Ad Hoc Coal. of Am. SAP Producers v. United States Public version: 03/08/2024., No. 23-00010 (Ct. Int'l Trade 2024)

Annotate this Case

Download PDF

Slip Op. 24-26 UNITED STATES COURT OF INTERNATIONAL TRADE - - - - - - - - - - - - - - - - - - - -x THE AD HOC COALITION OF AMERICAN SAP PRODUCERS, Senior Judge Aquilino : : Plaintiff, : v. Court No. 23-00010 : UNITED STATES, PUBLIC VERSION : Defendant, : -and: LG CHEM, LTD., : Intervenor-Defendant. : - - - - - - - - - - - - - - - - - - - -x Opinion & Order [Plaintiff’s motion for judgment on the agency record granted; remand to defendant for reassessment.] Decided: March 1, 2024 Stephen J. Orava, Jamieson L. Greer, and Daniel L. Schneiderman, King & Spaulding LLP, Washington, D.C., for the plaintiff. Kyle S. Beckrich, Trial Attorney, Commercial Litigation Branch, Civil Division, U.S. Department of Justice, Washington, D.C., for the defendant. With him on the brief Brian M. Boynton, Principal Deputy Assistant Attorney General, Civil Division, Patricia M. McCarthy, Director, and L. Misha Prehiem, Assistant Director. Of counsel on the brief Rachel Bogdan, Senior Attorney, Office of Chief Counsel for Enforcement and Compliance, U.S. Department of Commerce, Washington, D.C. Court No. 23-00010 PUBLIC VERSION Page 2 J. David Park, Henry D. Almond, Kang Woo Lee, Gina M. Colarusso, and Archana Rao P. Vasa, Arnold & Porter Kaye Scholer LLP, Washington, D.C., for the intervenor-defendant. AQUILINO, Senior Judge: From the government’s supposedly sober double-take into the extent of dumped chemicals in diapers, the Ad Hoc Coalition of American SAP Producers (“Coalition”1) contests the Superabsorbent model matching Polymers from methodology the Republic used in of Korea: Certain Final Determination of Sales at Less Than Fair Value, 87 Fed.Reg. 65035 (Dep't Commerce Oct. 27, 2022), as explained in its accompanying issues and decision memorandum (Dep't Commerce Oct. 20, 2022) ("IDM") (together, the “Contested Determination”). Jurisdiction herein is pursuant to 28 U.S.C. §1581(c). The plaintiff has interposed a motion for judgment on the agency record pursuant to USCIT Rule 56.2.2 1 Herein, the collective noun “Coalition” is treated as a plural for ease of clarification among party references. 2 See Pl’s Mot. J. on Agency Rec. (“Pl’s Br”), ECF No. 20; Def’s Resp. to Pl’s Mot. J. on Agency Rec. (“Def’s Resp.”), ECF. No. 25; Int-Def’s Resp. to to Pl’s Mot. J. on Agency Rec. (“LG Chem Resp.”), ECF No. 26; Pl’s Reply Br. (“Pl’s Reply”), ECF No. 28. Court No. 23-00010 PUBLIC VERSION Page 3 I The Coalition petitioned the International Trade Administration (“ITA”) of the U.S. Department of Commerce in November 2021 to investigate whether superabsorbent polymers (“SAP”) from Korea are being or likely to be imported into the United States at less than fair value. See 19 U.S.C. §1673. SAP retains large amounts of water and other aqueous liquids. It is made of sodium polyacrylic acid, commonly of granular powder, also pellets, powder fibers, flakes, liquids, or gel. Petition Volume I (Nov. 2, 2021) at 1 (C.R. 2, P.R. 2); see Final Determination at 65037. SAP can also incorporate additives for anti-caking, anti-odor, and other similar purposes. Id. It is primarily used in downstream hygiene products that require fluid absorption. In its notice of initiation, ITA solicited comments and information general from physical proceeding’s potential interested characteristics control numbers in parties order (“CONNUMs”) that regarding to define would SAP’s the identify identical or similar merchandise for comparison when calculating dumping margins. Certain Superabsorbent Polymers From the Republic Court No. 23-00010 PUBLIC VERSION Page 4 of Korea: Initiation of Less-Than-Fair-Value Investigation, 86 Fed.Reg. 67915, 67916 (Dep’t Commerce Nov. 30, 2021) (P.R. 34). The Coalition and LG Chem, Ltd. (“LGC” or “LG Chem”3) provided separate characteristic hierarchies that they believed most important to distinguish SAP products. See Petitioner Model Match Comments (Dec. 13, 2021) (P.R. 42); LG Chem Model Match Comments (Dec. 13, 2021) (P.R. 43); see also Petitioner Rebuttal Comments on Model Match Product Characteristics (Dec. 23, 2021) (P.R. 49); LG Chem Rebuttal Comments on Model Match Product Characteristics (Dec. 23, 2021) (P.R. 54-55). All parties agreed that the model match criteria should include a characteristic for the ability of super-absorbent polymer products to hold liquid, recognized within the industry as “centrifugal retention capacity” or “CRC”. See id. It is measured in grams of saline solution retained per gram of SAP (“g/g”). Interested parties argued for different CRC ranges. The Coalition proposed that the model match hierarchy should consist of CRC divided into “low,” “intermediate,” and 3 A Korean manufacturer of SAP products, LG Chem was selected as the sole mandatory respondent in ITA’s investigation. See Respondent Selection Memorandum (Dec. 21, 2021) (C.R. 23, P.R. 47). Court No. 23-00010 PUBLIC VERSION Page 5 “high” capacity ranges, and it requested reporting producers to identify the specific standard used to measure CRC based on a CRC range of 6 g/g between low-capacity and high-capacity grades, to wit: (1) less than 30 g/g; (2) greater than 30 g/g but less than 36 g/g; and (3) greater than 36 g/g. See Petitioner Model Match Comments at 2. LG Chem agreed that CRC is “[t]he first and most important criterion for distinguishing SAP products” but asserted that “the traditional classification in the market [for SAP] is to group by divisions of 4 g/g.” LG Chem Model Match Comments at 2. LG Chem thus recommended that ITA establish five ranges of 4 g/g increments: (1) minimum or no guaranteed CRC of less than 26 g/g; (2) minimum guaranteed CRC of 26 g/g or more and less than 30 g/g; (3) minimum guaranteed CRC of 30 g/g or more and less than 34 g/g; (4) minimum guaranteed CRC of 34 g/g or more and less than 38 g/g; (5) minimum guaranteed CRC equal to or more than 38 g/g. See id. at 3. LG Chem also requested that ITA adopt two additional product characteristics for model match purposes of guaranteed performance levels: (1) absorbency under pressure (“AUP”), including absorbency under load (“AUL”), and (2) permeability Court No. 23-00010 PUBLIC VERSION Page 6 (“PERM”). LG Chem claimed that AUP indicates how well SAP responds to stress, and that PERM indicates the ability to pass liquid between superabsorbent polymer particles. Id. at 6. LG Chem claimed that “these commercially meaningful characteristics have impacts on the effectiveness of downstream products and on customer preferences.” LG Chem Resp. at 8, citing id. AUP assesses SAP’s ability to absorb liquid under a certain amount of pressure, e.g., 0.7 psi. LG Chem proposed model match on codes for AUP and AUL depending the type of test performed, with a cut-off threshold of 15 g/g for each test-based division: (1) no minimum guarantee; (2) minimum guaranteed AUP (0.3 psi) less than 15 g/g; (3) minimum guaranteed AUP (0.3 psi) equal to or more than 15 g/g; (4) minimum guaranteed AUP (0.7 psi) less than 15 g/g; (5) minimum guaranteed AUP (0.7 psi) equal to or more than 15 g/g; (6) minimum guaranteed AUL (0.9 psi) less than 15 g/g; (7) minimum guaranteed AUL (0.9 psi) equal to or more than 15 g/g. Id. at 5. Permeability refers to the ability with which liquid passes between SAP particles, where “g/g” refers to the weight of water that each gram of SAP can retain. Id. LG Chem proposed divisions into model match codes depending on the test used by the Court No. 23-00010 PUBLIC VERSION Page 7 producer to measure permeability: (1) no minimum guarantee; (2) minimum guaranteed “Gel Bed Permeability” (“GBP”4) less than 40 (u.o.m. = Darcy (10-8 cm2)); (3) minimum guaranteed GBP equal to or more than 40 10-8 (Darcy cm2); (4) minimum guaranteed Gel Permeability Under Pressure (“GPUP”) or Saline Flow Conductivity (“SFC”) less than 15 (10-7 cm3 sec/g); (5) minimum guaranteed GPUP or SFC equal to or more than 15 (10-7 cm3 sec/g); (6) minimum guaranteed Permeability Dependent Absorbency Under Pressure (“PDAUP”) less than 10 (g/g); (7) minimum guaranteed PDAUP equal to or more than 10 (g/g). See id. at 6. LG Chem also explained that AUP and permeability are “generally inversely related” to CRC; increases, AUP and permeability decrease. for example, as CRC Id. at 4.5 4 “GBP is a measurement under which no pressure is placed on the SAP in the swelling stage. GPUP and SFC are measurements under which 0.3 psi pressure is placed on the SAP in the swelling stage. Finally, PDAUP is a measurement under which 0.7 psi pressure is placed on the SAP in the swelling stage.” LG Chem Resp. at 8 n.2, referencing its Model Match Comments at 6 (P.R. 42). 5 Not relevant here, LG Chem also proposed that ITA include a fourth physical characteristic distinguishing between SAP with raw materials ultimately sourced from crude oil and SAP with raw materials sourced from biodiesel and other bio materials. LG Chem Model Match Comments at 6–7 and Attachment 1. Court No. 23-00010 PUBLIC VERSION Page 8 Another Korean producer and interested party, Sumitomo Seika Polymers Korea Co., Ltd., also submitted rebuttal comments regarding the model match hierarchy.6 Sumitomo Rebuttal Comments on Model Match Product Characteristics (Dec. 23, 2021) (P.R. 53). Like the Coalition and LG Chem, Sumitomo Seika agreed that CRC is the most relevant characteristic of SAP, but it requested that ITA not adopt pre-established gram-to-gram ranges of CRC and instead require respondents to explain in narrative how they define low, medium, and high capacity when making sales of SAP in the normal course of business. Alternatively, Sumitomo Seika requested that, if ITA determined to use pre-established ranges to define the CRC characteristic, then it should adopt the ranges that Sumitomo Seika uses in its normal course of business: (1) < 30 g/g; (2) > 30 g/g but < 42 g/g; and (3) > 42 g/g. Id. at 2-3. Regardless, Sumitomo Seika “d[id] not believe the inclusion of AUP or permeability is 6 It was not selected as a mandatory respondent so it sought to participate in the review as a voluntary respondent. See 19 C.F.R. §351.204(d). Towards that objective, it submitted preliminary information on its organization, accounting practices, markets, and merchandise, see SSPK’s Section A Response (Jan. 19, 2022) (C.R. 24-27; P.R. 67-70), but before providing a Sections B-D response of cost and sales data it withdrew its request to further participate. See Letter from SSPK to ITA, “Withdrawal of SSPK’s Request for Voluntary Respondent Treatment” (Jan. 25, 2022) (P.R. 91). Court No. 23-00010 PUBLIC VERSION Page 9 necessary to differentiate between its different models of [SAP].” Id. Sumitomo Seika proposed a broader divisional increment (12 g/g versus Coalition’s 6 g/g); in doing so, it agreed with distinguishing CRC as low, intermediate, and high capacity. Pl’s Mot. for J. on the Agency Record, ECF No. 20, at 6. See In their rebuttal comments, the Coalition also reiterated emphasis on the importance of classifying CRC in appropriate low, intermediate, and high categories, and that optimizing SAP for a certain CRC level generally will permeability. lead to trade-offs in the levels of AUP and Petitioner Rebuttal Comments on Model Match Product Characteristics (Dec. 23, 2021) at 4–6 (P.R. 49). The Coalition also emphasized that LG Chem’s proposal to assign divisions within AUP and permeability characteristics by a respondent’s chosen test methodology rather than physical characteristics would introduce distortions and allow a respondent to manipulate the dumping margin analysis, arguing that a given product could be categorized in multiple divisions within the AUP or PERM product characteristics depending on the selected testing method, which would result in more than one CONNUM being possible for such a product. Id. at 5–6. Court No. 23-00010 PUBLIC VERSION Page 10 After considering comments and rebuttal from interested parties, ITA announced its model match hierarchy in early 2022. The agency rejected LG Chem’s proposal and chose to use only CRC as the key physical characteristic for SAP, divided into three measurement ranges for low (less than 30 g/g), intermediate (30 to 36 g/g), and high Less-Than-Fair-Value Polymers from the (more than 36 g/g) of Certain Superabsorbent Korea: Product Characteristics Investigation Republic of Hierarchy (ITA Jan. 21, 2022) (P.R. 89). CRC. Memorandum, ITA did not address any of the arguments raised by the Coalition or LG Chem in support of their respective positions in its memorandum. Acknowledging that “[ITA] has historically been hesitant to revise the CONNUM once established,” LG Chem requested that the agency reconsider its model match hierarchy, asserting that it was “simplistic” and that the views of Sumitomo Seiko should be disregarded. LG Chem Request for Reconsideration (Jan. 28, 2022) at 2 and 4 (C.R. 53, P.R 93). ITA did not modify its decision in response at that time. LG Chem reported in its questionnaire response the CRC characteristic as defined by ITA. It also voluntarily provided alternative sales and cost data applying the model match criteria Court No. 23-00010 as outlined measuring in CRC in PUBLIC VERSION its affirmative increments of model 4 g/g Page 11 match and including permeability as key physical characteristics). Chem reported alternative “CONNUM2s” comments using (i.e., AUP and Specifically, LG its proposed CRC characteristic (“CRC1") as well as its permeability (“PERM”) and absorbency-under-pressure (“AUP”) characteristics, based on testing protocols chosen by LG Chem. LG Chem’s Sections B-D Questionnaire Response (Feb. 11, 2022) at B-10--B-12 (C.R. 60-61, 99, P.R. 97). In the preliminary determination, ITA based its antidumping analysis on the standard sales and cost files using the January 21, 2022 model match hierarchy (i.e., did not rely on LG Chem’s alternative CONNUM2s) and calculated LG Chem’s dumping margin as 28.74 percent. Certain Superabsorbent Polymers From the Republic of Korea: Preliminary Affirmative Determination of Sales at Less Than Fair Value, Postponement of Final Determination, and Extension of Provisional Measures, 87 Fed.Reg. 34647 (June 7, 2022) (P.R. 154). ITA afterwards conducted responsive cost and sales files. verification of LG Chem’s During verification, it did not explicitly verify LG Chem’s alternative cost and sales files or any associated values in the CRC1, AUP, or PERM characteristics put Court No. 23-00010 forth by LG Chem. PUBLIC VERSION Page 12 Sales Verification Report (Sept. 1, 2022) (C.R. 257, P.R. 173) and Cost Verification Report (Aug. 29, 2022) (C.R. 256, P.R. 171). Following requested that verification, ITA re-evaluate LG Chem and in its revise case the brief product characteristics to include its proposed AUP, permeability, and CRC ranges of 4 g/g increments. See LG Chem Case Brief (Sept. 9, 2022) (P.R. 174). The Coalition’s rebuttal case brief requested that the agency continue to adopt CRC as the sole product characteristic. See Petitioner’s Rebuttal Brief (Sept. 19, 2022) (P.R. 178). Upon review of the briefing and reexamination of the record, ITA revised the model match hierarchy to include LG Chem’s proposed AUP and permeability, amended CRC reporting to increments of 4 g/g, and recalculated LG Chem’s margin using LG Chem’s alternative sales and cost files based on LG Chem’s volunteered CRC1, PERM, and AUP characteristics. Specifically, ITA found that “reporting CRC in 4 g/g increments, as well as including AUP and permeability, are commercially significant.” IDM at 14 (P.R. 184). It claimed that replacing the CONNUMs used in the investigation with LG Chem’s CONNUM2s “recognizes the significant physical and Court No. 23-00010 PUBLIC VERSION Page 13 price differences in SAP produced with certain guaranteed levels of the physical characteristics.” Id. For the Final Determination, ITA ultimately asserted that price differences between products based on LG Chem’s proposed characteristics were “meaningful from a commercial perspective.” Id. at 9. It stated, on the one hand, that “cost differences were not instructive in determining that the differences were related to the additional physical differences.” Id. On the other hand, ITA found that each of the proposed characteristics were commercially meaningful because “the price differences highlighted by LG Chem and observed by [it] appear to have a commercial basis.” Id. at 10. ITA found that, of the many characteristics that can be identified for SAP, those of AUP and PERM, as proposed by LG Chem along with its 4 g/g increments significant with respect to price. asserted that differences of CRC, are commercially IDM at 11–12 (P.R. 184). associated with these It three characteristics, as presented by LG Chem, “are reflected in the sales price to LG Chem’s customers.” Id. at 12. Following ITA’s replacement of the model match hierarchy, the dumping margin for LG Chem decreased from 28.74 percent to Court No. 23-00010 17.64 percent. PUBLIC VERSION Page 14 Final Determination at 65036 (P.R. 188). After an affirmative final injury determination by the International Trade Commission (“ITC”), ITA published an antidumping-duty order based upon the Contested Determination. Certain Superabsorbent Polymers From the Republic of Korea: Antidumping Duty Order, 87 Fed.Reg. 77794 (Dec. 20, 2022) (P.R. 196). This appeal ensued. II In an action such as this, the standard of judicial review is whether such a final determination is “unsupported by substantial evidence on the record, or otherwise not in accordance with law.” 19 U.S.C. §1516a(b)(1)(B). The statute requires ITA to conduct a “fair comparison” between a product’s “normal” value and either export price or constructed export price for its dumping margin analysis. 19 U.S.C. §1677b(a). “Normal” value is typically based on the adjusted home-market price of the foreign like product, which is defined in 19 U.S.C. § 1677(16)(A) in relevant part as “[t]he subject merchandise and other merchandise which is identical in physical characteristics.” To implement the statute’s requirements, ITA is authorized to compare merchandise that is “identical in physical Court No. 23-00010 PUBLIC VERSION Page 15 characteristics” by establishing a model match hierarchy7 of the physical characteristics of the subject merchandise. Koyo Seiko Co. v. United States, 66 F.3d 1204, 1209 (Fed.Cir. 1995). Physical characteristics must be “commercially differences are to be disregarded. significant”; minor Pesquera Mares Australes Ltda. v. United States, 266 F.3d 1372, 1384 (Fed.Cir. 2001). Deciding upon the model match hierarchy is a “fact-intensive inquiry”8, over which ITA has discretion. 1369, 1381 (Fed.Cir. SKF USA Inc. v. United States, 263 F.3d 2001). And, while the “commercial significance” of a difference in physical characteristics is a determination made on a case-by-case basis, “at the very least it is a feature that is recognized in the broader industry of the subject merchandise.” Bohler Bleche GmbH & Co. KG v. United States, 42 CIT ___, ___, 324 F.Supp.3d 1344, 1350 (2018)(“Bohler Bleche”) (citing Pesquera Mares, 266 F.3d at 1385). 7 This hierarchy of characteristics is used to create a control number, or CONNUM, in this instance for each unique SAP product. CONNUMs are comprised of digits, and each digit is a code for a physical characteristic of the product. 8 La Molisana S.p.A. v. United States, 47 CIT ___, ___, 633 F.Supp.3d 1266, 1271 (2023) (citation omitted). Court No. 23-00010 PUBLIC VERSION Page 16 III The Coalition contends the Final Determination is unlawful because ITA: (A) erred by departing from its established practice of using the model match hierarchy that was decided early in the proceeding; (B) did not rely on substantial evidence in changing that hierarchy for the Final Determination; (C) relied on unverified alternative sales and cost information; and (D) did not address their argument regarding the potential for manipulation of the dumping margin. A The parties dispute whether ITA has an “established practice” of using the model match hierarchy chosen at the outset of an investigation for its final determination. The Coalition insists that ITA does, and it points to several agency statements from other proceedings indicating a practice of adhering to product matching criteria developed early in the investigation.9 9 They also Pl’s Reply at 1-2, referencing Steel Propane Cylinders from Thailand: Final Determination of Sales at Less Than Fair Value, 84 Fed.Reg. 29168 (Dep’t Commerce June 21, 2019), and accompanying issues and decision memorandum (“I&D Memo”) at Comment 1; Carbon and Alloy Steel Wire Rod From the Republic of Korea: Final Affirmative Determination of Sales at Less Than Fair Value . . ., 83 Fed.Reg. 13228 (Dep’t Commerce March 28, 2018) (“SWR from Korea”), I&D Memo at Comment 3 (“Commerce has a long-standing practice of developing product characteristics and a model match (continued...) Court No. 23-00010 PUBLIC VERSION Page 17 point to ITA’s own statement in this proceeding that it has “historically been hesitant to revise the CONNUM once established.” See Final Determination at 13. The defendant and LG Chem point to AB Rubber from France10 as an instance where ITA modified the model match framework at the final stage of an investigation. The court considers that, to the extent the agency’s model match hierarchy practice can be said to be “established”, once announced, the practice indicates reticence to revise the hierarchy after investigation. it is decided at the early stage of an Yet, it is settled that ITA has the authority to reconsider any decision on any aspect of an investigation made earlier in a proceeding prior to reaching a final determination. 9 (...continued) methodology in the early stages of each proceeding, and in consultation with the interested parties”); Notice of Final Determination of Sales at Less Than Fair Value . . .: Diamond Sawblades and Parts Thereof from the Republic of Korea, 71 Fed.Reg. 29310 (Dep’t Commerce May 22, 2006), I&D Memo at Comment 1 (“[ITA] should not make any changes to the product characteristics or model match criteria at this time. We find that the appropriate time to consider comments with respect to the physical characteristics and model match criteria is at the beginning of the proceeding”). 10 Acrylonitrile-Butadiene Rubber From France: Final Affirmative Determination of Sales at Less Than Fair Value . . ., 87 Fed.Reg. 37833 (Dep’t Commerce June 24, 2022). Court No. 23-00010 PUBLIC VERSION Page 18 See, e.g., Tokyo Kikai Seisakusho, Ltd. v. United States, 529 F.3d 1352, 1360 (Fed.Cir. 2008) (“[t]he power to reconsider is inherent in the power to decide”); Hyundai Steel Co. v. United States, 42 CIT ___, ___, 319 F.Supp.3d 1327, 1343 (2018), quoting NTN Bearing Corp. v. United States, 74 F.3d 1204, 1208 (Fed.Cir. 1995) (“[p]reliminary determinations are ‘preliminary’ precisely because they are subject to change”). Implicitly acknowledging this state of the law, the Coalition contend that ITA has provided no “compelling” reason for altering its model match hierarchy in this investigation. Pl’s Reply at 4. See supra n.9. ITA has explained that it will find that “compelling reasons” exist if a party proves by “compelling and convincing evidence” that the existing model-match criteria “are not reflective of the merchandise in question,” that there have been changes in the relevant industry, or that “there is some other compelling reason present which requires a change.” Fagersta Stainless AB v. United States, 32 CIT 889, 894, 577 F.Supp.2d, 1270, 1277 (2008), quoting Notice of Final Results of the Twelfth Administrative Review of the Antidumping Duty Order on Certain Corrosion–Resistant Carbon Steel Flat Products from the Republic of Korea, 72 Fed.Reg. 13086, and accompanying I&D Memo at Comment 1(b) (March 20, 2007). Court No. 23-00010 PUBLIC VERSION Page 19 In AB Rubber from France11, concerning which the parties differ as to interpretation, ITA did not explicitly invoke the phrase “compelling reason” for altering its model-match hierarchy in the end, but it is apparent from the issues and decision memorandum of that proceeding that the agency found differences among product produced with and without certain stabilizers to be commercially significant, which it therefore considered a “compelling reason” to adopt a respondent’s proposed alternative CONNUMs for the final determination. In this proceeding, similarly, ITA did not explicitly claim a “compelling reason” in altering its model match hierarchy, but it is apparent that, when it examined the record as a whole, it concluded it had a compelling reason to do so. Whether the Final Determination adequately examines and articulates the substantial evidence and reasoning to support that conclusion (to adopt a new model match hierarchy at the tail-end of the proceeding) is considered below, but the court declines to hold as a matter of law that ITA was required to adhere to the model match hierarchy that it constructed during the early stage of the proceeding. 11 Supra, n.10. Court No. 23-00010 PUBLIC VERSION Page 20 B The IDM purports that LG Chem supported arguing the existence of commercially significant differences between SAP with certain guaranteed levels of AUP and permeability as well as SAP with CRC ranges reported in 4 g/g increments by providing separate analyses showing cost and price variations in comparison to ITA’s existing CONNUMs; that those analyses showed that the price and cost differences could be substantial; that because they appeared significant on their face, ITA performed its own analysis of the prices reported in LG Chem’s home market database and found that the price differences could be as much as 20 percent; that ITA’s analysis of LG Chem’s reported costs found their differences were not instructive in determining that they were related to the additional physical characteristics. Because price and cost IDM at 9. differences alone are an insufficient basis for changing its model match methodology, ITA requires such differences to be “meaningful from a commercial perspective.” observation Id. that at 9-10. relevant Noting differences the court’s linked to previous product characteristics are those which “customers would view . . . as distinct in utility and value”, Bohler Bleche, 42 CIT at ___, 324 F.Supp.3d at 1350, ITA “examined the evidence on the record and Court No. 23-00010 PUBLIC VERSION Page 21 found that the price differences highlighted by LGC and observed by [it] appear to have a commercial basis which is ‘recognized by the broader industry of subject merchandise.’” Id. at 10. That evidence included the Coalition’s own marketing materials, ITA stating: (1) CRC, AUP, and permeability are typical current SAP characteristics and an integral part of modern diapers; (2) SAP producers are able to break the normal restrictions of CRC and AUP independently; and (3) SAP products are marketed as distinct due to a small percentage difference in CRC. . . . [A] technical paper from BASF . . . lists CRC, AAP[12], and permeability as “Typical current SAP Characteristics;” an information sheet from BASF describes CRC, AAP, and permeability as “an integral part of modern diapers”; and a brochure from Evonik Superabsorber LLC discusses its entire FAVOR® product range as featuring “different basic characteristics” which includes “absorption, retention, absorption under pressure, {and} permeability.” Further, BASF’s marketing materials emphasize that one product is meaningfully distinct from another product due to a ten percent improvement in CRC and results in a seven percent improvement of the downstream diaper product. Id. (citations omitted). The Coalition characterize this evidence as thin, sparse, and tenuous. Pl’s Br. at 16, 18. They criticize that the marketing materials provided by LG Chem were obtained from the public domain and mostly undated; that the slide presentation from 12 “AAP” is short for Absorbency Against Pressure and refers to the same physical characteristic as AUP. IDM at 5 n.26. Court No. 23-00010 PUBLIC VERSION Page 22 BASF Corporation (a member of the Coalition claiming a difference in CRC of 10 percent between two SAP products in its “HySorb” product line13 –- based upon which LG Chem apparently claimed that a ten percent difference is approximately equivalent to 4 g/g) was dated from 2015 and concerned a legacy product; that ITA relied on those marketing materials for the purpose of including only AUP and PERM in model matching; yet, the materials identify numerous physical characteristics of SAP, including CRC, absorption speed, odor control, haptics properties, SFC, raw materials purity, flow rate, bulk density, particle size distribution, absorption, pH, absorption under pressure, GBP, residual monomer, extractables, and color14; and they criticize ITA’s lack of explanation for finding it appropriate to accept and rely on only two of these (in addition to narrower CRC increments) as providing “commercial significance;” and that ITA’s reference to the CRC ranges they suggested in the ITC’s injury investigation were submitted for the purpose of identifying product groups for the underselling analysis of domestic SAP, with CRC ranges of 6, 7, and 8 g/g (and which ranges 13 14 I.e., HySorb 9030 and HySorb 9900. Pl’s Br. at 10-11, referencing LG Chem Model Match Rebuttal Comments (Dec. 23, 2021) at: Attachment 1, pp. 3, 13, and 18-21; Attachment 2, Section II “SAP Properties;” Attachment 3, p. 1 and Table 3; Attachment 4; and Attachment 5, pp. 9–10 and 13–14 (P.R. 54-55). Court No. 23-00010 PUBLIC VERSION Page 23 were broader, not narrower, in any event). Pl’s Br. at 11, referencing IDM at 11, n.83; see also Pl’s Reply at 12-13. The dispute, at this point, is over ITA’s interpretation of what the foregoing evidence implies. Defendant’s central argument is that “the record contains information from both LG Chem and the Coalition showing that th[e AUP and permeability, in addition to CRC,] characteristics are commercially significant and not merely inversely related to one another.” Def’s Resp., ECF No. 25, at 7. The Coalition contend that that is not this case, because CRC, AUP and permeability parameters necessarily involve trade-offs such that optimizing performance for one characteristic tends to decrease performance in the other characteristics, and since it is not possible to optimize all three parameters in a single product,“[t]his means that no SAP type is inherently superior or inferior to any other; it simply depends on the customer’s preference for balancing the SAP parameters” -- which implies that there is no inherent relation between costs, price, and the characteristics proposed by LG Chem. Pl’s Br. at 18-19. This means, in other words, that the AUP and permeability product characteristics “have no commercial significance or ‘utility’ that Court No. 23-00010 PUBLIC VERSION Page 24 is not already captured by the CRC product characteristic.” Pl’s Reply at 11. The court observes that ITA’s price analysis of LG Chem’s home market database compared the original CONNUMs to LG Chem’s alternate CONNUMs and “found that the price differences could be as much as 20 percent.” IDM at 9; see also Def’s Resp. at 13. However, as the Coalition argue, this does not actually support the claim of “commercial significance” because changes to weighted-average prices are an unsurprising and expected result of changing the CONNUMs assigned to products: ITA observed that there are also cost differences between databases but in contrast to its treatment of commercially price the agency insignificant characteristics. and dismissed unrelated such to differences the new as physical See, e.g., Pl’s Br. at 12-13. The Coalition submitted for ITA’s consideration a table demonstrating that the new CONNUMs, when controlled for each new physical characteristic, show no correlation between price and the characteristic, i.e., as the values in LG Chem’s “alternative” data fields (CRC1, AUP, or PERM) increase, the average unit value (AUV) for price does not have a correlating increasing (or decreasing) trend. See Pl’s Br. at Attachment 1. Further, they point out, Court No. 23-00010 PUBLIC VERSION Page 25 there is no indication in LG Chem’s sales documentation that certain minimum guaranteed performance levels for a given grade of SAP [[ ]]. Pl’s Reply at 9, referencing, e.g., LG Chem Section A Response (Jan. 19, 2022) (“LGCA”) at Exhibit A-18(A)’s Exhibit C, and Exhibit A-18(B)’s Exhibit (P.R. 77, C.R. 36-39). ITA’s practice is to rely on price and cost correlations as evidence of commercially significant differences among product characteristics and to reject product characteristics that do not show this. See Pl’s Reply at 9, referencing SWR from Korea15 I&D Memo at Comment 3 (“[w]ith regard to pricing differences, POSCO has not demonstrated that pricing differences arose as a result of differences in the three product characteristics it proposes, or even the extent to which such pricing differences are correlated with variations in those three proposed characteristics”). But, ITA dismissed the Coalition’s analysis on the ground that the methodology did not “accurately reflect” the foreign like product, which in the agency’s view consists of all three product characteristics “collectively”. IDM at 11 n.82. ITA reasoned that 15 Supra, n.9. Court No. 23-00010 PUBLIC VERSION Page 26 its response in SWR from Korea is not inconsistent with taking a “collective” view of product characteristics when it examines pricing or cost correlation, and it here avers that it was persuaded that AUP and permeability, in addition to CRC in 4 g/g increments, are commercially meaningful CONNUMs for the foreign like product, as its IDM also explained: LGC’s marketing materials show that LGC also views CRC, AUP, and permeability as basic properties of SAP. For example, LGC provided a product brochure which contains information regarding raw materials, specifications, and applications of SAP. The brochure provides a graphic separating the three major categories of SAP properties and it clearly states that CRC, AUP, and permeability are basic properties. Further, and in line with its marketing materials, LCG sells SAP by grade, which is based on the physical characteristics of the products. LGC did not provide a key to its grade codes and, thus, we cannot determine to what extent LGC accounts for these characteristics in its grades. Therefore, we turned to LGC’s home market sales database to analyze this question; our review showed that the grade codes and CONNUM2s proposed by LGC tracked perfectly (i.e., each grade code fell into only one CONNUM), suggesting that . . . LGC defines its SAP grades using both AUP and permeability (and that customers purchase SAP with expectations related to these characteristics). In this regard, . . . the question before us is not whether it is possible to optimize each of the three characteristics; rather, it is whether, in combination, those characteristics create distinct products which are regarded as meaningfully different from a commercial perspective. IDM at 11. Court No. 23-00010 PUBLIC VERSION Page 27 The agency’s analysis of the record appears flawed to the extent the Coalition show that it does not actually evince any material correlation between LG Chem’s reported alternative CONNUMs (i.e., CRC1, PERM, and AUP) and either prices or costs. ITA appears to have largely premised the commercial significance of AUP, permeability, and 4 g/g CRC increments on a relatively small set of unverified and mostly undated marketing materials. at 14-15. Those substantial materials evidence of the in isolation commercial do not See id. represent significance of AUP, permeability, and 4 g/g CRC increments. The defendant argues that AUP and permeability must be included in the model match because they are “basic properties of SAP”, and it supports this claim by citing to lists of SAP physical characteristics in three slide decks discussing SAP, contending that the mention discussion of significance. of price Id. AUP or at and permeability cost, 14. is But enough mere in to them, show references with no commercial to AUP and permeability in those materials do not demonstrate the existence of commercially significant differences, which must be manifest on the record. As discussed earlier, they identify a large number of physical characteristics of SAP products, e.g.: CRC; absorption speed; odor control; haptics properties; saline flow conductivity; Court No. 23-00010 PUBLIC VERSION Page 28 raw materials purity; flow rate; bulk density; particle size distribution; absorption; pH; absorption under pressure; GBP; residual monomer and extractables; color, and so forth. See, e.g., LG Chem’s Rebuttal Model Match Comments at Attachment 1, pp. 3, 13, 18-25; Attachment 2, Section II “SAP Properties;” Attachment 3, p. 1 and Table 3; Attachment 4; Attachment 5, pp. 9-10 and 13-14; and Attachment 6, p. 101 (P.R. 54-55). Apart from the primary physical characteristic of CRC, upon which the parties agree, there is nothing particularly evident in terms of commercial significance about any one of the other physical characteristics that would make it stand out from the rest. Similarly, the defendant points to the discussion during the preliminary staff conference of the ITC’s injury investigation, where a domestic industry official listed some examples of SAP properties, including CRC, speed, AUP, and permeability. See Def’s Resp. at 14-15. Yet the ITC’s final report lists the fuller, greater number of quality-related characteristics of SAP identified by purchasers than only AUP and permeability in addition to CRC, including: absorption speed; free swell; hydroxyl value; iodine value; permeability; particle size distribution; color; color stability; residual monomer; moisture content; foreign material; pH; and odor. Superabsorbent Polymers from South Korea, USITC Pub. Court No. 23-00010 5388 (Dec. 2020) at II-20. PUBLIC VERSION Page 29 Out of all such characteristics, many of which are discussed at length in the marketing materials, ITA summarily concluded that AUP and permeability drive commercially significant price differences. Def’s Resp. at 14. But it provided no data or analysis demonstrating that AUP and permeability have greater commercial significance than the other characteristics cited in the materials that served as the primary basis for the agency’s decision. Fundamentally, the defendant claims instead that, because AUP and permeability are “an integral part of modern diapers,” they are commercially significant, ipso facto, in addition to CRC. id. at 13-15. See When viewed in the context of the many SAP characteristics identified throughout the investigation, however, mere reference to AUP and permeability in briefing and marketing materials is scant support on which to base a finding of commercial significance before ITA. The Coalition argue that considering the CRC1, AUP and PERM CONNUM2s “collectively” means ITA looked at whether the new CONNUMs reflected a price correlation instead of examining the new product characteristics themselves. They contend that ITA provides no explanation why this should be the case, and, in so doing, fails Court No. 23-00010 PUBLIC VERSION Page 30 in its obligation to consider evidence provided by them that fairly detracts from its conclusion. See, e.g., CS Wind Viet. Co. v. United States, 832 F.3d 1367, 1373 (Fed.Cir. 2016). The court concurs. ITA defines product control numbers in order to capture those physical characteristics that have a meaningful impact on costs or prices. The fact that there are price differences among the CONNUM2s reported by LG Chem is irrelevant if it cannot be demonstrated that those differences relate to particular physical characteristics that have commercial significance. §351.411(b) (permitting the agency to See 19 C.F.R. consider only those differences among products associated with “physical differences”) and ITA Policy Bulletin 92.2 (July 29, 1992) (prohibiting ITA from attributing price or cost differences related to “extraneous factors” to physical differences). By regarding LG Chem’s CONNUM2s “collectively,” and looking at price differences among CONNUM2s -rather than looking among specific product characteristics within those CONNUM2s -- ITA has effectively “flipped the script” and assumed the conclusion. In this case, under ITA’s “collectively”- considered methodology, the source of any price differences among CONNUM2s remains unknown: price differences may result from random variations or non-random factors having nothing to do with the AUP, Court No. 23-00010 PUBLIC VERSION Page 31 permeability, and CRC1 characteristics as reported by LG Chem. Thus the fact that CONNUM2s may show different prices says nothing about whether those differences are attributable to physical characteristics. Absent finding that prices move specifically in relation to each of LG Chem’s reported AUP, permeability, and CRC1 characteristics (the Coalition demonstrably showing that they do not), ITA’s decision to define CONNUMs based on those characteristics is unsupported by substantial evidence and not in accordance with law. The Coalition appear correct that viewing the CONNUM2s “collectively” masks the commercial significance of each of the product characteristic codes selected by ITA -- the very question ITA is attempting to answer in considering whether to include AUP and permeability in addition to CRC. Here, the product codes differ according to testing methodologies for the product characteristics; they are not based on the underlying physical characteristics themselves. Several CONNUMs have product codes that reflect [[ ]], making it impossible to even articulate a priceproperty relationship. See Final Determination Margin Calculation (Oct. 20, 2022) at Attachment 3 (P.R. 187, C.R. 260) (showing that Court No. 23-00010 [[ ]] of [[ PUBLIC VERSION Page 32 ]] of LG Chem’s CONNUM2s have product characteristic code [[ ]]). The defendant claims that AUP, permeability, and CRC in narrow ranges create CONNUMs that are “commercially distinct in both utility and value,” citing Bohler Bleche, 42 CIT ___, 324 F. Supp. 3d 1344. Def’s Resp. at 12, 14, and 18. However, the new product characteristics do not fit this description of commercial significance. First, as explained above, the new characteristics and increments show no correlation when analyzed with SAP value, i.e., price. Second, the Coalition’s evidence demonstrates that AUP and permeability have an inverse relationship with CRC. This means that these two product characteristics have no commercial significance or utility that is not already captured by the CRC product characteristic. LG Chem’s product brochure itself highlights the existing “correlation between properties” among CRC, AUP, and proportion” permeability, or “direct explicitly proportion” describing between the them. “inverse LG Chem’s Rebuttal Model Match Comments at Attachment 2, Section II “SAP Properties” (P.R. 54-55). Other LG Chem submissions confirm that “[t]here is a trade off between main absorption properties of SAP.” LGCA at Exhibit A-25, pp. 22-23 (P.R. 77-78). Court No. 23-00010 PUBLIC VERSION Page 33 The defendant diminishes this underlying facet of SAP chemistry but highlights a marketing claim in undated BASF materials for a legacy SAP product of “break[ing] the normal restrictions of the CRC & AUP interdependency.” Def’s Resp. at 14. Apparently based solely on this, ITA “concluded that LG Chem [also] uses more advanced technologies to achieve SAP with the right balance of CRC, AUP, and permeability.” never makes this claim. Id. at 16. But LG Chem In fact, it concedes in its brief that “‘AUP’ is a trait that is generally inversely related to CRC.” Chem’s Br. at 23, 28. LG ITA appears to have made its assumption about LG Chem’s products without any evidence, data, or supporting information. Its model match determination is therefore premised upon a conclusion with no basis. With respect to the narrower 4 g/g CRC increments, ITA’s determination relies on (1) a single graphic in an eight-year-old presentation discussing the relative CRC levels of legacy products no longer sold in the U.S. market and (2) the broad CRC ranges that the Coalition suggested the ITC use for its underselling analysis. See Def’s Resp. at 15-16. Regarding the graphic, nothing in it links a narrower CRC level to the price (or cost) of SAP. Whether narrower CRC ranges caused commercially significant price or cost differences cannot reasonably be concluded from this graphic. Court No. 23-00010 PUBLIC VERSION Page 34 ITA’s reliance on it does not satisfy the substantial evidence standard. Regarding the ITC’s underselling analysis, the pricing products ITA used for its purpose must be tethered to that context, if the analysis is to have any relevance in the context of an ITA investigation. The pricing products before the ITC are used to compare U.S. sales of imported and domestic SAP; whereas ITA’s investigation compares a foreign producer’s foreign like product normal values and export sales. Regardless, the pricing products before the ITC, in CRC ranges of 6, 8, and 7 g/g, establish broader CRC increments, not narrower ones. Moreover, the only other foreign producer in the investigation before ITA agreed that broad increments for high, medium, and low CRC levels are appropriate, and it explicitly opposed the inclusion of AUP and permeability. See Sumitomo Seika Rebuttal Comments on Model Match (Dec. 23, 2021) at 2-3 (P.R. 53). Such evidence of record does not bolster the agency’s rationale for changing its model match hierarchy to a narrower CRC increment at the eleventh hour. LG Chem did not produce any evidence from its own records to support its proposed CRC increments. LG Chem’s own materials differentiate SAP as “high capacity” and “low capacity.” See LGCA Court No. 23-00010 PUBLIC VERSION Page 35 at Exhibit A-25, p. 7 (P.R. 77). This record evidence does not support 4 the specific claim that g/g CRC increments are a “traditional” industry practice. In addition to repeating much of defendant’s arguments, LG Chem’s brief criticizes the early “settled” model match used throughout the period of the investigation, because it “only established three CONNUMs”; LG Chem implies that no other antidumping proceedings relied on a small number of CONNUMs. LG Chem’s Br. at 9, 10, and 21. The number of CONNUMs, however, is not relevant to whether the agency’s determination is based on substantial evidence or otherwise in accordance with law. necessary only when products have Different CONNUMs become commercially significant differences, not for the sake of creating unnecessary complexity. For that matter, there have been a number of proceedings with limited numbers of CONNUMs, particularly with respect to chemical products that have a basic molecular makeup like SAP. In Glycine from China, for example, there were no reportable product characteristics, and all glycine was effectively treated as being within a single CONNUM. Request for Information, Glycine from China (ITA May 18, 2017) at C-5, ACCESS Barcode Number 3573743-01. Court No. 23-00010 PUBLIC VERSION Page 36 Similarly, in a Urea Ammonium Nitrate Solutions investigation, the CONNUM was based on one physical characteristic (nitrogen content) and also a second binary characteristic to distinguish product that also contained corrosion inhibitors. Memorandum, Product Charac- teristics for the Less-Than-Fair-Value Investigations of Urea Ammonium Nitrate Solutions from the Russian Federation and the Republic of Trinidad and Tobago (ITA Sept. 1, 2021), Attachment, ACCESS Barcode Number 4157013-01. with past ITA practice to have Thus, it is not inconsistent a limited number of product characteristics and CONNUMs. To summarize, the agency apparently relied for the most part on a few pieces of anecdotal information as the sole factors weighing in favor of finding commercial significance among LG Chem’s preferred product characteristics. That is hardly a “robust” evidentiary basis for replacing the model match hierarchy. The court thus concurs with the Coalition that this evidence would neither convince nor compel a reasonable person to conclude that AUP and permeability are commercially significant or conclude that CRC levels are more appropriately narrowed to 4 g/g ranges from the initial model match hierarchy. The justifications offered by the defendant and LG Chem for altering the model match Court No. 23-00010 PUBLIC VERSION Page 37 hierarchy at the tail-end of the investigation fail to demonstrate that ITA based that decision on substantial evidence. C The Coalition bolster their position in arguing that ITA’s eleventh-hour change in the model match hierarchy meant that it failed to verify LG Chem’s alternative sales and cost files based on LG Chem’s CONNUM2 definitions and used such unverified information in the Final Determination, contrary to the statute. See, e.g., Pl’s Br. at 14. The statute requires the agency to “verify all information relied upon in making . . . a final determination in an investigation.” 19 U.S.C. §1677m(i). No party contests this basic requirement. See, e.g., Def’s Resp. at 24; Pl’s Reply at 14. The defendant claims that ITA “verified the sales and cost data that LG Chem provided, including the AUP and permeability levels that LG Chem also provided in the alternative databases.” Def’s Resp. at 6. LG Chem takes that claim a step further, insisting that ITA “specifically verified [its] product characteristics at both the sales and cost verifications, and did so on the basis of both the initial control number and the control number [it] proposed.” Court No. 23-00010 LG Chem’s Resp. at 35. PUBLIC VERSION Page 38 The legal and factual premises of both of these assertions appear to be flawed. Courts generally do, as the defendant notes, provide ITA “latitude” in its approach to executing the statute’s requirement to verify all information. Def’s Resp. at 25. The court has also, on more than one occasion, explained that “[v]erification is like an audit, the purpose of which is to test information provided by a party for accuracy and completeness.” See, e.g., Dalian Meisen Woodworking Co. v. United States, 45 CIT ___, ___, 571 F.Supp.3d 1364, 1371 (2021); Bomont Indus. v. United States, 14 CIT 208, 209, 733 F. Supp. 1507, 1508 (1990). For its cost verification, ITA sought to verify “the cost data file submitted on April 04, 2022.” This was cost database lgccop02.sas7bdat, not lgccop02_alt.sas7bdat, the latter database being the one with the alternative model match data. See LG Chem’s First Section D Supplemental Questionnaire Response (April 5, 2022) at 2 (P.R. 116).16 16 ITA informed LG Chem it would review costs for original CONNUM [[ ]] and CONNUM [[ ]] “in detail.” Cost Verification Report (Aug. 29, 2022) at 2 (C.R. 256, P.R. 171). Court No. 23-00010 PUBLIC VERSION Page 39 ITA reviewed the reported per unit costs for the selected CONNUMs, i.e., based on the initial model match methodology. It reported that it “traced the physical characteristics of the grade [[ ]] (i.e., centrifuge retention capacity) to the COA (i.e., certificate of analysis) data from the global supplier quality assurance system . . . and confirmed that the product had been appropriately classified as CONNUM [[ ]].” Id. at 13. But ultimately, those CONNUMs were not used in the Final Determination. ITA did not trace CONNUM2s for AUP, permeability, or replacement CRC ranges to the certificate of analysis or otherwise verify their accuracy, even on a “spot check” basis, as part of the cost verification. The agency did not ask to review -– and did not review -– costs for any CONNUM2s created for the altered model match hierarchy. The fact that AUP and permeability coincidentally may appear on occasion on the same document sheet as CRC does not appear to be relevant to the process ITA set out for verification. See LG Chem’s Resp. at 37. Similarly, for the sales verification, the agency planned to “[r]eview the product matching criteria listed in the Appendix to the questionnaire,” i.e., the original model match criteria. Sales Verification Report (Sept. 1, 2022) at 10 (C.R. 257, P.R. 173). LG Chem officials informed ITA officials that “the values of Court No. 23-00010 PUBLIC VERSION Page 40 centrifuge retention capacity (CRC) are based on the guaranteed values of the certificate of analysis.” ITA “compared that to the information reported in the [home market] sales database.” Id. Again, ITA officials verified the CRC characteristics, but not with the new characteristics or CONNUM2s. The court therefore doubts defendant’s and LG Chem’s claims that the agency conducted an adequate verification. For example, the defendant never asserts that ITA actually verified those product characteristics or new CONNUMs, instead asserting that “it verified LG Chem’s cost accounting data, the basis of the alternative databases that reflect its proposed product characteristics.” Def’s Resp. at 25. But that is not equivalent to the verification program that ITA apparently set out to accomplish, which was to audit the product characteristics and CONNUMs used in the margin program. The defendant states that ITA “reviewed the certificate of analysis for each selected sale,” but it only did so with respect to CRC (in the original ranges). at 26. Id. ITA was clear about what it checked –- CRC –- and was silent about what it did not check –- AUP, permeability, and CRC1 in narrow ranges. Court No. 23-00010 PUBLIC VERSION Page 41 The agency does not appear to have verified –- even on a spot-check basis -- the new product characteristics or new CONNUMs used to calculate LG Chem’s margin. And ITA verified reported costs only for CONNUMs [[ ]] and [[ ]], which were not used for the final margin analysis. The first order of business for its cost verification was to verify the cost buildups and allocations of CONNUMs, but it did not do this for any particular CONNUM2. Remand on the basis of issue B, above, may moot further discussion of this issue C; but, of course, on remand ITA has “latitude” to provide further explanation on the foregoing, at its discretion. D Concerning the Coalition’s last issue, they argued to ITA that the CRC1, PERM, and AUP characteristics as defined by LG Chem were distortive and unusable because the same SAP product could be classified into multiple categories at LG Chem’s discretion based on its chosen testing protocol, creating a significant risk of manipulation. That is an obvious problem, the Coalition contend, that ITA ignored without addressing or attempting to ameliorate it. In the investigation, the Coalition argued to the agency that the new model match hierarchy promoted by LG Chem permits it Court No. 23-00010 PUBLIC VERSION Page 42 to categorize identical products in more than one CONNUM, allowing manipulation, and reducing the accuracy of the dumping margin. Specifically, they asserted that LG Chem’s model match hierarchy allows it to choose among several codes to report a given product characteristic depending on how LG Chem decided to test for the relevant characteristic. This is the case, they contend, even though the type of testing does not affect or change the underlying physical characteristic. The Coalition thus argued that LG Chem could thus report a “unique” physical characteristic of a product in more than one way, based simply on the selected testing methodology, which in turn meant that products with identical physical characteristics could have different CONNUMs – or a single product could be classified in more than one CONNUM – based on the testing methodology for such characteristics. See Pl’s Br. at 39-40. This approach allows for manipulation, they argued, and can reduce the accuracy of the dumping margin. Coalition Rebuttal Model Match Comments at 5-6 (P.R. 49); Coalition Response to LGC’s Request for Reconsideration (Jan. 31, 2022) at 3 (P.R. 94); Coalition Rebuttal Brief at 6-7 (P.R. 178). The Coalition contend that this argument has been the core of their objection to LG Chem’s preferred model-match hierarchy, and is directly relevant to whether the antidumping margin is accurate. Pl’s Reply at 18. Court No. 23-00010 PUBLIC VERSION Page 43 The court concurs that ITA did not fully address this issue and therefore remands for such consideration. IV ITA did not engage with the Coalition’s argument in its Final Determination. It did not discuss the possibility of LG Chem’s testing changing characteristic the and its effect procedure on for accuracy. a given The physical Coalition’s argument, however, is fundamental, and the potential validity of it would undermine the effectiveness of the antidumping-duty order. Under the statute, ITA must “address[] relevant arguments, made by interested parties who are parties to the investigation or review.” See 19 U.S.C. §1677f(i)(3)(A). It “must address any arguments made by the parties that are material to [ITA]’s determination.” Suzano S.A. v. United States, 46 CIT ___, ___, 589 F.Supp.3d 1225, 1233 (2022); Bonney Forge Corp. v. United States, 46 CIT ___, ___, 560 F.Supp.3d 1303, 1312 (2022)(“a failure to address an essential argument in making a final decision is sufficient grounds for remand”); see also Hung Vuong Corp. v. United States, 44 CIT ___, ___, 483 F.Supp.3d 1321, 1367 (2020). The defendant claims that ITA did, in fact, address the Coalition’s concerns about manipulation. It claims that the agency Court No. 23-00010 PUBLIC VERSION Page 44 “discerned” in its Final Determination evidence with which it engaged and manipulation dismissed and the inaccuracy Coalition’s resulting argument from the framework introduced in the Final Determination. 23-24. regarding model match Def’s Resp. at Conceding that ITA did not mention “manipulation” or “distortion” in its Final Determination, the defendant argues this is immaterial to the court’s analysis. Although important than its the semantics substance, defendant’s ultimate assertion. of the Id. at 23. a IDM determination does not are less support the Neither the substance nor the language of the Final Determination addressed the Coalition’s argument. ITA’s analysis in the Final Determination only focused on whether LG Chem’s proposed model match reflected commercially significant differences; not on the possibility for manipulation, distortion, misuse, “gaming” the system, or a myriad of other terms that could describe this problem, as raised by the Coalition. The defendant also points to ITA’s analysis on the commercial significance of LG Chem’s preferred characteristics in a separate discussion, but this appears to be a post hoc attempt to find an implicit basis for any analysis purportedly relevant to the potential for manipulation. Id. at 23-24. The defendant Court No. 23-00010 PUBLIC VERSION Page 45 specifically claims that ITA made two points, regarding “customer preference” as part of its commercial significance analysis, that constitute consideration and reasoned analysis of the Coalition’s manipulation argument, namely (1) that “the home market sales database suggests that LG Chem defines its SAP grades using both AUP and permeability ‘and that customers purchase SAP with expectations related to these characteristics’”; and (2) “achieving different guaranteed levels of AUP and permeability, completed through relevant testing, is commercially significant to downstream customers.” Id. According to the defendant, these points show that there was a “reasonably ‘discernible path’ to conclude that ITA considered and addressed potential concerns of distortion or manipulation.” Coalition’s Id. at 24. core concern But these points do not address the at all, which is the ability of respondents to manipulate the applicable CONNUM by using different tests for a given physical characteristic. The considerations highlighted by ITA address customers’ expectations for AUP and permeability, and say nothing about LG Chem’s ability to use a variety of testing options to measure such physical characteristics. In any rationalizations event, offered the by court cannot government accept counsel. post See, hoc e.g., Court No. 23-00010 PUBLIC VERSION Page 46 Coalition for Fair Trade in Garlic v. United States, 44 CIT ___, ___, 437 F.Supp.3d 1347, 1356 (2020) ("[t]he court cannot sustain an agency determination based on findings the agency itself did not make")(citing Burlington Truck Lines, Inc. v. United States, 371 U.S. 156, 168–69 (1962)). Furthermore, defendant’s contention that “customer preferences” immunize LG Chem from manipulating the CONNUMs, i.e., because LG Chem “conducts tests to satisfy customers’ expectations and preferences, [such that] it would be difficult to manipulate testing”, Def’s Resp. at 23, cuts directly against the procedural posture of certain domestic interests: LG Chem’s customers actively opposed trade relief in the investigation below and have a very strong interest in obtaining SAP that is not subject to a cash deposit rate that captures the full margin of dumping. See, e.g., USITC Pub. 5388 at 3, B-4 (showing that customer Procter & Gamble appeared alongside LG Chem before the ITC “In Opposition” to the order and that Procter & Gamble Kimberly-Clark Corporation appeared as “respondent entities”). and In other words, LG Chem’s customers’ interests are aligned with its interests, which does not mitigate but potentially increases the risk of manipulation. In what may seem a tad ironic here, the court, sub silencio, has considered LG Chem’s arguments on this issue and Court No. 23-00010 PUBLIC VERSION Page 47 finds that they do not merit rejecting the Coalition’s plea for remand, as they argue in their reply brief. Cf. LG Chem Resp. at 40-43 with Pl’s Reply at 22-23. V In view of the foregoing, plaintiff’s motion for judgment on the agency record must be granted17, and the matter is hereby remanded to the ITA for further proceedings consistent with this opinion. Results thereof to be filed on or before May 31, 2024 and a joint proposal for scheduling comments to be filed on or before June 14, 2024. So ordered. Decided: New York, New York March 1, 2024 /s/ 17 Thomas J. Aquilino, Jr. Senior Judge The quality of the written submissions on all sides has obviated the need for oral argument, see ECF No. 33, and the order of January 9, 2024 therefor can be, and it hereby is, rescinded.

Some case metadata and case summaries were written with the help of AI, which can produce inaccuracies. You should read the full case before relying on it for legal research purposes.

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.