method to confirm reliability, so long as it is based on “more than a mere scintilla” of substantial

evidence to that effect. See Consolidated Edison Co. of New York v. NLRB, 305 U.S. 197, 217

(1938). And as a general proposition, determining reliability through corroboration is a valid

method. Cf. 19 U.S.C. §1677e(c); Watanabe Group v. United States, Slip Op. 10-139 at 11 (Dec.

22, 2010) (corroboration simply “requires the use of independent sources to confirm the validity of

[the] secondary information”). That said, the court agrees with the plaintiffs that by considering only

the number of matching data fields among the DAM website data, in comparison with the DAM

worksheets and notwithstanding the counter-analysis of a not-insignificant percentage of

mismatches, such methodology can hardly be concluded “robust,” in contrast with Commerce’s

expressed concern therewith. Even if it may be concluded to be “more than a mere scintilla”, the

I&D Memo’s “gloss” over the magnitude of the discrepancies between the DAM worksheets and the

DAM website data, characterized as “minute,” appears specious, because while it may be true that

a no-public-availability finding does not, necessarily, correlate to invalidity of the same data for other

purposes, the reliability determination with respect to the DAM website data appears here to rest on

the thinnest of reeds that also appears considerably bent by the fact that Commerce did not “carry

through” on what the plaintiffs purport as Commerce’s previously-evinced concerns regarding the

DAM collection methodology and the DAM’s non-responsiveness thereto in determining the

reliability of the DAM website data. Such concerns would seem to go beyond the DAM worksheets’

public availability, and the method employed to determine the DAM website data “reliable” does

not appear to have resolved the potential “GIGO” principle inherent in such method. See, e.g., Consol. Court No. 12-00087 Page 14

Mississippi v. EPA, 744 F.3d 1334, 1352 (2013). Remand at least for further explanation of the

data’s reliability in light of the above is therefore necessary, but Commerce also maintains the

discretion to reconsider the matter as a whole, should it so choose.

The plaintiffs also argue that there are no quantities associated with the weekly

wholesale prices stated on the DAM website, and that in the absence of quantities Commerce had

no basis to ascertain whether the DAM data provided values for a commercial level of sales of

Pangasius in the marketplace. The Final Results do not address the plaintiffs’ argument that both

the Indonesian and Philippine sources provided a stated quantity associated with the average unit

price and that no quantities are associated with the DAM website data. Since the court’s review

function does not entail fact-finding, on remand Commerce is respectfully requested to address the

plaintiffs’ commercial-quantities contention in determining whether the DAM website data are the

best available information.

On another tack, the plaintiffs also contest the I&D Memo’s statement that the

distinction between wholesale and farm-gate levels of pricing is “uncertain” because SV “seeks to

determine the price a respondent would pay for an input if it were to produce subject merchandise

in the surrogate country, not necessarily what producers/sellers of the input in the surrogate country

receive.” I&D Memo at 12. They argue this explanation ignores the fact that the Vietnamese

respondents purchased the whole live fish inputs directly from farmers and that farm-gate prices for

Indonesia and the Philippines were available on the record. The defendant responds that Consol. Court No. 12-00087 Page 15

Commerce’s farm-gate-versus-wholesale concern is only with respect to the potential inclusion of

taxes or duties and is inapplicable here because Commerce determined that the Bangladeshi

wholesale data were tax and duty free. Def’s Resp at 25-26, referencing I&D Memo at 8 and Certain

Frozen Fish Fillets from the Socialist Republic of Vietnam, 75 Fed. Reg. 12726 (Mar. 17, 2010) and

accompanying IDM at cmt. 2 (“one cannot decipher if the prices are derived from farm gate prices,

and are therefore tax and duty exclusive, or from market prices”).

If the defendant’s point is true, it is post hoc, and the I&D Memo’s explanation is

insufficiently responsive to the plaintiffs’ concerns. On remand of the prior administrative review,

Commerce tested the idea of transportation cost as an aspect of the type of measurable and

quantifiable distinction that would impact the wholesale versus farm gate price data of record for that

review and found only an insignificant difference. The plaintiffs here are not arguing that

quantifiable differences can be concluded from the evidence of record; they are instead essentially

arguing that Commerce should proceed from the assumption that qualifiable distinctions should be

presumed, e.g., condition of fish at wholesale versus farm-gate, in order to mirror more closely the

respondents’ actual production experience and avoid introducing pricing distortions at the wholesale

level, whether or not there is actual record evidence of qualifiable distinctions. The reviewing

standard and appellate precedent might preclude imposing that kind of presumption on Commerce’s

methodology. See 19 U.S.C. §1516a(b)(1)(B)(i); see, e.g., Nation Ford Chemical Co. v. United

States, 166 F.3d 1373, 1377 (Fed. Cir. 1999) (Commerce need not duplicate the exact production

experience at the expense of choosing a surrogate value that most accurately represents the fair

market value of the input in a hypothetical market-economy). But it would not be inappropriate to Consol. Court No. 12-00087 Page 16

request on remand, since remand is being ordered in any event, further explication from Commerce

on the validity of the plaintiffs’ concerns, summarized above, in the determination of the best

available information for SV of the whole live fish input. Commerce is therefore requested to do

so.

Lastly on this issue, the plaintiffs argue Commerce failed to consider the totality of

available data for all FOPs in its surrogate country selection. See, e.g., Folding Metal Tables and

Chairs from the People’s Republic of China, 76 Fed. Reg. 2883 (Jan. 18, 2011) and accompanying

IDM (Jan. 10, 2011) at cmt. 1C (“consistent with the Department practice, we evaluated data

considerations for the purposes of surrogate country selection as a whole, including availability of

surrogate financial ratio data and availability of surrogate values for direct material inputs and other

FOPS, rather than dissecting the elements”).

Whenever the record includes multiple economically comparable countries that are

also significant producers of comparable merchandise, “the country with the best factors data is

selected as the primary surrogate country.” Policy Bulletin 04.1. In this instance, the data sets for

Bangladesh, Indonesia, and the Philippines each possess merits and demerits. The plaintiffs do not

minimize the importance of the whole live fish factor, but they argue Commerce failed to “move on”

to consider the evidence of record that, they contend, demonstrated that the Philippines and

Indonesia offered more contemporaneous and specific data for secondary material inputs, labor,

energy, by-products, and packing materials than were available from Bangladesh. Consol. Court No. 12-00087 Page 17

It is merely arguable that the I&D Memo evinces improper a priori selection of

Bangladesh as the primary surrogate country before consideration of the non-whole-fish factors and

financial data,10 and the court is not persuaded that the I&D Memo, as it stands, does not express

therein or thereby a sufficient degree of consideration of the totality of available data in compliance

with Policy Bulletin 04.1.11 For the current Final Results, Commerce found the data for Bangladesh

to represent the best surrogate for calculating whole live fish, farming inputs, labor, additives, diesel,

packing materials, overhead, SGA expenses, and profit, in accordance with its selection of

Bangladesh as the primary surrogate country. See generally I&D Memo at cmt. II; Final Surrogate

Value Memo, APDoc 113, at 4-5. Pursuant to 19 C.F.R. §351.408(c)(2), Commerce normally values

factors in a single surrogate country if such useable information exists on the record.

Be that as it may, reconsideration of the totality of available data might be necessary

depending upon how Commerce proceeds on remand in further explaining or redetermining the best

available information for the whole live fish SV, which would implicate the non-fish FOP SV

10 Cf. I&D Memo at 15 (regarding financial ratios, “[a]s noted above in Comment I, we have selected Bangladesh as the primary surrogate country”), & id. at 21 (regarding fish meal, “[a]s we explained in Comment I above, we have selected Bangladesh as the primary surrogate country”), & id. at 21 (regarding fingerlings, etc., “[w]e have selected Bangladesh as the primary surrogate country”), with Slip Op. 13-63 at 34 (remanding for consideration the impact of the secondary factors upon the determination of surrogate country as a whole, along with the court’s concerns over Commerce’s interpretation of the DAM and Philippine data submitted for that review, since a priori determination was not in accordance with Policy Bulletin 04.1). 11 See I&D Memo at 13 (“As described above, the Bangladeshi DAM data offer the best option for valuing the whole fish. Moreover, Bangladesh also has multiple viable surrogate financial companies as discussed below. Therefore, given the totality of the facts, we find that Bangladesh is the most suitable primary surrogate. Both Indonesia and the Philippines are suitable secondary surrogate countries.”); see also id. at 14-17 (financial ratios). Consol. Court No. 12-00087 Page 18

selections since the ultimate selection of the primary surrogate country has not yet been finally

resolved. Certain issues pertinent thereto are more fully addressed in the following section.

II. Plaintiffs’ and Vinh Hoan’s Motions for Judgment -- Surrogate Valuations of By-products

When calculating normal value, Commerce interprets 19 U.S.C. §1677b(c) to permit

offsetting incurred production costs with SVs calculated for by-products generated during the

production process,12 which interpretation the parties do not dispute. The I&D Memo restates: that

it is Commerce’s preference to value all FOPs with data from the primary surrogate country; that

when data are not available from the primary surrogate, Commerce will look to secondary sources;

and that Commerce declines to use price quotes when other “more reliable” data are on the record.13

12 Selection of the best available information of record for valuing by-products focuses on yielding accurate dumping margins. See Rhone Poulenc, Inc. v. United States, 899 F.2d 1185, 1191 (Fed. Cir. 1990). Commerce’s stated practice is to evaluate data for reliability, public availability, quality, specificity, and contemporaneity. See Magnesium Metal from the People’s Republic of China, 69 Fed. Reg. 59187, 59195 (Oct. 4, 2004) (preliminary determination); see also 19 C.F.R. §351.408(c)(1). Of those considerations, specificity has been held of primary importance for surrogate by-product valuation. See, e.g., Jinan Yipin Corp. v. United States, 35 CIT ___, ___, 800 F. Supp. 2d 1226, 1304 (2011) (“[i]f a set of data is not sufficiently ‘product specific,’ it is of no relevance whether or not the data satisfy the other criteria”); see also Taian Ziyang Food Co. v. United States, 35 CIT ___, ___, 783 F. Supp. 2d 1292, 1330 (2011) (“‘product specificity’ logically must be the primary consideration in determining ‘best available information’” and “[i]f a set of data is not sufficiently ‘product specific,’ it is of no relevance whether or not the data satisfy the other criteria set forth in Policy Bulletin 04.1”) (citation omitted). 13 See, e.g., I&D Memo at 18, referencing Certain Frozen Warmwater Shrimp from the Socialist Republic of Vietnam, 74 Fed. Reg. 47191 (Sep. 15, 2009) (final results) (“Frozen Warmwater Shrimp”) and accompanying IDM at cmt. 7 (Commerce’s “general practice is to not use price quote information if other publicly available data is on the record, because [price quotes] do not represent actual prices, broad ranges of data, and [Commerce] does not know the conditions under which these were solicited and whether or not these were self-selected from a broader range of quotes”). The defendant also argues Commerce’s reasons for selecting the Indonesian data correlate with the agency’s long-standing practice with regard to surrogate price data: Commerce prefers to use surrogate price data that is (1) an average non-export value, (2) representative of a Consol. Court No. 12-00087 Page 19

In the final analysis, Commerce granted SV offsets for respondents’ fish waste, fish oil, fresh broken

fillets, and frozen broken fillets by-products using import data maintained in the Global Trade Atlas

(“GTA”) for Indonesia from the Harmonized Tariff Schedule (“HTS”). See I&D Memo at cmt II.B.

The plaintiffs contest each by-product SV selection, arguing that the selected SVs are

demonstrably aberrant, both in absolute terms and relative to the value for the whole live fish input,

because they significantly overstated the by-product offset, thus rendering the margin calculations

inaccurate and unreasonably reduced. More precisely, the plaintiffs complain that the Indonesian

import data are far less specific to Pangasius by-products than the available price quotes, and that

Commerce unreasonably cites merely to its preference for import statistics without further

explanation, which they claim is legally defective because it presupposes that price quotes will never

constitute the “best available information” for valuing a reported factor of production contrary to

established judicial precedent. Vinh Hoan contests Commerce’s treatment of its fresh broken fillets,

and it also raises an imprecise point regarding Commerce’s treatment of the fresh broken fillets in

their frozen state. With the caveat that Commerce’s response to section I, supra, might necessitate,

in its discretion, a reconsideration of these by-products’ SV, the court offers the following

observations.

1. Unprocessed Fish Waste

Agreeing with Vinh Hoan that Indonesian GTA data for HTS 0511.91.90.00 appeared

to be the more specific evidence of record to value fish waste on the ground that the provision does

range of prices within, and contemporaneous with, the POR, (3) product-specific, and (4) tax-exclusive. Def’s Resp. at 30. See Taiyuan Heavy Machinery Import & Export Corp. v. United States, 23 CIT 701, 706 (1999); accord Hebei Metals & Minerals Import & Export Corp., 28 CIT 1185, 1191 (2004). Consol. Court No. 12-00087 Page 20

not include whole fish, Commerce valued respondents’ unprocessed fish waste thereby, to wit,

“Animal Products Nesoi[14]; Dead Animals (of Ch. 3), Unfit for Human Consumption, Other

Products of Fish or Crustaceans, Moluscs or Other Aquatic Invertebrates.” I&D Memo at cmt. II.B.1.

In so doing, Commerce rejected a 2010 commercial price quote for Pangasius fish waste from

Vitarich, a Philippine Pangasius processor,15 two 2006 price quotes from Indian companies Aditya

Udyong and Ram’s Cold Storage,16 and four published Indonesian prices for unprocessed fish waste

or other seafood waste products. Commerce rejected the plaintiffs’ argument that the import

statistics are overly broad, and it did “not find the price quotes to be any more specific” because,

among the several general price quotes provided by the plaintiffs for fish waste from India,

Indonesia, and the Philippines, only the Vitarich quote was species specific -- and even then,

Commerce did not “believe that the species from which the fish waste originated is necessarily

important because there is no information on the record indicating there are meaningful distinctions

among the species”. Commerce also found only two of the price quotes contemporaneous, which

the defendant contends were from among the Indian price quotes.

Contending that these findings are at odds with the record, the plaintiffs contest

Commerce’s finding that the price quotes were not “any more specific” than the import statistics

because the Philippine, Indian, and Indonesian price quotes are all specific to unprocessed fish waste,

14 I.e., “not elsewhere specified or included.” 15 Use of this quote in the prior review was sustained on appeal here. 16 Use of these quotes in a prior new shipper review, covering the period August 1, 2006 through January 31, 2007, was sustained in Vinh Quang Fisheries Corp. v. United States, 33 CIT 1277, 637 F. Supp. 2d 1352 (2009). Consol. Court No. 12-00087 Page 21

and the Vitarich quote was also species specific (as Commerce itself acknowledges). After

considering the plaintiffs’ various contentions, the court is persuaded as to the correctness17 of their

argument that HTS 0511.91.90.00 is a “catch-all” provision for aquatic invertebrates that is not

facially restricted to species’ “waste”. Therefore, it was erroneous for Commerce to have articulated

that the price quotes were not “any more specific” than the import statistics for valuing fish waste,

because that statement proceeds from either ipse dixit or circular reasoning that the import data for

that provision are “other more reliable data . . . on the record”. See I&D Memo at 18.

In addition, the plaintiffs argue that continuing to use Indonesian HTS 0511.91.90.00

import-statistics to value the fish waste at $0.59/kg after Commerce switched the primary surrogate

country from Indonesia to Bangladesh resulted in a near-doubling of the relative fish-waste-to-

whole-live-fish-input ratio, from 33% to 60%. See Final Surrogate Value Memo (Mar. 7, 2012) at

Exhibit 1, APDoc 113. The plaintiffs’ evidence of record indicates that fish waste is a wholly

unprocessed by-product that is collected from the factory floor and sold to local buyers for a nominal

(and, the plaintiffs argue, more realistic) amount between $0.01/kg to $0.09/kg, according to their

17 The court finds the other arguments raised by the plaintiffs on this issue less persuasive. They contend, for example, that Commerce acknowledged from the prior Sixth Review that “the import statistics in question” -- meaning those pertinent to this review -- include products other than unprocessed fish waste, particularly processed by-products. Pls’ Br. at 32, referencing Sixth Review I&D Memo at cmt. IV.I.i. The argument fails because those “import statistics in question” were with respect to HTS 0304.90 (other fish meat of marine fish) for the Philippines, not HTS 0511.91.90.00 import statistics for Indonesia. (Commerce rejected Philippine HTS 0304.90 during remand of that issue from the Sixth Review because “valuing fish waste using import statistics results in a fish waste SV which is higher than that of the whole fish”; Sixth Review Remand Results at 19.) A second argument that fails is that Commerce needs to explain with greater clarity its “divergence” from Vinh Quang Fisheries and other cases. Commerce’s statement that there was “no information on the record” as to any difference in species’ waste provides a reasonably clear explanation for departing from (or the inapplicability of) its general requirement of “prices specific to the input in question” in Policy Bulletin 04.1. Consol. Court No. 12-00087 Page 22

recitation of SVs from each of the other potential surrogate countries and the independent research

studies on fish waste values in the Asia-Pacific region they submitted for the record. See Petitioners’

Case Brief at 25-34, APDoc 83; see also Certain Fresh Cut Flowers from Colombia, 61 Fed. Reg.

42833 (Aug. 19, 1996) (final results) at cmt. 25 (“the distinguishing feature of a by-product is its

relatively minor sales value in comparison to that of the major product or products produced”). No

party points to any contradictory evidence of record. The plaintiffs argue such a pricing disparity

to the Indonesian import data is obvious indication of inclusion of higher-priced value-added

products among those data, and therefore reliance on them is inappropriate because the result is

improper assignment of an inflated value to the input in question and an inaccurate antidumping

calculation. See, e.g., Globe Metallurgical Inc. v. United States, 32 CIT 1070, 1079 (Oct. 1, 2008).

The defendant responds that Commerce reasoned that the Indonesian data were

“preferable” to the competing price quote data in the record for valuing the fish waste by-product

because: (i) only one price quote on the record was species specific; (ii) only two price quotes were

contemporaneous; (iii) the Indonesian import statistics reflected broad market averages; and (iv) the

Indonesian import statistics were more specific to fish waste because they did not include whole fish.

Def’s Resp. at 30. With the possible exception of contemporaneity, however, and about which

Commerce barely commented, none of these reasons actually impugns the price quotes’ reliability,

since Commerce itself took the position that the record evinced no distinction among the waste of

“species”.

Obviously, an administrative preference for a particular kind of data must yield if the

relevant facts of record so compel. It would be unreasonable, for example, to continue to “prefer” Consol. Court No. 12-00087 Page 23

import data simply on the basis of their breadth if they are in fact overinclusive or overvaluative in

relation to the specific circumstances of the input or by-product in question. See, e.g., Jinan Yipin

Corp., supra, 35 CIT at ___, 800 F. Supp. 2d at 1304 (“[i]f a set of data is not sufficiently ‘product

specific,’ it is of no relevance whether or not the data satisfy the other [SV] criteria”).18 And with

respect to any substantial evidence of record that calls into question continued reliance on the

administrative preference, Commerce needs to reasonably resolve the conflict, because

administrative determinations must be explained “with sufficient clarity to permit ‘effective judicial

review’” as a matter of law. See Motor Vehicle Mfrs. Ass’n of the U.S., Inc., 463 U.S. 29, 43; see,

e.g., Timken U.S. Corp. v. United States, 421 F.3d 1350, 1355 (Fed. Cir. 2005). The function of the

court, of course, is not to weigh the evidence, but to determine whether the agency has reasonably

considered the record. On that basis, this record needs deeper articulation. Whatever the appeal of

certain import data in meeting the “broad market” criterion, Commerce needs to explain why they

are not overinclusive and more reliable than other information of record for the SV valuation of the

Pangasius by-product in question beyond merely providing a conclusory statement that they are,

particularly if the import data’s HTS provision is facially overbroad, which appears to be the case

here. Otherwise, this aspect of the Final Results simply rests on the fallacy of argumentum ad

populum.

18 See also, e.g., Vinh Quang Fisheries, supra, which sustained Commerce’s reliance upon the two Indian price quotes on the record pertinent to that review. After throughly detailing the background and proposed alternatives for valuing the fish waste considered in that decision, the court observed that the respondent in that case “never c[ame] to terms with Commerce’s finding that [the respondent]’s alternative processed fish waste data were an inappropriate match for the unprocessed fish waste factor.” 33 CIT at 1282, 637 F. Supp. 2d at 1357 (italics in original). Consol. Court No. 12-00087 Page 24

More precisely, Commerce in this instance did not comment on what the Indonesian

GTA data for HTS 0511.91.90.00 actually encompasses beyond accepting Vinh Hoan’s contention

that it does include fish waste. Commerce only indicated its preference is not to use price quotes

“when other more reliable data [are] on the record”, but that still begs the question, and Commerce

provided little else by way of analysis on the reliability of the Philippine, Indian, or Indonesian price

quotes, or on the evidence of record on other fish waste prices pertinent to the Indonesian region

vis-à-vis the import statistics in question, or on the arguments set forth by the plaintiffs regarding

satisfaction of the SV selection criteria. See Petitioners’ Case Brief at 23-34, APDoc 83. Still,

approximately eighteen years ago Commerce articulated that “the distinguishing feature of a by-

product is its relatively minor sales value in comparison to that of the major product or products

produced.” Certain Fresh Cut Flowers from Columbia, 61 Fed. Reg. 42833 (Aug. 19, 1996) (final

admin. results), at cmt. 25. See also Charles Horngren, George Foster, Cost Accounting: A

Managerial Emphasis (7th ed. 1991), p. 527 (“[a] by-product is a product that has a low sales value

compared with the sales value of [the principal product]”). Although there may be instances where

a process that produces waste to the degree implicated by the SV of the waste by-product calculated

for the Final Results may not seem unreasonable, an SV for “fish waste” that amounts to 60% of the

SV for the primary input of a process rather leaves a difficult bone of contention to swallow, and at

least requires Commerce to “explain its action with sufficient clarity to permit ‘effective judicial

review.’” Timken, 421 F.3d at 1355. Further explanation thereof is therefore requested in addition

to the foregoing.

2. Fish Oil Consol. Court No. 12-00087 Page 25

Commerce’s valuation of the respondents’ fish oil by-product suffers from similar

defect. In rejecting the Pangasius-specific price quote in favor of import statistics, Commerce again

stated only that it has “a preference not to use price quotes when other data is available” and that

“this HTS category includes fish oil and is therefore[ ] not overly broad.” I&D Memo at 19.

Commerce thus valued the fish oil using Indonesian GTA data for HTS 1504.20: “Fish Fats & Oils

& Their Fractions Exc Liver, Refined or Not, Not Chemically Mod., solid fractions, not chemically

modified, other” (italics added).

Similar to its fish waste SV, supra, Commerce’s SV calculation of $1.95/kg for fish

oil using Indonesian import data represents nearly double the price of the whole fish input, using the

plaintiffs’ calculation.19 See Final Surrogate Value Memo (Mar. 7, 2012) at Exhibit 1, APDoc 113.

The fact that HTS 1504.20 facially covers refined and further-processed fish oil is at odds with the

administrative record of respondents’ unprocessed fish-oil by-product and of the respondents’ actual

production experience, as well as at odds with Commerce’s position in Vinh Quang Fisheries, albeit

as argued in the context of fish waste. As mentioned, specificity has been held of primary

importance in determining whether particular import statistics are even appropriate for surrogate

valuation. See, e.g., Jinan Yipin Corp., supra, 35 CIT at ___, 800 F. Supp. 2d at 1304.

The plaintiffs further add that Commerce also failed to respond to their various

arguments that the respondents only sold their fish oil in the domestic market, that the record

contains a reliable and specific price quote for unprocessed Pangasius fish oil from an Indonesian

supplier, Yahdi, and that record evidence shows that Pangasius fish oil prices in Indonesia and India

19 At such valuation, one wonders whether fish fillet or fish oil is the primary product. Consol. Court No. 12-00087 Page 26

ranged from $0.13/kg to $0.84. E.g., Pls’ Br. at 34 referencing Petitioners’ Case Brief at 26-29,

APDoc 83. In view of such unresolved questions, Commerce’s determination here has also not been

explained with sufficient clarity to permit “effective judicial review,” thus requiring remand. On

remand, also bearing in mind the plaintiffs’ averment that the combined effect of Commerce’s fish

waste and fish oil surrogate valuations adds over 150% of the cost of the whole fish input, if

Commerce continues to value the unprocessed fish waste and fish oil by-products using Indonesian

import statistics or any other similar such import statistics, it shall clearly explain why any such

elevated results, as compared with the whole fish input, are reasonable for fish waste and fish oil by-

products, and also briefly explain how, or if, its views have evolved since Fresh Cut Flowers.

3. Fresh Broken Fillets

Commerce valued the respondents’ fresh broken fillets “by-product” based on

Indonesian GTA data for HTS 0304.19.00.00 “Fish Fillets and Other Fish Meat (Whether or Not

Minced), Fresh, Chilled or Frozen, Fresh or Chilled, Other”. I&D Memo at 20. Those data produced

a value of $2.89/kg value for the broken fillets. But, because “[b]roken fillets are not value added

by-products, they are simply meat from the fish that is broken or torn”, Commerce “capped” the

fresh broken fillets at the value for whole live fish of $0.98/kg, reasoning that it is illogical to

generate a deduction from normal value that is higher than the value of the whole live fish input.

I&D Memo at 20. See also Final Surrogate Value Memo (Mar. 7, 2012) at Exhibit 1, APDoc 113.

The plaintiffs contest this determination on the ground that the record contained, in

their opinion, contemporaneous, reliable, Pangasius-specific, Philippine price quotes from Vitarich

of $0.65/kg for broken Pangasius meat. See Petitioners’ Surrogate Value Submission (May 10, Consol. Court No. 12-00087 Page 27

2011) at Exhibit 16, PDoc 101; see also Petitioners’ Rebuttal Brief at 95-97 (Jan. 27, 2012), APDoc

93. In the context of Commerce’s repeated preference for import statistics over price quotes, the

plaintiffs argue that Commerce did not find their Vitarich price quote to be unreliable as a source for

valuing fresh broken fillets, and that Commerce failed to address their arguments that the import

statistics were “clearly aberrational” because they included values for both by-products and the

finished frozen fish fillets, a fact Commerce implicitly recognized insofar as it “capped” the value

for broken fillets derived from this HTS category at the $0.98/kg value for whole live fish.

Vinh Hoan contests the determination to “cap”. It argues that fresh broken fillets are

indeed value-added products that take labor, energy, equipment and other resources to produce and

are rendered “more valuable” than the unprocessed whole fish from which they have been produced.

Vinh Hoan also contests Commerce’s conclusion that Vinh Hoan cannot sell fresh broken fillets as

frozen fish fillets because of their appearance on the ground that Commerce failed to cite any

evidence in the record to support this factual assertion. The defendant responds that this is

contradicted by the record itself, specifically, Vinh Hoan’s questionnaire response. Def’s Resp. at

35-36, referencing I&D Memo at 20 n.62 and Vinh Hoan’s Supplemental Section D Response, PDoc

99, Question 23 (“[t]he company confirms that other than fish oil and fish meal, there were no other

value added by-products”).

The court need not opine on the issue at this point, since Commerce needs to resolve

in the first instance whether this issue will need revisiting as a result of the remand of the whole live

fish input and selection of primary surrogate country issues.

4. Frozen Broken Fillets Consol. Court No. 12-00087 Page 28

Commerce valued the respondents’ frozen broken fillets with Indonesian GTA data

for HTS 0304.29.00.00 “Fish Fillets and Other Fish Meat (Whether or Not Minced), Fresh, Chilled

or Frozen, Fresh or Chilled, Frozen Fillets, Other.” I&D Memo at 20. In selecting Indonesian import

statistics, Commerce stated that no party other than Vinh Hoan (et al.) commented on “this issue.”

Id. The plaintiffs argue they “clearly” addressed the issue of frozen broken fillets in their rebuttal

brief in advocating for the use of Vitarich price quotes to value this by-product, see Petitioners’

Rebuttal Brief at 95-97, PRII-93, and the defendant responds that the I&D Memo statement that “[n]o

other party commented on this issue” referred to the issue of whether to value frozen broken fillets

using HTS category 0304.29.00.00 or another Indonesian HTS category -- not whether the Philippine

price quote should be used to value the by-product. I&D Memo at 20. Vinh Hoan does not appear

to contest Commerce’s choice of SV for frozen broken fillets, but it does take exception to

Commerce’s treatment of its fresh broken fillets in their frozen state, as aforesaid.

At any rate, the court again need not opine at this point, since Commerce needs to

resolve in the first instance whether this issue will need revisiting as a result of remand of the whole

live fish input and selection of primary surrogate country issues.

III. Plaintiffs’ Motion for Judgment -- Partial Adverse Facts re Vinh Hoan

Commerce also found that Vinh Hoan had not misrepresented its consumption of

whole fish in the production of subject merchandise during the period of review. I&D Memo at

35-36. Commerce determined that Vinh Hoan’s reporting methodology was reasonable, that Vinh

Hoan had provided reasonable explanations as to why its fish consumption ratio might have

decreased relative to the last review, and that the record did not support plaintiffs’ claim that Vinh

Hoan understated its fish consumption. I&D Memo at 13. Consol. Court No. 12-00087 Page 29

The plaintiffs argue record evidence demonstrates the contrary, that Vinh Hoan

misreported its whole fish consumption factor and failed to substantiate the reasons for the reported

decline in its whole fish consumption factor, and therefore Commerce erred in determining not to

apply partial adverse facts available to Vinh Hoan pursuant to 19 U.S.C. §1677e. Pls’ Br. at 39.

According to the defendant, Commerce properly considered Vinh Hoan’s explanations for the

decline in its whole fish consumption ratio, the lack of record evidence demonstrating underreporting

by Vinh Hoan, and the satisfactory reconciliation of Vinh Hoan’s total quantity of fish entered into

production with its financial statements. Id. Thus, the defendant argues, Commerce properly

determined that the record did not demonstrate the need for application of 19 U.S.C. §1677e.

The court agrees with the defendant that the plaintiff’s estimations of underreporting

are speculative and do not constitute direct record evidence demonstrating that Vinh Hoan

misrepresented its whole fish consumption. See Pls’ Br. at 41-44. The criticism that Commerce did

not conduct a verification in the underlying review, see id. at 44, is unavailing because by statute

Commerce does not have to verify every review. See 19 U.S.C. §1677m(i)(3) (noting verification

of a final determination in an administrative review is only required when timely requested by an

interested party and when no verification was made during two immediately preceding reviews,

unless good cause for verification is shown).

In considering Vinh Hoan’s explanations for the decline in its reported whole fish

consumption factor from the preceding review, Commerce noted that although plaintiffs had

challenged each of Vinh Hoan’s explanations individually, “Vinh Hoan counters that collectively

[the explanations] account for the decline.” I&D Memo at 35. Commerce further observed that

although Vinh Hoan did not submit detailed record evidence tying each explanation to an “exact Consol. Court No. 12-00087 Page 30

percentage” of change, nevertheless, “when taken together, these explanations appear reasonable,

particularly in light of Vinh Hoan’s reconciliation [of its total fish entered into production with its

audited financial statements].” Id. at 36. Commerce also rejected the speculative nature of

plaintiffs’ assertion that Vinh Hoan understated its fish consumption, noting that “[t]he record does

not contain any evidence of underreporting.” Id.

Although the plaintiffs allege that Vinh Hoan failed to substantiate with

documentation the reasons for the decline in its whole fish consumption ratio, Pls’ Br. at 38-39,

Commerce did not require Vinh Hoan to submit such documentation in responding to its

supplemental questionnaire. See Commerce’s Supplemental Questionnaire to Vinh Hoan, APDoc

77 (“[p]lease explain what may have caused the total fish consumption ratio to decree . . . between

this review and the last”). The plaintiffs’ claim that Commerce “accepted [Vinh Hoan’s]

explanations simply because they ‘appeared reasonable’”, Pls’ Br. at 40, appears to be an

oversimplification of the analysis applied by Commerce. That is, the plaintiffs attempt to create the

appearance that Commerce simply accepted Vinh Hoan’s reported whole fish consumption factor

based on Vinh Hoan’s individual explanations rather than consider their collective weight, see Pls’

Br. at 40-41; however, it appears that Commerce based its decision on the totality of the evidence,

and therefore the determination not to apply partial facts adverse facts to Vinh Hoan pursuant to 19

U.S.C. §1677e is supported by substantial evidence and is hereby sustained.

IV. Defendant-Intervenors’ Rule 56.2 Motions

Vinh Hoan, QVD Food Company, Ltd., and collectively Anvifish Joint Stock

Company, Bien Dong Seafood Company, Ltd., and Vinh Quang Fisheries Corporation, filed separate

motions for judgment appealing Commerce’s application of zeroing methodology in the Final Consol. Court No. 12-00087 Page 31

Results. Vinh Hoan’s motion also contests Commerce’s rejection of its request for revocation of the

review as well as Commerce’s determination to cap the surrogate value for Vinh Hoan’s fresh

broken fish fillets at the surrogate value for whole live fish. Anvifish’s motion also alleges that

Commerce failed to calculate Anvifish’s margin as accurately as possible.

A. Vinh Hoan’s Motion for Judgment -- Revocation Request

Commerce rejected a request for revocation from Vinh Hoan because the request was

filed 232 days after the regulatory deadline established in 19 C.F.R. §351.222(e), stating that

revocations “require additional analysis beyond the requirements of an administrative review,

including conducting verification, and determining if sales of subject merchandise were made in

commercial quantities.” I&D Memo at 37. See Vinh Hoan’s Request for Revocation, PDoc 89;

Preliminary Results, 76 Fed. Reg. at 55873. Commerce noted that interested parties must be allowed

an opportunity to comment on the preliminary results of the agency’s revocation analysis, id., and

it distinguished the cases cited by Vinh Hoan, specifically stating that its determination to reject the

revocation request was based on considerations of fairness and accuracy. Id.

The defendant contends that Commerce acted reasonably and consistently with its

regulations and that Vinh Hoan has failed to demonstrate that Commerce’s rejection of the

revocation request was not supported by substantial evidence or an abuse of discretion and not in

accordance with law. See Vinh Hoan Br. at 8-24. Although Vinh Hoan claims that Commerce

abused its discretion in “fail[ing] to address” Vinh Hoan’s fairness and accuracy arguments,

Commerce, in rejecting the untimely request, determined that the request was not fair to the other

parties and the agency. Vinh Hoan has also failed to demonstrate that its 232-days late submission Consol. Court No. 12-00087 Page 32

is factually analogous to the five-day delay in Carbon Steel Flat Products and the timely submission

in PET Film from Korea. See I&D Memo at 38; Vinh Hoan Br. at 21-23 (citing Certain

Corrosion-Resistant Carbon Steel Flat Products and Certain Cut-to-Length Carbon Steel Plate

From Canada, 64 Fed. Reg. 2173 (Jan. 13, 1999); Polyethylene Terephthalate Film from Korea, 61

Fed. Reg. 36032 (July 9, 1996)).

Vinh Hoan’s brief appears to blur the distinction between the submission of factual

information and the submission of a revocation request. With the sole exception of NSK Corp. v.

United States, 33 CIT 1185, 637 F. Supp. 2d 1311 (2009), which Vinh Hoan cites for the general

proposition that Commerce must consider “an important aspect of the problem,” none of the cases

cited in Vinh Hoan’s brief involve revocation requests. See Vinh Hoan Br. at 8-24.20 Rather, the

cases upon which Vinh Hoan relies concern the timeliness of factual information submissions under

19 C.F.R. §351.301, not revocation requests made under 19 C.F.R. §351.222, and the court

previously recognized the “significant distinction” between Commerce’s administration of the two

provisions in upholding Commerce’s interpretation of the predecessor to 19 C.F.R. §351.222(e) as

a “mandatory, bright line rule requiring that a producer submit its revocation request during the

anniversary month of an antidumping duty order.” Samsung Electronics Co. v. United States, 20 CIT

1306, 1310, 946 F. Supp. 5, 9 (1996), aff’d, 129 F.3d 135 (Fed. Cir. 1997). See also Exportaciones

20 The other cases cited in Vinh Hoan’s brief for its proposition include the following: Timken U.S. Corp. v. United States, 434 F.3d 1345 (Fed. Cir. 2006); NTN Bearing Corp. v. United States, 74 F.3d 1204 (Fed. Cir. 1995); Grobest & I-Mei Industrial (Vietnam) Co. v. United States, 36 CIT ___, 815 F. Supp. 2d 1342 (2012) (“Grobest I”); Fischer S.A. v. United States, 34 CIT ___, 700 F. Supp. 2d 1364 (2010); Yantai Timken Co. v. United States, 31 CIT 1741, 521 F. Supp. 2d 1356 (2007); Maui Pineapple Co. v. United States, 27 CIT 580, 601, 264 F. Supp. 2d 1244, 1262 (2003); Usinor Sacilor v. United States, 18 CIT 1155, 872 F. Supp. 1000 (1994). Consol. Court No. 12-00087 Page 33

Bochica/Floral v. United States, 16 CIT 670, 671, 802 F. Supp. 447, 448 (1992) (sustaining

Commerce’s decision not to consider an untimely revocation request), aff’d, 996 F.2d 317 (Fed. Cir.

1993).

In addition to the distinction between factual information and revocation requests, the

cases Vinh Hoan cites are factually distinguishable from the present proceeding. Vinh Hoan relies

heavily on Grobest I, see Vinh Hoan Br. at 9, 12, 14, 19, however that case involved a separate rate

certification that was 95 days late, unlike the 232-day late request for revocation in this case. The

decisions in Maui Pineapple Co. and NTN Bearing Corp., see Vinh Hoan Br. at 12, are also

distinguishable because both involved minor clerical errors, not a revocation request. See Maui

Pineapple Co., 27 CIT at 599, 264 F. Supp. 2d at 1261 (“the number of cans per case for product

code 38900-72475 was mistakenly listed as eight rather than four and the conversion factor . . . was

incorrectly indicated as 0.33”); NTN Bearing Corp., 74 F.3d at 1207-08 (“[f]or 23 U.S. sales, one

digit in a 13-digit code . . . was entered incorrectly into the database . . . [and] NTN . . . had

mistakenly included in its U.S. sales database four sales which were actually sales to a Canadian, not

a U.S., customer”).

Vinh Hoan’s discussion of Commerce’s consideration of other revocation requests,

see Vinh Hoan Br. at 10-12, is not relevant, because the circumstance at bar involves Commerce’s

receipt of what the defendant not unreasonably characterizes as “an extremely untimely request”.

Def’s Resp. at 41. Moreover, in light of the uniqueness of each review, see, e.g., Peer Bearing

Co.-Changshan v. United States, 32 CIT 1307, 1310, 587 F. Supp. 2d 1319, 1325 (2008), Vinh

Hoan’s references to facts and determinations in the eighth administrative review, Vinh Hoan Br. Consol. Court No. 12-00087 Page 34

at 16, 18-19, are not germane to determining the reasonableness of Commerce’s findings in the

present review. Further, Vinh Hoan’s assertion that conducting a verification would not have created

any undue burden for Commerce is unconvincing because “[i]t is not the role of the court to second

guess Commerce’s allocation of its resources. Any assessment of Commerce’s operational

capabilities or deadline rendering must be made by the agency itself.” Grobest & I-Mei Industrial

(Vietnam) Co. v. United States, 36 CIT ___, ___, 853 F. Supp. 2d 1352, 1362-63 (2012) (Grobest

II) (citation omitted); see also Samsung Electronics Co. v. United States, 20 CIT 1306, 1309-10, 946

F. Supp. 5, 9 (1996) (“In response to a request for revocation, Commerce must initiate and conduct

an entire investigation . . . . If the plaintiff could command Commerce to conduct such an

investigation at its whim rather than only once per year, Commerce’s administrative efficiency would

be adversely affected.”) (citation omitted).

Citing Grobest I, 36 CIT at ___ n.36, 815 F. Supp. 2d at 1366 n.36, Vinh Hoan

claims that the rejection of its revocation request would work a substantial hardship on the company

because of the costs associated with its participation in the proceeding. Vinh Hoan Br. at 12-13. But

it is well-settled that ordinary participation in administrative proceedings is not a “substantial

hardship” or a “draconian penalty,” despite claims to the contrary. See Vinh Hoan Br. at 13; see also

Daido Corp. v. United States, 16 CIT 681, 685, 796 F. Supp. 533, 537 (1992) (“[o]rdinary burdens

and expenses associated with antidumping procedures . . . do not constitute irreparable injury”)

(internal quotation omitted); Matsushita Electric Industrial Co. v. United States, 823 F.2d 505, 509

(Fed. Cir. 1987) (finding of irreparable injury was “clearly erroneous” when based on “having to

comply with Commerce’s demands for data and verification” because “ordinary consequences of Consol. Court No. 12-00087 Page 35

antidumping duty procedures do not constitute irreparable harm”) (internal quotation marks omitted);

Nissan Motor Corp. v. United States, 10 CIT 820, 823, 651 F. Supp. 1450, 1454 (1986) (“[w]hile

this Court can appreciate that plaintiffs seek to avoid the expenditure of time and resources which

may ultimately prove unnecessary, this cannot be equated with the threat of immediate and

irreparable harm”). As such, Vinh Hoan has failed to show that Commerce’s rejection of its

untimely revocation request was an abuse of discretion or otherwise improper.

B. Defendant-Intervenors’ Motions for Judgment -- “Zeroing”

In the Final Results, when comparing the respondents’ normal value and net United

States price for purposes of calculating dumping margins, Commerce used the “AT”

(average-to-transaction) comparison method that included the use of zeroing. Under zeroing

methodology, for each control number or “CONNUM,”

Commerce uses the average normal value and on a month-to-month basis compares it to the individual United States transaction prices in that month. If the result is a transaction which is not dumped, i.e., there is no margin, Commerce sets the margin for that transaction at zero. Nonetheless, the sale price for the transaction goes into the denominator in calculating the final weighted-average dumping margin percentage, thereby lowering the percentage margin.

Union Steel v. United States, 36 CIT ___, ___, 823 F. Supp. 2d 1346, 1350 (2012) (internal citations

and footnote omitted), aff’d, 713 F.3d 1101 (Fed. Cir. 2013). Commerce did not alter these findings

in the Final Results. See I&D Memo at 27-34, 37-38.

Vinh Hoan, QVD, and Anvifish argue that Commerce acted unlawfully when it used

its zeroing methodology to calculate the final dumping margin in this case because Commerce

employed contradictory interpretations of the same statute (i.e., for the review as compared with

Commerce’s abandonment of that practice in investigations). See Vinh Hoan Br. at 26-31; QVD Br. Consol. Court No. 12-00087 Page 36

6-11; Anvifish Br. 6-19. The court disagrees. Commerce provided the same kind of detailed and

reasonable explanation for its interpretation of 19 U.S.C. §1677(35) as permitting the application

of its zeroing methodology in the administrative review at issue in this case that was recently upheld

by the appellate court in Union Steel, supra, and numerous decisions of this court. E.g., Thai Plastic

Bags Industries Co., Ltd. v. United States, 37 CIT ___, Slip Op. 13-21 (Mar. 19, 2013); Fischer S.A.

v. United States, 36 CIT ___, 885 F. Supp. 2d 1366 (2012); Camau Frozen Seafood Processing

Import Export Corp. v. United States, 36 CIT ___, 880 F. Supp. 2d 1348 (2012); Far Eastern New

Century Corp. v. United States, 36 CIT ___, 867 F. Supp. 2d 1309 (2012); Grobest II, supra; Union

Steel, supra. These decisions hold that Commerce has reasonably explained why it may use a

methodology offsetting positive and negative dumping margins in the context of investigations using

average-to-average comparisons, while continuing to employ zeroing in administrative reviews using

average-to-transaction comparisons, such as the one in this case. Commerce similarly explained in

the Final Results that employing offsets when using an average-to-average comparison methodology

“allows for a reasonable examination of pricing behavior, on average,” whereas employing zeroing

when making average-to-transaction comparisons is appropriate because Commerce is examining

“the pricing behavior of an exporter or producer with respect to individual export transactions.” I&D

Memo at 33.

Contrary to the contentions raised by Vinh Hoan, QVD, and Anvifish, 19 U.S.C

§1677f-1(d) provides for different comparison methods, and Commerce has analyzed the differences

between these distinct methodologies to determine that different interpretations of 19 U.S.C.

§1677(35) reasonably account for the inherent differences between the distinct methodologies. In Consol. Court No. 12-00087 Page 37

addition, the Federal Circuit in Union Steel recognized that Commerce’s explanation should not be

scrutinized in fragments, holding that Commerce’s rationale of making a limited change in certain

investigations for purposes of implementing a World Trade Organization (“WTO”) finding

“[c]ertainly . . . is relevant when considered in conjunction with the other explanations offered by

Commerce.” 713 F.3d at 1109-10; see also Union Steel, 36 CIT at ___, 823 F. Supp. 2d at 1357-58

(WTO compliance was valid part of total rationale for Commerce’s choice).

In sum, Commerce in the Final Results explained the differences between the

statutory comparison methodologies and why the differences justified its reasonable interpretation

of 19 U.S.C. §1677(35). Commerce’s application of zeroing is thus hereby sustained.

D. Anvifish’s Arguments re “Final Modification for Reviews”

Anvifish avers that Commerce abused its discretion when it declined to apply its

notice of Final Modification for Reviews21 to the subject review. Anvifish Br. at 17-19. In that

notice, Commerce announced, pursuant to section 123 of the Uruguay Round Agreements Act, that

it would cease to apply zeroing in reviews “pending before [Commerce] for which the preliminary

results are issued after April 16, 2012.” 77 Fed. Reg. at 8113. In the underlying review, Commerce

published the Preliminary Results in September 2011, i.e., seven months prior to the deadline

established in the Final Modification for Reviews. 76 Fed. Reg. 55872. This court has recognized

that Commerce has “no legal authority” to apply a section 123 determination “in a manner that

ignores the express legal directive set forth therein.” See Advanced Tech. & Materials Co. v. United

21 Antidumping Proceedings: Calculation of the Weighted-Average Dumping Margin and Assessment Rate in Certain Antidumping Duty Proceedings; Final Modification, 77 Fed. Reg. 8101 (Feb. 14, 2012). Consol. Court No. 12-00087 Page 38

States, 35 CIT ___, Slip Op. 11-105 at 16 (Aug. 18, 2011); see also Corus Staal BV v. United States,

32 CIT 1480, 1492-93, 593 F. Supp. 2d 1373, 1384-85 (2008) (holding that a section 123

determination is limited to its express terms). Thus, contrary to Anvifish’s claim, Commerce

properly declined to apply the Final Modification for Reviews to the Final Results.

Anvifish also alleges that Commerce failed to calculate Anvifish’s margin as

accurately as possible. As the basis for its claim, Anvifish contends that QVD’s weighted average

dumping margin, the rate upon which Anvifish’s separate rate was calculated, would have been zero

in the absence of zeroing. See Anvifish Br. at 17. Anvifish also points to the zero margins

calculated in the subsequent administrative review of the antidumping order as support for its

allegation. Id. at 17-18. However, as noted above, the Federal Circuit recently affirmed

Commerce’s use of zeroing in administrative reviews, and Anvifish’s reliance upon events that

occurred in the subsequent review is unpersuasive. See U.S. Steel Corp. v. United States, 33 CIT

984, 1003, 637 F. Supp. 2d 1199, 1218 (2009) (“each agency determination is sui generis, involving

a unique combination and interaction of many variables, and therefore a prior administrative

determination is not legally binding on other reviews before this court.”); accord Nucor Corp. v.

United States, 414 F.3d 1331, 1340 (Fed. Cir. 2005). Anvifish does not persuade that its preferred

methodology necessarily results in a more “accurate” margin, since for decades zeroing has been

held to be a proper philosophical interpretation of U.S. law, and it is only recently that a

“mathematical” interpretation has been imposed from without, due to the accretive process of U.S.

abdication of its national sovereignty to the vagaries of unaccountable foreign trade “organizations”. Consol. Court No. 12-00087 Page 39

Conclusion

For the above reasons, the matter must be, and hereby is, remanded for

reconsideration and further explanation in accordance with the foregoing.

The results of remand shall be filed with the court by April 20, 2015. Within seven

days thereafter, the parties shall confer and the plaintiffs shall file a joint proposed scheduling order

for comments, if any, thereon.

So ordered.

Dated: December 18, 2014 /s/ R. Kenton Musgrave New York, New York R. Kenton Musgrave, Senior Judge