DYK, Circuit Judge.
This case presents the question whether the Department of Commerce’s (“Commerce”) regulation providing for the use of entered value of imported merchandise in the assessment rate formula (19 C.F.R. § 351.212(b)(1)) is consistent with the anti-dumping statute (codified in pertinent part at 19 U.S.C. § 1675) and, if so, whether it is reasonable. We conclude that Commerce’s interpretation is not foreclosed by the statute, nor have the appellants shown that the interpretation is unreasonable. We accordingly defer to that interpretation under
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), and we affirm the decision of the Court of International Trade in
Koyo Seiko Co., Ltd. v. United States,
110 F.Supp.2d 934 (Ct. Int’l Trade 2000).
BACKGROUND
This case concerns the determination of antidumping duties for imports of tapered roller bearings with an outside diameter of four inches or less and parts thereof from Japan (“0-4” “TRBs”) by Koyo Seiko Co., Ltd., and its sole United States subsidiary, Koyo Corporation of U.S.A. (collectively, “Koyo”).
Under the antidumping statute, Commerce is required to impose antidumping duties on imported merchandise that “is being, or is likely to be, sold in the United States at less than fair value” to the detriment of a domestic industry. 19 U.S.C. § 1673. Commerce determines those duties by first calculating the “dumping margin” for the subject merchandise,
i.e.,
the total amount by which the price charged for the subject merchandise in the home market (the “normal value”) exceeds the price charged in the United States (the “United States price”). 19 U.S.C. § 1677(35)(A).
The statute obligates Commerce to use the dumping margin as “the basis for the assessment of countervailing or antidumping duties on entries of merchandise covered by the [antidumping] determination and for deposits of estimated duties.” 19 U.S.C. § 1675(a)(2)(C).
Commerce uses the dumping margin to assess antidumping duties on merchandise imported during the review period, and also to calculate “cash deposits of estimated duties for future entries” of the subject merchandise.
Torrington Co. v. United States,
44 F.3d 1572, 1575 (Fed.Cir.1995). A calculational problem arises, however, in the determination of antidumping duties. Although the “dumping margin” is calculated on “sales” during the review period, the duty is imposed upon “entries,”
i.e.,
imports during the review period. Since sales and imports are typically not the same during any particular review period, a method for determining the dumping duty on imports is required. Commerce has adopted two different calculational approaches — one for cash deposits and one for final duties. Commerce requires importers to make cash deposits in an amount based, in pertinent part, on the “estimated weighted average dumping margin” for the merchandise. 19 U.S.C. § 1673b(d)(l)(B). Commerce calculates this “estimated weighted average dumping margin,”
ie.,
estimated duty, by “dividing the aggregate dumping margins determined for a specific exporter or producer [here, Koyo] by the aggregate export prices or constructed export prices of such exporter or producer.” 19 U.S.C. § 1677(35)(B). In other words, the cash
deposit rate is calculated “as a percentage of United States [sales] price.”
Torrington Co. .,
44 F.3d at 1576. This rate is then applied to estimated imports (“estimated entries”).
Commerce has devised a different methodology for use in calculating the final amount of the duties to be imposed on merchandise already imported into the United States. “When an antidumping duty is imposed upon imported merchandise, Commerce calculates an assessment rate for each importer by dividing the dumping margin for the subject merchandise [here, the 0-4” TRBs] by the entered value of such merchandise for normal Customs purposes.”
Koyo Seiko Co.,
110 F.Supp.2d at 938.
This methodology has been codified in a regulation that states, in pertinent part, that: 19 C.F.R. § 351.212(b)(1).
The assessment rate is then applied “uniformly on all entries each importer made during the [period of review.]”
Tapered Roller Bearings and Parts Thereof, Finished and Unfinished, From Japan, and Tapered Roller Bearings, Four Inches or Less in Outside Diameter, and Components Thereof, from Japan: Final Results of Antidumping Duty Administrative Reviews,
63 Fed. Reg. 63,860, 63,875 (Nov. 17, 1998) (“Final Results”). In other words, the assessment rate is calculated “as a percentage of entered value” of the subject merchandise sold in the United States during the review period.
Torrington Co.,
44 F.3d at 1576. That rate is then applied to the merchandise imported (“actual entries”) during that review period.
[T]he Secretary [of Commerce] normally will calculate an assessment rate for each importer of the subject merchandise covered by the review. The Secretary [of Commerce] normally will calculate the assessment rate by dividing the dumping margin found on the subject merchandise examined by the entered value of such merchandise for normal customs duty purposes.
In simple terms, the formula for cash deposit rates uses
sales
during the review period as the denominator; the formula for the final duty uses
imports
as the denominator. Both formulae use sales figures in the numerator. Both apply the formula to imports (“entries”).
Commerce has recognized that the methodology for calculating the final duty (assessment rate) allows Commerce to collect only a “reasonable approximation” of
the duties that would be imposed if it had calculated the specific duties due on particular sales of merchandise imported during the review period. In
Tapered Roller Bearings and Parts Thereof, Finished and Unfinished, From the People’s Republic of China; Preliminary Results of 1996-1997 Antidumping Duty Administrative Review and New Shipper Review,
63 Fed. Reg. 37,339 (July 10, 1998) (“Tapered Roller Bearings”), for example, Commerce stated in pertinent part that:
While [Commerce] is aware that the entered value of sales during the [review period] is not necessarily equal to the entered value of entries during the [review period], use of entered value of sales as the basis of the assessment rate permits [Commerce] to collect a reasonable approximation of the antidumping duties which would have been determined if [Commerce] had reviewed] those sale of merchandise actually entered during the [review period].
Id.
at 37,344.
Commerce’s methodology for calculating the assessment rate was applied in the present case. In November 1997, Commerce initiated an antidumping administrative review of the antidumping duties on 0-4” TRBs covering the period from October 1, 1996, through September 30, 1997.
See Initiation of Antidumping and Countervailing Duty Administrative Reviews and Request for Revocation in Part,
62 Fed.Reg. 63,069 (Nov. 26,1997).
On July 10, 1998, Commerce issued the preliminary results of its administrative review, assigning to Koyo a 7.62% dumping margin for 0-4” TRBs for the review period.
Tapered Roller Bearings and Parts Thereof, Finished and Unfinished, From Japan, and Tapered Roller Bearings, Four Inches or Less in Outside Diameter, and Components Thereof, from Japan: Preliminary Results of Anti-dumping Duty Administrative Reviews,
63 Fed.Reg. 37,344, 37, 348 (July 10, 1998) (“Preliminary Results”).
On November 17, 1998, Commerce issued the final results of its administrative review, once again assigning to Koyo a 7.62% dumping margin for 0-4” TRBs for the review period. Final Results, 63 Fed. Reg. at 63,875. In those Final Results, Commerce stated in pertinent part that:
In accordance with 19 CFR 351.212(b)(1), we will calculate importer-specific
ad valorem
assessment rates for the merchandise based on the ratio of the total amount of antidumping duties calculated for the examined sales made during the [period of review] to the total customs value of the sales used to calculate those duties
[i.e.,
the entered value]. This rate will be assessed uniformly on all entries each importer made during the [period of review].
Id.
at 63,875 (emphasis in original). In other words, Commerce used the
entered value
of the subject merchandise sold during the period of review as the denomina
tor when calculating an importer-specific assessment rate. The issue on appeal is whether it was proper for Commerce to do so, or whether it was required to use the
sales value
of merchandise imported during the review period in the denominator, as is done for the cash deposit calculation.
Koyo appealed the assessment rate imposed by Commerce to the Court of International Trade, first arguing that Commerce’s methodology contravened the plain terms of the antidumping statute. The court disagreed, concluding that “neither the statute nor its legislative history provides an ‘unambiguously expressed intent’ with regard to the precise question at issue,” namely, the designation of a particular denominator for the assessment rate formula.
Koyo Seiko Co.,
110 F.Supp.2d at 940 (citing
Chevron,
467 U.S. at 843, 104 S.Ct. 2778).
Koyo next argued that Commerce’s use of the entered value was unreasonable. Koyo noted that the numerator of the assessment rate — the dumping margin — is calculated using two sales values, namely, the normal value of the subject merchandise and its United States price. Koyo argued that because the numerator is based on sales values, so too must the denominator be based on sales value, namely, the United States price of the merchandise. The Court of International Trade disagreed, noting in part that “the entered value of the subject merchandise is an adjusted purchase price of the merchandise, clearly a sales-related concept.”
Id.
The court also rejected two other arguments regarding the alleged unreasonableness of Commerce’s methodology. First, the court rejected Koyo’s contention that the assessment rate calculation was meant to allocate the dumping margin among different importers, and that its use of a single importer in the United States accordingly rendered the assessment rate formula inapplicable to its imports. The court noted that Commerce’s adoption of the entered value as the denominator for the assessment rate calculation ensured “administrative ease, accuracy, promptness and efficiency” as well as a distribution of the dumping margin among the responsible importers.
Id.
at 942, 104 S.Ct. 2778. The court further noted that “Commerce is not expected to cater to Koyo’s private circumstances based on Koyo’s preferences in doing business, its corporate structure or business needs.”
Id.
at 942, 104 S.Ct. 2778.
Second, the court rejected Koyo’s argument that the methodology is unreasonable because it differed from the methodology used to calculate cash deposit rates. Pointing to our decision in
Torrington Co. v. United States,
44 F.3d 1572 (Fed.Cir.1995), the Court of International Trade noted that this court has already held that the antidumping statute “does not require the same method of calculation for assessment rates and cash deposit rates.”
Koyo Seiko Co.,
110 F.Supp.2d at 942 (quoting
Torrington Co. .,
44 F.3d at 1578). It accordingly sustained Commerce’s methodology as a reasonable interpretation of the antidumping statute.
Id.
at 943. This timely appeal followed.
DISCUSSION
I
This court has jurisdiction over this appeal pursuant to 28 U.S.C. § 1295(a)(5). “When reviewing anti-dumping determinations made by Commerce, this court applies anew the standard of review applied by the Court of International Trade in its
review of the administrative record.”
F.LLI De Cecco Di Filippo Fara S. Martino S.p.A. v. United States,
216 F.3d 1027, 1031 (Fed.Cir.2000). In doing so, we uphold Commerce’s determination unless it is “unsupported by substantial evidence on the record, or otherwise not in accordance with law.” 19 U.S.C. § 1516a(b)(1)(B)(i).
We review questions of statutory interpretation without deference.
U.S. Steel Group v. United States,
225 F.3d 1284, 1286 (Fed.Cir.2000). In reviewing an agency’s construction of a statute that it administers, this court addresses two questions, as required by the Supreme Court’s decision in
Chevron U.S.A., Inc. v. Natural Resources Defense Council, Inc.,
467 U.S. 837, 842-43,104 S.Ct. 2778, 81 L.Ed.2d 694 (1984). The first question is “whether Congress has directly spoken to the precise question at issue.”
Id.
at 842, 104 S.Ct. 2778. If so, this court and the agency “must give effect to the unambiguously expressed intent of Congress.”
Id.
at 843, 104 S.Ct. 2778. If, however, Congress has not spoken directly on the issue, this court addresses the second question of whether the agency responsible for filling a gap in the statute has rendered an interpretation that “is based on a permissible construction of the statute.”
Id.; see also United States v. Mead Corp.,
— U.S. —, 121 S.Ct. 2164, 2185, 150 L.Ed.2d 292, (2001);
Micron Tech., Inc. v. United States,
243 F.3d 1301, 1308 (Fed.Cir.2001).
In antidumping cases, this court has repeatedly recognized “Commerce’s special expertise,” and it has “accord[ed] substantial deference to its construction of pertinent statutes.”
Micron Tech., Inc. v. United States,
117 F.3d 1386, 1394 (Fed.Cir.1997).
II
There is certainly nothing in the statute that requires Commerce to use Koyo’s proposed methodology, or forbids Commerce from using the methodology that it does use. The statute states, in pertinent part, that the amount by which normal value exceeds the United States price (the dumping margin) “shall be the basis for the assessment of ... antidump-ing duties on entries of merchandise.... ” 19 U.S.C. § 1675(a)(2)(C). Commerce uses the dumping margin as the numerator of the assessment rate formula to comply with section 1675’s requirement that the dumping margin must serve as the “basis” for the assessment of antidumping duties.
See
19 C.F.R. § 351.212(b)(1).
Pointing to section 1675, Koyo argues that the numerator of the assessment rate formula is calculated using two sales values of the subject merchandise: its sales value in the home market (normal value) and the corresponding sales value in the United States (the United States price).
See
19 U.S.C. § 1675(a)(2)(A);
see also
19 U.S.C. §§ 1677a(b), 1677b(a). Koyo urges that because the numerator is based on the sales value of the subject merchandise, so too must the denominator be based on sales value rather than entered value. We are unable to find any such requirement in the statute, and we implicitly rejected Koyo’s argument in our earlier decision in
Torrington Co. v. United States,
44 F.3d 1572 (Fed.Cir.1995), where we observed that the antidumping statute does not “specify a particular divisor when calculating either assessment rates or cash deposit rates. Rather, the statute merely requires that ... the difference between foreign market value and United States price serve[s] as the basis” for the assessment rate.
Id.
at 1578. The methodology at issue here is accordingly entitled to deference under
Chevron
if it is based on a reasonable interpretation of the antidump-ing statute.
Commerce’s methodology was adopted after notice-and-eomment rulemaking, and thus is entitled to maximum deference. In
United States v. Mead Gorp.,
— U.S. —, 121 S.Ct. 2164, 150 L.Ed.2d 292 (2001), for example, the Supreme Court made clear that:
[An] administrative implementation of a particular statutory provision qualifies for
Chevron
deference when it appears that Congress delegated authority to the agency generally to make rules carrying the force of law, and that the agency interpretation claiming deference was promulgated in the exercise of that authority.
Id.
at 2171. The Court also emphasized in turn that:
It is fair to assume generally that Congress contemplates administrative action with the effect of law when it provides for a relatively formal administrative procedure tending to foster the fairness and deliberation that should underlie a pronouncement of such force. Thus, the overwhelming number of our cases applying
Chevron
deference have reviewed the fruits of notice-and-comment rule-making or formal adjudication.
Id.
at 2172 (internal citations omitted).
Similarly, in
United States v. Haggar Apparel Co.,
526 U.S. 380, 119 S.Ct. 1392, 143 L.Ed.2d 480 (1999), the Supreme Court held that the government’s customs classifications regulations are entitled to
Chevron
deference because, in part, the agency responsible for promulgating the regulations “utilized the notice-and-comment rulemaking process before issuing the regulations....”
Id.
at 390, 119 S.Ct. 1392. The appellants bear the burden of showing that the agency’s approach is arbitrary or otherwise unreasonable.
The appellants have failed to carry that burden, and thus we cannot conclude that Commerce’s methodology is unreasonable. There is some logic to Koyo’s contention that the sales value for the imported merchandise, rather than its entered value, should have been used in the denominator of the assessment rate formula. As noted above, the numerator of that formula — the dumping margin — is calculated by subtracting the United States sales price for the subject merchandise from the normal value of corresponding sales in the country of origin, here, Japan, during the review period. Calculating the dumping margin using sales during the review period in the denominator accordingly would make some sense. The assessment rate per unit would be more accurately calculated by using the same sales base in the numerator and denominator. Commerce’s suggestion that its methodology serves the interests of administrative convenience is hardly persuasive. After all, Commerce already uses the sales value of the merchandise in the numerator, and indeed also uses that figure when calculating the cash deposit rate.
We also disagree with Commerce’s suggestion that the approach Koyo urges is a variation on the previous master list assessment method, which Commerce abandoned as impractical when it adopted the methodology at issue. The master list method required Commerce as part of the calculation of the assessment rate to list “the appropriate amount of duties to assess
for each entry of subject merchandise separately
.... “
Antidumping Duties; Countervailing Duties";
Final Rule, 62 Fed.Reg. 27,296, 27,314 (May 19, 1997) (emphasis added). In other words, that method required Commerce, when calculating the assessment rate, to link specific entries of imported merchandise to the
subsequent sale of that merchandise in the United States. In contrast, the approach urged by Koyo does not require such entry-by-entry comparison. Nor is there any reason to believe that Commerce’s methodology ensures a more accurate result than Koyo’s proposed methodology.
If the assessment rate were applied to a base of CEP sales, we would likely hold that Commerce’s methodology is arbitrary, since an accurate assessment rate using sales figures would recover the dumping margin, and no more and no less. But the assessment rate is not applied to sales. Congress has expressly provided that the assessment rate is to be applied to “entries of merchandise” for the review period. 19 U.S.C. § 1675(a). Calculating the duties on entered value will, as Commerce admits, necessarily result in collection of either more or less than the total dumping duties.
See Tapered Roller Bearings,
63 Fed.Reg. at 37,344 (noting that the current methodology only allows Commerce “to collect a reasonable approximation of the antidumping duties”). And since the assessment rate is applied to entries and not sales, there is at least a certain symmetry in using entered value as the denominator. A high level of deference is paid to Commerce’s interpretations, and Commerce’s methodology is not rendered unreasonable simply because it lacks a particularly compelling logic, or because Commerce has been unable to explain to us why the methodology serves a particular goal of the antidumping statute. As noted above, the burden is on the party challenging Commerce’s interpretation of the antidumping statute to show that that interpretation is unreasonable, and appellants have failed to meet their burden.
Appellants’ most plausible argument is that Commerce’s methodology is invalid because Commerce has used a different methodology for cash deposits. But we have already held that this difference in approach is permissible, as the antidump-ing statute “does not require the same method of calculation for assessment rates and cash deposit rates.”
Torrington Co.,
44 F.3d at 1578.
Finally, Koyo urges that the assessment rate methodology is designed to allocate the dumping margin among multiple importers, and that it accordingly should not be used where, as here, the foreign exporter’s merchandise is imported into the United States by a single importer. But the fact that Koyo uses only a single importer in the United States hardly renders Commerce’s methodology unreasonable. Indeed, as the Court of International Trade observed in the decision on appeal here, “[i]t would be unreasonable, if not anomalous, for Commerce to devise an assessment rate formula for importers enjoying exclusivity with manufacturers different from the formula applied to all other importers. This approach would allow manufacturers and importers to manipulate assessment rates by changing the number of importers.”
Koyo Seiko Co.,
110 F.Supp.2d at 942.
CONCLUSION
This is a close case, but we are not convinced that Koyo has shown Commerce’s methodology for the calculation of importer-specific assessment rates to be contrary to the antidumping statute or unreasonable. We accordingly affirm the decision of the Court of. International Trade.
AFFIRMED.
COSTS
No costs.