s 
V-11–V-14, V-33; [[




                          ]].

       The Commission also excluded data from two [[              ]] importers, [[

                                                        ]] on bases that Plaintiffs claim were

unreasonable. See Pls.’ Br. at 26–28. The data of one importer, [[                   ]] were excluded

based in part on the Commission’s conclusion that the importer “imports products and transfers to

affiliates” rather than “engag[ing] in arm’s length transactions.” Views at 34 n.136. That

importer’s questionnaire response states otherwise, namely, that the importer [[



                                                                                     ]]; FSR at V-19

n.24. Though the Commission staff report treated [[                  ]] as a direct importer (rather

than indirect), the Commission ultimately excluded the company [[                             ]] data

altogether. See Views at 34 n.135. The Commission also relied upon the fact that this importer,

[[             ]], was unable to identify exactly which product type it imported. See
 id. at 36
n.141, C.R. 342, P.R. 209; [[                                        ]]. However, the Commission

did include data in the pricing information from another company, [[                        ]], which
Court No. 18-00238                                                                           Page 9
PUBLIC VERSION
was also unable to identify exactly which product it imports and had data showing overselling [[

                ]]. See Views at 36 n.141; FSR at Tables V-11–V-14; [[



]].

       Importer [[              ]] submitted data at the preliminary stage of the investigation, but

not the final stage. See FSR at I-6 n.10. Petitioners urged the Commission to consider the

preliminary stage data of this importer, [[            ]], in its final analysis, but the Commission

did not address this in its Final Determination. See generally Views (omitting any mention of [[

             ]]). The Commission’s Final Staff Report states the pricing data were excluded

because they were [[             ]] and because there “may” have been errors. FSR at V-9 n.18.

Pricing data from a different importer, [[        ]], were included, and Plaintiffs allege that they

show similar errors, particularly [[          ]]. See Importer Questionnaire of [[



               ]].

                       2.    The Commission’s Domestic Supply Shortages Finding

       The Commission relied on several sources in finding that domestic supply shortages

partially explained subject imports’ market gains.      First, fourteen of eighteen responses of

purchasers that switched from domestic to foreign PET resin stated that a lower price was not the

primary reason for purchase of subject imports. See Views at 47; FSR at Table V-12. Purchasers,

responding to an inquiry regarding the entire POI, also reported delayed deliveries, allocations

imposed by domestic producers, and disruptive weather as factors in the domestic supply shortage.

See Views at 47. And domestic producers’ data showed that in 2016 they increased imports from
Court No. 18-00238                                                                            Page 10
PUBLIC VERSION
affiliated producers in subject and non-subject countries by [[       ]] percent. See
 
id. at 46
; 
FSR

at Table IV-3.

       The Commission does not cite to specific record evidence that shows supply constraints

prior to October 2016. None of the twenty-one narrative responses of U.S. purchasers and

importers reference any specific pre-October 2016 causes or instances of supply shortages. See

FSR at Table II-4. [[       ]] reference supply shortages without identifying any specific time they

occurred. See
 id. By contrast: 
[[       ]] limit their identification of supply constraints to the last

year, last six to nine months, or “currently”; an additional [[    ]] reference the M&G shutdown

in September 2017; 7 and an additional [[      ]] reference “hurricanes”, “floods,” “force majeure,”

and “adverse weather.”
 Id. Thus, [[ 
             ]] importers and purchasers effectively state that

supply constraints were limited to October 2016 onward (and [[         ]] of those to September 2017

onward).
 Id.

       The Commission 
found that, despite Petitioners’ argument that supply constraints were

limited to a “brief period at the end of the POI,” the evidence of supply constraints mentioned

above -- in conjunction with the Commission’s finding that overselling predominated -- “all

indicate that low-priced subject imports were not the basis for the market share changes . . . [and]

that subject imports have not had a significant adverse impact.” Views at 47–48.




7
  Shortly prior to Plaintiffs’ filing of their petition with the Commission in September 2016, M&G
Polymers (which accounted for under 10 percent of the domestic industry) announced it would be
ceasing production. See Conf. Final Staff Report of the Commission in Polyethylene
Terephthalate Resin from Brazil, Indonesia, Korea, Pakistan, and Taiwan: Investigation. Nos. 731-
TA-1387-1391 (Final) at III-7 (“FSR”). In October 2017, M&G announced it would be filing for
bankruptcy and stated its financial difficulties were caused by competition from subject imports
as well as cost overruns at the new facility M&G had been building in Corpus Christie, Texas. See

id. at III-7; 
Tr. of ITC Staff Conference at 33–35, Oct. 17, 2017, P.R. 81.
Court No. 18-00238                                                                          Page 11
PUBLIC VERSION
               B.    The Commission’s Finding of No Significant Adverse Impact on Domestic
                     Industry Under 19 U.S.C. § 1677(7)(c)(iii).

       The Commission ultimately concluded that subject imports “have not had a significant

adverse impact on the domestic industry.” See
 id. at 48. 
In reaching this conclusion, the

Commission emphasized several factors, including that “the domestic industry’s financial

condition improved in 2016, the year with the most significant increase in subject imports.”
 
Id. at

45. 
The Commission also stated that its finding that subject imports did not have “any significant

adverse price effects on the domestic industry” led to its conclusion that subject imports do not

“explain[] the decline in the industry’s financial performance during the POI.”
 Id. Likewise, the
Commission noted that “subject imports largely oversold the domestic product . . . so any market

share shifts . . . were due to factors other than price.”
 Id. at 46. 
The Commission further explained

that even if “supply constraints were limited to a brief period at the end of the POI,” several other

considerations (i.e. U.S. producers’ increase in imports, purchasers’ narrative responses, and

subject imports gaining market share despite generally overselling) indicate that subject imports

being lower-priced was not the cause of market share changes following Certain Polyethylene

Terephthalate from Canada, the People's Republic of China, India, and the Sultanate of Oman:

Amended Final Affirmative Antidumping Determination (Sultanate of Oman) and Antidumping

Duty Orders, 81 Fed. Reg. 27,979 (Dep’t Commerce May 6, 2016).
 Id. at 46–47.
               C.    Procedural History of this Case

       On December 26, 2018, Plaintiffs filed a complaint with the court to contest the Final

Determination. See Compl. on Behalf of DAK Americas LLC, Indorama Ventures USA, Inc., and

Nan Ya Plastics Corp. America Against United States, Dec. 26, 2018, ECF No. 9. On May 16,

2019, Plaintiffs moved for judgment on the agency record, seeking review of the above

determinations and not contesting other Commission findings. Pls.’ Mot. for J. on Agency R.,
Court No. 18-00238                                                                         Page 12
PUBLIC VERSION
May 16, 2019, ECF No. 42; Pls.’ Mem. of L. in Supp. of Mot. for J. on Agency R., May 16, 2019,

ECF No. 43 (“Pls.’ Br.”); Def.’s Resp. in Opp’n to Pls.’ Mot. for J. on the Agency R. at 7, Aug.

16, 2019, ECF No. 53 (“Def.’s Br.”) (identifying eight uncontested aspects of the Commission’s

determination). 8 The Commission filed a brief in opposition to Plaintiffs’ motion on August 16,

2019. See Def.’s Br. Novatex Limited, G-Pac Corporation, iResin, LLC, and Niagara Bottling,

LLC (collectively, “Defendant-Intervenors”) all participated in the Commission’s investigation

and also filed briefs in opposition to Plaintiffs’ motion for judgment on the agency record. See

Def.-Inter. iResin, LLC’s Resp. to Pls.’ Mot. for J. on the Agency R., Aug. 16, 2019, ECF No. 55

(“iResin’s Br.”); Def.-Inters. Novatex Limited and G-Pac Corporation’s Resp. to Pls.’ Mot. for J.

on the Agency R., Aug. 16, 2019, ECF No. 52 (“Novatex’s Br.”); Def.-Inter. Niagara Bottling,

LLC’s Resp. to Pls.’ Mot. for J. on the Agency R., Aug. 16, 2019, ECF No. 50 (“Niagara’s Br.”).

Plaintiffs filed a reply brief on September 20, 2019. See Pls.’ Reply Br., ECF No. 63 (“Pls.’

Reply”). All parties filed responses to the court’s oral argument questions on March 30, 2020.

See Def.-Inter. Niagara Bottling, LLC’s Conf. Resp. to Oral Arg. Questions, ECF No. 95; Def.’s

Conf. Resp. to Oral Arg. Questions, ECF No. 93 (“Def.’s Conf. Resp. to Questions”); Def.-Inter.

iResin, LLC’s Conf. Resp. to Oral Arg. Questions, ECF No. 91 (“iResin’s Conf. Resp. to

Questions”); Pls.’ Conf. Resp. to Oral Arg. Questions, ECF No. 89; Def.-Inters. Novatex Ltd. &

G-Pac Corp.’s Conf. Resp. to Oral Arg. Questions, ECF No. 87. The court held oral argument via

teleconference on April 1, 2020. ECF No. 97.




8
 Many citations are to confidential filings for clarity in explaining the timeline of events. Public
versions, often filed at later dates, are available on the public docket with corresponding
pagination.
Court No. 18-00238                                                                           Page 13
PUBLIC VERSION
                      JURISDICTION AND STANDARD OF REVIEW

       The Tariff Act of 1930 requires the court to review the Commission’s AD and CVD

determinations for whether they are “unsupported by substantial evidence on the record, or

otherwise not in accordance with law.” 19 U.S.C. § 1516a(b)(1)(B)(i). Determinations are

presumed correct; it is the burden of the party challenging the determination to demonstrate

otherwise. 28 U.S.C. § 2639(a)(1) (2018).

       Substantial evidence requires “more than a mere scintilla”; it must be “such relevant

evidence as a reasonable mind might accept as adequate to support a conclusion.” Suramerica de

Aleaciones Laminadas, C.A. v. United States, 
44 F.3d 978
, 985 (Fed. Cir. 1994) (quoting

Consolidated Edison Co. v. NLRB, 
305 U.S. 197
, 229 (1938)). The record as a whole must be

examined, including evidence that “fairly detracts from the substantiality of the evidence.” Nippon

Steel Corp. v. United States, 
337 F.3d 1373
, 1379 (Fed. Cir. 2003) (quoting Atl. Sugar. Ltd. v.

United States, 
744 F.2d 1556
, 1562 (Fed. Cir. 1984)). An agency determination may not be

sustained without considering “contradictory evidence or evidence from which conflicting

inferences could be drawn.” Universal Camera Corp. v. NLRB, 
340 U.S. 474
, 487 (1951); cf. CS

Wind Vietnam Co. v. United States, 
832 F.3d 1367
, 1373 (Fed. Cir. 2016) (“The substantiality of

evidence must take into account whatever in the record fairly detracts from its weight.”).

       The Tariff Act requires that the Commission provide “an explanation of the basis for its

determination that addresses relevant arguments.” 19 U.S.C. § 1677f(i)(3)(B). In this context,

Motor Vehicle Mfrs. Ass’n v. State Farm requires the Commission to “articulate a satisfactory

explanation for its action including a ‘rational connection between the facts found and the choices

made.’” 
463 U.S. 29
, 50 (1983) (quoting Burlington Truck Lines, Inc. v. United States, 
371 U.S.

156
, 168 (1962)). An agency, though presumed to have considered all evidence on the record,
Court No. 18-00238                                                                        Page 14
PUBLIC VERSION
must respond to “significant arguments and evidence which seriously undermine its reasoning and

conclusions.” Altx. Inc. v. United States, 
25 CIT 1100
, 1117–18, 
167 F. Supp. 2d 1353
, 1374

(2001). The court must defer to the Commission’s role as trier of fact and not disturb its

“considerable discretion in evaluating information obtained from questionnaires.” Int’l Indus. v.

United States, 42 CIT __, __, 
311 F. Supp. 3d 1325
, 1336 (2018) (quoting NSK Corp. v. United

States, 
32 CIT 966
, 978, 
577 F. Supp. 2d 1322
, 1336–37 (2008)); see also Goss Graphics Sys. v.

United States, 
22 CIT 983
, 1004, 
33 F. Supp. 2d 1082
, 1100 (1998), aff’d, 
216 F.3d 1357 
(Fed.

Cir. 2000). The court’s role is limited to reviewing the agency’s decisions for reasonableness. See

U.S. Steel Grp. v. United States, 
96 F.3d 1352
, 1357 (Fed. Cir. 1996).

                                         DISCUSSION

       Plaintiffs argue that: (1) even taking the Commission’s indirect pricing data as is, the

Commission failed to explain why the underselling it showed was not significant in light of prior

Commission and court decisions finding mixed underselling significant; (2) the Commission’s

determination that subject imports predominantly oversold domestic product was not supported by

substantial evidence because the Commission selectively excluded several points of conflicting

evidence; (3) the Commission’s finding that domestic supply constraints explain the subject import

surge is not supported by substantial evidence because the record indicates supply constraints

solely manifested from October 2016 onwards; and (4) the aforementioned errors, whether

individually or collectively, make the Commission’s ultimate no adverse impact determination

flawed and unsupported by substantial evidence.

       The Government responds that: (1) the Commission was under no obligation to address

prior determinations finding mixed underselling was significant because each investigation is sui

generis; (2) the overselling determination was reasonable, good reason supported the
Court No. 18-00238                                                                       Page 15
PUBLIC VERSION
Commission’s exclusion of certain data, and the court must respect the significant discretion the

Commission holds in assessing questionnaire responses and weighing evidence; (3) the

Commission’s finding that domestic supply constraints explain in part the subject import surge is

supported by substantial evidence, as many questionnaire responses indicated price was not the

primary factor for domestic purchasers and many responses provided to inquiries regarding the

POI as a whole cited supply constraints as a factor in switching to imports; and (4) even setting

aside the domestic supply constraints and overselling findings, other considerations supported

finding that subject imports had no adverse impact on domestic industry.

       The court grants Plaintiffs’ motion for judgment on the agency record with respect to three

issues. First, the court determines that the Commission’s conclusion that there was no significant

underselling is not supported by substantial evidence because the Commission failed to address or

distinguish applicable findings of a general nature from prior determinations and the Commission

did not address conflicting evidence. Second, the court determines that the Commission’s finding

that supply constraints, rather than underselling, were the cause of market share shifts is

unsupported by substantial evidence because there was a lack of evidence of supply constraints

prior to October 2016 and the Commission failed to reasonably address this. Third, the court

determines that the Commission’s ultimate no adverse impact determination is unsupported by

substantial evidence because, as is clear from the Commission’s own language, the two other

findings remanded here were integral to the no adverse impact determination. The court also

instructs the Commission to address and distinguish on remand any applicable findings of a general

nature from prior determinations that conflict with its ultimate determination regarding

underselling and supply constraints.
Court No. 18-00238                                                                          Page 16
PUBLIC VERSION
         I.    The Commission’s Failure to Address Prior Determinations Finding “Mixed”
               Underselling Significant in Price-Sensitive Markets for Fungible Products Must
               be Addressed by the Commission on Remand.

       The court, in reviewing the Commission’s underselling finding, first addresses the issue of

findings from prior determinations. Plaintiffs argue that the Commission failed to explain why the

underselling that it found in this case was not significant in light of prior Commission

determinations finding that mixed underselling could be significant in highly price-sensitive

markets for substitutable products. Pls.’ Br. at 30–33. The Government concedes the Commission

did not address this aspect of prior determinations, characterizing Plaintiffs’ argument that the

Commission should even address it as “astonishing,” and reiterating that each determination is sui

generis. Def.’s Br. at 29. Plaintiffs answer that they are not challenging the Commission’s

methodology, but rather its failure to consider or address this argument at all. Pls.’ Reply at 6.

               A.    Case Law Supports Finding that the Commission Was Required To Address
                     or Distinguish Applicable Findings of a General Nature from Prior
                     Determinations.

       As the court stated in Usinor v. United States, “[a]lthough each sunset review must be

based on the particular set of facts before the Commission, the Commission may not disregard

previous findings of a general nature that bear directly upon the current review.” 
26 CIT 767
, 792

24 I.T.R.D. (BNA) 1711 
(2002) (emphasis added). Usinor, persuasive but unpublished and non-binding

authority for the court, thus holds that if there are distinguishing factors between the prior

determinations in question and the instant case, the burden is on the Commission to reasonably

address them.
 Id. (“Although the 
court is aware that the [European Union] market dynamics for

[the products at issue] may be vastly different from its dynamics in the cited determinations, the

Commission’s discussion fails to make any such distinctions . . . .”).

       In Cleo v. United States -- the primary authority that the Government cites in opposing any

obligation to address the prior determinations -- the Federal Circuit implicitly confirmed Usinor’s
Court No. 18-00238                                                                            Page 17
PUBLIC VERSION
approach, though it did not expressly adopt it. 
501 F.3d 1291
, 1298–99 (Fed. Cir. 2007). After

reiterating the sui generis nature of each case, the Federal Circuit discussed how the Commission

had addressed and distinguished the prior determinations. See
 id. (noting that 
each of the prior

determinations “involved significantly different facts,” and that the Commission expressly

distinguished the case before it -- as well as the industry more generally -- to justify departure from

otherwise similar contrary prior determinations). Such a discussion would be unnecessary if the

Government bore no duty to address or distinguish contrary prior determinations. See also Usec.

Inc. v. United States, 
25 CIT 49
, 64–65, 
132 F. Supp. 2d 1
, 14–15 (2001) (sustaining Commission’s

decision and rejecting plaintiff’s argument regarding inconsistency with a prior determination

because Commission had “offered several reasonable explanations for [its] decision to depart

from” the prior determination’s analysis), aff’d per curiam, 34 Fed. App’x 725 (Fed. Cir. 2002).

The recent Federal Circuit decision Hitachi Metals, Ltd. v. United States, cited by the Government

in support of the sui generis nature of Commission investigations, did not reach the issue of

findings of a general nature and thus does not influence the court’s analysis. See 
949 F.3d 710
,

718 (Fed. Cir. 2020); Def.’s Conf. Resp. to Questions at 2 & n.1.

       The Government next invokes Mexichem Fluor Inc. v. United States, 40 CIT __, __, 
179

F. Supp. 3d 1238 
(2016), for the principle that each case is sui generis. Def.’s Br. at 36 (arguing

that “facts underlying how certain information is used in one investigation do[] not automatically

transfer to another investigation”). In Mexichem, however, not in issue are Usinor’s discussion of

“previous findings of a general nature” or Cleo’s implicit recognition that the Commission should

distinguish prior determinations that could fairly undermine its 
conclusion. 179 F. Supp. 3d at

1255
. The sui generis nature of each case addressed in Mexichem does not conflict with the
Court No. 18-00238                                                                           Page 18
PUBLIC VERSION
requirement that the Commission address and distinguish “previous findings of a general nature.”

Cleo, moreover, engages more directly with this issue and is the precedent binding upon the court.

                 B.   Administrative Law Considerations Support Requiring the Commission To
                      Address or Distinguish Applicable Findings of a General Nature from Prior
                      Determinations.

         Scrutinizing whether an agency has departed from previously established “findings of a

general nature” without reasonable explanation is buttressed by several other considerations rooted

in principles of administrative law. First, agencies generally have a duty to “address significant

arguments and evidence which seriously undermine its reasoning and conclusions.” 
Usinor, 26

CIT at 784 
(quoting ALTX, Inc, 25 CIT at __, 
167 F. Supp. 2d 
at 1373)). The court sees no

compelling reason why it should not be a “significant argument[]”, that an agency has failed to

explain seeming departures from findings of a general nature. The second consideration buttresses

the first: consistency has long been a core interest of administrative law, and inconsistent treatment

is inherently significant.    Cf. Skidmore v. Swift, 
323 U.S. 134
, 140 (1944) (holding that

“consistency with earlier and later pronouncements” is one factor that weighs in favor of deference

to an agency’s decision); Chisholm v. Def. Logistics Agency, 
656 F.2d 42
, 47 (3d Cir. 1981)

(agencies “acting as quasi-judicial bodies . . . have an obligation to render consistent opinions and

to either follow, distinguish or overrule their own precedent”). 9 Thus, though past agency


9
    For example, Yoav Dotan provides a description of consistency’s significance:

         The requirement of consistency . . . -- that like cases be treated alike -- is
         fundamental both for bureaucratic decisionmaking and for legal systems at large.
         It has strong intuitive appeal to our sense of justice, and is intertwined with the
         notion of fairness. It is a due process value. It is fundamental to the notions of
         prompt administrative order, rationality in administrative decisionmaking, and
         impartiality in adjudicative proceedings. . . . Inconsistency in administrative
         decisionmaking (that is, where agencies fail to treat similar cases alike) defies the
         values of the rule of law. Such inconsistency may signal serious flaws in the
         administrative process and provide several grounds for judicial intervention to
         rectify such flaws.
Court No. 18-00238                                                                              Page 19
PUBLIC VERSION
decision-making may not be precedential in the same way as case law through stare decisis, it

remains of great importance. 10 Finally, nearly seventy years of Supreme Court and Federal Circuit

precedent recognize an agency’s duty to address departure from prior norms and policies. See

Atchison, Topeka & Santa Fe Ry. v. Wichita Bd., 
412 U.S. 800
, 808 (1973) (restating “the

agency’s duty to explain its departure from prior norms” (citing Sec’y of Agric. v. United States,

347 U.S. 645
, 653 (1954))); see also
 id. at 808–09 
(“Whatever the ground for departure from prior

norms . . . it must be clearly set forth so that the reviewing court may understand the basis of the

agency’s action and so may judge the consistency of that action with the agency's mandate . . .

.”). 11




Making Consistency Consistent, 57 Admin. L. Rev. 995, 1000 (2005) (citations omitted).
10
     Several leading authorities in administrative law concur:

          It is often said that, in contrast [to caselaw], stare decisis does not apply to agency
          adjudicatory decisions. Whether this is true in the abstract is a nice jurisprudential
          question but not an important practical one. First, even if agency precedents are not
          binding in theory, in practice they do carry weight, just like any previous decision
          by any decisionmaker . . . . Second, even if agency precedents are binding in theory,
          an agency can surely distinguish (perhaps disingenuously) or overrule a prior
          decision, just like a court can. Third, and most important, courts have consistently
          held that while an agency can reject a prior adjudicatory decision, it must
          acknowledge and explain its change.

Stephen G. Breyer, Richard B. Stewart, Cass R. Sunstein, Adrian Vermuele, & Michael Herz,
Administrative Law & Regulatory Policy: Problems, Text, and Cases 435 (8th ed. 2017).
11
  Although it dealt with a policy decision rather than prior quasi-judicial determination from an
independent agency, Motor Vehicle Manufacturers Association v. State Farm illustrates that a
similar duty to explain attaches to prior policies as it does to arguments more generally. 
463 U.S.
29
, 44 (1983) (finding that the National Highway Traffic Safety Administration had failed to
provide “clear and convincing reasons” for rescission of its prior policy). Similarly, the Federal
Circuit has consistently found agencies have a duty to explain departures from prior policy. See
British Steel PLC v. United States, 
127 F.3d 1471
, 1475 (Fed. Cir. 1997) (quoting M.M. & P.
Maritime Advancement, Training, Educ. & Safety Program v. Dep’t of Commerce, 
729 F.2d 748
,
755 (Fed. Cir. 1984)).
Court No. 18-00238                                                                         Page 20
PUBLIC VERSION
       The considerations elaborated above illustrate that such concerns are in fact multiplied in

a situation where an agency departs from applicable findings of a general nature from prior

determinations without reasonably distinguishing them or justifying the departure. Not only does

the court’s generalized concern with ensuring reasoned decision-making come to bear, but also

administrative law’s long-standing and constant concern for consistency.           Neither binding

precedent nor persuasive reasons support the Government’s contention that the court should not

consider whether the agency has failed to address applicable findings of a general nature. See

Def.’s Br. at 29 & n.11. Thus, where an agency departs from prior determinations, it is appropriate

to compel the agency to explain whether: (1) good reasons prompt that departure; or (2) the prior

determinations are inapposite such that it is not in fact a departure at all. Cf. 
Atcheson, 412 U.S.

at 808
; British Steel PLC v. United States, 
127 F.3d 1471
, 1475 (Fed. Cir. 1997); 
Chisholm, 656

F.2d at 47
.

               C.    The Finding Here Is General in Nature, and the Commission Failed To
                     Address It.

       The instant case is one where the finding is truly general in nature and relevant factual

determinations are on point. Specifically, Plaintiffs argue that the Commission failed to address

prior findings that mixed underselling is more significant in price-sensitive markets for

substitutable goods. Pls.’ Br. at 30–33. The Commission has articulated these findings since at

least 1982. See Certain Carbon Steel Products from Spain at 16–17, USITC Pub. 1331 (Final)

(1982) (noting that “[t]he impact of seemingly small import volumes . . . is magnified in the

marketplace” in price-sensitive industries producing fungible products).           The court has

consistently affirmed these findings, and the Commission has likewise reiterated them. See, e.g.,

Hynix Semiconductor, Inc. v. United States, 
30 CIT 1208
, 1213, 
431 F. Supp. 2d 1302
, 1308–09

(2006) (“The presence of an increase in absolute volume of subsidized imports in a market
Court No. 18-00238                                                                         Page 21
PUBLIC VERSION
characterized by product fungibility is significant because such evidence tends to prove that

purchasers were acquiring subject imports in lieu of domestically produced DRAMS.”); USX

Corp. v. United States, 
11 CIT 82
, 85, 
655 F. Supp. 487
, 490 (1987) (same); Certain Hot-Rolled

Steel Flat Products from Austria, et al. at 35–36, USITC Pub. 4638 (Final) (2016) (finding

underselling in 49.5 percent of quarterly comparisons significant); Certain Nails From China at

18–19, USITC Pub. 4022 (Final) (2008) (finding underselling in circa 48 percent of quarterly

comparisons significant); Stainless Steel Plate from Belgium et al. at 18, USITC Pub. 3188 (Final)

(1999) (finding underselling in 41.5 percent of comparisons was significant).

       The court is not persuaded by the Government’s argument that such a finding is narrowly

delimited by its factual circumstances and the product at issue. See Def.’s Conf. Resp. to Questions

at 6. Rather, the court finds that it constitutes a general conclusion regarding how two key

characteristics of a market -- analysis of which the Commission expressly stated to be part of its

general practice -- affect the significance of underselling. See Views at 6 n.15, 16–17 (identifying

price and interchangeability as factors the Commission generally considers (citing 
Nippon, 19 CIT

at 455 
n.4)). No aspect of this finding from prior determinations cabins its significance to

investigations regarding the same product. Thus, where the Commission made the same two

determinations (i.e. price-sensitivity and substitutability), upon which this finding of a general

nature is predicated, the Commission must address why the underselling that it found was not

significant. See 
Cleo, 501 F.3d at 1298
–99; 
Usinor, 26 CIT at 792
.

       The court does note that distinguishing factors from previous “mixed” underselling cases

could perhaps be found in the Commission’s Final Determination. The asserted supply constraints

may distinguish the instant case from those above. That the portion of underselling found by the

Commission (circa one-third) is lower than other cases of “mixed” underselling (which fell in the
Court No. 18-00238                                                                              Page 22
PUBLIC VERSION
40 to 49 percent range as discussed above) might adequately distinguish the case before the court.

But it is not the court’s role to attempt to discern in the first instance what distinguishes the factors

of this case nor to accept post hoc rationalizations of the agency’s analysis. See Burlington Truck

Lines, 371 U.S. at 168
–69 (reiterating that “courts may not accept . . . counsel’s post hoc

rationalizations for agency action”). Rather, as Usinor states and Cleo indicates, the burden lies

with the Commission to address arguments based on findings of a general nature from prior

determinations that fairly undermine its present conclusion where applicable factual

determinations were the same. 
Usinor, 26 CIT at 792
; 
Cleo, 501 F.3d at 1298
–99.

        The Commission’s failure to explain why the underselling it found was not significant in

light of these prior determinations warrants remand. As such, the court orders the Commission in

considering its underselling finding on remand to reasonably distinguish its conclusions from

findings of a general nature in prior determinations in the event of any conflict.

        II.     The Commission Unreasonably Relied on Flawed Indirect Pricing Report Data
                to the Exclusion of Other Conflicting Evidence in its Underselling Analysis.

        The court further finds that the underselling determination is unsupported by substantial

evidence because of unaddressed conflicting evidence. Plaintiffs argue that (1) the Commission’s

quarterly indirect importer pricing analysis either included or excluded several companies’ data,

without adequate explanation from the Commission; and (2) that several other aspects of the record

undermine the Commission’s conclusion and were not adequately addressed. Pls.’ Br. at 15–30.

The Government generally responds that the Commission conducted its analysis in a reasonable

way that was consistent with its long-standing practice and did not ignore or fail to address these

aspects of the record. Def.’s Br. at 14–18, 31–37.

        An agency must “examine the relevant data and articulate a satisfactory explanation for its

action.” State 
Farm, 463 U.S. at 43
. Per Altx, Inc., the Commission “must address significant
Court No. 18-00238                                                                            Page 23
PUBLIC VERSION
arguments and evidence which seriously undermine its reasoning and conclusions.” 
167 F. Supp.

2d 
at 1374. The court finds that the Commission has not done so here. Several aspects of the

record demonstrate that there are significant arguments and evidence that undermine the

Commission’s reasoning and conclusions. The court does not minimize the substantiality of the

evidence from indirect purchase reports upon which the Commission primarily relied; if no other

evidence existed, then the price data upon which the Commission relied would be substantial

evidence. However, “[t]he substantiality of evidence must take into account whatever in the record

fairly detracts from its weight.” CS Wind Vietnam 
Co., 832 F.3d at 1373
. Here, there is a

significant amount of “contradictory evidence or evidence from which conflicting inferences could

be drawn” that the Commission did not adequately address. 
Suramerica, 44 F.3d at 985 
(quoting

Universal Camera 
Corp, 340 U.S. at 487
).

               A.    Some of the Asserted Errors in Indirect Pricing Data Undermine the
                     Commission’s Determination.

       Plaintiffs argue that Commission wrongly excluded the data of three companies, [[

                                  ]], and wrongly included the data of two companies, [[

                            ]], from its quarterly indirect importer pricing analysis. See Pls.’ Br. at

25–30. The Government argues that the Commission’s decisions were reasonable, non-arbitrary,

and within the Commission’s considerable discretion, and that Plaintiffs failed to administratively

exhaust some of their claims. See Def.’s Br. at 31–37.

                       1.     Exclusion of [[                   ]] Data and Inclusion of [[
                                        ]] Data

       With regard to exclusion of importer [[                   ]] data, the Government argues that

the decision falls within the agency’s broad discretion regarding information from questionnaires

particularly because “the weight to be assigned a particular piece of evidence, lies at the core of

that evaluative process.” Int’l 
Indus., 311 F. Supp. 3d at 1333 
(quoting U.S. Steel Grp., 96 F.3d at
Court No. 18-00238                                                                              Page 24
PUBLIC VERSION
1357); Def.’s Br. at 12. International Industries involved a plaintiff attempting to cast doubt upon

the Commission’s reliance on one purchaser’s data -- the converse of the instant situation, but

nonetheless 
applicable. 311 F. Supp. 3d at 1333
. There is a difference, however, between

contesting the relative weight afforded a piece of evidence and contesting its complete and utter

exclusion from the evidence on which the Commission predominantly relied. The court finds that

the latter demands a somewhat stronger rationale, particularly where the excluded data -- [[

                                                                     ]] -- would have represented a

substantial portion of comparisons. Indeed, the latter borders upon failure to address conflicting

evidence. Thus, the court finds that the Commission failed to explain exclusion of the data from

this importer, [[              ]], and orders the Commission to address this on remand.

        Plaintiffs further argue that the data of [[

                                      ]] was included despite a similar question with the data of the

other importer, [[              ]]. Pls.’ Br. at 27. Defendants argue that Plaintiffs failed to raise

this issue below, and thus it is not administratively exhausted. See Def.’s Br. at 34–35. Plaintiffs

counter that they in fact identified the error in their Post-Hearing Brief. See Pls.’ Reply at 13 n.6;

Pet’rs’ Post-Hr’g Br. at Ex. 7, bullet 3, Sept. 21, 2018, C.R. 311, P.R. 188 ( [[

                                             ]] ). Because Plaintiffs raised the issue at an appropriate

time and the Commission had the chance to consider it, the court agrees, and thus finds that

administrative exhaustion does not bar Plaintiff’s challenge to inclusion of this importer [[

           ]] data. See Mittal Steel Point Lisas Ltd. v. United States, 
548 F.3d 1375
, 1383–84

(Fed. Cir. 2008) (“[A]s a general rule . . . courts should not topple over administrative decisions

unless the administrative body not only has erred but has erred against objection made at the time

appropriate under its practice.” (quoting United States v. L.A. Tucker Truck Lines, Inc., 344 U.S.
Court No. 18-00238                                                                            Page 25
PUBLIC VERSION
33, 37 (1952))); Palladian Partners, Inc. v. United States, 
783 F.3d 1243 
(Fed. Cir. 2015) (same).

Even if administrative exhaustion did bar claims that this importer [[                         ]] data

should not have been included, administrative exhaustion would not bar use of this similar

inconsistency in [[                       ]] included data to question whether the exclusion of the

other importer [[                    ]] data was supported by substantial evidence, given that the

Government does not dispute that Plaintiffs exhausted administrative remedies with respect to

exclusion of that importer [[                 ]] data. See Def.’s Br. at 33–34. As such, if on remand

the Commission again decides to exclude the data [[                          ]], the court orders the

Commission to address this argument and explain its reasoning.

                       2.       Exclusion of the Pricing Data of Importer [[                         ]]

       [[

                                                      ]] FSR at V-9 n.18. The Commission did not

incorporate the pricing data submitted from the preliminary stage into the final phase’s price

comparisons. The Government acknowledges that the Commission could have done so as

Plaintiffs had requested but states it was not required to do so. See Pls.’ Reply at 15; Pet’rs’

Posthr’g Br. at 7 (arguing for inclusion of [[                     ]] data); Def.’s Br. at 36. The

Government contends that the sui generis nature of each investigation places such a decision firmly

within the Commission’s discretion, while Plaintiffs contend that it is “still incumbent upon the

[C]ommission to explain why it is not using data that, it admits, it often relies upon in other cases.”

Def.’s Br. at 36; Pls.’ Reply at 15. The Government cites to some portions of the Final Staff Report

-- but nothing in the Commission’s Views -- to show that the Commission possibly thought [[

               ]] data contained errors. See Def.’s Br. at 36 (citing FSR at V-9 n.18).
Court No. 18-00238                                                                          Page 26
PUBLIC VERSION
       The court finds that the Commission’s failure to address, in any way, in its Views the

exclusion of data from a significant importer is insufficient to constitute substantial evidence on

the record. Though the Commission is presumed to have considered the full record and “does not

need to address every piece of evidence presented by the parties,” the significant nature of the data

excluded here demonstrates that the Commission failed to adequately address an important aspect

of the issue. Siemens Energy, Inc. v. United States, 38 CIT ___, ___, 
992 F. Supp. 2d 1315
, 1324

(2014), aff’d, 
806 F.3d 1367 
(Fed. Cir. 2015); cf. SKF USA, Inc. v. United States, 
263 F.3d 1369
,

1382–83 (Fed. Cir. 2001) (stating that the AD statute is “highly complex” and “[t]he more complex

the statute, the greater the obligation on the agency to explain its position with clarity.”). Thus,

the court orders the Commission to address exclusion of [[                        ]] data on remand.

       The court notes that Plaintiffs also contend that the Commission “effectively precluded

itself from obtaining the data by not asking any questions until [very late] in the case and by not

using data that it only found ‘possibly’ to contain errors.” Pls.’ Reply at 15–16. The record

supports this contention: the Commission did not contact [[                ]] regarding any potential

errors until over eleven months after it submitted its data at the preliminary stage and only twenty-

four days before the record closed. See [[

                                      ]]; Email from Amelia Preece to [[                ]], Sept. 17,

2018, C.R. 332. Case law, however, does not provide a clear answer as to whether the Commission

failed to meet its obligation in the instant situation, where the Commission technically did seek

additional information, but belatedly and on a very short time-frame. 12 The court, having found




12
   Case law holds that it is contrary to law “where the ITC actively precludes itself from receiving
relevant data or takes no effort to seek relevant data.” Mitsubishi Elec. Corp. v. United States, 
12
CIT 1025
, 1058, 
700 F. Supp. 538
, 564 (1988). The Commission, moreover, “is obligated to make
active, reasonable efforts to obtain relevant data.” Allegheny Ludlum Corp. v. United States, 287
Court No. 18-00238                                                                             Page 27
PUBLIC VERSION
that the Commission’s failure to address the exclusion of [[                      ]] data in its Views

warrants remand, need not fill this lacuna in the case law as to the reasonability of the

Commission’s timeline.

                       3.    Inclusion of [[                           ]] Data

       Plaintiffs acknowledge that they did not raise claims regarding the inclusion of [[

]] data to the Commission below and focus instead on the inclusion of [[                  ]] data, as it

was a much larger importer. See Pls.’ Reply at 13 n.7 (“Plaintiffs focused their argument on [[

        ]].”). Thus, the court will solely address [[             ]] to the extent it is relevant to the

Commission’s inclusion of [[                 ]] data. Regarding [[               ]], the Government’s

argument that the question regarding reasons for purchasing subject imports solely reflected the

firm’s “general impressions” is unpersuasive, and, regardless, constitutes a post hoc rationalization

that the court cannot accept as support for the Commission’s finding. Def.’s Br. at 32; see

Burlington Truck 
Lines, 371 U.S. at 168
–69 (“The courts may not accept . . . counsel’s post hoc

rationalizations for agency action; Chenery requires that an agency's discretionary order be upheld,

if at all, on the same basis articulated in the order by the agency itself.” (citing SEC v. Chenery

Corp., 
332 U.S. 194
, 196 (1947))).          While the Government contends that the purchaser

questionnaire merely solicited the firm’s “general impressions”, the straight-forward text of the

question (i.e. “was price a primary reason for purchasing subject imports rather than domestic

product”) and prefatory language to the questionnaire 13 frustrate this claim. Cf. Def.’s Br. at 
28



F.3d 1365
, 1373 (Fed. Cir. 2002); see also Bethlehem Steel Corp. v. United States, 
27 CIT 1662
,
1673, 
294 F. Supp. 2d 1359
, 1370 (2003) (same).
13
  Specifically, the questionnaire requires an authorized official to certify as follows: “I certify that
the information herein supplied in response to this questionnaire is complete and correct to the best
of my knowledge and belief and understand that the information submitted is subject to audit and
verification by the Commission.” U.S. Purchasers’ Questionnaire at 1, June 7, 2018, P.R. 103.
Court No. 18-00238                                                                                Page 28
PUBLIC VERSION
(describing the same question as “direct and unambiguous” in an earlier section of its brief). As

the Commission does not explain this discrepancy itself, the Government’s argument here is a post

hoc rationalization that the court cannot consider in reviewing agency action. See Burlington

Truck 
Lines, 371 U.S. at 168
–69; Timken Co. v. United States, 
20 CIT 1115
, 1119, 
937 F. Supp.

953
, 955 (1996) (“[P]ost hoc rationalization . . . is insufficient.”).

          Despite Plaintiffs’ repeated requests for the Commission to attempt to reconcile the

inconsistencies between the U.S. purchaser’s [[              ]] questionnaire response and its pricing

data, there may be a reasonable explanation for the Commission’s determination not to have done

so. See Pet’rs’ Posthr’g Br. at 8–9 & Ex. 7 (raising concerns regarding this U.S. purchaser’s data).

The Government relies upon the significant deference courts grant to the Commission in making

such findings. Def.’s Conf. Resp. to Questions at 14 n.42 (citing U.S. Steel Grp. v. United States,

96 F.3d 1352
, 1357 (Fed. Cir. 1996)). The general deference granted to agency determinations is

inapplicable here, however, because the Commission failed to offer any such explanation, and the

Government has offered only post hoc rationalizations.            As such, remand is necessary --

particularly given that [[                                               ]]. Cf. Pls.’ Reply at 12. 14

                 B.   Other Data and Questionnaire Responses Undermine the Commission’s
                      Conclusion.

          Beyond the asserted errors in the indirect purchasing data, a variety of other data and

questionnaire responses undermine the Commission’s conclusion that overselling predominated




14
     [[


                                                                                  ]]. See Pls.’ Reply Br.
at 12, Sept. 20, 2019, ECF No. 63 (“Pls.’ Reply”).
Court No. 18-00238                                                                           Page 29
PUBLIC VERSION
and any underselling was not significant. The Commission failed to adequately address some or

all of them.

                       1.    Purchaser Price Rankings from Questionnaires

       Reports from sixteen purchasers showed prices of imports were lower than U.S. producer

prices. See Views at 33, 38 n.150; FSR at V-9. Asked to compare U.S.-produced PET resin with

that of other subject countries, the significant majority of purchasers ranked imports as lower- or

comparably-priced. See FSR at Table II-11. For all subject countries except Pakistan, fifty percent

or more of responding purchasers reported U.S. PET resin was higher-priced. 15 Aggregating

responses for each subject country, there were twenty purchaser responses of underselling and only

nine of overselling. See
 id. This fairly 
undermines the Commission’s conclusion that overselling

predominated.

       The Commission did note this evidence:

       [T]he number of responding purchasers for each country comparison was relatively
       small (from 4 to 11) compared to the total number of responding purchasers (25).
       While we have considered this information, we do not believe it outweighs the
       actual price data we have collected that show that subject imports are typically
       higher priced.

See Views at 26 n.150 (citations omitted). Plaintiffs argue that this is inaccurate and inconsistent

with the Commission’s weighing of evidence elsewhere in its determination. Pls.’ Br. at 17.

Specifically, Plaintiffs note that there were in fact a total of sixteen purchasers that provided this

data, though not all purchasers provided data for every country.
 Id. at 17–18. 
They further argue




15
 The comparison table below provides a breakdown by country and a total tally.
                           Brazil     Indonesia Korea Pakistan Taiwan                      TOTAL
  U.S. lower-priced           2           1         2          2            2                 9
     Comparable               2           1         1          4            3                11
  US higher-priced            4           2         5          3            6                20
Adapted from FSR Table II-11.
Court No. 18-00238                                                                          Page 30
PUBLIC VERSION
that there is “no justification” for the claim that these numbers were so small as to warrant

discounting them given that the quarterly pricing data the Commission relied heavily upon was in

fact from a smaller total number. See
 id. at 18; 
FSR at V-9 (“Four U.S. producers and eight

importers provided usable pricing data . . . although not all firms reported pricing data for all

quarters.” (emphasis added)). On the whole, Plaintiffs’ arguments and the record evidence cast

significant doubt on the reasonability of the Commission’s decisions.

                       2.      Reported Prices from Purchasers Who Purchased Subject Imports
                               Instead of Domestic

       Eleven of nineteen purchasers shifting from buying the U.S. product reported the imports

were lower-priced. See FSR at V-34. The Government and Commission focus on the fact that

only five of the purchasers in fact reported that price was the primary reason for the shift. See
 id.

This evidence 
on which the Government relies is of greater relevance to other determinations (i.e.

the cause of market share shifts) than to the question of whether underselling occurred.

Nonetheless, most purchasers’ reported import prices were lower, thus constituting evidence that

fairly undermines the Commission’s finding that underselling was not significant and that

overselling predominated. The court finds the Commission failed to address this evidence and

explain why it was outweighed by other evidence.

                       3.    Quarterly Pricing Data for “Direct Imports” Showing Underselling
                             in [[ ]] Percent of Comparisons

       Pricing data solicited by the Commission from importers for “imports used for internal

consumption (direct imports)” showed underselling in [[       ]] percent of comparisons. FSR at V-

19, V-21–V-24. Plaintiffs assert this as an additional piece of evidence that undermines the

Commission’s conclusion. Pls.’ Br. at 22. The Government does not directly counter this, again

focusing on narrative responses from the importers stating non-price reasons for switching to

subject imports. See Def.’s Br. at 16 (citing FSR at V-19). The court finds that direct pricing data
Court No. 18-00238                                                                           Page 31
PUBLIC VERSION
comparisons constitute evidence that fairly undermines the Commission’s overselling finding --

though perhaps not related findings -- and that the Commission failed to address.

                        4.   Direct Quarterly Pricing Data from [[                       ]] Showing
                             Underselling

          The Commission excluded quarterly pricing data of [[                                       ]]

that showed underselling in [[                ]] of comparisons. See FSR at V-9. Even assuming

that [[               ]] data should not have been included in the direct or indirect analyses, [[

            ]] data still constitute evidence that fairly undermines the Commission’s conclusion. [[



                              ]] See supra Part 2.a.i; [[

      ]]; FSR at V-11–V-14. By excluding [[                    ]] data entirely and not weighing the

data at all in its overselling analysis, the Commission failed to address conflicting evidence.

                        5.   Significant Market Share Gains in a Price-Sensitive Market
                             Characterized by Product Fungibility

          The Commission found that the U.S. PET resin market was highly price-sensitive and

characterized by product fungibility, that subject imports’ market share gains came at the same

time as domestic industry’s lost market share, and that the domestic industry suffered a steep

decline in financial condition during the POI. See Views at 29–30, 45, 46. Plaintiffs argue that

market share gains of imports in a price-sensitive market for fungible products are evidence of

underselling (absent supply constraints or other causes). See Pls.’ Br. at 11; Pet’rs.’ Posthr’g Br.

at 3–9. Though the Government characterizes Plaintiff’s argument as postulation, the court finds

Plaintiffs’ argument is in fact well-supported by the Commission’s prior findings.                The

Commission, therefore, should have considered the argument rather than dismissed it out of hand.

See supra Part 1.
Court No. 18-00238                                                                           Page 32
PUBLIC VERSION
        III.    The Commission’s Finding that Supply Constraints, Rather than Underselling,
                Were the Primary Cause of Domestic Industry Market Share Loss Is Not
                Supported by Substantial Evidence.

       As the Government notes, the Commission concluded that “primarily nonprice reasons, in

particular domestic supply constraints, led to purchases of subject imports rather than the domestic

like product . . . .” Def.’s Br. at 4. This finding is unsupported by substantial evidence because:

(1) no specific supply constraints are identified prior to October 2016; (2) the majority of narrative

responses solely cite post-October 2016 supply constraints; (3) the Commission’s determination

here is inconsistent with its prior determinations that later-in-time supply constraints cannot

explain earlier-in-time subject import increases; and (4) the Commission failed to adequately

respond to this conflicting evidence. The court thus holds that the Commission’s conclusion --

that supply constraints rather than underselling were the cause of market share shifts -- is

unsupported by substantial evidence and remands this issue for reconsideration consistent with

this opinion.

                A.    Specific Events that the Commission Found Contributed to Supply
                      Constraints Occurred in the Middle or Towards the End of the Period of
                      Investigation.

       The Commission identified several specific events as causing supply constraints: (1)

Hurricane Matthew in October 2016, see FSR at II-15–II-16; (2) Hurricane Harvey in August

2017, see
 id. at II-16; 
and (3) M&G’s bankruptcy in September 2017 and the related failure to

open the Corpus Christi plant on schedule, see
 id. at III-7. 
All these events occurred late in the

POI. The timing of these events undermines the Commission’s conclusion that they explained the

surge in subject imports because the surge was already in full force by the first of these events.

Record evidence, moreover, shows that Hurricanes Matthew and Harvey did not cause significant

supply disruptions.
Court No. 18-00238                                                                        Page 33
PUBLIC VERSION
       The first specific event that the Commission identified as a cause of supply constraints was

Hurricane Matthew in October 2016. As the Commission noted: [[




                                                        ]] There is conflicting record evidence as

to whether this hurricane had any significant impact. [[



                                                                               ]] Other evidence,

however, suggests any impact was small or minimal, including record evidence proffered by

respondent Novatex. 16 See Novatex Post-Hr’g Br. at Ex. 8, Sept. 21, 2018, C.R. 311, P.R. 188,

(“Novatex’s Posthr’g Br.”) (industry report shortly post-Hurricane Matthew stating that [[



                                    ]]). Even deferring to the Commission as to whether Hurricane

Matthew had any non-minimal effects, the Hurricane could still not explain market share shifts

that began prior to October 2016.

       The second specific event that the Commission identified as a cause of supply constraints

was Hurricane Harvey.      The Commission stated that “flooding from Hurricane Harvey in

September to November 2017 disrupted raw material supplies.” FSR at II-15. Evidence on the



16
  Petitioners offered uncontested testimony that any impact was minimal. See Pet’rs’ Post-Hr’g
Br. at Ex. 3. ¶ 22, Sept. 21, 2018, C.R. 311, P.R. 188 (noting that [[


                                                           ]]);
 
id. at Ex. 
5 ¶ 8 (stating that [[
                                                                                               ]]
post-Hurricane Matthew);
 
id. at Ex. 
8 ¶ 24 (stating that Hurricane Matthew [[

                                    ]] ).
Court No. 18-00238                                                                             Page 34
PUBLIC VERSION
record -- including that placed on the record by Defendant-Intervenor Novatex -- demonstrates

that Hurricane Harvey had little if any impact on PET resin supply, despite the Commission’s

finding to the contrary. See Novatex’s Posthr’g Br. at Ex. 8 (industry report dated [[

            ]] noting that [[

                                                                                                   ]] );

cf. Pls.’ Br. at 42–43 (listing contrary record evidence).

          Third, M&G’s bankruptcy occurred in September 2017. See Views at 45 n.173. At the

final hearing before the Commission, Defendant-Intervenor iResin acknowledged that any supply

constraints stemming from M&G’s bankruptcy did not emerge until September of 2017. See Tr.

of ITC Hr’g at 171–72, Sept. 13, 2018, P.R. 213. No parties contest that M&G’s bankruptcy

substantially contributed to a supply shortage after it occurred. But as this bankruptcy occurred in

September 2017, it could not have been a significant cause of subject imports’ market share gains,

which began over a year prior.

                 B.    No Narrative Responses Identify Specific Supply Constraints that Occurred
                       Before October 2016, and Most Cite Post-October 2016 Supply Constraints.

          Twenty-one U.S. importers and purchasers provided narrative responses concerning supply

shortages. See FSR at Table II-4 (providing responses of the four importers and seventeen

purchasers). Of these importers, four limited their identification of supply constraints to the last

year (or less) of the POI, or to their current situation at the time. 17 An additional six reference the

M&G shutdown, which occurred in September 2017. 18 An additional three reference weather



17
     [[
                                . ]]
18
  The six importers or purchasers reporting the M&G bankruptcy as a cause of supply constraints
are: [[                                                              . ]] See
 
id. at Table 
II-4.
Court No. 18-00238                                                                          Page 35
PUBLIC VERSION
events or force majeure -- which the record indicates refer to Hurricane Matthew (and potentially

Hurricane Harvey) and associated flooding. 19 This leaves four that do not identify any specific

event that caused supply constraints (let alone one prior to October 2016) or a specific time period

in which they experienced supply constraints. [[            ]] of these remaining four [[

                            ]] See FSR at Table II-4 ( [[                      ]] ).

          Though iResin attempts to draw a distinction between supply constraints (which occurred

throughout the POI) and supply shortages (which solely occurred late in the POI), the Commission

did not articulate such a distinction. See iResin’s Conf. Resp. to Questions at 3. Indeed, the

Commission appears to use the terms interchangeably. The only place in which the Commission

used the term “supply shortages” was in reference to what was elsewhere classified as supply

constraints. Compare Views at 38 (referring to “supply shortages” reported by fourteen U.S.

purchasers), with
 id. at 47 
(discussing the same purchaser reports as evidence of “supply

constraints”). Thus, this is a post hoc rationalization that the court may not consider. See

Burlington Truck 
Lines, 371 U.S. at 168
–69.

                 C.    The Commission’s Finding Here Is Inconsistent with its Prior
                       Determinations.

          Prior Commission determinations clearly establish that supply constraint events that follow

subject import increases cannot be the cause of them. In Certain Corrosion-Resistant Steel

Products from China, et al., the Commission found that supply constraints (including disruptive

weather) that did not match the timing of subject imports’ increase “cannot explain the magnitude




19
     [[


]] See
 
id. at Table 
II-4.
Court No. 18-00238                                                                            Page 36
PUBLIC VERSION
and duration of the increase in subject imports.” USITC Pub. 4620 (Final 2016) at 23–24. 20

Similarly, in Cold-Rolled Steel Flat Products from China and Japan, the Commission found that

alleged supply constraints could not explain the subject imports’ increase because their timing was

not corroborated by market participant reports. USITC Pub. 4637 (Final) 2016; Cold-Rolled Steel

Flat Products from China and Japan; Determinations, 81 Fed. Reg. 45,305 (ITC July 7, 2016).

Much the same conditions hold here: the only specific supply constraints supported by record

evidence occurred well into the POI, after subject imports had already significantly increased. See

supra Part 3.a. Moreover, market participant reports do not corroborate any earlier supply

constraint findings. See supra Part 3.b. The court thus orders the Commission to address these

prior determinations on remand.

               D.    Evidence that Supply Constraints Occurred Primarily, if Not Solely, in
                     October 2016 Onwards Undermines Purchaser Responses Citing Supply
                     Constraints as a Reason for Switching to Subject Imports’ Relevant to
                     Subject Imports’ Increase.

       The Commission cited purchaser responses that supply constraints were a reason for

switching to subject imports as evidence that supports its conclusion -- even if Plaintiffs are right

that supply constraints solely occurred from October 2016 onwards. See Views at 47; supra Part

3.a. However, the Commission did not identify any specific instances of supply constraints prior

to October 2016. This constitutes evidence that fairly undermines the Commission’s conclusion

that supply constraints were a main motivating factor for switching to subject imports. If supply

constraints caused switching to subject imports prior to October 2016, the Commission must point

to evidence of this. Essentially, this is an issue of specificity: the Commission did not ask questions




20
  For a list of Federal Register citations to determinations regarding individual countries in this
investigation, see Certain Corrosion-Resistant Steel Products from China, et al. at A-5, USITC
Pub. 4637 (Final) 2016.
Court No. 18-00238                                                                            Page 37
PUBLIC VERSION
that distinguished between times in the POI. Thus, the court concludes that the Commission’s

finding -- that supply-related decisions reported by purchasers occurred anytime but in that post-

October 2016 (and particularly post-September 2017) period -- is not supported by substantial

evidence.

                 E.   The Commission Did Not Adequately Address Conflicting Arguments and
                      Evidence.

         The lack of evidence of any instances of (or specific events causing) supply constraints

prior to October 2016 substantially undermines the Commission’s conclusion that supply

shortages, rather than underselling, were the primary cause of the subject import surge that began

in 2015. The Commission’s response to this is unavailing in several regards. 21 The Commission

put forward several pieces of evidence in support of its conclusion: (1) U.S. producers’ decision

to increase imports; (2) fourteen of eighteen purchasers citing supply constraints as a cause for

purchasing subject imports; and (3) the increase in subject imports despite being higher-priced.

See Views at 46–47.

         Substantial evidence does not support the conclusion that U.S. producers’ decision to

increase imports was indicative of supply constraints throughout the period. [[

                                                                      ]] See FSR at III-23. M&G’s

imports were [[                                                                          ]] Novatex’s




21
     The Commission stated in its Views at 47:

         Therefore, although Petitioners argue that supply constraints were limited to a brief
         period at the end of the POI, U.S. producers' decisions to increase their imports to
         supplement U.S. supply, the responses by purchasers regarding their decisions to
         purchase subject imports, and the increase in subject imports despite being higher
         priced, all indicate that low-priced subject imports were not the basis for the market
         share changes that occurred when the volume of nonsubject imports from Canada,
         China, India, and Oman declined substantially in 2016 and thereafter.
Court No. 18-00238                                                                            Page 38
PUBLIC VERSION
Br. at 26; Pls.’ Br. at 45. Regardless, U.S. producers’ imports represented a very small proportion

of subject imports. See FSR at III-15, III-22–III-23 (showing that [[



                            ]]).

       The Government’s argument based on purchasers’ narrative responses and importers’

assertion of supply constraints is unavailing, as it fails to address the pre-October 2016 period. As

discussed in supra Part 3.b, only four of the seventeen purchasers and importers reporting supply

constraints did not limit their response to the last year (or less), tie their answer directly to M&G’s

bankruptcy in late fall 2017, or tie their answer to weather events and weather-related force

majeures that solely occurred from October 2016 onwards. See FSR at Table II-4. Thus, rather

than diminishing the significance of Plaintiffs’ argument, the responses the Commission references

in fact buttress Plaintiffs’ argument. Moreover, the overselling finding, which the Commission

relied on in support of its supply constraints finding, is not supported by substantial evidence, as

discussed supra
. The court thus holds that that the Commission did not adequately address

conflicting evidence regarding its supply constraints finding.

        IV.    The Commission’s Ultimate no Adverse Impact Determination Was Flawed due
               to the Errors Discussed Above.

       Plaintiffs argue that (1) the asserted errors in the Commission’s analysis of underselling

and supply constraints led to a flawed analysis on impact determination; (2) the Commission failed

to address extensive contrary record information; and (3) the Commission failed to explain how

the limited pricing data it relied upon outweighed all contrary record data. See Pls.’ Br. at 46–50.

The Government (1) reiterates its earlier arguments that the Commission’s analysis of underselling

and supply constraints was supported by substantial evidence; and (2) emphasizes that the

Commission noted that in 2016 -- the year in which subject imports increased most significantly -
Court No. 18-00238                                                                         Page 39
PUBLIC VERSION
- the domestic industry’s financial condition (as measured by several different indicators) in fact

experienced period highs. Def.’s Br. at 38–41 (citing Views at 44–45). The Government also

notes the broad range of factors that the Tariff Act of 1930 mandates the Commission to consider

in its significant adverse impact analysis.
 Id. at 38–39 
(analysis must be conducted “within the

context of the business cycle and conditions of competition that are distinctive to the affected

industry” (quoting 19 U.S.C. § 1677(7)(C)(iii))).

          Setting aside for a moment the contested issues of underselling and supply constraints,

industry conditions were mixed. Domestic industry’s net sales revenues declined from 2015 to

2017 22 and other financial indicators (i.e. gross profit, 23 operating income, operating income

margin, and net income) declined from 2015 to 2017. See Views at 44–45. But, some financial

indicators showed strong financial performance of domestic industry in 2016. See
 id. Regardless,

further 
evidence, not adequately addressed by the Commission, casts doubt on the Commission’s

no adverse impact finding, including that U.S. industry increased capacity utilization by 20 percent

(from 69 percent to 89 percent) from the first quarter of 2017 (after the petition was filed) to the

first quarter of 2018 (after subject imports began receding from the market). See
 id. at 31–32; 
FSR

at III-15; cf. Pls.’ Reply at 28–29.

          Ultimately, this is a question of weighing various pieces of evidence. If the Commission

incorrectly weighted substantial pieces of evidence -- as it did here -- then remand is warranted

even if the result might not change. The Government effectively argues that the Commission’s

component findings are severable; in other words, even if one or more other findings were

unsupported by substantial evidence, that would not make the ultimate no adverse impact finding



22
     [[                                                                                           ]]
23
     [[                                                                                      ]]
Court No. 18-00238                                                                          Page 40
PUBLIC VERSION
unsupported by substantial evidence. Def.’s Br. at 38–41. However, the language of the Final

Determination, stating that the Commission’s overselling and supply constraints findings were the

cause of other findings, necessitates a contrary conclusion. Views at 45–47.

          The Commission relied upon its overselling finding to support its conclusion regarding

market share changes: “Subject imports increased significantly both absolutely and relative to

consumption, but did not significantly undersell the domestic like product. Therefore, although

the domestic industry ceded some market share to subject imports over the POI, this occurred for

non-price reasons . . . .”
 
Id. at 45. 
The Commission explained that “[w]hile the domestic industry’s

operating income experienced periodic declines, subject imports generally were higher priced

throughout the POI.”
 Id. Similarly, the 
Commission relied upon its overselling finding to explain

the domestic industry’s inability to realize the benefits of imposition of orders on Canada, China,

India, and Oman. See
 id. at 46 
(“[S]ubject imports largely oversold the domestic product in 2016,

so any market share shifts in that year were due to factors other than price.”).

          The Commission’s supply constraints finding likewise buttresses many of its other

conclusions -- including its conclusion that price was not the cause of market share shifts. 24 The

Commission articulated the same evidence of supply constraints that it found persuasive as also

being evidence that market share shifts following orders against subject countries in the prior PET

resin commission investigation were not caused by price. See Views at 47. 25 Thus, given the



24
   If no non-price factors explained market share shifts, then that undermines the conclusion that
price was not a factor in market share shifts.
25
     In full, the Commission stated in its Views at 47:

          [A]lthough Petitioners argue that supply constraints were limited to a brief period
          at the end of the POI U.S. producers' decisions to increase their imports to
          supplement U.S. supply, the responses by purchasers regarding their decisions to
          purchase subject imports, and the increase in subject imports despite being higher
Court No. 18-00238                                                                           Page 41
PUBLIC VERSION
Commission’s own explanation of its decision, and a lack of substantial evidence supporting either

or both of the Commission’s overselling and supply constraint findings, the court holds that the

Commission’s adverse impact determination is unsupported by substantial evidence and remands

for reconsideration consistent with this opinion.

                                           CONCLUSION

        The court remands the case to the Commission for further proceedings consistent with this

opinion. Specifically, the court remands on the following issues: (1) the Commission’s

determination that overselling predominated and underselling was not significant in light of

aspects of the record showing underselling and to address prior determinations that even

underselling that occurred in less than 50 percent of price comparisons could be significant; (2)

the Commission’s finding that supply constraints were the cause of market share shifts rather than

underselling in light of conflicting evidence that supply constraints solely occurred from October

2016 onwards; and (3) the Commission’s no adverse impact determination in light of the

insufficiencies in its overselling and supply constraint findings. While the Commission’s failure

to consider and address these significant arguments and evidence warrants remand, the court

expresses no view on the ultimate appropriate determination of this matter on remand.

        The Commission shall file with this court and provide to the parties its remand results

within 90 days of the date of this order; thereafter, the parties shall have 30 days to submit briefs

addressing the revised final determination with the court, and the parties shall have 30 days

thereafter to file reply briefs with the court.

        SO ORDERED.



        priced, all indicate that low-priced subject imports were not the basis for the market
        share change that occurred when the volume of nonsubject imports from Canada,
        China, India, and Oman declined substantially in 2016 and thereafter.
Court No. 18-00238                             Page 42
PUBLIC VERSION
                            /s/   Gary S. Katzmann
                                   Judge
Dated: June 4, 2020
       New York, New York