MICHEL, Chief Judge.
This is an antidumping duty case relating to wooden bedroom furniture imported from China. In 2005, the Department of Commerce (“Commerce”) imposed an anti-dumping duty order on this merchandise. Both the foreign producers and the domestic interested parties brought suit in the Court of International Trade (“CIT”), raising issues with Commerce’s calculation of the duty rate. The foreign producers argued that Commerce’s calculation of Chinese labor rates was not in accordance with the governing statute, both because the calculation used data from countries that are not economically comparable to China, and because the calculation used data from countries that are not significant producers of merchandise comparable to the wooden bedroom furniture at issue here. The CIT affirmed Commerce’s labor rate calculation method. Because we hold that Commerce’s method for calculating wage rates uses data not permitted to be used by the governing statute, we invalidate the regulation establishing that calculation method, vacate the CIT’s affirmance of the application of that regulation, and remand for further proceedings consistent with the governing statute.
The domestic interested parties take issue with the CIT’s treatment of Commerce’s method for calculating non-pro
duction factors, such as profit, in determining the fair value of the imported Chinese merchandise. Commerce had used data from seven surrogate companies from India to determine average ratios used to calculate the value of several non-production factors. The CIT ultimately required Commerce to eliminate four of the seven companies from its calculation. Because we find that Commerce’s original calculation method, using all seven surrogate companies, represented a permissible application of the governing statute, we find that the CIT erred by rejecting Commerce’s original calculation, vacate the CIT’s affirmance of Commerce’s eventual calculation using only three companies, and remand for further proceedings using Commerce’s original calculation method.
Finally, the foreign producers and the domestic interested parties each raise a separate alleged problem with Commerce’s calculation of the antidumping duty margin. In both cases, these problems were raised before the CIT, but the CIT held that it was unable to address the issues because they had not been appropriately raised first before Commerce. Because we find that these issues needed to be raised before Commerce at an appropriate time and were not so raised, we affirm the CIT’s refusal to decide these issues on the merits.
I. BACKGROUND
Under the antidumping duty statute, Commerce may “determine[ ] that a class or kind of foreign merchandise is being, or is likely to be, sold in the United States at less than fair value.” 19 U.S.C. § 1673. If Commerce finds that this activity, known as “dumping,” is occurring, and if the corresponding domestic industry is materially injured or is materially hampered from forming by the dumping, then Commerce is required to impose a duty on imports of the dumped foreign merchandise. The amount of the duty is set at “an amount equal to the amount by which the normal value [of the merchandise] exceeds the export price.”
Id.
This case is about the process by which Commerce determines the “normal value” of the dumped merchandise.
Normal value is usually determined as the price at which the merchandise in question is sold in the exporting country, but this does not apply to merchandise exported from countries with non-market economies. In these countries, “sales of merchandise ... do not reflect the fair value of the merchandise,” because those sales “do[ ] not operate on market principles of cost or pricing structures.”
Id.
§ 1677(18). For merchandise from non-market economy countries, the procedure for determining normal value has been established by statute.
Id.
§ 1677b(c). In these cases, normal value is determined “on the basis of the value of the factors of production utilized in producing the merchandise^] to which shall be added an amount for general expenses and profit plus the cost of containers, coverings, and other expenses.”
Id.
§ 1677b(c)(l). The factors of production “include, but are not limited to ... hours of labor required, ... quantities of raw materials employed, ... amounts of energy and other utilities consumed, and ... representative capital cost, including depreciation.”
Id.
§ 1677b(c)(3). In valuing these factors of production, Commerce is required to “utilize, to the extent possible, the prices or costs of factors of production in one or more market economy countries that are ... at a level of economic development comparable to that of the non-market economy country, and ... significant producers of comparable merchandise.”
Id.
§ 1677b(c)(4).
Commerce has promulgated regulations that prescribe how it carries out these statutory duties. For most of the factors
of production, Commerce uses the values that prevail in a single market economy country (called a “surrogate country”) that Commerce finds is both (a) economically comparable to the non-market economy country in question and (b) a significant producer of the merchandise in question. 19 C.F.R. § 351.408(c)(2). One factor of production, labor, is treated differently from the other factors. To value the labor used to produce subject merchandise, Commerce uses “regression-based rates reflective of the observed relationship between wages and national income in market economy countries.”
Id.
§ 351.408(c)(3). In addition to these methods of valuing the direct inputs to production, Commerce also values selling, general, and administrative expenses (“SG & A”), factory overhead, and profit. These are determined using financial ratios derived from publicly-available financial statements of producers of comparable merchandise in the surrogate country.
Id.
§ 351.408(c)(4). Generally, if more than one producer’s financial statements are available, Commerce averages the financial ratios derived from all the available financial statements.
Here, in 2003, Commerce opened an investigation into whether wooden bedroom furniture from China was being dumped.
Notice of Initiation of Antidumping Duty Investigation: Wooden Bedroom Furniture from the People’s Republic of China,
68 Fed.Reg. 70,228 (Dec. 17, 2003). In June 2004, Commerce preliminarily determined that dumping was occurring.
Notice of Preliminary Determination of Sales at Less Than Fair Value and Postponement of Final Determination: Wooden Bedroom Furniture from the People’s Republic of China,
69 Fed.Reg. 35,312 (June 24, 2004)
(“Preliminary Determination
”). After taking briefing from interested parties, Commerce published its final determination on November 17, 2004.
Notice of Final Determination of Sales at Less Than Fair Value and Postponement of Final Determination: Wooden Bedroom Furniture from the People’s Republic of China,
69 Fed.Reg. 67,313 (Nov. 17, 2004)
(“Final Determination
”). This was amended to correct some ministerial errors in January 2005.
Notice of Amended Final Determination of Sales at Less Than Fair Value and Postponement of Final Determination: Wooden Bedroom Furniture from the People’s Republic of China,
70 Fed.Reg. 329 (Jan. 4, 2005)
(“Amended Final Determination
”).
In these original proceedings, Commerce determined that wooden bedroom furniture from China was being sold below fair value. Because Commerce determined that China is a non-market economy country, to determine the normal value of the dumped merchandise, Commerce had to (1) select a surrogate country on which to base the cost of non-labor factors of production, (2) determine the wage rate applicable to production of wooden bedroom furniture based on the regression analysis of wages and national incomes, (3) determine the appropriate financial ratios for companies producing wooden bedroom furniture in the surrogate country, and (4) use these determinations to calculate the normal value. Commerce selected India as the surrogate country. In accordance with 19 C.F.R. § 351.408(c)(3), Commerce determined the value of the labor factor of production for China using its regression analysis. To determine financial ratios, Commerce averaged the ratios obtained from the financial statements of seven Indian surrogate companies.
There are two groups of appellants here. The first is a group composed of Chinese manufacturers: Dorbest Limited, Rui Feng Woodwork (Dongguan) Co., Ltd., and Rui Feng Lumber Development (Shenzhen) Co., Ltd. These companies are subject to the antidumping duty, and they
are referred to here collectively as “Dorbest.” The second is a group of domestic interested parties: the American Furniture Manufacturers’ Committee for Legal Trade and Vaughan-Bassett Furniture Company, Inc. These companies are referred to here collectively as “AFMC.” Both Dorbest and AFMC filed suit in the CIT to challenge the results of the original proceedings before Commerce. AFMC opposes Dorbest on each issue Dorbest appeals, and Dorbest opposes AFMC on each issue AFMC appeals.
On March 23, 2005, the initial complaint was filed in the CIT. There was an initial remand to Commerce that resulted in an August 2005 set of remand results, referred to here as
“Remand I.”
In October 2006, the CIT remanded certain additional issues to Commerce. In May 2007, Commerce issued its second set of findings on remand, referred to here as
“Remand II.”
In February 2008, the CIT considered the
Remand II
results and again remanded some issues to Commerce. In July 2008, Commerce issued its third set of findings on remand, referred to here as
“Remand III.”
In January 2009, the CIT affirmed the results of
Remand III.
This appeal followed.
Dorbest and AFMC present several issues on appeal. They are as follows.
First, Commerce determined that a regression-based method provides the “best available information” for calculating the value of labor in China’s non-market economy. 19 C.F.R. § 351.408(c)(3). Commerce has designated India as the primary surrogate market economy for the Chinese non-market economy, but rather than using the Indian surrogate wage rate, Commerce’s regression-based method estimates a wage rate for China that is over 300% higher than the Indian surrogate wage rate. The CIT sustained Commerce’s use of this method, and Dorbest
argues that this conflicts with the statutory requirement that Commerce value factors of production using surrogate values from market economy countries that are economically comparable with China, “to the extent possible.” 19 U.S.C. § 1677b(c)(4). The issue is whether Commerce’s valuation method is in accordance with law and supported by substantial evidence.
Second, Dorbest
also argues that the regression-based method discussed above violates the statutory requirement that Commerce value factors of production using surrogate values from market economy countries that are significant producers of merchandise comparable to the merchandise at issue in a given antidumping duty order. 19 U.S.C. § 1677b(c)(4). The United States argues that this issue was not raised before the CIT, so Dorbest is barred from raising it here. If Dorbest is permitted to raise this argument, then the issue is whether Commerce’s regression method for calculating wage rates comports with the statutory requirement to use countries that are significant producers of comparable merchandise.
Third, while Commerce used a regression analysis to determine surrogate wage rates (as discussed above), it did not use a regression analysis to determine surrogate values of other factors (SG
&
A, factory overhead, and profit) for the Chinese non-market economy. Instead, consistent with its normal practice, Commerce used an average of the costs from seven Indian surrogate companies. The CIT suggested that Commerce should have excluded the four smallest of the seven companies, because the CIT believed there might be a correlation between company size and fi
nancial results, and those four companies were too small to be comparable to Dorbest. On remand, Commerce refused to conduct a regression analysis to determine whether the CIT was correct that company size and financial results were correlated, but Commerce did exclude the four smallest surrogate companies in order to comply with the CIT’s order. AFMC
argues that Commerce’s acquiescence to the CIT’s order, combined with its failure to use a regression analysis to support its action, violated the statutory requirement that Commerce use the “best available information” to determine surrogate values for non-market economies. The issue is whether the CIT erred by requiring Commerce to abandon its original approach in favor of excluding the four smallest companies.
Fourth, Dorbest
argued at the CIT that Commerce failed to properly add excise duty expenses as an income item when calculating the surrogate value for the profit of one particular surrogate company, Indian Furniture Products. Dorbest had not raised this issue in its administrative case brief before Commerce, although it did raise the issue in a footnote in its rebuttal brief before Commerce and again during the ministerial comment period before Commerce’s adoption of its original final determination. The CIT held that Dorbest’s failure to raise this argument in its administrative case brief before Commerce constituted a failure to exhaust administrative remedies and thereby rendered the CIT unable to review the issue on the merits. The issue is whether the CIT erred by holding that it was unable to review this issue due to Dorbest’s failure to exhaust its administrative remedies.
Fifth, during the administrative process leading to Commerce’s
Remand II results,
AFMC
raised for the first time the issue of whether Commerce had committed a clerical error that caused it to use the incorrect value for the expense of a raw material called rubberwood when determining the cost of production in the Chinese non-market economy. AFMC had not raised the issue either during the ministerial comment period at Commerce or in its complaint or earlier briefing to the CIT. Commerce refused to consider the issue during the remand proceedings, and the CIT affirmed Commerce’s decision, holding that AFMC had waived the issue by not raising it during its multiple opportunities to do so in a timely manner. The issue is whether Commerce abused its discretion by holding that it was unable to review this issue due to AFMC’s waiver.
II. DISCUSSION
The five issues on appeal can be sorted into three groups: two issues relate to Commerce’s valuation of the labor factor of production, one issue relates to Commerce’s valuation of non-production factors, and two issues concern whether the failure to raise an issue before Commerce in a timely fashion results in waiver of the issue.
A. Labor Value Calculation
We first examine Dorbest’s argument that Commerce’s regulation governing the calculation of labor value for products imported from non-market economy countries, 19 C.F.R. § 351.408(c)(3), fails to comply with the requirement of 19 U.S.C. § 1677b(c)(4)(A) that Commerce value fac
tors of production using “to the extent possible, the prices or costs of factors of production in one or more market economy countries that are ... at a level of economic development comparable to that of the nonmarket economy country.”
We “uphold Commerce’s determinations, findings, and conclusions unless they are ‘unsupported by substantial evidence on the record, or otherwise not in accordance with the law.’ ”
Wheatland Tube Co. v. United States,
495 F.3d 1355, 1359 (Fed.Cir.2007) (quoting 19 U.S.C. § 1516a(b)(1)(B)(i)). In determining whether Commerce acted in accordance with the law, we must determine whether Commerce’s compliance with its own regulation, 19 C.F.R. § 351.408(c)(3), satisfies the requirements of 19 U.S.C. § 1677b(c)(4)(A), the statute under which that regulation was promulgated. When deciding whether Commerce properly interpreted a governing statute, we review Commerce’s actions de novo, but we conduct the review within the framework established by
Chevron U.S.A., Inc. v. Natural Resources Defense Council, Inc.,
467 U.S. 837, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984).
Agro Dutch Indus. Ltd. v. United States,
508 F.3d 1024, 1029-30 (Fed.Cir.2007).
The governing statute, 19 U.S.C. § 1677b(c)(4)(A), provides that surrogate data used to calculate the value of factors of production, including labor, must, to the extent possible, come from market economy countries with “a level of economic development comparable to that of the non-market economy country,” China in this case. In promulgating its labor factor calculation regulation, Commerce interpreted this statute to permit the use of “regression-based wage rates reflective of the observed relationship between wages and national income in market economy countries.” 19 C.F.R. § 351.408(c)(3). In constructing its regression analysis, Commerce has used data on wage rates and national income from 61 market economy countries, with gross national incomes ranging from $420 per capita per year (Pakistan) to $39,470 per capita per year (Luxembourg). Commerce does not simply average the labor rates prevailing in all these countries. Instead, Commerce determines a linear trend that best fits the data, providing a way to predict the labor rate for a country with any given gross national income. Commerce argues that this method allows the calculation to take into account the generally positive correlation between labor rates and gross national income, while smoothing out the trend so as not to have the calculation overly influenced by outlier countries whose wage rates are abnormally high or low given their national incomes. Given that the gross national income used for China in determining wage rates in this case was $960 per capita per year, Dorbest argues that a calculation using data from countries with widely-varying national incomes ranging as high as $39,470, as required under the Commerce regulation, does not comply with the statutory requirement to use data from only economically comparable countries to the extent possible.
We agree with Dorbest. Where the intent of Congress is clear from the language used in a governing statute, neither the agency interpreting the statute nor this court may interpret the statute in such a way as to deviate
from
Congress’s intent.
Chevron,
467 U.S. at 843, 104 S.Ct. 2778;
Food & Drug Admin. v. Brown & Williamson Tobacco Corp.,
529 U.S. 120, 132, 120 S.Ct. 1291, 146 L.Ed.2d 121 (2000). Here, the statute requires the use of data from economically comparable countries “to the extent possible.” 19 U.S.C. § 1677b(c)(4)(A). This seems to us to be a clear statement that Congress intended to require use of data from eco
nomieally comparable countries except in situations where such data were not available or were irretrievably tainted by some statistical flaw. In promulgating its regulation, however, Commerce has decided to use data from many market-economy countries, regardless of their economic comparability to China, without any finding that data from economically comparable countries were unavailable or otherwise unusable.
The statute requires Commerce to use data from economically comparable countries “to the extent possible.” 19 U.S.C. § 1677b(c)(4)(A). Commerce has not shown that using only these data is impossible.
This is not to say that Commerce could not show in an appropriate situation that using the data Congress has directed Commerce to use is impossible. For instance, there might be cases in which no data from economically comparable market-economy countries are available. In these cases, Commerce would be free to use whatever data it felt were appropriate to use to determine labor rates, presuming that Commerce remained within the bounds of 19 U.S.C. § 1677b(c)(l), which requires Commerce to use the “best available information regarding the values of’ the factors of production. But this is not such a case. Here, there were five market-economy countries with gross national incomes less than that of China and an additional eleven countries with gross national incomes between one and two times that of China. Although we need not resolve which of these countries, or which additional countries, could properly be considered economically comparable to China, some subset of these countries must surely fit the bill.
A similar analysis pertains to Commerce’s use of data from countries that are not significant producers of merchandise comparable to the Chinese wooden bedroom furniture at issue here. The governing statute requires Commerce to use, to the extent possible, data from countries that are “significant producers of comparable merchandise.” 19 U.S.C. § 1677b(c)(4)(B). Dorbest argues that the inclusion in Commerce’s regression analysis of some countries that are not significant producers of such merchandise violates this requirement. We agree. It may be that no country satisfying both criteria in 19 U.S.C. § 1677b(c)(4) — that is, being both economically comparable to China and a significant producer of merchandise comparable to wooden bedroom furniture — exists. If so, though, Commerce has not shown this to be the case, and the inclusion of countries that do not produce comparable merchandise is therefore in error. We also disagree with the United States that Dorbest has waived this argument by failing to raise it below. Dorbest timely raised this argument both before the CIT and before Commerce.
Given that the governing statute requires the use of data from economically comparable market-economy countries that are significant producers of comparable merchandise unless such data are not available, 19 C.F.R. § 851.408(c)(3) does not comply with the statutory requirements. It improperly requires using data from both economically comparable and economically dissimilar countries, and it improperly uses data from both countries that produce comparable merchandise and countries that do not. These improper data are required to be used despite the lack of any finding by Commerce that proper data was unavailable or otherwise unusable. We therefore invalidate the regulation, vacate the portion of the CIT’s judgment affirming Commerce’s use of the regulation, and remand to the CIT with instructions to require Commerce to recalculate the normal value using a method
wholly in compliance with 19 U.S.C. § 1677b(c)(4).
B. Exclusion of Small Companies from Non-Production Factor Calculation
Consistent with its normal practice, Commerce used an average of the publicly available cost ratios for seven Indian surrogate companies to determine the value of SG & A, factory overhead, and profit for the Chinese non-market economy. The CIT remanded, suggesting that Commerce should have excluded the four smallest of the seven companies, because there appeared to be a strong correlation between company size and financial results, and those four companies might be too small to be comparable to the Chinese manufacturers at issue. The CIT did not reverse Commerce’s decision to include these companies, but it required Commerce either to exclude them or to use a regression analysis to defend its decision to include them. On remand, Commerce refused to conduct a regression analysis to determine whether the CIT was correct that company size and financial results were correlated; instead, Commerce simply asserted that there was no relationship between company size and financial ratios. The CIT disagreed with this analysis and once again remanded to Commerce to recalculate the financial ratios without using the four smaller companies or to explain rigorously why including those companies was reasonable. On remand, Commerce continued to disagree with the CIT, arguing that no analysis supported by the available data showed a relationship between company size and financial ratios. Despite this argument, Commerce acquiesced to the CIT’s order and excluded the four smaller companies from its recalculation of surrogate financial ratios. AFMC argues that Commerce’s use of less than the full complement of surrogate companies, coupled with Commerce’s failure to use a regression analysis to support its action, violated the statutory requirement that Commerce use the “best available information” to determine normal values for non-market economies. 19 U.S.C. § 1677b(c)(l).
The governing statute directs Commerce to determine normal value for non-market economies using the “best available information.” 19 U.S.C. § 1677b(c)(l). Commerce is also directed to “utilize, to the extent possible, the prices or costs of factors of production in one or more market economy countries that are ... at a level of economic development comparable to that of the non-market
economy
country, and ... significant producers of comparable merchandise.” 19 U.S.C. § 1677b(c)(4). No party to this litigation has suggested that either approach used by Commerce here — using all seven surrogate Indian companies, or using only the three remaining surrogate companies once the four smallest companies are eliminated — fails to meet the requirements of 19 U.S.C. § 1677b(e)(4). We agree that either approach satisfies the statute’s requirements. In either case, the companies chosen are Indian manufacturers of merchandise similar to the Chinese wooden bedroom furniture at issue. India is a market economy country that is economically comparable to China, and India is a significant producer of merchandise comparable to wooden bedroom furniture. Thus, any selection of Indian furniture companies should comply with the requirements of 19 U.S.C. § 1677b(c)(4), and this statute does not speak directly to the issue here. To the extent that 19 U.S.C. § 1677b(c)(l) speaks directly to this issue, its requirement that Commerce use the “best available information” is ambiguous and unhelpful in resolving the issue.
To select financial statements, Commerce “normally will use nonproprietary information gathered from producers of
identical or comparable merchandise in the surrogate country.” 19 C.F.R. § 351.408(c)(4). To narrow the list of financial statements meeting this criterion, Commerce considers the quality and specificity of the statements, as well as whether the statements are contemporaneous with the data used to calculate production factors. Notably missing from this list is any mention of the size of the surrogate company in comparison to the size of the target non-market economy company or the size of other potential surrogate companies. It is certainly possible that including surrogate companies much smaller than the target company might result in higher-than-warranted financial ratios, because the larger target company might benefit from economies of scale not present in the much smaller surrogate companies. But Commerce explained that its preference is “to use multiple financial statements in order to eliminate potential distortions that may arise from using those of a single producer,” as long as those financial statements “are not distortive or otherwise unreliable.”
Once Commerce’s list of available financial statements was limited by the exclusion of companies whose exclusion is not challenged here, there were seven companies remaining. Indian Furniture Products, Akriti, and Raghbir, whose inclusion survived the CIT’s repeated remands to Commerce, had SG & A ratios of 24.38%, 13.53%, and 10.44%, respectively. The four smaller companies, Nizamuddin, Fusion Design, Swaran, and DnD, had SG & A ratios of 31.51%, 34.39%, 47.30%, and 15.66%, respectively. There certainly is a difference in the average SG & A ratios of these two groups, but there is some overlap, with the highest ratio in the first group (Indian Furniture Products, at 24.38%) being higher than the lowest ratio in the second group (DnD, at 15.66%). Given that the four excluded companies are smaller than the three included ones, the generally higher SG & A ratio for the excluded companies is consistent with SG & A ratio being affected by economies of scale. But we think the CIT went beyond the available evidence when it concluded that, because the average SG & A ratio for the larger companies was lower than the average SG & A ratio for the smaller companies, SG & A ratio must be completely determined by company size in the absence of any mathematically-supported finding by Commerce as to another cause.
If nothing else, the fact that the largest company (Indian Furniture Products) had an SG & A ratio higher than that of the smallest company (DnD) suggests that some other factor or factors beyond just company size must be at work in determining a company’s SG & A ratio. By demanding that Commerce discover and explain what these factors were, merely because there was some evidence to support the hypothetical economy-of-scale theory, the CIT did not give appropriate deference to Commerce’s application of the statute’s requirement that Commerce use the “best available information.” 19 C.F.R. § 351.408(c)(4).
In addition, forcing Commerce to eliminate the four smallest surrogate companies when none of the seven companies used by Commerce was even one-quarter as large as the target Chinese company makes very little sense. If the four smallest companies are too small to be appropriately comparable to Dorbest, then presumably none of the three larger surrogate companies should be compared to Dorbest either. We hold that the CIT erred by requiring Commerce to justify its calculation method or exclude the smallest surrogate companies. While Commerce in this appeal does not challenge the CIT’s decision in this respect, this is also not a situation in which Commerce on remand has voluntarily changed its policy.
See Tung Mung Dev.
Co., Ltd. v. United States,
354 F.3d 1371, 1378-79 (Fed.Cir.2004) (holding that “any error in the remand orders is irrelevant because Commerce’s redetermination decisions represent new, independent agency interpretations”). Rather, it is a situation in which Commerce has simply decided as a matter of litigation strategy not to pursue the matter further. The issue on appeal was properly raised by AFMC. We vacate the judgment of the CIT and remand with instructions to recalculate the financial ratios using Commerce’s original data set of seven surrogate companies.
C. Waived Arguments
Dorbest argued at the CIT that, when calculating normal value, Commerce failed to properly add excise duty expenses as an income item when calculating the surrogate value for profit for one particular surrogate company, Indian Furniture Products. Dorbest had not raised this issue in its administrative case brief before Commerce, although it did raise the issue in a footnote in its rebuttal brief before Commerce
and again
during the ministerial comment period before Commerce’s adoption of its original final determination. The CIT held that Dorbest’s failure to raise this argument in its administrative case brief before Commerce constituted a failure to exhaust administrative remedies and thereby rendered the CIT unable to review the issue on the merits. Dorbest appeals, urging that the CIT erred by holding that it was unable to review this issue due to Dorbest’s failure to exhaust its administrative remedies.
We disagree with Dorbest. Commerce regulations require the presentation of all issues and arguments in a party’s administrative case brief. 19 C.F.R. § 351.309(c)(2) (“The case brief must present all arguments that continue in the submitter’s view to be relevant to the Secretary’s final determination.”). Further, Commerce regulations state that the rebuttal brief is an inappropriate place to raise an issue for the first time. 19 C.F.R. § 351.309(d)(2) (“The rebuttal brief may respond only to arguments in the case briefs.... ”). We have held before that parties are “procedurally required to raise the[ir] issue before Commerce at the time Commerce [is] addressing the issue.”
Mittal Steel Point Lisas Ltd. v. United States,
548 F.3d 1375, 1383 (Fed.Cir.2008). This is because “[s]imple fairness to those who are engaged in the tasks of administration, and to litigants, requires as a general rule that 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.”
United States v. L.A. Tucker Truck Lines,
344 U.S. 33, 37, 73 S.Ct. 67, 97 L.Ed. 54 (1952). Thus, we agree with the CIT that Dorbest’s failure to raise its issue in its administrative case brief constituted a failure to exhaust administrative remedies. The CIT did not err by refusing to consider an issue on which the complaining party had not exhausted its administrative remedies. Accordingly, we affirm that portion of the CIT’s decision holding that Dorbest was precluded from raising its argument that Commerce erred in its treatment of the excise duty expenses of Indian Furniture Products.
In addition, during the administrative process leading to Commerce’s
Remand II
results, AFMC raised for the first time the issue of whether Commerce had committed a clerical error that caused it to use the incorrect value for the expense of a raw material called rubberwood when determining the cost of production in the Chinese non-market economy. AFMC had not raised the issue either during the ministerial comment period at Commerce or in its complaint or earlier briefing to the CIT. Commerce refused to consider the issue during the remand proceedings, and the
CIT affirmed Commerce’s decision, holding that AFMC had waived the issue by not raising it during its multiple opportunities to do so in a timely manner. AFMC appeals, arguing that Commerce abused its discretion by holding that it was unable to review this issue due to AFMC’s waiver.
Commerce certainly has the authority to act to correct ministerial errors in the course of judicial review of the final results of its determinations, though it lacks such discretion after judicial review is completed.
See Am. Signature v. United States,
598 F.3d 816, 826, 827-28 (Fed. Cir.2010) (noting that “Commerce has long claimed the authority to correct ministerial errors during judicial review of final results even when no request to correct the error has been made by an interested party pursuant to the regulations” and holding that “Commerce’s sua sponte corrections must be made before the final determination is no longer subject to judicial review [under 19 U.S.C. § 1516a]”). The issue here is whether Commerce is
required
to correct ministerial errors when they are only pointed out to Commerce in an untimely fashion. AFMC argues that, given Commerce’s duty to accurately calculate antidumping duties, even though Commerce has discretion as to whether or not to correct clerical errors, the failure to correct an error can — and did, in this case — amount to an abuse of that discretion.
AFMC cites several cases as supporting its argument that Commerce’s refusal to correct a clerical error is an abuse of discretion, even when that error is raised late. AFMC first cites
Alloy Piping Products v. Kanzen Tetsu,
334 F.3d 1284, 1293 (Fed.Cir.2003), as holding that, “where an error is apparent on the face of Commerce’s determination or from the calculations underlying that determination, it requires correction after the final determination.” But
Alloy Piping
in fact held that Commerce was not required to correct these errors unless “the respondent ... exhaust[s] its administrative remedies ... [by] applying to Commerce to correct the error within five days of the release of the final calculations or, if an extension is granted, within five days after the publication of the final determination.” 334 F.3d at 1293. Here, AFMC did not so apply to Commerce; instead, AFMC only raised the issue years later after remand (on an unrelated issue) from the CIT to Commerce. Thus,
Alloy Piping
does not help AFMC.
AFMC also cites
Cemex, S.A. v. United States,
133 F.3d 897, 904 (Fed.Cir.1998), as demonstrating that Commerce should “correct[ ] a clear calculation error discovered on remand.” In that case, during a remand, petitioners discovered a clerical error affecting the dumping margin, and the CIT allowed Commerce to correct the error. This court affirmed. But
Cemex
is also distinguishable from the present case. First, the clerical error in
Cemex
was not discoverable until after the ministerial comment period had already elapsed. 133 F.3d at 904. Here, AFMC could have discovered the error before Commerce issued its Final Determination. Thus, AFMC does not have the same excuse that the petitioners in
Cemex
had for failing to raise the clerical error issue in a timely fashion. Second, in
Cemex,
Commerce agreed with the need to fix the clerical error, so Commerce and the CIT exercised the discretion they have to do so. But there is no dispute here that Commerce has discretion to fix the error; instead, the question is whether Commerce’s failure to fix the error is an abuse of that discretion. Because
Cemex
is a case in which Commerce was not forced to fix an error, it does not shed any light on this question. Precisely the same problem afflicts AFMC’s citation to
Hyundai Electronics v. United States,
395 F.Supp.2d 1231 (CIT
2005), where Commerce fixed a late-identified clerical error and the CIT affirmed. The same is true in
Federal-Mogul Corp. v. United States,
872 F.Supp. 1011, 1014 (Ct. Int’l Trade 1994), and in
Aramide Maatschappij v. United States,
901 F.Supp. 353, 361 (Ct. Int’l Trade 1995), as well as in
Koyo Seiko Co. v. United States,
746 F.Supp. 1108 (Ct. Int’l Trade 1990).
AFMC has pointed to no decision of either the CIT or this court that holds that Commerce’s refusal to fix a clerical error in the calculation of an anti-dumping duty margin is an abuse of discretion when the clerical error was discoverable during the original proceedings but was not pointed out to Commerce during the time period specified in the regulations. We have found no such case. While it is certainly true that Commerce has discretion to correct clerical errors when they are discovered during judicial review of its final determinations, discretion implies that Commerce also has authority not to fix clerical errors in these situations. Because of AFMC’s inexcusable untimeliness in raising the issue and its failure to show that Commerce abused its discretion under the circumstances of this case, Commerce’s refusal to fix the error here does not constitute reversible error. Therefore, we affirm the CIT’s holding that Commerce acted appropriately by declining to fix the clerical error that AFMC only pointed out during remand.
III. CONCLUSION
Those portions of the CIT’s decision that affirm Commerce’s refusal to fix the clerical error in the valuation of rubberwood that AFMC brought to Commerce’s attention only on remand are affirmed. Those portions of the CIT’s decision holding that Dorbest was precluded- from raising its argument that Commerce erred in its treatment of the excise duty expenses of Indian Furniture Products because its earliest presentation of the issue was untimely are also affirmed. Those portions of the CIT’s decision that required Commerce either to justify using all seven Indian surrogate companies for calculating non-production factors or to eliminate from the calculations the four smallest surrogate companies are vacated, and the case is remanded to the CIT for further proceedings to recalculate the non-production factors using all seven surrogate companies. To the extent that 19 C.F.R. § 351.408(c)(3) requires or at least permits the use of labor value data from countries that are not economically comparable to the non-market economy country in question or are not significant producers of merchandise comparable to the merchandise in question when data from countries meeting both criteria are available, the regulation is facially invalid as noncompliant with 19 U.S.C. § 1677b(c)(4)(A). Those portions of the CIT’s decision that affirm Commerce’s reliance on 19 C.F.R. § 351.408(c)(3) are vacated, and the case is remanded to the CIT for further proceedings to determine the labor value using a method that complies with the requirements of 19 U.S.C. § 1677b(c)(4)(A).
IV. COSTS
No costs.
AFFIRMED-IN-PART, VACATED-IN-PART, AND REMANDED