LOURIE, Circuit Judge.
Huvis Corporation (“Huvis”) appeals from the judgment of the United States Court of International Trade affirming the United States Department of Commerce’s (“Commerce’s”) valuation of Huvis’s imports.
See Huvis Corp. v. United States,
525 F.Supp.2d 1370 (Ct. Int’l Trade 2007)
(“Huvis
/”);
Huvis Corp. v. United States,
No. 06-00380, 2008 WL 2977890, 2008 Ct. Int'l. Trade Lexis 82 (Aug. 5, 2008)
(“Huvis II
”). Because Commerce’s decision to use a constructed market price in valuing Huvis’s imports was supported by substantial evidence and not contrary to law, we affirm.
BACKGROUND
Huvis is a Korean producer and exporter of polyester staple fiber, a synthetic fiber that is used for stuffing products such as clothing and pillows. It is subject to an antidumping duty order, under which a dumping margin is applied to all of Huvis’s imports. Polyester staple fiber is made from terephthalic acid. Huvis uses three grades of terephthalic acid in manufacturing its fiber: qualified-grade, middle-grade, and purified. Huvis’s importation of polyester staple fiber into the United States had been reviewed by Commerce four times in periodic administrative reviews, prior to the administrative review it now appeals. During the fifth period of review, which is the subject of the present appeal, Huvis purchased all of the terephthalic acid it used from affiliated companies.
Commerce assigns a “value” to an import to determine the dumping margin applied to it. When an importer purchases a “major input” from an affiliated party, instead of assuming the input was purchased at arm’s length, Commerce applies the “major input rule” to determine the value of the input. 19 U.S.C. § 1677b(f)(2), (3). The parties agree that terephthalic acid constitutes a major input to polyester staple fiber. Commerce’s interpretation of the major input rule requires Commerce normally to “determine the value of a major input purchased from an affiliated person based on the higher of: (1) [t]he price paid by the exporter or producer to the affiliated person for the major input; (2)[t]he amount usually reflected in sales of the major input in the market under consideration; or (3)[t]he cost to the affiliated person of producing the major input.” 19 C.F.R. 351.407(b). Thus, for the fifth period of review, as with each of the previous review periods, Commerce requested that Huvis submit three measures for its purchases of each grade of terephthalic acid made during the period of review: (1) the price Huvis paid to its affiliated suppliers (the “transfer price”); (2) the price at which the affiliated producer made sales of the input to unaffiliated parties (the “market price”); and (3) the affiliated producer’s cost of production for the input (the “cost of production”). For middle-grade terephthalic acid, Huvis submitted all three measures. For qualified-grade and purified terephthalic acids, Huvis submitted only the transfer price and cost of production measures because it did not have access to market price data, as its supplier for those two products, Samnam Petrochemical Company, Ltd. (“Samnam”), considered that data proprietary.
In each of its three previous reviews, Huvis had similarly submitted information that was missing certain market price data. Commerce had, in all cases but one, applied the major input rule for those products by comparing only the two measures Huvis had provided, transfer price and the cost of production,
but in the fifth administrative review, Commerce chose to include in the comparison a market price that was constructed from “facts available.”
“If ... necessary information is not available on the record,” Commerce is required to “use the facts otherwise available in reaching the applicable determination under this title.” 19 U.S.C. § 1677e(a). However, under 19 U.S.C. § 1677e(c), Commerce must corroborate such secondary information “from independent sources that are reasonably at [its] disposal.” Commerce must also
19 U.S.C. § 1677m(e). Commerce finally may not draw “adverse inferences” against an importer unless the importer fails to cooperate in the investigation. 19 U.S.C. § 1677e(b). The parties agree that Huvis cooperated with Commerce in the fifth review.
consider information that is submitted by an interested party and is necessary to the determination but does not meet all the applicable requirements ... if
(1) the information is submitted by the deadline established for its submission,
(2) the information can be verified,
(3) the information is not so incomplete that it cannot serve as a reliable basis for reaching the applicable determination,
(4) the interested party has demonstrated that it acted to the best of its ability in providing the information and meeting the requirements established by the administering authority or the Commission with respect to the information, and
(5) the information can be used without undue difficulties.
Commerce applied the major input rule by first constructing the missing market prices for Huvis’s qualified-grade and purified terephthalic acids using facts available. It constructed the market prices by first calculating an average profit rate taken from the supplier’s submitted financial statements. It then increased the costs of production by that profit rate to form constructed market prices for each of the qualified-grade and purified terephthalic acids. Applying the major input rule, Commerce found that the constructed market prices were the highest of the three measures for both the qualified-grade and purified terephthalic acids, so it determined that the value of each of those products was its market price, as constructed. For middle-grade terephthalic acid, the highest of the three measures was the transfer price, so Commerce determined that the value of the middle-grade terephthalic acid was its transfer price. The following chart demonstrates the relative values for each product. In the chart, “r” represents the profit rate that Commerce applied.
Qualified-Grade Middle-Grade Purified _Terephthalic Acid Terephthalic Acid Terephthalic Acid
Highest Measure (QTA transfer (MTA transfer (PTA transfer
Provided by Huvis price) price) price)
Value (Highest Measure (QTA constructed (MTA transfer (PTA constructed
After Constructing market price) = price) market price) =
Missing Measures, the (QTA cost of (PTA cost of
Result of Applying production) x (1 + r) production) x (1+r)
Major Input Rule)_
Huvis challenged Commerce’s application of the major input rule in the Court of International Trade, arguing that substantial evidence did not support the constructed market prices that Commerce had calculated to fill in the missing market prices for qualified-grade and purified terephthalic acids. The court held that the prices Commerce had constructed were consistent with the statutory requirements and supported by substantial evidence, as Commerce had used Huvis’s own data in calculating the market prices and had not applied adverse inferences against Huvis. But the court held that Commerce’s construction of the market prices was not consistent with Commerce’s established practice, which had previously been, according to the court, to use the highest available measure rather than constructing a measure,
i.e.,
market price, that was otherwise unavailable. Thus, the court remanded for Commerce either to use only the available measures provided by Huvis,
ie.,
transfer price and cost of production, in applying the major input rule, or to adequately explain its reason for treating Huvis differently in this administrative review than it had in prior administrative reviews.
On remand, Commerce explained its change in treatment, stating that it had recognized for the first time that it had evidence on the record that could be used to construct a market price, providing “a more complete analysis under the major input rule, and resulting] in a more accurate calculation of Huvis’s dumping margin.”
Huvis II,
2008 WL 2977890, at *3, 2008 Ct. Intl. Trade Lexis 82, at *5. When
Huvis returned to the Court of International Trade, the court found that the explanation was reasoned, explaining that Commerce is free to change its methodology to improve accuracy and that Huvis did not detrimentally rely on the prior practice. Thus, the court affirmed Commerce’s application of the major input rule.
Huvis timely appealed the Court of International Trade’s decision. We have jurisdiction pursuant to 28 U.S.C. § 1295(a)(5).
DISCUSSION
“We review the Court of International Trade’s judgment, affirming or reversing the final results of an administrative review,
de novo.” Fag Kugelfischer Georg Schafer AG v. United States,
332 F.3d 1370, 1372 (Fed.Cir.2003). In so doing, “[w]e apply anew the same standard used by the court, and will uphold Commerce’s determination unless it is unsupported by substantial evidence on the record, or otherwise not in accordance with law.”
Yancheng Baolong Biochem. Prods. Co. v. United States,
337 F.3d 1332, 1333 (Fed.Cir.2003) (citation and internal quotation marks omitted). However, “we give great weight to the informed opinion of the Court of International Trade ..., [and] it is nearly always the starting point of our analysis.”
Nippon Steel Corp. v. United States,
458 F.3d 1345, 1351 (Fed.Cir.2006) (quotation marks and citation omitted). Substantial evidence means “more than a mere scintilla” and “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
Consol. Edison Co. v. NLRB,
305 U.S. 197, 229, 59 S.Ct. 206, 83 L.Ed. 126 (1938)). “[T]he substantiality of evidence must take into account whatever in the record fairly detracts from its weight,” including “contradictory evidence or evidence from which conflicting inferences could be drawn.”
Suramerica,
44 F.3d at 985 (quotation marks omitted).
Huvis argues that Commerce’s decision to use a constructed market price in applying the major input rule was both unsupported by substantial evidence and contrary to law. Huvis asserts that Commerce’s standard practice was to use only the available measures, cost of production and transfer price, in applying the major input rule when market price data was unavailable. Thus, according to Huvis, this practice became “law of the proceeding,” and Huvis reasonably expected Commerce to adhere to it. Huvis argues that such law of the proceeding does not require a showing of detrimental reliance on the standard practice, and, even if Huvis did have to show reliance, it never had a chance to present such evidence to the Court of International Trade. Huvis also argues that, in general, Commerce may not depart from a consistent past practice without explanation, and it did not explain what was different in this review from the four preceding reviews. Huvis argues that Commerce claims only to have recognized for the first time in this review that it could construct a market price based on existing evidence, so, according to Huvis, Commerce did not actually find any new evidence that would justify a departure from past practice.
Huvis also argues that Commerce’s method of constructing a market price unlawfully violated Huvis’s statutory rights. Huvis contends that Commerce unlawfully penalized Huvis for failing to provide information that it was unable to provide. According to Huvis, Commerce disregarded the cost of production, which was available information that Huvis provided and was necessary to its determination, violating 19 U.S.C. § 1677m(e). Huvis further asserts that Commerce’s methodology for constructing a market price was both adverse
to Huvis and demonstrably inaccurate, in violation of 19 U.S.C. § 1677e(b) and (c). According to Huvis, even though the record establishes that qualified-grade terephthalic acid is, on average, of lower quality and less expensive than middle-grade terephthalic acid, the relative values were switched after Commerce constructed the market price for qualified-grade terephthalic acid and applied the major input rule, resulting in a higher value for qualified-grade acid than middle-grade acid. Moreover, since Commerce constructed the market price by marking up the cost of production by an average profit margin, Huvis asserts that the resulting market price will always be higher than the cost of production. Thus, according to Huvis, Commerce’s method effectively eliminates the cost of production from the comparison and will always be adverse to importers, contravening the statutory scheme.
The government responds that Commerce’s decision to construct a market price was both supported by substantial evidence and not contrary to law. According to the government, Commerce had no consistent prior practice that it was required to continue to follow, and Huvis never submitted or identified evidence of detrimental reliance on any prior methodology. The government adds that Commerce had previously advised Huvis that it would use facts available to fill in missing measures, as long as there were reasonable, non-adverse data on the record to do so. Indeed, according to the government, Commerce had done the same thing in reviews of other imports. The government also argues that Commerce may deviate from its past practice when it reasonably explains its reasons for departure, and, according to the government, Commerce did so here. The government asserts that Commerce adequately explained why it treated Huvis differently in the fifth review than it did in earlier reviews, as it only then realized that record evidence would support a constructed market price and improve its accuracy.
The government also responds that its method for constructing market prices was consistent with Congress’s statutory scheme. According to the government, market price data was necessary to apply the major input rule, as the rule requires a comparison of three measures, one of which is market price. Because market price data were not available, the government argues that Commerce permissibly relied upon the facts available to determine market prices for the major inputs. According to the government, Commerce did not ignore the submitted costs of production in violation of 19 U.S.C. § 1677m(e); indeed, it used cost of production as an element to construct a market price and it used cost of production for the ultimate comparison required by the major input rule. Thus, according to the government, Commerce complied with the statute. The government also asserts that the data upon which Commerce relied were reasonable, not adverse, and supported by substantial evidence, in compliance with 19 U.S.C. § 1677e(b) and (c). The government argues that, to construct market prices, Commerce used verified information submitted by Huvis, including its supplier’s audited financial statements. Even though qualified-grade terephthalic acid is normally less expensive than middle-grade terephthalic acid, according to the government, Huvis acknowledges that sometimes the relationship between prices actually is reversed, with qualified-grade being more expensive than middle-grade. The government therefore asserts that Commerce was justified in producing an outcome wherein the qualified-grade value determined by applying the major input rule was greater than the middle-grade value determined by the rule. Thus, according to the government, Commerce applied a reasonable, statutorily based methodology.
We conclude that the Court of International Trade correctly found Commerce’s methodology for applying the major input rule to be supported by substantial evidence and adequately explained, in accordance with the law. In general, an agency may deviate from a past practice when “the new policy is permissible under the statute, ... there are good reasons for it, and ... the agency
believes
it to be better, which the conscious change of course adequately indicates.”
FCC v. Fox Television Stations, Inc.,
— U.S. --•, 129 S.Ct. 1800, 1811, 173 L.Ed.2d 738 (2009). Thus, Commerce need only show that its methodology is permissible under the statute and that it had good reasons for the new methodology.
A.
Commerce’s Methodology
Was
Permissible Under the Statute
We agree with the government that Commerce’s decision to construct a market price was in accordance with the governing statutes. We have found permissible Commerce’s interpretation of the major input rule, which uses as the value the highest of the three measures,
viz.
transfer price, cost of production, and market price.
See NTN Bearing Corp. of Am. v. United States,
368 F.3d 1369, 1374-76 (Fed.Cir.2004). Huvis argues that the constructed market price always consists of adding an amount to the cost of production, meaning the constructed market price ■will always be higher than the cost of production, and cost of production is therefore effectively ignored. That system, according to Huvis, violates 19 U.S.C. § 1677m(e) by ignoring available information that. Huvis provided and was necessary to Commerce’s determination. However, Commerce’s decision to construct a market price did not ignore the cost of production data. In fact, cost of production was taken into account twice in Commerce’s application of the major input rule. First, Commerce used the cost of production to construct the market price. Second, Commerce included the cost of production in the comparison of the three measures to calculate a value. Further, we agree with the government that the constructed market price will not always be higher than the cost of production, so even the comparison of the three measures does not ignore cost of production. If, for example, Huvis were to take a loss on a certain product, the profit rate would be negative, and the resulting market price would be lower than the cost of production; therefore, the value determined by applying the major input rule would be the cost of production. Thus, Commerce’s methodology did not violate 19 U.S.C. § 1677m(e).
Huvis also asserts that Commerce’s methodology for constructing a market price was adverse to Huvis and demonstrably inaccurate, in violation of 19 U.S.C. § 1677e(b) and (c). The parties agree that Huvis did not fail to cooperate in the investigation, so Commerce was not entitled to use adverse inferences against Huvis under § 1677e(b). Even when Commerce uses adverse inferences, it must use a “reasonably accurate estimate of the respondent’s actual rate” when using “facts available.”
F.lli De Cecco Di Filippo Fara S. Martino S.p.A. v. United States,
216 F.3d 1027, 1032 (Fed.Cir.2000).
We disagree with Huvis’s argument that the constructed market price is demonstrably inaccurate or constitutes an adverse inference against Huvis. As evidence, Huvis points to the relative market prices of qualified-grade and middle-grade terephthalic acids. Qualified-grade has more impurities, making it presumably of lower quality, than middle-grade acid. However, the constructed market price of qualified-grade acid is higher than the verified market price of middle-grade acid. Thus, Huvis argues that Commerce’s
method of constructing market prices is clearly inaccurate. We disagree. As the government points out and the Court of International Trade found, the higher level of impurities in qualified-grade acid does not necessarily mean that it should have a lower market price.
See Huvis II,
2008 WL 2977890, at *3-4, *5, 2008 Ct. Intl. Trade Lexis 82, at *7-9 n. 2, *11-12. Other factors affect market price. Indeed, the month-by-month transfer prices reveal that Huvis paid more for qualified-grade acid than it did for middle-grade acid in certain months, even though the average transfer price of qualified-grade acid was lower.
Further, Commerce reasonably chose to construct the market price by adding the cost of production to the profit margin, since there is no suggestion here that product sales were unprofitable or that the profit margins were unusually low.
At oral argument, Huvis pointed out that the profit margin that Commerce used was the average profit margin for all of Samnam’s products, rather than a more precise profit margin for the particular input,
e.g.,
qualified-grade terephthalic acid. Oral Arg. 11:17-11:44, May 4, 2009,
available at
http: //oralarguments.cafc.uscourts.gov/mp3/2009-1021.mp3. However, Huvis never argued, and presented no evidence, that the overall average profit margin did not accurately reflect the profit margin for the particular input. Huvis instead argued that “the profit margin used, it’s undisputed, was the profit margin for the supplier company as a whole. That
may or may not be accurate
in terms of reflecting correctly the profit on this particular input.”
Id.
(emphasis added). Thus, rather than prove the inaccuracy of the profit margin directly, Huvis argued that it had proven the inaccuracy of the profit margin as reflected in the relative market prices of qualified-grade and middle-grade terephthalic acids.
Id.
at 9:20-9:39 (“The basis for our position that it was unreasonable to use it for this input is the verified information that the particular input here, this QTA, qualified terephthalic acid, is the least valuable, ... the lowest quality of the three different terephthalic acids.”). As we demonstrated above, the mere fact that the relative market prices of qualified-grade and middle-grade terephthalic acids were different than might have been expected, based on their average transfer prices, does not prove that Commerce’s construction of market prices was either inaccurate or adverse to Huvis. We therefore agree with the government that Commerce’s constructed market price is not demonstrably inaccurate and does not constitute an adverse inference against Huvis.
B.
Commerce Had Good Reason for Its Change of Policy
We also agree with Commerce that it had good reason for changing its methodology. When it changes a past practice, an agency must also show that there are good reasons for its new policy.
See Fox Television,
129 S.Ct. at 1811;
Nippon Steel Corp. v. U.S. Int’l Trade Comm’n,
494 F.3d 1371, 1378 n. 5 (Fed.Cir.2007) (“[T]he Supreme Court requires us to give deference to agency decisions that embody policy changes. When an agency decides to change course, however, it must adequately explain the reason for a reversal of policy.” (internal citations omitted)). Sometimes an agency must provide a more detailed justification than what would suffice for a new policy created on a blank slate, such as when “its new policy
rests upon factual findings that contradict those which underlay its prior policy; or when its prior policy has engendered serious reliance interests that must be taken into account.”
Fox Television,
129 S.Ct. at 1811.
Here, Commerce stated that it changed its methodology to improve accuracy. Improving accuracy is generally a good reason for a change in methodology.
See SKF USA, Inc. v. United States,
537 F.3d 1373, 1380 (Fed.Cir.2008) (affirming a new methodology that was changed to improve accuracy and stating that, “[w]ith respect to Commerce’s decision to revise [its] methodology after 14 annual reviews utilizing [a different] methodology, we have specifically affirmed changes to ... methodologies by Commerce where reasonable”);
Fujian Mach. & Equip. Imp. & Exp. Corp. v. United States,
178 F.Supp.2d 1305, 1327 (Ct. Int’l Trade 2001) (“Commerce is free to discard one methodology in favor of another, the better to calculate more accurate dumping margins.”).
Although Commerce need not show that its “reasons for the new policy are
better
than the reasons for the old one,”
Fox Television,
129 S.Ct. at 1811, here, Commerce stated its reasons for the new policy as
improving
accuracy. Thus, we address whether substantial evidence supported Commerce’s belief that including a constructed market price in the major input rule would improve the accuracy of the value. We agree with the government that Commerce’s belief was reasonable. For the same reasons that we discussed above with respect to 19 U.S.C. § 1677e(b) and (c), Commerce reasonably believed that its constructed market price would improve the accuracy of the value. As market price is reasonably defined in the circumstances of this case as the sum of the cost of production and a profit margin, Commerce reasonably believed that it could accurately estimate market price, given that Huvis provided both cost of production and profit margin. Commerce also reasonably believed the supplier’s average profit margin adequately represented the profit margin for each specific product, as Huvis offered no evidence to the contrary. Commerce further reasonably believed that including a market price measure in the comparison would improve the accuracy of the value, given the statute’s requirement of a comparison of all three measures. Finally, Huvis’s argument concerning the relative market prices of qualified-grade and middle-grade acids is unavailing for the reasons discussed above,
viz.,
that the market prices of the two acids do not necessarily reflect their relative purities.
We also agree with the government that Huvis did not detrimentally rely on Commerce’s old methodology used in multiple preceding reviews.
See Fox Television,
129 S.Ct. at 1811;
Fujian,
178 F.Supp.2d at 1327 (“Commerce may not make minor but disruptive changes in methodology where a respondent demonstrates its specific reliance on the old methodology used in multiple preceding reviews.”);
Shikoku Chems. Corp. v. United States,
795 F.Supp. 417 (Ct. Int’l Trade 1992) (overturning Commerce’s use of a slightly improved methodology when the respondent demonstrated that it had set its prices in reliance on the old methodology). As the Court of International Trade found, Huvis did not and could not show that it had detrimentally relied on any consistent pri- or practice. Huvis did not report market price data to Commerce because “Huvis was powerless to force its affiliated supplier to provide such data, or at least Huvis represented as much to Commerce. Therefore, the absence of market price data from the record was not the result of Huvis’s reliance on Commerce’s prior practice; rather, it was due to its affiliated supplier’s intransigence.”
Huvis II,
2008
WL 2977890, at *4, 2008 Ct. Inti. Trade Lexis 82, at *14. We therefore agree with the government that Commerce’s change in methodology was reasonable and adequately explained and that no further explanation was required, such that Commerce was free to change its methodology.
Commerce has therefore shown that its new methodology of constructing a market price was permissible under the statute and that it had good reasons for the new methodology. We have considered Huvis’s remaining arguments and find them unpersuasive.
CONCLUSION
Accordingly, the judgment of the Court of International Trade affirming Commerce’s valuation of Huvis’s imports is affirmed.
AFFIRMED