Clearon Corp. v. United States

2015 CIT 91

• Court: United States Court of International Trade
• Decided: August 20, 2015
• Docket: Consol. 13-00073
• Status: Published

Opinion

Slip Op. 15 - 91

UNITED STATES COURT OF INTERNATIONAL TRADE

: CLEARON CORP., and OCCIDENTAL : CHEMICAL CORP., : : Plaintiffs, : : v. : Before: R. Kenton Musgrave, Senior Judge : UNITED STATES, : Consol. Court No. 13-00073 : Defendant, : : and : : ARCH CHEMICALS, INC., and HEBEI : JIHENG CHEMICAL CO., LTD., : : Defendant-Intervenors, : : and : : JUANCHENG KANGTAI CHEMICAL CO., : LTD., : : Defendant-Intervenor. : :

OPINION AND ORDER

[Remanding sixth (2010-2011) review of antidumping duty order on chlorinated isocyanurates from the People’s Republic of China a second time.]

Dated: August 20, 2015

James R. Cannon, Jr. and Thomas M. Beline of Cassidy Levy Kent (USA) LLP, of Washington, DC, for the plaintiffs.

Gregory S. Menegaz, J. Kevin Horgan, John J. Kenkel, and Alexandra H. Salzman, DeKieffer & Horgan, of Washington, DC, for the consolidated-plaintiff and defendant-intervenor Juancheng Kangtai Chemical Co., Ltd. Consol. Court No. 13-00073 Page 2

Peggy A. Clarke, Law Offices of Peggy A. Clarke, of Washington, DC, for the consolidated- plaintiff Hebei Jiheng Chemical Co., Ltd. and the consolidated-plaintiff and defendant-intervenor Arch Chemical Co., Ltd.

Jane C. Dempsey, Trial Attorney, Commercial Litigation Branch, Civil Division, U.S. Department of Justice, of Washington, DC, for the defendant. On the brief were Benjamin C. Mizer, Principal Deputy Assistant Attorney General, Jeanne E. Davidson, Director, and Patricia M. McCarthy, Assistant Director. Of counsel on the brief was David W. Richardson, Senior Counsel, Office of the Chief Counsel for Trade Enforcement and Compliance, U.S. Department of Commerce, of Washington DC.

Musgrave, Senior Judge: Before the court are the Final Results of Redetermination

Pursuant to Court Remand, Clearon Corp. and Occidental Chemical Corp., et. al., v. United States

(“Remand” or “RR”), Court No. 13-0018, RR-PDoc 69 (Dec. 11, 2014) submitted from the

defendant’s International Trade Administration of the U.S. Department of Commerce (“Commerce”

or “Department”). The matter covers the sixth (2010-2011) administrative review of the

antidumping duty order on chlorinated isocyanurates (“chlor-isos”) from the People’s Republic of

China (“PRC”). Familiarity with Clearon Corp. v. United States, 38 CIT ___, Slip Op. 14-88 (July

24, 2014) (“Opinion”) and the basis of that remand is presumed.

The defendant-intervenors Arch Chemicals, Inc. and Hebei Jiheng Chemical Co., Ltd.

(“Jiheng”) (together “Arch”) and Juancheng Kangtai Chemical Co., Ltd. (“Kangtai”) argue for

further remand. Plaintiffs Clearon Corp. and Occidental Chemical Corp. (together, “Clearon”) argue

for sustaining the remand results. For the following reasons, remand is again necessary.

I. Background

Briefly summarizing: after Chlor-Isos from the PRC, 78 Fed. Reg. 4386 (Jan. 22,

2013) (final 2010-2011 admin. review results), PDoc 169, and accompanying issues and decision

memorandum (“IDM”), PDoc 164 (together, “Final Results”) were summonsed here, the case was Consol. Court No. 13-00073 Page 3

voluntarily remanded on issues related to the determinations of surrogate factors of production

(“FOPs”), namely: (1) whether certain identified labor, retirement, and employee benefit expense

items among the selling, general and administrative (“SG&A”) items of a financial statement, upon

which Commerce relied for its financial ratios are inadvertently double-counted as a result of

Commerce’s recent change in policy to rely upon International Labor Organization (“ILO”) Chapter

6A data for valuing labor; (2) change in methodology for calculating intra-company transportation

costs; and (3) changes in the methodology employed for determining respondent’s by-product

offsets. Because the selection of the surrogate country was also remanded due to certain flaws in that

process, consideration of the parties’ further challenges to the surrogate valuation (“SV”) of urea,

hydrogen gas, chlorine, sodium hydroxide, and electricity was therefore deferred.

Upon remand, Commerce placed additional information on the record for comment

and issued questionnaires to Arch and Kangtai requesting further information on intra-company

transport of goods and on the by-product offset claims for ammonium gas and sulfuric acid.1 During

remand, Commerce again selected the Philippines as the primary surrogate country. RR at 31.

Commerce states that during remand it adjusted the normal value (“NV”) calculation by recalculating

the transportation cost of intermediate goods between factories for Jiheng, and by recalculating the

by-product offset using company specific information for Jiheng and Kangtai. Commerce also states

it revised the by-product calculation made to the draft remand calculations and clarified certain

sentences in its explanation of its decision not to adjust financial ratios to account for benefits

1 See RR at 3 (citations omitted). Consol. Court No. 13-00073 Page 4

included in the ILO surrogate value for labor. RR at 3. All other aspects of the Remand apparently

remained unchanged.

Regarding those remand results, Kangtai continues to contest Commerce’s

elimination of India in the surrogate country selection process.2 Kangtai and Arch both argue that

the labor FOP continues to double count certain indirect labor costs, and that Commerce’s

by-product methodology is unsupported by record evidence and is contrary to law.3

Clearon requests that the court accept the Remand “in its entirety.”4 The court

construes this to mean Clearon is satisfied with Commerce’s reconsideration of Arch’s and Kangtai’s

by-product offsets claims and that Clearon has therefore abandoned its own claims with respect

thereto. However, Clearon’s other claims concerning the surrogate valuation of urea and hydrogen

gas remain live, as do Arch’s and Kangtai’s claims regarding the surrogate valuation of chlorine,

sodium hydroxide, and electricity. As discussed herein, because the court must remand again

concerning Commerce’s selection of surrogate values for certain FOP’s and its by-product offset

methodology, the primary surrogate country selection remains an open question subject to

reconsideration as may be appropriate.

2 Kangtai’s Comments on Remand Results, RR-PDoc 76 (Jan. 28, 2015) at 1-23 (“Kangtai’s Cmts.”). 3 Kangtai’s Cmts. at 23-37; see also Arch’s Comments on Remand Results, RR-PDoc 75 (Jan. 28, 2015) at 3-17 (“Arch’s Cmts.”). 4 Clearon’s Comments on Remand Results, RR-PDoc 82 (Feb. 26, 2015) at 22 (“Clearon’s Cmts.”). Consol. Court No. 13-00073 Page 5

II. Jurisdiction and Standard of Review

The action was brought pursuant to Section 516A(a)(2)(B)(iii) of the Tariff Act of

1930, as amended, 19 U.S.C. §1516a(a)(2)(B)(iii). Clearon, Kangtai, and Arch have standing under

19 U.S.C. §1516a(d) and 28 U.S.C. §2631(c).

The party challenging a final administrative determination of the type at bar is

burdened with showing how it is “unsupported by substantial evidence on the record” or is not

“otherwise in accordance with law.” 19 U.S.C. §1516a(b)(1)(B)(i). Substantial evidence means

“more than a mere scintilla”, it must be “such relevant evidence as a reasonable mind might accept

as adequate to support a conclusion.” Universal Camera Corp. v. NLRB, 340 U.S. 474, 477 (1951),

citing Consol. Edison Co. v. N.L.R.B., 305 U.S. 197, 229 (1938). Commerce’s statutory

interpretations are considered pursuant to the familiar two-step analysis set forth in Chevron U.S.A.,

Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 842-43 (1984) (if “Congress has

directly spoken to the precise question at issue . . .” et cetera).

III. Discussion

A. Process of Selecting the Primary Surrogate Country

The selection of a primary surrogate country was remanded because Commerce had

not explained its GNI range selection process of country inclusion on its Surrogate Country List.

Opinion at 34. The Remand details how Commerce arrives at its list of surrogate countries and

provides historical context, particularly in the form of helpful charts and graphs showing the

widening GNI gap between India and the PRC over time. Commerce’s current practice involves

seeking GNI ranges that are “evenly distributed around the PRC’s GNI”. See RR at 9, referencing Consol. Court No. 13-00073 Page 6

Dongguan Sunrise Furniture Co. Ltd. v. United States, 38 CIT ___,____, 865 F. Supp. 2d 1216,

1238 (2012). From the annual release of the World Bank Development Report, Commerce looks

beyond that report’s “lower middle” income grouping of the PRC’s GNI when considering which

countries are economically comparable thereto.5 See id. at 9-14. On remand, Commerce again

selected the Philippines as its primary surrogate country and disregarded India not only for use as

the primary surrogate country but also for valuing certain FOP’s. Id. at 2, 31-39.

Kangtai argues the very scope and length of Commerce’s “first ever” explanation of

its country selection process demonstrates that Kangtai was prejudiced by the lack of explanation

while the segment was ongoing, and that had Kangtai been aware of the policy considerations it

would have been in a better position to research surrogate countries. It argues dropping India from

the list was “sudden” because it had been the country of choice for 25 years, and that the “mere

existence of a regulation, one that had not been used to change the primary surrogate in 25 years,

cannot justify or render reasonable the Department’s implementation of this regulation (GNI reliance

for economic comparability) in this instance.” Kangtai’s Cmts. at 2, referencing Opinion at 44-45

(Kangtai’s italics). Further emphasizing the point, Kangtai argues:

The number and size of the companies offered in India versus the Philippines demonstrates this emphatically. The existence of a weekly chemical reporter in India and the non-existence, on this record, of a domestic chemicals trading market in the Philippines demonstrates this emphatically. Procedurally, the Department has dropped India after Kangtai’s POR sales were made, based on GNI data that was not

5 In particular, Commerce states that it relies on its experience and professional judgment on a range of factors when considering to add or remove countries from the list, including the surrogate value requirements for the existing products under investigation, the data quality and availability of alternative surrogate countries, economic diversity of the manufacturing sector in the alternative countries, and the degree of specificity in the import data relied on to value the FOP’s. RR at 13. Consol. Court No. 13-00073 Page 7

available when the sales were made. This totally frustrates the compliance purpose of the statute and is not a problem that a market economy respondent would have to encounter. Accordingly, the Department’s procedure was arbitrary and unreasonable.

Id. at 3. Kangtai thus continues to argue that India’s data quality outweigh those of countries on the

Surrogate Country List, and that Commerce has unreasonably refused to even consider India’s data

quality when making its primary surrogate country selection.

Commerce’s response is to explain that its general rule is to select a surrogate country

from its list of surrogate countries but that it also considers countries that other parties propose. It

generally selects a surrogate country that is “at the same level of economic development” as the

NME

unless it is determined that none of the countries are viable options because (a) they either are not significant producers of comparable merchandise, (b) do not provide sufficient reliable sources of publicly available surrogate value (SV) data, or (c) are not suitable for use based on other reasons.

RR at 6 (internal citation omitted). Commerce only selects countries that are not at the “same level

of economic development as the NME country, but still at a level of economic development

comparable to the NME country” when data considerations outweigh the difference in levels of

economic development.6 India’s data are not “better,” Commerce maintains, because they are not

from a country “at a level of economic comparability” to the PRC, noting that to be selected for the

list over the other countries which are at the same level of economic comparability to the PRC, “the

data quality and availability from India must outweigh its per capita GNI disparity with respect to

the PRC”. RR at 15. Commerce maintains more importantly that because the Philippines has

6 Id.; see also RR at 35 (“Countries outside the implied GNI range are also considered, [but] are selected only to the extent that the data considerations outweigh the level of economic development factor (as indicated by disparate GNIs).”)(italics in original). Consol. Court No. 13-00073 Page 8

“reliable and useable” data (that Commerce also characterizes as “quality” data), Commerce did not

need to consider the Indian data’s quality. RR at 35-36, 38 (internal citations ommitted).

As discussed in the Opinion, Commerce’s primary reliance on per capita GNI to

identify economically comparable countries was not unreasonable and was in accordance with law.

See Opinion at 25.7 In the Remand, Commerce provided a reasonable explanation of how it

generated the Surrogate Country List and selected the range of GNI’s that qualify countries as

proximate and “economically comparable” to the PRC. See RR at 4-19. However, Commerce’s

selection of the Philippines as the primary surrogate country relies heavily on its determination that

the Philippine data for valuing chlorine and hydrogen gas is the “best available information”8 on the

record, and as discussed infra, that determination cannot be sustained at this time; therefore, the

primary surrogate country selection remains an open question, to be addressed on remand as

appropriate.

7 Kangtai re-argues several points in its motion for judgment regarding the methodology Commerce applies when creating the surrogate country list that the court addressed and rejected in that opinion and which it will not again consider here. Specifically, that it was prejudiced by Commerce’s “dropping India from the Surrogate Country List”, that Commerce’s use of GNI in determining economic comparability was unreasonable, and that the statute requires Commerce to equally weigh economic comparability and significant production of comparable merchandise. Kangtai’s Cmts. at 1-23, and Kangtai’s Motion for Judgment on the Agency Record, ECF No. 30 (Aug. 14, 2013) (“Kangtai’s Br.”) 6-11. See Intergraph Corp. v. Intel Corp., 253 F.3d 695, 697 (Fed. Cir. 2001) (noting that the law of the case doctrine “generally bars retrial of issues that were previously resolved”) (internal citations ommitted). 8 IDM cmt. 2 at 7 (“the Department is selecting the Philippines as the surrogate country given its superior data availability”), and at 8 (“[s]ince the Department has usable information from the Philippines on the record to value all inputs, except for steam, the issues raised above [concerning “Surrogate Values if the Philippines is Not Selected as the Surrogate Country”] are moot”), and at 11 (“[t]herefore, we can find no basis to consider the Philippines GTA value for liquid chlorine to be unreliable and find no reason to consider information from a non- Philippine source”). Consol. Court No. 13-00073 Page 9

B. Consideration of India as Surrogate Country for Valuing FOP’s

While in the Remand Commerce added India’s per capita GNI to the record, provided

a reasonable explanation of the methodology it applied when determining that India was “less

economically comparable” than the PRC and that its GNI did not qualify it for the Surrogate Country

List, Commerce did not provide the court with the data analysis it claims to undertake, or it did not

adequately articulate the analysis if it did in fact undertake it, when considering India and the data

therefrom should be used for valuing certain FOP’s. RR at 4-17. Commerce’s adherence to it’s

regulatory “preference” to value all FOP’s from one surrogate country, in this instance the

Philippines, and to completely disregard Indian data for consideration for certain FOP’s was

unreasonable.

It must first be pointed out that the statutory standard for Commerce valuing FOP’s

is not whether the surrogate data are merely “usable”, as Commerce categorized the Philippine data,

but whether they are the “best available”. Jiaxing Brother Fastener Co. v. United States, 38 CIT

___, ___, 961 F. Supp. 2d 1323, 1333 (2014) (“Jiaxing Brother”).9 While it is reasonable for

Commerce to prefer to use data from a surrogate country that is at a comparable level of economic

development over one that is at a less comparable level of development, when presented with a “less

economically comparable” country off the list it must still provide an analysis of how the data from

the less comparable country presented does not outweigh its economic disparity. RR at 36.

9 Commerce declaring that the Philippine data are “quality” data, on the other hand, indicates that they satisfy Commerce’s five-factor test of “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 [are] publicly available”. Import Policy Bulletin 04.1. Consol. Court No. 13-00073 Page 10

In the selection of primary surrogate country, Commerce found the fact that India may

have a significant chemical industry comparable to the PRC was “irrelevant” to its analysis of India’s

level of economic development because the point only addresses whether the country is a “significant

producer” of comparable merchandise. RR at 38.10 On the one hand, it is unreasonable for

Commerce to acknowledge that the level of economic comparability and the quality of a country’s

data are two separate considerations, and then refuse to undertake a comparative analysis, of the type

Commerce here implies it must undertake, in order to determine whether data quality outweighs the

fact that a country is not on the surrogate country list. See RR at 6, 14-15, 35. The fact that India’s

data originates from a country not inside the GNI band does not implicate those data’s availability

or quality. On the other hand, requiring a full comparative evaluation of the data quality of a country

not on the surrogate country list, as compared with the data of those that are, would be a pointless

exercise if in the final analysis the non-listed country’s data quality is in fact insufficient to overcome

the fact that the country is not on the surrogate country list and substantial evidence of record can

support the conclusion that data for another country thereon are the “best available information” for

purposes of selecting surrogate values for FOP’s. Commerce therefore acts not unreasonably in

burdening the party proposing a non-listed country with demonstrating that no country on the

10 It also observed that “[i]f a country is a significant producer of comparable merchandise, then the economy of the surrogate country is developed enough to support an industry in the comparable merchandise.” RR at 38. This is confusing. Is the point here that because Commerce finds India’s GNI “not comparable” to the PRC’s, India cannot, therefore, be a “significant producer” of comparable merchandise or cannot have a significant chemical industry comparable to that of the PRC, in contradiction of the roughly 25 years of prior proceedings in which India has been relied upon as an appropriate primary surrogate country or even a secondary surrogate country, as the record of the preliminary determination of this proceeding shows? Or is the point that India’s economy is “developed enough” that its economy provides a fruitful comparison (i.e., is “economically comparable”) to the NME economy under consideration? Consol. Court No. 13-00073 Page 11

surrogate country list provides the scope of “quality” data that it requires in order to make a primary

surrogate country selection.

However, if that threshold is met, then Commerce must consider the quality of the

data on the country not on the list that a party proposes. Towards that end, determining the interstice

of Commerce’s five-factor data quality checkbox vis-à-vis particular datasets is a start, but the

analysis does not end there if there are other relevant qualities of the datasets that require

consideration. Commerce “must consider the record as a whole, including evidence that supports

as well as evidence that fairly detracts from the substantiality of the evidence”, Nucor Corp. v.

United States, 32 CIT 1380, 1384,594 F. Supp. 2d 1320, 1332 (2008) (internal quotation marks

omitted), aff’d, 601 F.3d 1291 (Fed. Cir. 2010), as it is the facts that drive the law, not the other way

around.

Commerce has steadfastly refused to address Kangtai’s arguments concerning the

quality of the data of record with respect to India’s chemicals industry and Kangtai’s conclusion that

those data are far superior the Philippines’ chemicals industry data, on the ground that the Philippine

data satisfy Commerce’s five-factor test. The court therefore deems this as an admission on

Commerce’s part as to the quality of the data covering India’s chemicals industry. The validity of

both parties’ positions (Kangtai’s as well as Commerce’s) on the issue of resort to India as the

primary or even secondary surrogate country for valuing FOP’s is therefore dependant on the

reasonableness of Commerce’s conclusions as to the quality of each challenged element of the

Philippines data. Cf. Preliminary Results (selecting South Africa’s primary surrogate country but

resorting to India as secondary surrogate for certain data). Consol. Court No. 13-00073 Page 12

At this point, after considering the Remand and the parties positions, the court must

here conclude that Commerce’s selection of the Philippines as the primary surrogate country from

the Surrogate Country List has general support in the record. However, as discussed below, because

the choice of the Philippines was expressed in the Remand as largely dependant upon Philippine

import data for chlorine and hydrogen gas, during which Commerce ignored a previously well-

articulated preference for domestic data for these types of chemicals (even including reliance upon

a domestic source from a non-surrogate-list country; see, e.g., Preliminary Results), and because the

latter determinations have not been reasonably explained, the choice of the Philippines as the primary

surrogate country11 remains an open question. See RR at 36-39.12

C. Surrogate Valuation of Hydrogen Gas and Chlorine

For the Preliminary Results, Commerce looked to its second surrogate country, India,

and used the values from three and four Indian producers, respectively, of hydrogen gas and chlorine

producers to value those chemicals. See Prelim. SV Memo, at 12-13 and Appx. III.39 & III.40, PDoc

104. In doing so, Commerce explicitly recognized the previous reviews (as well as the Preliminary

11 Or “countries” -- as provided by 19 U.S.C. §1677b(c)(1)(B). 12 Cf. also, Kangtai’s Surrogate Country Cmts., PDoc 58 (Dec. 19, 2011) (acknowledging that India is no longer on the Surrogate Country List but urging flexibility about what Commerce “will consider ‘comparable’ production in the countries it now does list as comparable”); Jiheng’s Prelim. SV Submission, PDoc 65 (Jan. 9, 2012) (urging reliance upon data from India for valuation of steam and water consistent with Multilayered Wood Flooring, in which the Philippines was also selected as the primary surrogate country); Clearon’s Prelim. SV Submission, PDoc 66 (Jan. 9, 2012) (arguing for use of Indian financial statements for calcium hypochlorite and stable bleaching powder values due to lack of publicly available financial statements for producers thereof in South Africa, the petitioners’ preferred choice of primary surrogate country); Kangtai’s Prelim. SV Submission, PDoc 70 (Jan. 9, 2012) (urging reliance upon Indian financial statements’ value of chlorine notwithstanding argument that Commerce “should look to the Philippines and/or to Thailand for industries producing comparable product”) Consol. Court No. 13-00073 Page 13

Results) in which it had found that both hydrogen and chlorine are not only infrequently traded on

an international basis, but that due to the very nature of those chemicals they face special concerns

both in transporting and in packaging, which are exacerbated over longer distances, greatly adding

to their costs. See id.13; see, e.g., Chlor-Isos from the PRC, 76 Fed. Reg. 40689, 40695 (July 11,

2011) (prelim. admin. review), unchanged in Chlor-Isos from the PRC, 76 Fed. Reg. 70957 ( Nov.

16, 2011) (final admin. review). For the Final Results, however. Commerce used GTA import data

from the Philippines to value chlorine and hydrogen gas. IDM cmts 7 & 8, at 11-16.

Clearon and Kangtai both argue that the domestic Indian prices are better alternatives,

not only because the volumes of those chemicals in the Philippines import data are small, and the

prices contained therein unreliable, but because Commerce did not articulate what had changed about

those chemicals’ “nature” since the Preliminary Results to suddenly render import data for them

reliable surrogate values. Clearon’s Motion for Judgment on the Agency Record, ECF No. 31 (Aug.

15, 2013) (“Clearon’s Br.”) at 17-21; Kangtai’s Br. at 11-27. Clearon and Kangtai also present

myriad other arguments for their respective positions,14 which the court has considered, but it will

13 The Preliminary Results specifically acknowledge that “chlorine gas and hydrogen gas are not frequently traded on an international basis,” Prelim. SV Memo at 4, and that “due to the very nature of chlorine, it faces special concerns both in transporting and in packaging, which are exacerbated over longer distances, greatly adding to the cost of chlorine”, id. at 12 (italics added); therefore, due to “these reasons, the Department continues to find that the GTA does not provide the best surrogate value for chlorine”, id. With respect to hydrogen gas, Commerce stated that it “has previously determined that the GTA does not provide the best representative surrogate value for hydrogen because hydrogen, like chlorine, is not frequently traded on an international basis, and incurs special transport costs over long distances.” Id. at 13 (italics added). 14 Commerce raises a litany of points in response, interalia : (1) the court has recognized that a small volume of imports, by itself, does not establish that the import data are aberrational, Trust Chem Co. v. United States, 35 CIT ___, ___, 791 F. Supp. 2d 1257, 1265-66 (2011); (2) interested (continued...) Consol. Court No. 13-00073 Page 14

follow the path of least resistance to focus only on those points pertinent to the ultimate conclusion,

to wit, that there are certain flaws in Commerce’s determinations with respect to surrogate valuations

of chlorine and hydrogen gas, which must therefore be remanded for further analysis,

reconsideration, or explanation.

As indicated, the IDM does not articulate direct responses to a number of Clearon’s

and Kangtai’s points, in particular those concerning the higher transportation and packaging costs

14 (...continued) parties bear the burden of creating an adequate record, QVD Food Co. Ltd. v. United States, 658 F.3d 1318, 1324-25 (Fed. Cir. 2011); (3) at the time of the Preliminary Results the record lacked Philippine production data, and Commerce accordingly could not determine whether the Philippines was a significant producer of comparable merchandise, IDM cmt. 7 at 11, 13; (4) after the Preliminary Results Commerce examined data availability for both South Africa and the Philippines to determine which country had superior data availability, IDM cmt. 2 at 7; (5) once it had selected the Philippines as the primary surrogate country, Commerce applied its regulatory preference to value all factors in a single country where possible and examined the record to identify Philippine data, 19 C.F.R. § 351.408(c)(2); (6) the only Philippine surrogate value on the record for chlorine was the Philippine GTA import data for chlorine, IDM cmt. 7 at 11-13; (7) there was no evidence on the record to value hydrogen using a source from one of the other economically comparable countries, IDM cmt. 8 at 14; (8) the Philippine GTA import data for hydrogen and chlorine satisfied its criteria of being product-specific, representative of a broad-market average, publicly available, contemporaneous with the period of review, and free of taxes and duties, e.g., IDM cmt. 7 at 12; (8) Commerce selects data from outside of the primary surrogate country only when data sources from the primary surrogate country cannot provide reliable surrogate values, Certain Frozen Fish Fillets from the Socialist Republic of Vietnam, 68 Fed. Reg. 37116 (June 23, 2003) (final LTFV determin.), and accompanying I&D Memo at cmt. 14; (9) because India is not on the economically comparable Surrogate Country List “any comparison to data from India is inappropriate”, Defendant’s Response to Plaintiff’s Motions for Judgment on the Agency Record, ECF No. 49 (Feb. 2, 2014) (“Def’s Resp.”) at 32 (citation omitted); (10) using data from a country that is not on the list of economically comparable countries could result in distortions in Commerce’s calculations, Clearon Corp. v. United States, 37 CIT ___, ___, Slip Op. 13-22 (Feb. 20, 2013) at 12; (11) the record contained no data demonstrating that chlorine and hydrogen gas are rarely traded internationally because the parties failed to place price data on the record from either the domestic industries therefor or from other economically comparable countries, and “Commerce cannot make a finding that Philippine GTA import prices vary significantly when compared to other GTA import data without this information”, See Def’s Resp. at 33. Consol. Court No. 13-00073 Page 15

associated with movement of those chemicals and why Commerce’s expressed preference for

valuing all FOPs from a single country should trump its other (presumably co-equal) preference for

using domestic prices over import prices especially where these chemicals are concerned.

Commerce’s articulation in the IDM is essentially that it finds that the Philippine GTA import data

for those chemicals are reliable because they constitute “commercial quantities” and their prices have

not been shown to be aberrant or distorted.15 Rather than abide by its previous statements concerning

the “nature” of hydrogen gas and chlorine, Commerce also shifts the burden onto Clearon and

Kangtai to provide proof for the record thereof. Hence, whether substantial evidence of record moots

the points Clearon and Kangtai raise, thereby rendering Commerce’s improper burden-shifting16

harmless error, depends upon the validity of Commerce’s ultimate conclusion.

As a “preliminary” matter, Kangtai disagrees with Commerce’s implication that

Commerce was unable to consider the Philippines import data for the Preliminary Results at the time

thereof. Kangtai contends that import data for chlorine pertaining to South Africa and the

Philippines were on the record before the Preliminary Results, and that nothing in the record has

15 Commerce, supported by Clearon (thus undercutting its own argument with respect to hydrogen gas), makes the point that the record of this review does not indicate that the cost of shipping chlorine is so burdensome that it is not a “frequently” traded international good, and that Kangtai identifies no record evidence indicating that the containers used to store chlorine domestically could not be used to ship chlorine internationally, such that those “special . . . concerns” of transport and packaging exist regardless of whether those chemicals are moved domestically or internationally. Whether that was rather Commerce’s burden -- to explain in the Final Results why it was reversing its precedent on the “nature” of chlorine and hydrogen gas this post hoc rationalization here will not carry the day. SEC v. Chenery Corp., 332 U.S. 194, 196 (1947) (agency action is to be upheld, if at all, only on the grounds articulated by the agency itself). 16 Final findings in prior reviews become the law of the case and “agency action is arbitrary when the agency offers insufficient reasons for treating similar situations differently”. See Dongbu Steel Co., Ltd. v. United States, 635 F.3d 1363, 1371 (Fed. Cir. 2011). Consol. Court No. 13-00073 Page 16

changed since those results that would cause Commerce to select Philippine import data to value

chlorine. Kangtai’s Br. at 15-16. The point is somewhat at odds with Kangtai’s overall contention.

Commerce responds that Kangtai’s point ignores the fact that at the time of the Preliminary Results,

the record lacked Philippine production data, and that Commerce accordingly could not determine

whether the Philippines was a significant producer of comparable merchandise, and therefore

whether it could use Philippine data for surrogate values. Def’s Resp. at 23, referencing IDM cmt.

7 at 12-13. But that does not appear to be the case, as Clearon also makes the point that Mabhuay

Vinyl Corporation (“MVC”), upon whose financial statements Commerce determined to rely, “was

a major producer of chlorine in the Philippines”, Clearon’s 56.2 Response Brief, ECF No. 47 (Feb.

24, 2014)( “Clearon’s Resp.”) at 32, referencing Jiheng Final SV Submission , PDoc 122 (Sep. 5,

2012), Att. 1 at 14, and that MVC’s 2010 financial statement was of record at the time of the

Preliminary Results. See also Jiheng’s Prelim. SV Submission at Tab 4. Arguably, therefore, there

were at least sufficient data of record at the time of the preliminary determination to have determined

if the Philippines was a “significant producer” of comparable merchandise, just as Commerce did

when “conceding” (according to Kangtai) that India was a significant producer of comparable

merchandise and selecting the Indian data for chlorine at the time of the Preliminary Results.

Kangtai’s Br. at 16. Be that as it may, “Preliminary Results are just that -- preliminary”, and parties

may “not presume Commerce would not adopt a different approach in determining the Final

Results.” Changshan Peer Bearing Co., Ltd. v. United States, 38 CIT ___, ___, 953 F. Supp. 2d

1354, 1363 (2014) (italics removed) (“Changshan Peer”). Consol. Court No. 13-00073 Page 17

All parties acknowledge that there is some degree of international trade in chlorine

and hydrogen gas. The issues appear to be (1) whether substantial evidence of record supports

finding that the import data represent “commercial quantities” and (2) whether the import prices can

be concluded non-aberrant. As to these questions, the court requested further briefing on what

constitutes “commercial quantities” and the standard for establishing that an import statistic for a

particular input represents or does not represent a “commercial quantity.” See Supplemental Briefing

Request, ECF No. 89 (May 8, 2015).

Clearon responded that whether a quantity is a “commercial quantity” depends upon

the product itself and the manner in which it is traded in the market; thus a commercial quantity of

a gas might be an ISO tank, a 100-kg pressurized cylinder, or a container holding a large number of

smaller volume cylinders.17 Kangtai responded similarly, stating that its understanding of the

agency’s practice is that the definition of a commercial quantity is “contextual and somewhat

flexible.” Kangtai also argued that at a minimum the term “must be understood and interpreted in

the context of the quantities previously considered by the agency itself in the previous review

segments of this very order.”18

17 Clearon’s Resp. to Court Questions at 8-9, referencing Ferrosilicon from Russia and Venezuela, Inv. No. 731-TA-1224-1225 (Preliminary), USITC Pub. 4426 (Sep. 2003) (ferrosilicon shipped in super sacks, pallet boxes, drums, and 25 and 50 pound bags); Certain Stilbenic Optical Brightening Agents from [the PRC] and Taiwan, Inv. Nos. 731-TA-1186-1187 (Final), USITC Pub. 4322 (May 2012) (aqueous solutions shipped in bulk by tank truck or rail cars or in non-bulk by drums or intermediate bulk containers; powder shipped in bulk bags); Certain Potassium Phosphate Salts from [the PRC], Inv. Nos. 701-TA-473 and 731-TA-1173 (Final), USITC Pub. 4171 (July 2010) (sales to distributors typically made in truckloads). 18 Kangtai’s Resp. to Court Questions at 5-6. Consol. Court No. 13-00073 Page 18

Commerce’s response to the court’s questions indicated that it “generally” compares

the total quantity of the input imported by the primary surrogate country against the total quantities

for the same input imported by other potential surrogate countries on the surrogate country list.19

Arch, however, provided references to administrative comparisons of an imported quantity to the

quantity of the domestically produced product in the surrogate country under consideration.20

As indicated below, it appears Commerce does both, which also appears to be

appropriate.21 On that note, however, Commerce here maintains that “[t]he appropriate comparison

19 Def’s Resp. to Court Questions at 8. 20 Arch’s Resp. to Court Questions at 11, referencing Certain Seamless Carbon Alloy Steel Standard, Line, and Pressure Pipe From the PRC, 75 Fed. Reg. 57449 (Sep. 9, 2010) (final LTFV determ., inter alia) and accompanying I&D Memo at cmt. 8 (comparing imported quantity to quantity of domestically produced product); Certain Steel Threaded Rod From the PRC, 74 Fed. Reg. 8907 (Feb. 27, 2009) (final LTFV determ.) and accompanying I&D Memo at cmt. 3 (“the Department finds that there is no record evidence to indicate that any of the reported values are aberrant or unrepresentative of commercial quantities”); Lightweight Thermal Paper From the PRC, 73 Fed. Reg. 57329 (Oct. 2, 2008) (final LTFV determ.) and accompanying I&D Memo at cmt. 10 (addressing a commercial quantity challenge by stating that the challenged surrogate value was not aberrational and represented the best information available). 21 Full analysis of all data of record, e.g., cross-country comparisons of import-with-import, import-with-domestic, domestic-with-domestic, is to be encouraged to the extent it paints the fullest picture of whether particular data are appropriate for purposes of surrogate valuation and produces the greatest accuracy in the attempt to reflect a surrogate’s commercial appropriateness to a respondent’s actual production experience. Cf., e.g., Fuwei Films (Shandong) Co., Ltd. v. United States, 36 CIT ___, ___, 837 F. Supp. 2d 1347, 1355 (2012) (noting Commerce’s reason for rejecting import statistics in that case, to wit, that they “contained an insignificant quantity of imports not representative of the DuPont Group’s PET chip purchase volume or consumption experience”); Freshwater Crawfish Tail Meat From the PRC, 75 Fed. Reg. 79337 (Dec. 20, 2010) (final rev. and new shipper results) and accompanying I&D Memo at cmt. 3 (“the Spanish import prices may or may not, arguably, constitute information that is directly representative of the production experience of the respondents in these reviews”); Circular Welded Austenitic Stainless Pressure Pipe from the PRC, 73 Fed. Reg. 51788 (Sep. 5, 2008) (prelim. LTFV determ.) and accompanying I&D Memo at “Selection of Surrogate Country” (“because India better represents the experience of producers of (continued...) Consol. Court No. 13-00073 Page 19

for the prices represented in the Philippine GTA import data are import prices from other countries

on the comparable countries list, or prices from Philippine domestic companies, none of which are

on the record”. Def’s Resp. at 27. At this point, the court fails to understand why that is

“appropriate” or why the prices represented in the Philippine GTA import data cannot be compared

with domestic price data from a country (or countries) not on the Surrogate Country List, when the

comparison would be at least for the purpose of showing those prices -- for that one FOP -- in relief,

e.g., through a August 17, 2015comparison of import price data with Indian domestic price data.

See, e.g., Jiaxing Brother, supra, 38 CIT at ___, 961 F. Supp. 2d at 1332-35 (significantly lower

domestic Indian price as compared to import price and fluctuation in the Indian import volumes and

prices revealed comparable fluctuation in the Thai import volumes and prices, implying that the

“only reasonable inference one could draw from the administrative record is that the Thai import

values are similarly affected and thus do not reflect domestic Thai HCL prices”). Cf. Blue Field

(Sichuan) Food Indus. Co., Ltd. v. United States, 37 CIT ___, ___, 949 F. Supp. 2d 1311, 1317

(2013) (discussing use of benchmark data, which “need not come from an economy comparable to

the foreign producer’s”) (citation omitted) with Hebei Metals & Minerals Imp. & Exp. Corp. v.

United States, 29 CIT 288, 299-300, 366 F. Supp. 2d 1264, 1273-74 (2005) (“[D]omestic price is

preferred for the calculation of surrogate values by prior practice, policy, and logic. All else being

equal, tax-and duty-free domestic data is clearly preferable over imports data”); & Yantai Oriental

Juice Co. v United States, 26 CIT 605, 617 (2002) (rejecting more contemporaneous import data

21 (...continued) subject merchandise and provides better financial data[,] we have selected India as the surrogate country”). Consol. Court No. 13-00073 Page 20

because the Department failed to explain why the industry would purchase more expensive imported

coal over domestic coal); & Rhodia, Inc. v. United States, 25 CIT 1278, 1287, 185 F. Supp. 2d 1343,

1352 (2001) (“Rhodia”) (“Commerce has a stated preference for the use of the domestic price over

the import price, all else being equal”). Explanation from Commerce would therefore assist, in

accordance with the following.

The entirety of Commerce’s support for finding that the Philippines data represent

internationally traded commercial quantities for hydrogen and chlorine rests on its decision in

Glycine from the PRC, 77 Fed. Reg. 64100 (Oct. 18, 2012) (final rev. results) (“Glycine from PRC”).

IDM cmt. 7 at 10-12, cmt. 8 at 13-14. See Glycine from PRC, I&D Memo cmt. 1 at 3-9. Issued

subsequent to the Preliminary Results, in Glycine from PRC Commerce selected Indonesian GTA

import data for chlorine, claiming that they represent commercially significant quantities.22

Commerce emphasizes here that the Philippine GTA import data show over 1,000 metric tons of

chlorine imported into the Philippines, that the quantity of imports in this review is higher than

imports in the previous reviews, see Kangtai’s Admin. Rebuttal Br., PDoc 159 (Dec. 10, 2012) at

16, and that since Glycine from the PRC determined 2,000 MT of chlorine to be a commercially

22 See Glycine from PRC, accompanying I&D Memo cmt. 1 at 6-9. Specifically therein, Commerce determined that Indonesian GTA import prices for chlorine were not aberrational because Indonesia’s average unit value was within the range of values of imports from countries on the economically comparable list and because Indonesia had the highest volume of chlorine among the countries on the list. Id. at 6-7. Commerce therein stated that in previous reviews for glycine it had rejected import prices because the import volume of chlorine was just one metric ton, and because the import volume for Indonesia during the relevant review period exceeded 2,000 metric tons, Commerce determined that the Indonesian GTA import data “show[ed] that liquid chlorine is shipped frequently on an international basis and in substantial commercial quantities, thus undermining the notion that high transportation costs are prohibitive of a robust international trade in chlorine.” Id. at 9. Consol. Court No. 13-00073 Page 21

representative quantity, the 1,000-plus MTs of chlorine imports into the Philippines cannot be

dismissed as “commercially insignificant.” Def. Resp. at 26.

On the issue of the surrogate valuation of chlorine, Glycine from the PRC is

essentially ipse dixit, and Commerce here neglects to mention that the issue is among those of that

determination that are under appeal. See generally, Baoding Mantong Fine Chemistry Co., Ltd. v.

United States, Court No. 12-00362; see also id., ECF No. 30, Motion for Judgment (July 22, 2013)

at 20-23. The court will therefore accord Glycine from the PRC only a Skidmore level of deference

at this time. See Skidmore v. Swift & Co., 323 U.S. 134, 140 (1944) (“power to persuade, if lacking

power to control”).

Kangtai also stresses that Commerce “had consistently rejected the use of imported

chlorine values for two reasons[ ], one of [which] was that ‘chlorine is not frequently traded on an

international basis’”, a finding on a record that contained imports into a potential surrogate country

that exceeded 2,000 MT, in stark contrast to this record of only 1,000 MT of chlorine imported into

the Philippines being found commercially significant. Kangtai’s Resp. to Court Questions at 5-6,

referencing AR09-10 Chlor-Isos Prelim. SV Memo at 12 & id. at Att. XXXII(a). Kangtai argues that

the frequency23 of imports into the Philippines does not equate to commercial amounts thereof,

because Commerce had to aggregate an entire year’s volume to reach 1,000 MT and the actual

volumes underlying the transactions are far smaller.

23 The court understands “frequent” “frequently” and “frequency”, in the sense used by the parties, to refer to the number of transactions in a given period. In that sense, “frequent” produces a greater number of data points during a specific period, thus imparting breadth to the average, but that does not, in itself, indicate anything with respect to the prices underlying the transactions. On the other hand, given a total known volume, then the higher the frequency (i.e., number of transactions), the lower the average volume. Consol. Court No. 13-00073 Page 22

Commerce, however, points out that the record at bar contains comparable GTA

import data, albeit pointing to the only other country on the list of economically-comparable

countries with GTA import data for chlorine -- South Africa (see Clearon’s Prelim. SV Submission

at Ex. 23), and when Commerce compared the 4.6 MT import volume for South Africa against the

1,062.3 MT import volume for the Philippines, Commerce found that the import volume for the

Philippines exceeded those of the other import data on the record pertaining to a country on the

Surrogate Country List (i.e., “economically comparable”), and therefore, as in Glycine from the PRC,

Commerce determined that the Philippine imports represented a significant commercial quantity

during the period of review.24 See Def’s Resp. at 25, referencing IDM cmt. 7 at 14; Final SV Memo,

PDoc 167 (Jan. 18, 2013) at Appx. III.39.

In other words, as Kangtai argues, Commerce is simply saying that because a

particular quantity of chlorine was imported into the Philippines, the import data therefor was per

se superior, See Kangtai’s Br. at 16. Such a response, of course, ignores the data for India of record.

Kangtai also emphasizes that in contrast to Glycine from the PRC, in the prior 2009-2010 review

of chlor-isos (“AR09-10 Chlor-Isos”), Commerce rejected GTA import data for chlorine representing

approximately 2,000 MT, finding that chlorine was not only not frequently traded into India, but not

frequently traded into any country on an international basis. Kangtai’s Reply at 15 (Kangtai’s

italics), referencing AR09-10 Chlor-Isos Prelim. SV Memo at 12 & at Att. XXXII(a).

Responding, Commerce implies that it “only” rejected Indian GTA import data in

AR09-10 Chlor-Isos because of the “wide range of import volumes” reported in the Indian GTA

24 This also appears to have largely informed Commerce’s decision to no longer rely on South Africa as the primary surrogate country. Consol. Court No. 13-00073 Page 23

import data compared to other economically comparable countries. Def’s Resp. at 26-27 (citation

omitted). That is inaccurate.25 The response leads to discussion of the second issue noted above.

As to that issue, Commerce states that it found nothing of record concerning the

domestic prices in the Philippines that could be compared to the Philippine GTA data. The only

substantive consideration of price in the IDM is the statement where Commerce found “that record

evidence does not support a finding that the average unit value from any of the other countries, when

compared with that of the Philippines, either is more specific to the input or demonstrates that the

value from the Philippines is aberrational.” IDM cmt. 7 at 12. This is a weak comparison for that

inference. As Commerce itself points out, there was only one other country for that comparison --

South Africa -- which had relatively minuscule imports of chlorine that Commerce itself rejected as

appropriate for surrogate valuation purposes in the Preliminary Results, and there is no indication

in the IDM of what South Africa’s average import unit value for chlorine is for the purpose of that

comparison. Commerce’s statement, in other words, exists in a vacuum as far as the reader is

concerned.

In any event, averages mask variance, and Kangtai pointed out that the average unit

value of chlorine imports into the Philippines for the review at bar ranged from 3.4 Philippine pesos

(“PhP”) to 134 PhP -- an extraordinarily wide range for a purported chemical commodity. The court

agrees that the IDM’s reasoning for finding the average unit value of the Philippines GTA data

25 The IDM itself explains that the decision “was partly based on the wide range of import volumes reported in the Indian GTA data as compared to other potential surrogate countries, and partly attributed to the various means and costs associated with transporting chlorine over long distances”, IDM cmt. 7 at 13 -- but, as above mentioned, if in fact the amount imported into the Philippines during the POR at bar can be concluded representative of a price that a producer would pay for the inputs, then that fact would moot those special cost concerns associated with those inputs. Consol. Court No. 13-00073 Page 24

reliable as a surrogate for the chlorine input is undercut by a lack of consideration of the apparent

extraordinarily wide range of Philippine import values in light of the fact that Commerce used such

a variation to explain why, in part, it was opting for Indian domestic data in the prior review,

therefore requiring remand. Cf. Trust Chem, supra, 35 CIT at ___, 791 F. Supp. 2d at 1264-65

(aberrancy is demonstrated through juxtaposition, i.e., “relative”, of data) (italics in original).

Commerce’s ultimate conclusion, essentially, is that there were in fact sufficient

commercial quantities of chlorine and hydrogen gas imported into the Philippines that enabled it to

adhere to its “preference” for valuing from a single surrogate country “where possible.” All things

are “possible,” but that does not make their realization reasonable. Commerce argues its preference

does not “require [it] to use domestic price in all circumstances,” Def’s Resp. at 23, quoting Rhodia,

supra, 25 CIT at 1287, 185 F. Supp. 2d at 1352 (this court’s bracketing; italics omitted), and that

Clearon and Kangtai did not urge Commerce to use domestic data instead of import data from

countries on the economically comparable list, but rather urged Commerce to use domestic data from

a country no longer on the Surrogate Country List, rather than import prices from the primary

surrogate country. That in no way imparts anything of relevance to the reader concerning the quality

of the Indian domestic data versus the quality of the GTA import data of record, and Commerce’s

only apparent redoubt, once again, is that Clearon’s and Kangtai’s argument is “contrary” to its

regulatory preference. Id. at 23 and 49, referencing Kangtai’s Br. at 12-13; and cf. Clearon’s Br. at

13 (domestic prices are preferred “all else being equal”).

Commerce is required to use the best information available in choosing surrogate

values. E.g., Blue Field, supra, 37 CIT at ___, 949 F. Supp. 2d at 1317, 1326, citing 19 U.S.C. Consol. Court No. 13-00073 Page 25

§1677b(c)(1). The relevant regulation, 19 C.F.R. §351.408(c)(2), expresses leeway in providing that

Commerce “normally will value all factors in a single surrogate country” (italics added).

Commerce’s response contradicts its own position in the Preliminary Results, where it relied on data

from India -- a country not on the Surrogate Country List -- as its “second” surrogate country to value

chlorine. Prelim. SV Memo at 4. For that reason, Commerce’s stance in the Final Results -- that

not only is reliance upon Indian data improper but even reference thereto (let alone comparison

therewith) improper, even for the purpose of enlightening as to what is the best surrogate value for

a specific individual FOP (i.e., because “India is not on the Surrogate Country List”) -- rings hollow,

even if it is the case that “Preliminary Results are just that -- preliminary”. Changshan Peer, supra.

Clearon and Kangtai maintain it is unreasonable for Commerce to consider import

data for hydrogen gas and chlorine as “reasonable” surrogate values in part due to the “very small”

quantities represented by the import data. Kangtai in particular argues that the import quantities

represented by the Philippine GTA import data were small as compared to MVC’s production, the

only apparent domestic producer of chlorine in the Philippines. Kangtai’s Br. at 20. For this

comparison, Kangtai relies on the fact that MVC reported that the company produced 5,000 MT of

chlorine in 2010, representing approximately 60% of the purported 8,000 MT chlorine market in the

Philippines. Kangtai’s Admin. Rebuttal Br. at 19.

Commerce, however, here points out that the stockholder’s meeting minutes from

MVC’s financial statement indicate that import prices for chlorine are “competitive” with prices for

domestically produced chlorine, albeit at the time of the international financial crisis,26 to wit:

26 Kangtai also interprets a statement from MVC’s financial statements as indicating that it (continued...) Consol. Court No. 13-00073 Page 26

MVC remains to be a regular supplier of Manila Water and Maynilad, which are the main market[s] for chlorine in the Philippines. However, both companies had signified their preference to use imported material for economic reasons. Apparently, they are able to procure the imported material at prices lower than what MVC can offer.

See Clearon’s Final SV Submission, PDoc 128 (Sep. 5, 2012), Ex. 15 at Annex C at 3 of 9. In other

words, Commerce argues, the cost of importing chlorine at the time was not so high as to not allow

importers of chlorine to compete with domestic production; therefore, Commerce argues, Kangtai’s

argument that the price of chlorine reflected in the Philippine GTA import data are significantly

higher than domestic prices because of transportation concerns is unsupported by the record.

The IDM did not, however, advance the foregoing heresay as a reasoned part of the

analysis. If such statements were to be used, then in any event they need to be considered in the

context of any record evidence addressing and MVC management’s explanation of the state of the

Philippines economy as a whole, including the impact of the international financial crisis and Thai

plant disruptions, whether there were other producers of chlorine in the Philippines or whether MVC

was able to command monopolistic pricing, whether MVC is an efficient producer of chlorine, et

cetera, because the meaning of “economic reasons” and “imported material” are not immediately

apparent, nor is it entirely clear whether Manila Water and Maynilad’s “imported material”

source(s), assuming for the sake of argument the truth of MVC management’s statement, is or are

from a market economy source or privately transacted.

26 (...continued) does not export chlorine because the export of chlorine is cost prohibitive. See Kangtai’s Br. at 19. Commerce points out that the precise statement therefrom is that MVC “is not engaged in export sales,” but MVC does not indicate that this is due to the expense of exporting chlorine. See Jiheng Resubmission of SV for FOP, PDoc 118 (Sep. 5, 2012), at Att. 1, MVC SEC Form 20-IS at 14. The record thus does not support the extent of Kangtai’s construal. Consol. Court No. 13-00073 Page 27

All in all, the issue of selecting the best surrogate value for chlorine requires remand

and reconsideration, as the IDM’s rationale does not reflect a full consideration of the parties’

arguments, it instead reflects inconsistent logic as compared with Commerce’s treatment of the

chlorine surrogate value in the Preliminary Results and prior reviews, and it does not approximate

a surrogate country with comparable production or Kangtai’s actual production experience. Cf. note

21, supra. Whether other data of record might be more suitable for that purpose, no opinion is here

offered.27

D. Surrogate Valuation of Urea

Clearon argues the value selected for the urea FOP for the Final Results, based on

GTA import data under the relevant Harmonized Tariff Schedule (“HTS”) heading, also “violated

the well established preference for published domestic price data”, all other things being equal,28 and

that publicly available, POR-contemporaneous, and reliable domestic prices from the Philippines’

Bureau of Agricultural Statistics were available on the record. Clearon’s Br. at 11-16; Clearon’s

Reply at 2-6.

27 The court notes in passing that Kangtai also claims that the GTA import price for chlorine makes no commercial sense because it is valued at approximately six times the value of the sodium salt, the input consumed to produce chlorine. See Kangtai’s Br. at 25-27. The court finds little merit in the argument, but notes that Commerce’s apparent nominalization of all off-shoots of production as “by-products” (or so it would seem) is unnecessarily obfuscating. See infra. Commerce succinctly stated that the value of a “by-product is the value it can obtain in the market”, IDM cmt. 7 at 14, which is true, but it is unclear whether the price therefor is skewed by the above concerns. At any rate, insofar as Kangtai’s arguments are concerned, the surrogate value for chlorine includes values of the additional inputs and processing costs and is not analogous to the products in Paslode Division or Wood Flooring from PRC, and Kangtai cites to no evidence of record showing that chlorine is not a value-added product as a result of the processes that produce it. 28 See, e.g., Ferrovanadium and Nitrided Vanadium from the Russian Federation, 62 Fed. Reg. 65656, 65661 (Dec. 15, 1997) (final rev. results) (Commerce has “articulated a preference for a surrogate country’s domestic prices over import values”) (italics added). Consol. Court No. 13-00073 Page 28

Commerce stated in the Final Results that it was unsure what the BAS data actually

represent. Citing to a submission to the record from Arch, the Final Results state “there is record

evidence that urea is not produced in the Philippines.” IDM cmt. 5 at 10 (citation omitted).

Commerce therefore opted for the Philippines’ GTA data as the best available data. However, since

the record shows that urea is in fact produced domestically in the Philippines, as the defendant

concedes,29 and since this contrasts with the apparent reason Commerce gave for selecting the

Philippines’ GTA data, the selection of the surrogate value for this FOP must be remanded for

reconsideration notwithstanding the defendant’s characterization that “the underlying concern of the

data is that the price of urea in domestic Philippine market is not really representative of the

domestic price of urea because domestic production had dropped significantly over time and imports

comprised 92 percent of Philippine demand.” Def’s Resp. at 49. See Motor Vehicle Mfrs. Ass’n v.

State Farm, 463 U.S. 29, 50 (1983) (“an agency’s action must be upheld, if at all, on the basis

articulated by the agency itself”).

E. Surrogate Valuation of Sodium Hydroxide

For the Final Results, Commerce selected Philippine import data collected by the

GTA for the HTS number for sodium hydroxide to value “sodium hydroxide.”30 Final SV Memo

at 6. Kangtai argues that the record reflects that it consumed sodium hydroxide at a concentration

of 32 percent, which is lower than the 50 percent concentration produced commercially and reflected

29 See Def’s Br. at 49 (“the evidence cited shows there is production [of urea] in the Philippines”). 30 The parties sometimes refer to sodium hydroxide (NaOH) as lye or caustic soda. The court notes that the common names would also cover potassium hydroxide (KOH). Consol. Court No. 13-00073 Page 29

in the GTA Harmonized Tariff Schedule (“HTS”) data and thus, Commerce should have made a

downward adjustment to the surrogate value for sodium hydroxide in accordance with Synthetic

Indigo.31 See Kangtai’s Br. at 27-31.

Commerce declined, explaining that the record contains information that sodium

hydroxide is commercially traded at different levels of concentration, not just 50 percent as asserted

by Kangtai, and there is no information on the record regarding the concentration level reflected in

the Philippine GTA import data for sodium hydroxide. IDM cmt. 18 at 26-27. Clearon adds that

the evidence submitted by it established that there was no correlation between prices for 100%

sodium hydroxide flakes (HTS 2815.11) and liquid sodium hydroxide (HTS 2815.12). Clearon’s

Resp. at 37 (citation omitted).

Kangtai submitted Philippine GTA import data before the Preliminary Results and

there was apparently only one source of Philippine surrogate value data on the record for sodium

hydroxide. See Kangtai’s Prelim. SV Submission at Ex. SV-13. In the Final Results, Commerce

selected contemporaneous Philippines’ GTA import data under the HTS number for sodium

hydroxide to value sodium hydroxide. Final SV Memo at 6.

Kangtai argues that it presented in full that as a general matter of international

commerce sodium hydroxide is normally traded at a 50% concentration, and that, based on the

typical concentration level for sodium hydroxide, “it is unreasonable to speculate that imports are

made at concentrations other than the standard commercial concentration.” Kangtai’s Br. at 29.

31 See Synthetic Indigo from the PRC, 68 Fed. Reg. 53711 (Sep. 12, 2003) (final rev. results) (“Synthetic Indigo”) and accompanying I&D Memo at cmt. 5, referencing Saccharin from the PRC, 68 Fed. Reg. 27530 (May 20, 2003) (final LTFV determination) (“Saccharin”) and accompanying I&D Memo at 2. Consol. Court No. 13-00073 Page 30

Based on this assumption, Kangtai argues that the surrogate value for sodium hydroxide should be

adjusted downward from 50 percent concentration to 32 percent concentration to account for

Kangtai’s consumption of sodium hydroxide at the 32 percent concentration. Id. at 29-30.

Commerce responds that although the record indeed contains evidence that sodium

hydroxide is sold at a 32 percent concentration level,32 Kangtai cites to no evidence that the

Philippines’ GTA import data in fact reflect prices for sodium hydroxide at only the 50 percent

concentration level. Kangtai’s claim that Commerce unreasonably speculated that imports are made

at concentrations other than the standard commercial concentration is undermined by Kangtai’s

concession that it “purchases and consumes [sodium hydroxide] at a lower 32% concentration.”

Kangtai’s Br. at 27-28. More importantly, regardless of whether 50 percent concentration is the

typical concentration level for commercially-traded sodium hydroxide, Kangtai presented no

evidence demonstrating that the Philippines’ GTA import data actually reflect prices for sodium

hydroxide at 50 percent.33 See IDM cmt. 18 at 26-27. Absent any positive evidence on the record

establishing the concentration level of the Philippines’ GTA import data, any adjustment made by

Commerce to the surrogate value would have been arbitrary. And regarding Synthetic Indigo,

Commerce contends Kangtai’s argument ignores the fact that therein Commerce had been able to

determine that the surrogate value data source in question (the “Monthly Statistics of the Foreign

32 See Kangtai’s Section D Questionnaire Resp., Nov. 30, 2011, at Ex. D-2. 33 Commerce further argues that in a marketplace where sodium hydroxide is sold at various concentration levels, Kangtai is requesting a downward adjustment from 50 percent to 32 percent when the record simply does not establish that the starting point of the downward adjustment is 50 percent, and the public materials that Kangtai placed on the record do not suggest that the Philippines’ GTA import data actually reflect prices for sodium hydroxide at 50 percent. Def’s Resp. at 35, referencing Kangtai’s Br. at 27-28. Consol. Court No. 13-00073 Page 31

Trade of India”) represented prices for chemicals at commercially traded concentration levels,

whereas there is no such information regarding the Philippines’ GTA import data on this record.

Kangtai does not persuade that Commerce’s surrogate valuation of sodium hydroxide

was unsupported by substantial evidence or not in accordance with law, although judgment thereon

will need to abide reconsideration of the selection of primary surrogate country, which may

necessarily result in altering the surrogate valuation of this chemical.

F. Surrogate Valuation of Electricity

For the Final Determination’s surrogate valuation of electricity in the Philippines,

Commerce analyzed Camarines Sur rate data and Manila Electric Company (“Meralco”) rate data

pursuant to the usual factors of public availability, broad market average, product specificity,

contemporaneity, and freedom from taxes and duties.34 Commerce selected the Camarines Sur

electricity rate data as the surrogate value for the electricity FOP because the “electricity rate matches

the factor rate in kilowatt hours for industrial users, is publicly available from the primary surrogate

country, represents electricity rates from two cities in the Philippines, does not appear to include

taxes or duties, and does not suffer from the unknown variability factors of the MERALCO rate”.

IDM cmt. 10 at 18-19.

Kangtai and Arch argue that the Meralco data set is better than the Camarines Sur

data. Kangtai’s Br. at 40-42; Arch’s Motion for Judgment on the Agency Record, ECF No. 27 (Aug.

15, 2013) (“Arch Br”) at 15-21. Commerce responds that all the plaintiffs are asking this court to do

34 See, e.g., See Certain Polyester Staple Fiber From the PRC, 75 Fed. Reg. 1336 (Jan. 11. 2010) (final rev. results), and accompanying I&D Memo at cmt. 1(“Certain Polyester Staple Fiber”). Consol. Court No. 13-00073 Page 32

is re-weigh evidence, while referring to non-record material and administrative determinations that

do not support their arguments. Def’s Resp. at 36-46.

1. Relevant Facts

There are only two Philippine surrogate values on the record. One is the Camarines

Sur electricity rates in kilowatts hours from the “2009 Doing Business in Camarines a Sur” report

submitted by Kangtai. Kangtai’s Prelim. SV Submission at Exs. SV-15 & SV-16b. The Camarines

Sur data lists industrial electricity rates (with demand) for two cities in the Philippines, Naga and

Iriga. Id. at SV-16b.

The other Philippine electricity rate data, the Meralco data, were submitted by Jiheng.

Jiheng’s Prelim. SV Submission at Tab 5. The Meralco data consist of a single page chart for the

month of December 2010. Id. In the left hand column of the chart there are eight different main

categories of users each with several subparts, 39 in all. Id. The remaining 21 columns list various

charges, adjustments, discounts, and subsidies some in kilowatt hours others in kilowatts. Id. The

notes to the chart indicate that certain charges vary on a monthly basis. Id. Commerce contends that

neither Arch nor Kangtai provided any additional argument, explanation, or evidence as to within

which of the 39 categories of users their production would fall into or how to use 21 columns of

charges, adjustments, discounts, and subsidies to derive an industrial kilowatt hour electricity rate.

Def’s Resp. at 42-43.

To evaluate the quality of the two data sets to value electricity, Commerce applied

its standard surrogate value analysis, which is its five-factor test of public availability, product Consol. Court No. 13-00073 Page 33

specificity, whether they represented a broad market average, the contemporaneity of the data, and

whether the data were free of taxes and duties. See, e.g., Certain Polyester Staple Fiber at cmt. 1.

For the Final Results, Commerce found that both data sets were publicly available,

IDM cmt. 10 at 18-19, but that the Camarines Sur data were more specific to the kilowatt hour factor

than the Meralco data. Id. at 18-19. Commerce explained that the Meralco data contained some data

in kilowatts, not kilowatt hours, and that the record lacked sufficient information to make the

conversion. Id. at 18. Commerce determined that Arch’s suggested conversion methodology made

assumptions that were not supported by the record. Id. Commerce also determined that the Meralco

data indicated that several of the 21 different components of the electricity charge were variable on

a monthly basis, and there was only one month of Meralco data on the record. Id. at 18-19.

Commerce found that both data sets represented a broad market average but that

neither data set represented the entire Philippine market. Id. at 19. It noted that, although Arch

argued that the Meralco data represented a broader section of the Philippine market, the Camarines

Sur data represented industrial rates for two cities in the Philippines. Id. Commerce found that the

2009 Camarines Sur industrial electricity rates were sufficiently contemporaneous to the 2010-2011

period of review because utility rates apply forward and there was no record evidence that they had

changed since 2009. Id. at 19. Finally, Commerce found that the Meralco data specifically excluded

taxes and duties, and that there was no evidence that the Camarines Sur data included taxes and

duties. Id. After weighing all of these factors, Commerce determined that the Camarines Sur data

were the best available information on the record with which to value the electricity factor for both

Kangtai and Arch. Id. Consol. Court No. 13-00073 Page 34

As an initial matter, the court notes that no party challenges the public availability of

the Carmines Sur and the Meralco data sets. Kangtai and Arch express three shared contentions with

regards to flaws they perceive in the Camarines Sur data and Commerce’s analysis thereof. First,

Arch and Kangtai argue the Camarines Sur data is not broadly based, a fact Kangtai claims the

defendant concedes.35 Second, both parties claim that the Camarines Sur data is not

contemporaneous. See Kangtai’s Reply at 19; see also Arch’s Br. at 19-20. Third, Kangtai claims

that the data is not more specific than the Meralco data and Arch adds that the variability of the

Camarines Sur was unknown and unknowable. See Kangtai’s Reply at 19; see also Arch’s Br. at

19. Arch also claims that the absence of any evidence of the Camarines Sur data being tax exclusive

does not mean that the data actually were as Commerce claims. Arch’s Br. at 18.

With respect to Arch’s and Kangtai’s challenges to Commerce’s selection of the

Camarines Sur data, Commerce contends the parties impermissibly refer to non-record information

and misunderstand Commerce’s administrative determinations, and that the court’s role is not to re-

weigh the evidence but to determine whether Commerce’s weighing of the evidence is supported by

substantial evidence and is otherwise in accordance with law. Def’s Resp. at 39, referencing

Metallverken Nederland B.V. v. United States, 13 CIT 1013, 1017, 728 F. Supp. 730, 734 (1989).

2. Representative of a Broad Market

Commerce’s specific findings were that neither the Meralco data nor the Camarines

Sur covered the entire Philippine market and that the Camarines Sur data covered two cities. IDM

cmt. 10 at 19. Commerce made these findings by weighing the results of its analysis of the factors

35 See Kangtai’s Reply at 18, referencing Def’s Resp. at 44; see also Arch’s Br. at 18. Consol. Court No. 13-00073 Page 35

against each other. In the same paragraph, Commerce also discussed the specificity and the tax-and-

duties factors. Id. Kangtai avers that Meralco data is more representative of a broad market as it is

the largest electrical supplier in the Philippines and is one that covers all of the country’s “major

industrial zones”, while the Camarines Sur data covers only “two tiny cities in one non-industrial

province.” See Kangtai’s Reply Br. at 41-43. The choice Commerce made, it claims, does not take

into account a respondent’s production experience and is between “one source which is broadly

based and representative of the industrial experience in the Philippines and one source which is

neither.” Id. at 18. Arch echos Kangtai’s contentions stating that the record indicates that “the

Meralco rate applied to 60% of the industrial base of the country, including one of the facilities

making the comparable product”, and that “none of the surrogate product was manufactured within

the coverage of the Camarines Sur data, because the company had no facilities located in that

Province”. Arch’s Br. at 18.

Although the Meralco data may have broader market coverage than the Camarines

Sur data, when weighed with other factors, which Commerce concluded detract from the Meralco

data, Commerce concluded the Camarines Sur data have sufficiently broad coverage (two cities) to

be a reliable surrogate value of the Philippine market. Arch’s and Kangtai’s assertions that the

Meralco data cover a broader portion of the Philippines’ electricity market do not overcome or render

unreasonable that analysis, and the court cannot engage in re-weighing of this evidence of record.

3. Specificity

The production factor that Commerce valued was the kilowatt hours of electricity

used to make the subject merchandise (chlor-isos). Commerce observed that the Camarines Sur data Consol. Court No. 13-00073 Page 36

represent a single average industrial electricity rate in kilowatt hours. See Kangtai’s Prelim. SV

Submission at Ex. SV-15 & SV-16b. Also, that the Meralco data do not provide an industrial

electricity rate in kilowatt hours, but consist of a chart representing one month of 21 components of

an electricity rate, some of which are not in kilowatt hours, some of which are identified as varying

by month, with no explanation or evidence as to what should be included in an electricity rate and

no explanation of how to convert the components that are in kilowatts to kilowatt hours. See Jiheng’s

Prelim. SV Submission at Tab 5. Based on this, Commerce found that the record did not contain the

information to make the conversion. IDM cmt. 10 at 18.

Commerce’s decision that the Camarines Sur data were more specific to the

electricity factor, which was in kilowatt hours, therefore has support in the record. See id. at 18-19.

Arch and Kangtai argue that Commerce calculated kilowatt hour rates from one month of the

Meralco data in other cases and should have done so here. Arch’s Br. at 16-17; Kangtai’s Br. at 40.

However, the cases on which they rely point more towards the principle that the record facts in each

administrative review determine the analysis that Commerce will perform to determine the best

available information on the record.

The determinations in those cases provide no information other than that Commerce

used the Meralco data to value electricity. In the Steel Wire Hangers case,36 there is no discussion

of any alternative electricity rate data sources, nor is there any “best available information” analysis

using the five factors. The most that can be inferred from this determination is that there was only

36 Steel Wire Garment Hangers from the PRC, accompanying Steel Wire Garment Hangers From the PRC, 77 Fed. Reg. 66952 (Nov. 8, 2012) (prelim. rev. results) and accompanying I&D Memo, unchanged in the final results, 78 Fed. Reg. 25946 (May 16, 2013) (final rev. results) (“Steel Wire Hangers”). Consol. Court No. 13-00073 Page 37

one electricity data source on the record and that Commerce used the Meralco data as the “best

available information.” Arch’s reliance on the Hardwood Plywood decision is similarly unavailing.

See Arch’s Br. at 17 (citing Hardwood and Decorative Plywood from the PRC, 78 Fed. Reg. 25946

(May 3, 2013) (final LTFV investigation) (“Hardwood Plywood”). There is no indication that

Commerce had any other alternatives from which to choose. See generally id.

Arch attempts to support its challenge here by referring to a document from the Steel

Wire Hangers record. Arch’s Br. at 17, n.2. As it is not in the record of this case, it will be

disregarded. See 28 U.S.C. §1516a(b)(2)(A). Even assuming it can be considered, it does not

establish that Commerce had any alternative electricity rates from which to choose. See Arch’s Br.

at 17, n.2. As a result, Arch’s reliance on the Steel Wire Hangers and Hardwood Plywood cases do

not demonstrate that Commerce’s determination here should be set aside.

Next, as explained above, there is no evidence on the record of this proceeding with

which to make the conversions from kilowatts to kilowatt hours and Arch and Kangtai have not cited

any. Furthermore, there is no information or argumentation on the record which indicates how the

21 components for the electricity rate should be combined to form an actual electricity rate. Finally,

there is no information on the record to indicate in which of the 13 industrial user categories out of

39 categories, Kangtai and Arch fall. Because the record contains an average industrial rate in

kilowatt hours from Camarines Sur there was no reason to go through the speculative process of

converting and constructing an average industrial electricity rate from the Meralco data, even if the

components of that rate in kilowatts represent “a minuscule part of the overall Meralco rate” as

Kangtai claims. Kangtai’s Reply at 19. Consol. Court No. 13-00073 Page 38

To defend the record deficiencies in the Meralco data, Arch contends that, because

the website for the Meralco data is on the record, Commerce should have gone to the website and

allayed any concerns concerning the Meralco data and its variability. Arch’s Br. at 19. Kangtai

similarly argues that the website for the Camarines Sur data is on the record and argues that the data

on the website do not support Commerce’s determination. Kangtai’s Br. at 42. That the website

addresses are on the record does not mean that all of the data on the websites are on the record. If

a party wants evidence from a website on the record of a Commerce proceeding, it must submit the

appropriate pages from the website; otherwise, the information is not on the record of the

proceeding. Both Kangtai and Arch had the opportunity to put whatever aspects of these websites

on the record they chose. Kangtai put on selected portions of the Camarines Sur data. Kangtai’s

Prelim. SV Submission, at Ex. SV-15, SV-16b. Jiheng placed on the record a single chart from the

Meralco site but without explanation. Jiheng’s Prelim. SV Submission at Tab 5. Arch and Kangtai

cannot now rely on data that they never placed on the record.

Finally, Arch and Kangtai speculate that the Camarines Sur data lack detail to

determine whether they suffer from an unknown variability. See Arch’s Br. at 19; see also Kangtai’s

Reply at 19. Kangtai claims that half of the rate from the Camarines Sur is based off “completely

unknown” variables from an “on demand” electricity based system.37 The record shows that the

Camarines Sur data are an average industrial electricity rate from two cities in the Philippines.

Kangtai’s Prelim. SV Submission, at Ex. SV-15, SV-16b. There is no record evidence that the

average rates suffer from unknown monthly variability. In contrast, even if there were data to convert

37 Kangtai’s Reply at 19 (referring to the Naga City industrial rate which it claims depends on the individual customers daily demand or maximum usage). Consol. Court No. 13-00073 Page 39

the Meralco rate components from kilowatts to kilowatt hours, the Meralco chart itself indicates that

some of the Meralco components are subject to monthly variation. Jiheng’s Prelim. SV Submission

at Tab 5. To construct an annual average rate from the one month of data would require Commerce

to assume numerous variables did not vary during the year, which would have been an unreasonable

assumption given that the chart itself identifies monthly variation.

Kangtai also argues that the Camarines Sur data are listed as industrial “with

demand” which it contests Commerce “is certainly aware” means that the Camarines Sur data are

also variable. Kangtai’s Br. at 42. However, neither Kangtai nor Arch made this argument to the

agency. As a result, they failed to exhaust their administrative remedies and accordingly the issue

was not discussed in Commerce’s Final Results. See, e.g., Shandong Huarong Machinery Co., Ltd.

v. United States, 30 CIT 1269, 1305, 435 F. Supp. 2d 1261, 1292 (2006). In any event, Kangtai’s

statement that Commerce “is certainly aware” is not record evidence and thus the argument is not

supported by record evidence. In short, the Camarines Sur data are more specific to the factor of

production being valued, electricity usage in kilowatt hours, and Arch’s and Kangtai’s argumentation

on the record does not persuade that Commerce’s specificity determination is unsupported by

substantial evidence or not in accordance with law.

4. Exclusion of Taxes And Duties

The record demonstrates that the Meralco data do not include taxes and duties while

there is no record evidence that the Camarines Sur data do include taxes and duties, as Arch points

out and as Commerce specifically found in its Final Results. See IDM cmt. 10 at 19; see also Arch’s

Reply at 18-19. Given this identified and acknowledged difference in the record evidence on the

exclusion of taxes and duties, based on a weighing of all of the factors which includes the serious Consol. Court No. 13-00073 Page 40

problems with the specificity of the Meralco data, as discussed above, Commerce preferred to use

the more specific Camarines Sur data over the less specific Meralco data. This is not an

unreasonable decision. In Commerce’s view, the difference in the record data on taxes and duties

is not enough to justify rejecting the Camarines Sur data.

5. Contemporaneity

In the Final Results, Commerce found that the Camarines Sur data were from 2009

based on the copyright date on the publication “Doing Business in Camarines a Sur”. IDM at 19.

In addition, Commerce found that utility rates generally, “represent a current rate as indicated by the

effective date for each of the rates provided.” Id. It found further that the Camarines Sur rate was

sufficiently close to the period of review of 2010-2011 that it was likely to still be in effect. Id.

Arch distinguishes this review from the prior review. Arch’s Br. at 19-20. It claims

that Commerce used a “non-sequitur” concerning the effective dates of electricity rates because, in

the prior review, the surrogate country was India and not the Philippines. Id. at 19. There is no non-

sequitur. That the surrogate country in the last review was India and in this review the surrogate

country is the Philippines is irrelevant to a finding that generally utility rates represent a current rate

by the effective date, the proposition for which the case was cited.

Arch and Kangtai also both attempt to distinguish this case from the prior review by

arguing that the fact that the Camarines Sur data have a copyright date of 2009 does not mean that

their electricity rates are from 2009, whereas the India electricity rates in the last review have a

specific effective date, a date type which the Camarines Sur data do not contain. Arch’s Br. at 20-

21; see also Kangtai’s Reply at 18. This is again a distinction without a difference. The “Doing

Business in Camarines a Sur” publication is published to attract business to Camarines a Sur. The Consol. Court No. 13-00073 Page 41

section of the publication from which the electricity rates are derived contains, in relevant part, the

following description, “[t]his section provides investors with a clear perspective of what to consider

like fees and licenses and what to expect such as attractive incentive packages available before taking

a business venture of a lifetime [in Camarines a Sur].” Kangtai’s Prelim. SV Submission at Ex.

SV-16b. Based on the express purpose of the “Doing Business in Camarines a Sur” publication, it

is illogical to assume, as Arch does, that the 2009 publication would not contain electricity rates

effective in 2009. As a result, Arch and Kangtai fail to distinguish this case from the decisions in

the prior review.

Finally, once again, the discussion of the contemporaneity of the two sets of data in

Commerce’s Final Results is in the context of weighing the various factors against one another. IDM

at 19. Even assuming that the Camarines Sur data were 2009 data and not effective during the

2010-2011 period of review, this factor would not be enough to reject using the Camarines Sur data.

Commerce regularly indexes data to make it contemporaneous with the period of review.

In short, based on Commerce’s analysis and weight of the data factors, Commerce’s

selection of the Camarines Sur data was reasonable.

G. Accounting for Labor in the SG&A Financial Ratio

In the Final Results, Commerce relied on Philippine labor statistics as reported in the

International Labour Organization’s statistics for category 6A. As alluded in the prior opinion,

Commerce now employs a rebuttable presumption that this category 6A “better accounts for all

direct and indirect labor costs.” Antidumping Methodologies in Proceedings Involving Non-Market

Economies: Valuing the Factor of Production: Labor, 76 Fed. Reg. 36092, 36093 (June 21, 2011) Consol. Court No. 13-00073 Page 42

(“Labor Methodology”) (italics added).38 Because ILO Chapter 6A data are intended to be all-

inclusive of labor costs, Commerce further explained in Labor Methodology, that

[i]f there is evidence submitted on the record by interested parties demonstrating that the NME respondent’s cost of labor is overstated, the Department will make the appropriate adjustments to the surrogate financial statements subject to the available information on the record. Specifically, when the surrogate financial statements include disaggregated overhead and selling, general and administrative expense items that are already included in the ILO’s definition of Chapter 6A data, the Department will remove these identifiable costs items.

Id.

For the Final Results, Commerce relied upon the 2010 financial statement for

Mabuhay Vinyl Corporation (“Financial Statement”) for surrogate selling, general and administrative

(“SG&A”) expenses in the calculation of a surrogate SG&A financial ratio for the respondents. The

prior opinion remanded the issue of whether employee retirement and other benefits had been double

counted in that calculation, due to an indication in the Financial Statement of certain retirement and

employee benefits included among the SG&A portion of that statement.

On remand, Commerce concluded that since these itemized amounts relating to labor

had been included in the Financial Statement as operating costs rather than cost of sale, they were

38 Labor Methodology was the consequence of the Federal Circuit’s invalidation of Commerce’s prior regression-based analysis as provided in 19 C.F.R. §351.408(c)(3). See Dorbest Ltd. v. United States, 604 F.3d 1363, 1372 (Fed. Cir. 2010). As explained in Labor Methodology, Commerce first resorted to reliance upon ILO Chapter 5B labor cost data, but because those data only cover direct labor compensation and bonuses, Commerce became concerned that such data were underinclusive. Thus, going forward, Commerce announced in Labor Methodology that it would rely on ILO Chapter 6A instead, and whereas in the past Commerce distinguished between direct labor cost and indirect labor cost that was accounted either as a part of the surrogate value for factory overhead or as part of labor, in accordance with Labor Methodology it now appears the cost of “labor” as a whole is to be calculated simply by multiplying the labor hour input by the relevant ILO-based unit labor cost figure. Consol. Court No. 13-00073 Page 43

properly included in its surrogate SG&A calculation. Kangtai had argued that the Philippines ILO

data of record for labor costs include those for all paid employees, including managers, executives

and supervisors, but Commerce responded that it had only relied on the “Industrial/commercial

survey”, which made no such mention. RR at 39-40. Restating Arch’s argument, Commerce

interpreted it as claiming that the Financial Statement had “incorrectly accounted for” or

“misallocated” production-labor employee benefits in the administrative labor accounts,39 but

Commerce found that record evidence does not support such a finding. Id. at 40-41. The Financial

Statement indicated that employee and retirement benefits were provided to all “regular employees”

and Commerce stated that its conclusion on what “regular employee” means is based on its

understanding of the Philippines’ generally accepted accounting principles. Quoting Clearon,

Commerce stated that “there is no basis to assume that the ILO labor cost data would include

employee and retirement benefits associated with direct production workers, but that the same

employee benefits would be reported as operating expenses rather than costs of sales.”40

Actually, there is. Although the Philippine generally accepted accounting principle

upon which Commerce claims to have relied has not been made a part of the record (at least insofar

39 Arch had argued that, “[i]n its filing with the Philippine SEC authorities, MVC stated ‘The company has a registered, non-contributory retirement plan. All regular employees are covered from the President down to the rank and file.’” As quoted in RR at 41, referencing Arch’s Draft Remand Cmts. (Arch’s italics). 40 RR at 41-42, quoting Letter from Petitioner, “Remand of the 2010-2011 Administrative Review of the Antidumping Duty Order on Chlo-Isos from the PRC: Comments Regarding New Data Placed on the Record” (August 20, 2014). Consol. Court No. 13-00073 Page 44

as the court can discern from the papers submitted here41), the Financial Statement’s independent

auditors’ opinion letter of record asserts that the Financial Statement is in accordance with Philippine

Financial Reporting Standards. And insofar as the court could discern, by its own examination of

Philippine and international generally accepted accounting standards in existence in 2010-11, the

“proper” accounting treatment of employee retirement and other benefits, as those relate to

accounting for cost of goods sold, is not as clear-cut as Commerce assumes, except to the extent that

they must be accounted for and reported.

Be that as it may, Kangtai appears correct in arguing that Commerce’s assumption,

as to how labor is distinguished among the operating and period costs of the Financial Statement,

is all beside the point. Commerce faults Arch for “claiming that the ILO data includes SG&A type

labor for which an adjustment to the financial data is necessary” and that “like Kangtai’s argument,

[Arch] has also failed to demonstrate that such labor is included in the ILO data”, but Commerce’s

own Labor Methodology policy presumptively and apparently includes all costs relating to labor via

category 6A data. That policy pronouncement specifically stated that any labor item identified

among the SG&A (or “period”) items must be excluded from the surrogate SG&A ratio in order to

avoid overstatement, and it is not reasonably disputed that the employee and retirement benefits

described in the Financial Statement are a type of labor item that the Labor Methodology policy was

meant to address. If the determination in this matter implies that Commerce has discovered a

41 Therefore, Commerce’s assumption or projection of what is the “proper” way to account for these employee retirement and other benefits has no support in the record, unless that be by way of official notice. Consol. Court No. 13-00073 Page 45

problem with its policy, then it should address that by way of further notice and comment, rather than

attempting the type of tortured analysis in which it has engaged in here.

As Arch argues, the record shows that MVC has treated its employee and retirement

expenses as a coherent whole, as indicated in its Notes to the Financial Statement. Were Philippines’

Accounting Standards (“PAS”) a part of the record, in particular PAS 19, they would likely inform

that the Financial Statement has been prepared in accordance therewith and does not, as intimated

by Commerce, involve a “misallocation” of labor amounts. Note 19 of the Financial Statement

provides detail, including the exact calculation of year 2010’s retirement benefit costs. Note 17 also

itemizes “employee benefits,” although it does not provide as much information on these expenses

as for the retirement benefits. Note 24 breaks out the amounts of compensation of “key management

personnel” into “short-term employee benefits” and “retirement benefits”, which establishes that

these benefits, too, apply beyond administrative expenses. Thus, the Financial Statement has

itemized costs that would be included in the Chapter 6A labor rates, that are individually itemized,

and are not included in the labor costs in the financial statements. This appears to be precisely the

situation Commerce contemplated when stating that it would make adjustments in order to avoid

overstating labor costs, and the court must therefore conclude that Commerce has apparently failed

to interpret the record correctly, thereby inadvertently violating its Labor Methodology policy

without adequate justification.

On remand, in order to address the foregoing, Commerce should either remove the

labor items identified among MVC’s SG&A expenses or explain why adhering to its Labor

Methodology policy is inappropriate in this instance. Consol. Court No. 13-00073 Page 46

H. By-Product Offset Methodology

If a by-product resulting from production of subject merchandise has commercial

value, then the costs associated with its production must be allocated from the costs associated with

production of the subject merchandise. To date, apparently, Commerce has recognized by-products’

commercial value either by sales thereof or by reintroduction into the production of the subject or

non-subject merchandise.42 Arch and Kangtai challenge Commerce’s explanation of its by-product

offset determination.

1. Further Background on the Final Results

Commerce accepts that ammonia gas and sulfuric acid are the relevant by-products

cast out of the respondents’ production of subject merchandise at a certain “split off” point from the

production of chlor-isos. Kangtai avers that these by-products are then further processed into

ammonium sulfate, first into liquid form, and then into a powder, for sale primarily as fertilizer.

See Kangtai’s Resp. to Court’s Letter of May 8, 2015, ECF No. 98 (May 22, 2015).

In the original investigation, Commerce found that its “downstream by-products

practice” for determining by-product credits did not apply to the process of subject merchandise

42 See, e.g., DuPont Teijin Films China Ltd. v. United States, Consol. Court No. 13-00229, ECF No. 86 (Jan. 9, 2015) (“DuPont Teijin Films Redetermination”) at 5-6; Frontseating Service Valves From the PRC, 78 Fed. Reg. 35245 (June 12, 2013) (final 2010-2011 rev. results), and accompanying I&D Memo at cmt. 10; Frontseating Service Valves From the PRC, 76 Fed. Reg. 70706 (Nov. 15, 2011) (final 2008-2010 rev. results), and accompanying I&D Memo at cmt. 18; Certain Frozen Warmwater Shrimp From the Socialist Republic of Vietnam, 74 Fed. Reg. 47191 (Sep. 15, 2009) (inter alia, final rev. results), and accompanying I&D Memo at cmt. 7.A (“because by reintroducing the by-product into production, the material costs of the subject merchandise are directly reduced”). Complexities can arise depending upon whether the by-product is re-introduced into production of the subject merchandise or non-subject merchandise, and whether the by-product requires inventorying (different PORs) or is reintroduced immediately into a continuous production process. See, e.g., DuPont Teijin Films Redetermination at 5-8. Consol. Court No. 13-00073 Page 47

production. See, e.g., Arch’s Br. at 25. Therefore, in previous reviews of the Order as well as for

the Preliminary Results, Commerce had calculated the volume of the ammonia gas and sulfuric acid

by-products based upon the amounts of those products that are chemically (i.e., formulaically)

required to produce the amounts of ammonium sulfate reported by Arch and Kangtai as having been

actually produced (as opposed to the amount actually sold). Commerce would then calculate the

total of each respondent’s by-product offset based upon selected surrogate values for the ammonia

gas and sulfuric acid from among those the parties had submitted.43 See RR at 28.

After the Preliminary Results, in their administrative case brief the petitioners argued

that if Commerce continues to rely upon the Philippine import data, it should consider the fact that

the surrogate values for ammonia gas and sulfuric acid of record exceed that of ammonium sulfate,

leading to a “counterintuitive conclusion that respondents are combining two high-value by-products

. . . in order to produce a significantly lower value by-product in ammonium sulfate”, which is not

realistic. See RR at 47, quoting Clearon’s Admin. Case Br., PDoc 255 (Dec. 3, 2012) at 38.

Restating this argument, for the Final Results Commerce changed its methodology for determining

the by-product offsets, characterizing the change as necessary “to conform to the Department’s

recent practice,” and because the new methodology is both “more reasonable” than that employed

in the Preliminary Results and

is consistent with the information the Department requests in our questionnaire, which asks respondents: “[i]f the byproduct for which you are claiming an offset is

43 Neither respondent actually sold ammonia gas or sulfuric acid, and thus the calculations were based upon hypothetical “sales” derived from actual sales of ammonium sulfate. See Jiheng Resp. to Questionnaire Section D, PDoc 49 (Nov. 28, 2011), at D-32-33; Kangtai’s Resp. to Questionnaire Section D , PDoc 51 (Nov. 28, 2011) at 17. Consol. Court No. 13-00073 Page 48

a downstream by-product, in addition to responding to the items above,[44 ] please also: (i) Provide the per-unit usage rate of each input used to produce the downstream by-product.”

Consistent with this practice, the Department first starts with the value of the downstream product actually sold by the respondents, ammonium sulfate, produced during the POR. From this amount, the Department would normally deduct the costs associated with converting the by-products into the downstream product, such as labor and electricity. Since this information is not on the record of this review, the Department is not able to deduct such costs for these final results. In the future, the Department will require such information in order to grant this offset. But in this instance, we are using the full value of the ammonium sulfate as the by-product offset. We calculated this amount by multiplying the quantities of ammonium sulfate produced and sold by respondents during the POR by the surrogate value (Philippine GTA data) for ammonium sulfate.

IDM cmt. 14 at 23-24 (footnotes omitted; italics added).

The respondents challenged this change here, arguing, inter alia, that they had been

prejudiced by lack of notice concerning the new methodology and that they had relied upon the old

methodology in their pricing of subject merchandise. Agreeing with the parties that it had not

explained its new by-product methodology,45 Commerce voluntarily requested remand in order to

44 Earlier in precedence was this questionnaire request, which has been omitted from Commerce’s discussion in the IDM: “Please note: By-product/co-product offsets are only granted for merchandise that is either sold or reintroduced into production during the POR, up to the amount of that by-product/co-product actually produced during the POR. If you are claiming a by-product or co-product offset in your FOP database, please report each by-product or co-product in a separate field. Further, in your narrative response please: i. Provide a description of the by-product/co-product; ii. Provide an explanation why you have defined the products as by-products or co-products, as applicable; . . .” Questionnaire, Section D (Oct. 6, 2011) at D-9. 45 To the court, Clearon had argued that the value of the downstream by-product (ammonium sulfate) overstates the value of the input by-products (ammonia gas and sulfuric acid) and that Commerce should find that the respondents had “withheld” relevant information and that the matter should at least be remanded for its collection. See Clearon’s Br. at 21-25. Clearon had also argued for application of “adverse inferences” pursuant to 19 U.S.C. §1677e(b), which Commerce declined to apply. In addition to arguing that the change in methodology was unlawful, Kangtai and Arch had (continued...) Consol. Court No. 13-00073 Page 49

consider those arguments, provide explanation, and collect additional relevant information if

necessary. See Opinion at 18, referencing Def’s Resp. at 54.

2. Results of Remand

The Remand explains that in order to approximate a “market value” for the

ammonium gas and sulfuric acid by-products, RR at 29, Commerce is now determining a “net

realizable value”46 for the ammonium gas and sulfuric acid by-products. Commerce states that this

is “consistent with its practice” of “start[ing] with the value of a downstream product, in this case

ammonium sulfate that was actually sold by the respondents and produced during the POR”, which

leads to “an offset equal to the amount of value a company actually receives, less any processing

costs, and not a hypothetical value that is unrelated to a company’s financial books and records”47.

Commerce determines these values by deducting the costs of further processing the ammonium gas

and sulfuric acid by-products (e.g., labor and electricity) that are incurred from the split-off point in

the production of subject merchandise when transforming those by-products into the downstream

product ammonium sulfate. Commerce reasoned that

45 (...continued) argued and that the value of the ammonia gas and sulfuric acid by-products should be individually and directly determined by reference to surrogate values therefor, as Commerce had done in prior reviews. On remand, however, Commerce pursued applying its new methodology by issuing a questionnaire to Arch and Kangtai, obtaining data from the parties, and incorporating the data in its recalculation of the offset based on the new methodology. See RR at 30. 46 Id. at 29 (italics added), quoting Citric Acid and Certain Citrate Salts From the PRC, 79 Fed. Reg. 101 (Jan. 2, 2014) (final 2011-2012 rev. results) (“ Citric Acid and Certain Citrate Salts”), and accompanying I&D Memo at 12. 47 Id. at 28-29 (italics added), referencing Magnesium Metal from the Russian Federation, 73 Fed. Reg. 52642 (Sep. 10, 2008) (final rev. results), and accompanying I&D Memo at cmt 1.B & 1.C. Consol. Court No. 13-00073 Page 50

[t]he value of ammonium sulfate reflects the actual economic value of the byproducts generated through the respondents’ cyanuric acid production process and is accordingly an appropriate source to value the byproducts that are combined to produce ammonium sulfate. Thus [Commerce’s] methodology reflects the actual value that the company receives for the byproducts [that] are contained in the downstream product which Kangtai and [Arch] actually sell.

RR at 47. Commerce again noted that although it did not elaborate on its change in methodology

in the Final Results, the “policy is evident from our boilerplate questionnaire, used in the underlying

review, which asks parties to report the FOPs required to process the by-product into saleable

downstream product.”48 Commerce also noted again that this methodological “change” is in order

to comport with recent “agency-wide” policy and avoids overstating the value of the by-product

offsets. Id. at 28. As applied in the instant review, Commerce found that it “did not have the FOPs

to deduct, so we used the full value of the ammonium sulfate as the full value of the two by-products

combined as the by-product offset.” Id. at 28.

3. Arguments

Supporting Commerce’s determination, Clearon claims that the by-product offset

determinations for each respondent were appropriately limited to the downstream product actually

sold. Clearon’s Cmts. at 9-16. The respondents oppose, arguing that even on remand Commerce

has failed to provide a reasoned explanation for departing from its established methodology or

demonstrated why the new methodology is better, that the change was unsupported by substantial

evidence and was arbitrary, and they ask the court to remand the issue with instruction to apply the

original methodology. Arch’s Cmts. at 2-3, 17; Kangtai’s Cmts. at 33-34, 38.

48 Id. at 29, referencing Letter to Jiheng, “2010-2011 Administrative Review of the Antidumping Duty Order on Clor-Isos the PRC” (Oct. 6, 2011) at D-9. Consol. Court No. 13-00073 Page 51

Arch continues to argue that Commerce has not provided a reasoned explanation, and

without one they are not able to “respond to, or address, whatever concerns Commerce may have had

with the previous methodology” or comment substantively on the change. Arch’s Cmts. at 10, 13.

Arch argues that although the Draft Remand results stated that the change in practice was necessary

“to bring the calculation into conformity with agency-wide policy”,49 and that “this policy is evident

from [Commerce’s] boilerplate questionnaire used in the underlying review, which asks parties to

report the FOPs required to process the byproduct into saleable downstream product”, this is not, in

fact, what the underlying questionnaire asked, see supra, and that the Draft Remand was the first

time Commerce used the term “saleable product” in the proceeding.50

Because the foregoing was the only explanation provided in the draft remand results,

Arch argued to Commerce that beginning with “saleable product” was not an “agency-wide” practice

at the time of the underlying review, which was instead to require that a by-product have commercial

value.51 Pointing to the second administrative review of the Order, Arch notes that Commerce

defended its practice before the court and that the court upheld Commerce’s findings that ammonia

gas and sulfuric acid had commercial value and were the appropriate by-products for offset purposes.

49 Draft Remand Results, RR-PDoc 64. 50 Arch’s Cmts. at 10, referencing Questionnaire, Section D (Oct. 6, 2011) at D-9; see supra, note 44. Arch also argues that because Commerce had previously stated that the downstream by-product methodology did not apply to its ammonia gas and sulfuric acid by-products, Arch had no reason to think the boilerplate language in the questionnaire referred to its downstream by-product, a point bolstered by the fact that Commerce had accepted Arch’s response that it had no downstream by-products in the Preliminary Results of the review. Re-opening the record on remand and Arch’s responses to Commerce’s requests for additional information, however, moots the point. 51 Arch’s Cmts. at 11, citing the 2008-2010 and the 2010-2011 Frontseating Valves reviews, see supra note 42. Consol. Court No. 13-00073 Page 52

Arch’s Cmts. at 11-12, referencing their Cmts. on Draft Remand Results at 11-12, RR-PDoc 67. See

Clearon Corp., supra, 37 CIT at ___, Slip Op. 13-22 at 27-31. Arch avers that Commerce did not

explain why products that it had previously stated were not downstream by-products, and which over

the course of the investigation and previous reviews it did not treat as downstream by-products, were

now suddenly being treated as downstream by-products. Arch’s Cmts. at 10.

Kangtai avers that in making the change Commerce devalued its by-product offset

to production costs and created conditions where NV is determined by the manufacture of the non-

subject downstream ammonium sulfate, not by the manufacture of the subject merchandise itself,

while the original methodology captures the full costs and full measurable offsets most accurately

Kangtai’s Cmts. at 36. It avers that recent administrative decisions are evidence that its new

methodology is “no policy at all” but that it was instead applied arbitrarily in the review to increase

antidumping duty margins. Id. at 33-34. Kangtai also claims that because the argument for the new

methodology was only raised at the briefing stage by petitioners in their case brief they were not

provided sufficient opportunity to argue if the surrogate value for ammonium sulfate was artificially

low. Kangtai also supports Arch in arguing that the record demonstrates there are two methods

under which Commerce determines “commercial value” and the right to a by-product offset, that

Commerce has unreasonably ignored the latter method, and that Kangtai’s by-products are

reintroduced into production because they are piped “directly from where they are generated into a

centrifuge tank to make ammonium sulfate.”52

52 Kangtai’s Cmts. at 34-35, referencing DuPont Teijin Films Redetermination; see also Kangtai’s Br. at 39-40; Kangtai’s Reply at 9-11; Arch’s Br. at 24-30; Kangtai’s Cmts. on Draft Remand, RR-PDoc 68, at 20-22; Arch’s Cmts. on Draft Remand, RR-PDoc 67, at10-13. Consol. Court No. 13-00073 Page 53

4. Analysis

Because of uncertainty over how prices are determined in non-market economies, 19

U.S.C. §1677b(c) requires the calculation of “normal value” in these sorts of proceedings to be

achieved through FOP methodology. Although not directly addressed in the statute, Commerce’s

treatment of co-products and by-products apparently derives from the consideration that is required

of it (Commerce) with regard to generally accepted accounting principles pursuant to 19 U.S.C.

§1677b(f)(1)(A). Among those, accounting’s matching principle53 requires proper allocation to co-

and by-products of the costs of their production (i.e., all relevant FOPs). Since the accounting

objective therefor is to determine the impact of such products on income and inventory carrying

values for financial reporting purposes,54 Commerce’s apparently current by-product offset practice,

which focuses upon whether a by-product has commercial value (demonstrated either by the

respondent’s sales of the by-product or by re-introduction into production), accords with generally

accepted cost accounting principles’ income and inventory concerns.

53 I.e., matching expenses with the benefits derived therefrom. See, e.g., Live Swine From Canada, 70 Fed. Reg. 12181 (Mar. 11, 2005) (final LTFV determ.), and accompanying I&D Memo, at cmt. 57. 54 See, e.g., Steven M. Bragg, Wiley GAAP 2011, pp. 329-31 (2010) (discussing cost flow assumptions for determining inventory carrying cost); see also 26 U.S.C. §472. For financial and cost accounting purposes, the “split-off” point is the point at which co- and by-products become separate and identifiable and at which the joint costs of production to that point must be allocated based upon suitable methodology. See, e.g., Wayne J. Morse and Harold P. Roth, Cost Accounting, p. 147 (3rd ed. 1986) (“CA”). See generally id., pp. 147-62, and Charles T. Horngren, Cost Accounting: A Managerial Emphasis, 531-534 (5th ed. 1982). Cf. Ipsco, Inc. v. United States, 13 CIT 402, 405-06, 714 F. Supp. 1211, 1214-15 (1989) (discussing need for cost allocation but variable cost allocation methodologies), referencing id. Consol. Court No. 13-00073 Page 54

Commerce also has the discretion in antidumping and countervailing duty

proceedings to modify a given methodology in order to calculate a more accurate dumping margin

or for ease of use.55 Commerce may not, however, apply a new methodology if a respondent has an

expectation right in the application of existing methodology, e.g., demonstrated reliance upon the

methodology, in effect at the time of action taken, to avoid dumping.56 Commerce must also provide

a reasoned explanation for the change, and it must demonstrate that its explanation is in accordance

with law and supported by substantial evidence.57 In changing methodology, Commerce must also

provide parties with timely notice and sufficient opportunity to provide the information required by

the revised methodology.58

55 See, e.g., NSK Ltd. v. United States, 19 CIT 1013, 1027, 896 F. Supp. 1263, 1275 (1995) (noting that Commerce need not “adhere to its prior . . . methodology, especially where Commerce is striving for more accuracy”), aff'd in part and rev’d in part on other grounds, 115 F.3d 965 (1997); SeAH Steel Corp. v. United States, 34 CIT 605, 615, 704 F. Supp.2d 1353, 1361-62 (2010); Arch Chemicals, Inc. v. United States, 33 CIT 954, 963-64 (2009), referencing Fujian Machinery and Equipment Import & Export Corp. v. United States, 25 CIT 1150, 1169, 178 F. Supp. 2d 1305, 1327 (2001) (“Fujian Machinery”); SKF USA Inc. v. United States, 31 CIT 951, 958, 491 F. Supp. 2d 1354, 1362 (2007). 56 See, e.g., Shikoku Chemicals Corp. v. United States, 16 CIT 382, 386-89, 795 F. Supp. 417, 420-22 (1992) (“Shikoku Chemicals”). 57 Arch Chemicals, Inc. v. United States, 33 CIT 954, ___ (2009) (“Arch Chemicals”), referencing Fujian Machinery, 25 CIT at 1169-70, 178 F. Supp. 2d at 1327 (citation omitted); see also Nippon Steel Corp. v. U.S. Int’l Trade Comm’n, 494 F.3d 1371, 1378 n. 5 (Fed. Cir. 2007) (“[w]hen an agency decides to change course . . . it must adequately explain the reason for a reversal of policy”) (citation omitted). 58 See, e.g., Arch Chemicals, supra, 33 CIT at 963-64; Anshan Iron & Steel Co., Ltd. v. United States, 27 CIT 1234, 1241-42 (2003); Fujian Machinery, supra, 25 CIT at 1169-70, 178 F. Supp. 2d at 1326-27; Hussey Copper, Ltd. v. United States, 17 CIT 993, 998, 834 F. Supp. 413, 419 (1993); Shikoku Chemicals, supra, 16 CIT at 388, 795 F. Supp. 2d at 421. Consol. Court No. 13-00073 Page 55

Commerce explains in the remand results that between the preliminary and final

results of the matter at bar it modified its methodology to “net realizable value” in order to conform

with “agency wide policy” and “to avoid overstating the value of the by-product”. RR at 28. Net

realizable value, sometimes called “net sales value,” is common to both co- and by-product cost

accounting and is a recognized method for assigning income or inventory value to a co- or by-

product. See CA at 151. And yet, regarding Commerce’s attempted articulation of its new

methodology,59 it is the function of the court to sustain only on the basis of that articulation, and not

to impute reasoning that the agency itself did not raise. On that basis, the court has concerns over

Commerce’s reasoning and cannot sustain its change of by-product offset methodology.

As an initial matter, the first portion of Commerce’s explanation, that it modified its

methodology to conform with “agency wide policy”, is not supported by substantial evidence. In

supplemental briefing requested by the court, Commerce explains that it did not intend “agency

wide” as a term of art but only used it to signal both a departure from its past practice and an attempt

to conform with its then-recent practice.60 However, the Remand points to only one case, Citric Acid

and Certain Citrate Salts, to support the claim that the new methodology is an “agency wide”

practice, and as Arch rightly points out, one case does not qualify this methodology as “agency wide”

or even a “practice”.61 Moreover, the proceeding in question was published in 2014, a year after the

59 The court will uphold “a decision of less than ideal clarity” et cetera. See Bowman Transp., Inc. v. Ark.-Best Freight Sys., Inc., 419 U.S. 281, 285-86 (1974). 60 Def’s Resp. to Court Questions (May 29, 2015) at 7-8. 61 See RR at 29, referencing Citric Acid and Certain Citrate Salts, supra, 79 Fed. Reg. 101, and accompanying I&D Memo at 12; see also Arch’s Cmts. at 12. Consol. Court No. 13-00073 Page 56

results of the review at issue, and thus does not speak to what practice was in place or altered at the

time of the review.

The remainder of Commerce’s explanation for its change in methodology (to “avoid

overstating the value of the by-product”) also contains several deficiencies. First, the remand results

are unclear on whether Commerce in this matter is granting an offset to each respondent for the full

amount of the ammonia gas and sulfuric acid claimed as produced during the POR, in accordance

with Commerce’s general by-products practice, as opposed to limiting the offset to the value of the

amount of those by-products as embodied in the amount of ammonium sulfate actually sold during

the POR. Cf. RR at 28 (“it [i]s still the Department’s practice to first start with the value of the

downstream product (i.e., ammonium sulfate) that was actually sold by the respondents and

produced during the POR”) (italics added). At a minimum, the matter requires remand for

clarification thereof.

Second, if Commerce is only granting an offset based on the amount of ammonium

sulfate that was actually sold during the POR, cf. id (“we must grant an offset equal to the amount

of value a company actually receives, less any processing costs, and not a hypothetical value”), then

the new methodology is actually a “net realized value” standard (based upon the values of the

ammonium gas and sulfuric acid by-products in actual sales of the downstream product that occur

during a period of review), not a “net realizable value” standard, which would therefore be at odds

both with the generally accepted accounting principles’ cost accounting concerns for income and

inventory valuations as well as at odds with Commerce’s allegedly still-existing policy of

determining whether or not the by-product has commercial value by proof of sales or reintroduction Consol. Court No. 13-00073 Page 57

into production. Arch raised this last point in its comments on the Draft Remand, and it is an

argument of cogent materiality that Commerce failed to address, requiring remand for that reason

as well. See Altx, Inc. v. United States, 25 CIT 1100, 1103, 167 F. Supp. 2d 1353, 1359 (2001),

quoting United States v. Nova Scotia Food Products Corp., 568 F.2d 240, 252 (2d Cir.1977) (“[i]t

is not in keeping with the rational [agency] process to leave vital questions, raised by comments

which are of cogent materiality, completely unanswered”); see also Arch’s Cmts. at 12.62

Third, it is unclear how Commerce’s new methodology is an improvement over its

previously applied methodology and is a reasonable change. Although it is apparent that Commerce

perceived a need to adjust how it calculates the by-product offset in this proceeding due to concern

over the irrationality of record evidence of higher surrogate values for the by-products than for the

downstream product into which they were further processed, Kangtai correctly points out that “this

is an accident of the surrogate values in a particular surrogate country at a particular time period and

has nothing to do with the legitimacy of the methodology”63, and that Commerce’s concern is really

a “capping” argument. See, e.g., Multilayered Wood Flooring from the PRC, 76 Fed. Reg. 64318

(Oct. 18, 2011) (final LTFV determ), and IDM at cmt. 23. Further, it is difficult to fathom why

62 Citing DuPont Teijin Films China Limited, et al. v. United States, 38 CIT ___, Slip Op. 14-106 (Sep. 11, 2014), and final results of redetermination pursuant thereto (dated Jan. 9, 2015). “The Department grants an offset for by-products generated during the production of subject merchandise if evidence is provided that such by-product has commercial value. The Department considers that a byproduct has commercial value if it is sold, or if, as in this instance, it is reintroduced into production. Thus, the Department’s practice is to attribute the commercial value to a by-product by virtue of its reintroduction. Given that DuPont Group ultimately reintroduces the PETWASTEOUT into production, this demonstrates that this byproduct has commercial value.” Arch’s Cmts. on Remand at 12. 63 Kangtai’s Resp. to Court Questions at 4. Consol. Court No. 13-00073 Page 58

Commerce would opt for a more complex methodology over the simplicity of the earlier one in order

to address the problem identified by the petitioners. The former methodology simply determined the

volume of the ammonia gas and sulfuric acid by-products that must have been produced at the split-

off point and calculated their values based upon the per-unit surrogate values of record that were

deemed appropriate. Some form of “cap” to address the petitioners’ (and Commerce’s) concern

regarding overstatement would be appropriate, and would certainly comport with Occam’s Razor,

but simplicity itself, of course, does not render a more complex method unreasonable. Rather, the

reasonableness of increased complexity must be assessed on the basis of the increased accuracy it

purports to achieve.

From the first through the fifth administrative reviews Commerce determined that the

allowable by-product offset is for those by-products that are generated at the split-off point in the

production of subject merchandise. For those reviews, Commerce adhered to a methodology that

simply assigned a surrogate value to those by-products. For the matter at bar, the further-

manufacturing FOPs (labor and electricity) that Commerce states “must” be deducted appear to be

relevant only for purposes of determining income or inventory values. Yet in focusing on sales of

ammonium sulfate during the POR, both the new methodology and the previous one ignore the

under- or over-statements of ammonia gas and sulfuric acid production during the POR that occur

due to changes between beginning and ending inventories of ammonium sulfate64 at the close of the

POR.

64 Kangtai avers that it has a “continuous production” operation. Kangtai’s Resp. to Clearon’s Br. at 8 (Feb. 24, 2015). Consol. Court No. 13-00073 Page 59

Certainly the new methodology does not result in improved accuracy to that extent,

and if the concern is simply over a proper per-unit valuation, one is left wondering: what, exactly,

is the improvement of the new methodology over the old one? After all, by-product valuation is not

an exact science but is largely arbitrary, albeit with defined rules of varying complexity. See, e.g.,

Cost Accounting at 149; Wiley GAAP 2011, ch. 9. Further, the new methodology appears to be an

attempt to determine what the “actual” (i.e., “saleable”) value of the ammonia gas and sulfuric acid

is to the respondents, but the “reality” of those values are apparently tethered to the surrogate value

for the downstream by-product, based upon a Philippine value, and not what the “reality” of what

the respondents “actually” received as compensation for sales of ammonium sulfate. Commerce

does not, on that further basis, demonstrate or persuade that its new methodology actually produces

a more accurate result than the old methodology, even if it is an attempt to comport with accepted

cost-accounting methodology.

Fourth, in voluntarily requesting remand, Commerce has side-stepped the parties’

arguments concerning lack of notice and comment. Commerce claims to have modified its

methodology in order “to avoid overstating the value of the by-product”, but Kangtai points out that

in the petitioners’ administrative case brief following the Preliminary Results they only expressed

concern over the surrogate values used to value ammonia gas and sulfuric acid and did not voice

concern over the existing methodology, and no party otherwise expressed concerns over the existing

methodology. Where there is no reliance interest in a particular methodology, Commerce has the

discretion to reconsider the methodology on its own, sua sponte, but in this matter, in announcing

after the Preliminary Results that it would base the by-product offset on actual sales of the Consol. Court No. 13-00073 Page 60

downstream product during the POR, not only has Commerce not adequately explained what was

wrong with the old methodology (except to state a desire to avoid “overstating”, which can be

addressed via, e.g., capping, as argued by Kangtai), it has not addressed or apparently considered

Kangtai’s and Arch’s arguments that they relied on the old methodology for their pricing of subject

merchandise, and that applying the new methodology ex post facto is fundamentally unfair. See, e.g.,

Kangtai’s Reply to Court Questions, ECF No. 98 (May 29, 2015) at 8-9. The issue therefore requires

reconsideration via remand.

In passing, the court considers Kangtai’s argument regarding Commerce’s apparent

agreement with the petitioners that based on Kangtai’s record keeping in the normal course of

business there was no verifiable way to allocate to the downstream ammonium sulfate product

certain labor and electricity associated with its production as a basis for rejecting Kangtai’s proposed

method of allocating labor and electricity incurred after the split-off point, and allocating such FOPs

to the production of subject merchandise. On the one hand, the court agrees the record does not

demonstrate that any labor was involved in the respondents further processing of ammonium gas and

sulfuric acid into ammonium sulfate, and Commerce’s point is not in accordance with its own

acknowledgment that production of “[a]mmonium sulfate . . . involves a large amount of electricity”.

See RR at 50. Furthermore, Commerce’s point is not in accordance with the “net realizable value”

methodology that it claims it had to utilize.65 On the other hand, to the extent Kangtai argues that

this amounts, in effect, to an ex post facto adverse inference against Kangtai for not keeping the

65 See, e.g., CA at 151 (“[S]eparate production costs incurred after the split-off point are easier to identity with individual products . . .. When separate production costs exist, joint costs are allocated on the basis of relative net realizable values.”). Consol. Court No. 13-00073 Page 61

records that Commerce’s new methodology would require, the remand results state that Commerce

relied on Kangtai’s own contention that if “the Department does not agree with Kangtai’s allocation

methodology . . ., the Department should simply award the by-product offset with no additional

by-product FOPs and with the Direct Material FOPs as previously reported at Exhibit D-7 of Section

D response dated November 28, 2011, where Kangtai attributed all and total consumption to the

CYA production.” RR at 50, quoting Letter from Kangtai, “Certain Chlor-Isos from the PRC --

Remand Questionnaire Response,” August 18, 2014, at 3-4. In other words, since the remand results

are apparently in accordance with what Kangtai itself argues, they are therefore not unreasonable to

that extent.

Commerce has the discretion to adopt a new by-product valuation methodology with

prospective effect, so long as it provides the respondents time to adapt and comply. On remand of

this matter, Commerce might be able to offer a valid explanation of why the respondents had no

reliance interest in the then-existing methodology and why the new methodology results in greater

accuracy and “avoid[s] overstating the value of the by-product” as well as address the parties’

arguments concerning lack of notice and comment and the remainder of the foregoing. If on remand

it again determines to calculate Kangtai’s by-product offsets based on net realizable value

methodology, then Commerce is requested to consider using the facts available, notwithstanding

Kangtai’s apparent concession, above, in order to properly allocate and attribute all FOPs incurred

in the production of ammonium sulfate. Further, Commerce must either supply valid reasons to

support changing its methodology in this proceeding which amounts to a “sufficient, reasoned Consol. Court No. 13-00073 Page 62

analysis”,66 supported by substantial evidence, or it should revert to its “former” (apparently still-

existing) methodology, albeit with any appropriate modification (e.g. capping) to avoid the “illogical

conclusions that do not match the real world experience of [Arch] and Kangtai,” that Commerce

explained was its true concern. See RR at 47.

IV. Conclusion

For the above reasons, the matter must be remanded for further proceedings not

inconsistent with this opinion. The results of remand shall be due December 18, 2015, whereupon

by the fifth business day thereafter the parties shall file a joint status report as to a proposed

scheduling of comments, if any, on the remand results, as well as a proposed page limitation(s)

thereof.

So ordered. /s/ R. Kenton Musgrave R. Kenton Musgrave, Senior Judge Dated: August 20, 2015 New York, New York

66 See NMB Singapore Ltd., v. United States, 557 F.3d 1316, 1328 (Fed. Cir. 2009); see also Huvis Corp. v. United States, 570 F.3d 1347, 1353 (Fed. Cir. 2009).