Böwe Passat Reinigungs-Und Wäschereitechnik GmbH v. United States

OPINION POGUE, Judge: Plaintiff Bowe-Passat (“Bowe”), a German manufacturer of dry cleaning machinery, moves for judgment on the administrative record. See USCIT R. 56.2. Bowe asks the Court (1) to vacate both the final results of a United States Department of Commerce (“Commerce”) administrative review 1 and a corresponding remand decision dated August 5, 1993, and (2) to remand this case to the agency for further action. The Court has jurisdiction pursuant to 28 U.S.C. § 1581(c) and section 516A(a)(2) of the Tariff Act of 1930, 19 U.S.C. § 1516a(a)(2) (1988). Background Commerce’s administrative review 2 imposed an antidumping duty on Bowe’s dry cleaning machinery exported from Germany. See Drycleaning Machinery from Germany, 56 Fed.Reg. 66,838 (Dep’t Comm. Dec. 26, 1991) (final results admin, review). After Commerce published notice of its initiation of the administrative review, 3 Bowe submitted a *1140 compendium of documents (“compendium”) at a hearing held on October 23, 1991. The compendium detailed certain expenses that Bowe wanted subtracted from Commerce’s calculation of the foreign market value of its dry cleaning machinery, either as “level of trade” or “circumstances of sale” adjustments. In response, Commerce rejected the compendium as untimely; Commerce’s final determination rejected each of Bowe’s claimed expense adjustments to foreign market value. Drycleaning Machinery from Germany, 56 Fed.Reg. 66,838 (Dep’t Comm. Dec. 26, 1991) (final results admin, review). The final determination also addressed Bowe’s other objections: the use of constructed value rather than third country sales in calculating foreign market value, the inclusion of a discounted trade show machine in calculating United States price, and the failure to offset calculated positive dumping margins with negative margins. 56 Fed.Reg. at 66, 839-66, 840. In its subsequent action in the U.S. Court of International Trade, Bowe moved to include the compendium as part of the administrative record. The U.S. Court of International Trade (Musgrave, J.) so ordered, 4 and remanded the case to Commerce for redetermination based on the information contained in the compendium. Bowe-Passat v. United States, 17 CIT 335, 1993 WL 179269 (1993). Commerce issued its redetermination on August 5, 1993, and did not change its calculation of foreign market value. Remand at 3. Commerce reasoned that the information in the compendium did not fully satisfy Bowe’s burden of proof for the claimed circumstances of sale or level of trade adjustments. Remand at 4. Commerce stated that the information contained in the compendium, while providing adequate detail of expenses, did not adequately justify the claimed expenses. Remand at 7. Commerce, however, corrected certain ministerial errors which reduced the antidumping margin from .64 percent to .59 percent. Id. at 13. Both parties filed comments with the court on the remand determination. In an order dated April 21, 1994 the Court of International Trade (Musgrave, J.) ordered the parties to submit all substantive claims and prayers for relief in the form of a single motion for judgment on the agency record. In its motion Bowe argues that “if Commerce had accepted almost any of the data in the compendium, or had corrected any of the other issues contained in Plaintiffs Complaint ..., it is likely that the [.59 percent antidumping] margin would disappear, or at least drop below the de minimis level of .50 percent.” 5 Bowe’s motion and Commerce’s response present four issues: (1) Whether Commerce properly refused to make circumstances of sale adjustments for certain expenses: advertising, headquarters sales, order entry and control, technical publications, traffic shipment, sales administration and management, and legal and finance; or level of trade adjustments to foreign market value for certain expenses: advertising, headquarters sales, order entry and control, traffic shipment, and legal and finance? (2) Whether Commerce properly utilized constructed value methodology, as opposed to third country sales, in calculating foreign market value? (3) Whether Commerce properly included the sale of one of Bowe’s machines, imported for a trade show and subsequently sold at a reduced price, in calculating United States price? (4) Whether Commerce’s calculation of United States price properly assigned a less than fair value amount of zero to United States sales of subject merchandise which were at or above fair value? *1141 Discussion In reviewing Commerce's final antidumping determinations, the Court of International Trade decides whether they are supported by substantial evidence and in accordance with law. Section 516A(b)(1)(B)(i) of the Tariff Act of 1930, 19 U.S.C. § 1516a(b)(1)(B)(i) (1994). Substantial evidence is "such relevant evidence as a reasonable mind might accept as adequate to support a conclusion." Consolidated Edison Co. v. NLRB, 305 U.S. 197, 229, 59 S.Ct. 206, 217, 83 L.Ed. 126 (1938); Universal Camera Corp. v. NLRB, 340 U.S. 474, 477, 71 S.Ct. 456, 459, 95 L.Ed. 456 (1951); Matsushita Elec. Indus. Co., Ltd. v. U.S., 750 F.2d 927, 933 (Fed.Cir.1984). "[T]he possibility of drawing two inconsistent conclusions from the evidence does not prevent an administrative agency's finding from being supported by substantial evidence." Consolo v. Federal Maritime Comm'n, 383 U.S. 607, 619-20, 86 S.Ct. 1018, 1026, 16 L.Ed.2d 131 (1966). When Commerce's interpretation of the antidumping statute is challenged, this court applies the two step analysis set forth in Chevron, U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 842-43, 104 S.Ct. 2778, 2781-82, 81 L.Ed.2d 694 (1984): Using the traditional tools of statutory construction the court ascertains whether congressional intent on the disputed issue is clear, and, if clear, the court applies the statute in the manner Congress intended, regardless of the agency's position. 6 If the statute is ambiguous, the court, rather than interpreting the statute anew and rendering its own interpretation, must defer to an administrative agency's "permissible construction of the statute," 7 whether that construction manifests itself in the application of the statute, see, e.g., Daewoo Electronics Co., Ltd. v. International Union of Electronic, Elec., Technical, Salaried and Mach. Workers, 6 F.3d 1511, 1516 (Fed.Cir.1993), cert. denied, — U.S. —, 114 S.Ct. 2672, 129 L.Ed.2d 808 (1994), or in the promulgation of a regulation, see, e.g., Smith-Corona Group v. United States, 1 Fed.Cir. (T) 130, 136, 713 F.2d 1568, 1575 (1983), cert. denied, 465 U.S. 1022, 104 S.Ct. 1274, 79 L.Ed.2d 679 (1984). 1. Adjustments to Foreign Market Value Foreign market value 8 and United States price 9 determine whether dumping exists and ultimately, the amount of the antidumping duty. 10 Commerce adjusts its foreign market value and United States price calculations to account for factors unrelated to dumping that might distort the dumping margin. At issue in this case are level of trade 11 and circumstances of sale 12 adjustments to foreign market value. A. Level of Trade Adjustments to Foreign Market Value The statute in effect for the review period did not specifically provide for adjustments *1142 for differences in levels of trade. 13 Commerce, however, promulgated a regulation stating that price comparisons normally will be made at the same level of trade, and granting the Secretary the authority to adjust for differences affecting price comparability between the U.S. and foreign markets. Section 353.58 (1989) of the regulations stated: The Secretary normally will calculate foreign market value and United States price based on sales at the same commercial level of trade. If sales at the same commercial. level of trade are insufficient in number to permit an adequate comparison, the Secretary will calculate foreign market value based on sales of such or similar merchandise at the most comparable commercial level of trade as sales of the merchandise and make appropriate adjustments for differences affecting price comparability. 19 C.F.R. § 353.58 (1989). Bowe claims entitlement to level of trade adjustments because its U.S. sales were made to distributors at substantial discounts when compared with its home market sales to end users. Bowe alleges that the various functions which a distributor performs account for the difference in price. Specifically, Bowe claims Commerce should have made level of trade adjustments for the following-selling costs: advertising, headquarters sales, order entry and control, 14 traffic shipment, and legal and finance. For the review period in question, Commerce “allowed bad debt and indirect sales office expenses as level of trade adjustments.” Drycleaning Machinery From Germany, 56 Fed.Reg. 38, 112, 38,113 (Dep’t Comm. Aug. 12 1991) (prelim, results admin, review). Commerce disallowed Bowe’s other level of trade claims because Commerce did not believe them to be a measure of level of trade differences in this case. Id. On remand, Commerce rejected the claimed adjustments, stating: In this instance, Bowe sells at the distributor level of trade in the United States and at the end-user level of trade in the home market, but has no distributor level in the home market. Bowe has not provided an adequate explanation as to why the claimed expenses would not be incurred if it did sell to distributors in the home market. Rather, they merely attempt to quantify the alleged differences between level of trade. Remand at 5 (emphasis added). Commerce also denied the adjustments because Bowe relied on estimations in calculating the claimed adjustments. The Court turns first to the type of proof Commerce requires of Böwe. Böwe submitted data in its compendium which indicated that the differences in prices between the two markets was due in part to different levels of trade. Commerce found the data insufficient to provide a basis for the claimed adjustments because Böwe did not submit data for its own sales to distributors in the *1143 home market. 15 To require Böwe to submit data for sales to distributors in the home market, however, is to preclude Böwe from ever obtaining a level of trade adjustment because it does not sell to any distributors in the home market. Consequently, such a requirement renders Böwe powerless to produce evidence to satisfy the burden of proof that Commerce has laid down for these adjustments in this case. This requirement is not in the regulation; nor does Commerce provide a reasoned basis for it. See NEC Home Elecs. v. United States, 54 F.3d 736, 745 (Fed.Cir.1995) (holding that burden imposed to prove a level of trade adjustment was unreasonable because party could, under no practical circumstances, meet the burden); American Permac, Inc. v. United States, 12 CIT 1134, 703 F.Supp. 97 (1988). Moreover, if such a burden had been applied uniformly in this case, Commerce would not have made level of trade adjustments for bad debt expense and indirect sales office expenses, Drycleaning Machinery From Germany, 56 Fed. Reg. 38,112, 38,113 (Dep't Comm. Aug. 12 1991) (prelim. results admin. review); presumably, no level of trade adjustment would have been made at all. Commerce offers no rationale for this inconsistency. Reviewing a prior administrative determination involving the antidumping order in this case, the Court of International Trade discussed the appropriateness of level of trade adjustments for a foreign manufacturer (Bowe) that sells only to end-users in the home market and only to distributors in the U.S. market. American Permac, Inc. v. United States, 12 CIT 1134, 703 F.Supp. 97 (1988). In Permac, Bowe submitted detailed data in support of its claimed level of trade adjustments. Commerce rejected the claims on the ground that they were not adequately quantified. Commerce explained its rejection of a level of trade adjustment—despite the detailed cost data provided by Bowe measuring its distribution-related expenses— stating: Commerce has consistently rejected comparisons between cost of sales in the U.S. market and cost of sales in the home market as support for claimed adjustments for differences in levels of trade because they do not demonstrate that the differences in selling costs are due only to different levels of trade. The differences between the costs in the U.S. market and the costs in the home market are too disparate to attribute the differences in selling costs only to levels of trade. Thus, a level of trade adjustment can be measured accurately only when sales occur in the home market at two separate levels of trade. Only in that situation ... the number of variables that could affect selling price are reasonably minimized and the cost differences attributable to level of trade adequately isolated. American Permac, 12 CIT 1134, 1137, 703 F.Supp. 97. The defendant asserted that plaintiffs had failed to meet their burden of proof to establish an entitlement to a level of trade adjustment. Id. The Court held that the defendant’s reasons for denying the level of trade adjustments were “unsupported by substantial evidence on the record, not reasonable, and not in accordance with the law.” Id. The court stated: The Court can find nothing in the language of either the statute or the regulations that would require plaintiffs to trace precisely and conclusively the exact level of impact the difference in the levels of trade might have on their home market prices. To the contrary, 19 U.S.C. Sec. 1677b(a)(l) contains an economic presumption that certain major differences in commercial terms are bound to distort the price comparison. The implementing regulation, Sec. 19 C.F.R. 353.58, therefore, properly requires that the price comparison be made “at the same commercial level of trade,” and that adjustments be made if the comparison is not made at the same levels of trade. The Court does not find the burden of proof imposed on plaintiffs in this case to be reasonable. The ITA’s requirement *1144 that plaintiffs provide the actual price information with regard to home market sales to distributors, which the ITA knew did not exist, is clearly unreasonable, where the very absence of those sales is the legal prerequisite for the level of trade adjustment. The estimation of this adjustment is implicitly required by law, which prescribes this adjustment precisely because no actual sales at a distributors’ level of trade existed in Germany. The burden of proof imposed by the ITA in this case, and the ITA’s refusal to use any data other than actual, but nonexisting, wholesales prices traps plaintiffs in a vicious circle and is beyond any standard of reasonableness. American Permac, 12 CIT 1134, 1138, 703 F.Supp. 97,101. As in Permac, Commerce’s reasons for denying the claimed level of trade adjustments in the present case effectively preclude Bowe from obtaining a level of trade adjustment. Second, the Court turns to Commerce’s response to the evidence in the record. Commerce denied the adjustment for advertising expense because the data Bowe submitted “merely attempted to quantify the alleged differences between level of trade.” Remand at 7. For order entry and control expenses, Commerce disallowed the adjustment because “it [was] unclear whether Bowe would have incurred most or all of these expenses if it had sold through distributors instead of directly to end-users in the home market____ Bowe merely estimate[d] [a certain] percent of the expenses as an appropriate reduction.” Remand at 8. The Court has searched the record to ascertain why the estimations provided by Bowe are unacceptable. The Court can find no record evidence and Commerce has advanced no reason why the estimations provided by Bowe are suspect. The Court accordingly must remand the case to Commerce for reconsideration of the disallowance of the level of trade adjustments for advertising and order entry and control; these determinations are unsupported by substantial evidence on the record. Commerce’s suspicion about the nature of Bowe’s other claimed level of trade adjustments for headquarters sales department, traffic shipment, and legal and finance may be well-founded. In the absence of record evidence supporting Commerce’s reasons for denial, however, the Court must remand the case with instructions to substantiate the denial of these adjustments on the basis of record evidence. B. Circumstances of Sale Adjustments to Foreign Market Value The antidumping statute permits Commerce to make allowances for differences in the circumstances of sale when it is established to Commerce’s satisfaction that the amount of any price differential between the foreign and United States markets is wholly or partly due to differences in the circumstances of sale. 19 U.S.C. § 1677b(a)(4)(B) (1988); 16 19 C.F.R. 353.56 (1989). 17 “The *1145 statute does not expressly limit the exercise of the Secretary’s authority to determine adjustments, nor does it include precise standards or guidelines to govern the exercise of that authority. Additionally, the statute does not define the term “circumstance of sale” nor does it prescribe any method for determining allowances. Congress has deferred to the Secretary’s expertise in this matter.” Smith-Corona, 1 Fed.Cir. (T) at 137, 713 F.2d 1568; see also Zenith Elecs. Corp. v. United States, 77 F.3d 426, 431 (Fed.Cir. 1996). Commerce has limited the allowance to those circumstances of sale that bear a direct relationship to the sales compared. 19 C.F.R. § 353.56(a) (1989). 18 The regulation states that Commerce will make allowances for selling costs “normally only to the extent that such costs are assumed by the producer or reseller on behalf of the purchaser from that producer or reseller.” 19 C.F.R. § 353.56(a)(2) (1989) (emphasis added). For instance, if an exporter sells to distributors who then sell to end users, that exporter may properly claim an adjustment for the selling costs directed at the end users (or customers of the distributors). Commerce distinguishes between selling costs that are aimed at inducing the sale of a specific product, resulting in expenses directly related to that sale, and selling costs that are not aimed at inducing the sale of a specific product. The limitation for selling costs—designed to insure that the costs are sales-related—evolved to curb the abusive practice of parties attempting to convert every expense, whether direct or not, into a circumstances of sale adjustment to foreign market value. See Zenith, 77 F.3d 426, 432. 19 By requiring parties to establish that selling costs arose as part of a transaction and that were incurred to induce that transaction, Commerce created a methodology for insuring that the claimed selling costs were direct expenses traceable to specified sales. Id. The Court recognizes that the practice necessarily reduced the likelihood that Commerce would make adjustments based on dubious claims of direct selling costs. Bowe claimed entitlement to circumstances of sale adjustments because its sales to distributors in the U.S. were made at substantial discounts when compared with its sales to end users in the home market. Bowe alleges that the various functions which a distributor performs account for the difference in price. For these reasons the claimed circumstances of sale adjustments for certain selling costs 20 are largely duplicative of the claimed level of trade adjustments. 21 Bowe advanced claims for advertising, headquarters sales, order entry and control, technical publications, traffic shipment, sales administration and management, and legal and finance expenses. Salaries constituted a significant part of these costs. Bowe attempted to establish the requisite directness between the sales and salaries by parsing its employees’ hours into tasks performed in the *1146 home and foreign markets. For instance, the compendium data indicated that a large portion of the activities of employees in the order entry and control department was spent on orders in the home market which accounts for a comparatively smaller portion of Bowe’s sales. Similar data was submitted for the headquarters sales and traffic shipment departments. Commerce was not satisfied that the salary expenses led directly to specified sales. Commerce concluded that these expenses were overhead expenses— expenses that would have been incurred regardless of whether a specific sale was made (or alternatively, expenses that did not arise because of specific sales of the product). As overhead, Commerce treated the selling costs as indirect expenses and allowed an adjustment for the claimed categories of expenses to the extent that they did not exceed indirect selling expenses for exporter’s sales price transactions. See 19 C.F.R. § 353.56(b)(2). (1989). Based on the regulation, Commerce denied the requested adjustments because Bowe did not incur the selling costs on behalf of its purchasers. 22 Remand at 4-5. Commerce denied Bowe’s claim for advertising expense for the same reason. Remand at 6. The record evidence supports these conclusions leading the Court to conclude that Commerce’s rejection of Bowe’s claimed circumstance of sale adjustment was consistent with the regulation and therefore appropriate. See Zenith, 77 F.3d 426, 432 (1996). Bowe also submitted data claiming that expenses incurred for traffic shipment, sales administration and management, and legal and finance should be treated as direct expenses. Bowe has acknowledged that Commerce’s practice is to treat such expenditures as indirect but nevertheless requests the Court to require Commerce to change its practice in this case. The Court believes that Commerce’s application of the statute and regulation to these types of expenditures is reasonable. The record supports Commerce’s conclusion that these expenditures were of a general nature and not linked to specific sales. 2. Commerce’s Use of Constructed Value Methodology Bowe challenges Commerce’s utilization of constructed value instead of third country sales in calculating foreign market value for the P200 model dry cleaning machine. 23 The P200 machine is not sold in the home market because the machine does not meet German environmental standards. Other Bowe dry cleaning models are sold in the home market. The statute in effect at the time of the review specifies that foreign market value shall be the price of such or similar merchandise sold in the home market unless the quantity sold in the home market is so small as to form an inadequate basis for comparison. 19 U.S.C. § 1677b(a) (1988). 24 To determine whether the quantity of such or similar merchandise sold for home consumption is so small as to be inadequate, Commerce conducts a “viability” test of home market sales. 25 If the home market is viable, Com *1147 merce then attempts to find a comparable home market model against which to “match” each U.S. model sold. In the absence of a comparable home market model, Commerce calculates foreign market value based on constructed value, in accordance with 19 U.S.C. § 1677b(a)(2) (1988). 26 When the home market sales are inadequate and therefore not viable, Commerce normally will use third country sales pursuant to a regulatory preference. 19 C.F.R. § 353.48; see Certain Cut Flowers from Costa Rica, 52 Fed.Reg. 6852 (1987) (final admin, review). For situations in which the home market for such or similar merchandise is determined to be viable, but there are no home market sales of a particular model, Commerce uses constructed value to determine foreign market value, a methodology it has used where third country sales are present. See, e.g., Certain Small Business Telephone Systems and Subassemblies Thereof From Japan (“SBTs from Japan”), 57 Fed.Reg. 4949, 4949-4950 (1992) (final admin, review). On the basis of the information that Bowe provided, Commerce concluded that the German market was viable for the P200 dry cleaning machine. Final Results, 56 Fed. Reg. at 66,839; Preliminary Results, 56 Fed. Reg. at 38,113. Similar merchandise was sold in sufficient quantities to satisfy the viability test. Commerce subsequently determined, however, that there were no home market sales of the P200 model. Final Results, 56 Fed.Reg. at 66,839. The statute is clear that Commerce may use constructed value when (1) sales of such or similar merchandise in the home mai’ket are adequate and (2) the foreign market value cannot be determined using home market sales (because, e.g., there are no home market sales of the precise merchandise in question). Here, both conditions were met, and Commerce’s use of constructed value for the P200 machine was therefore appropriate. 3. Inclusion of Trade Show P200 in Calculation of United States Price For the administrative review in issue, Commerce’s calculation of United States price included the sale of a Model P200 machine that was imported for a trade show and subsequently sold at a discount. 27 Plaintiff challenges the inclusion of this machine in United States price and argues that it should have been excluded entirely from the calculation as unrepresentative and outside the ordinary course of trade, or in the alternative, that it warranted a circumstances of sale adjustment to reflect the particular events surrounding the sale. The statute defining foreign market value for the review period in question provides that only those sales “made in the ordinary course of trade” are included in the calculation. 19 U.S.C. § 1677b(a)(l)(A) (1988). Commerce is therefore required to exclude sales not made in the ordinary course of trade from its calculation of foreign market value. There is no parallel “ordinary course of trade” provision for the United States price component of the antidumping equation. See 19 U.S.C. § 1677a (1988). The statute and the underlying regulations are silent about excluding sales from United States price that are not made in the ordinary course of trade. Id.; 19 C.F.R. § 353.41 (1990). In IPSCO, Inc. v. United States, 12 CIT 384, 687 F.Supp. 633 (1988), the court interpreted the provision for United States price *1148 after considering the provision for foreign market value and concluded that the administering authority is not required to exclude sales made outside the ordinary course of trade from United States price. The court stated: Both the statute and the ITA’s regulations contain a detailed methodology for calculating foreign market value and United States price. Congress has provided for numerous adjustments, allowances and exclusions on each side of the fair value equation in order to insure the fairest possible comparison between markets. In light of-this detailed framework, the court may assume that if Congress intended to require the administering authority to exclude all sales made outside “the ordinary course of trade” from its determination of United States price it could have provided for such an exclusion in the definition of United States price, as it has in the definition of foreign market value. It has not done so. Id. at 394, 687 F.Supp. 633. See also Floral Trade Council v. United States, 15 CIT 497, 508 n. 18, 775 F.Supp. 1492 (1991) (“As the court has made quite clear, regular exclusion of sales not in the ordinary course of trade only occurs on the home-market side of the price comparison.” The court further stated in a footnote that “United States sales both in and out of the ordinary course of trade are included in the USP calculations.”). This Court agrees that Commerce’s interpretation of the United States price provision of the antidumping statute to include all United States sales whether in or out of the ordinary course of trade is a permissible construction of the statute. Accordingly, Commerce’s inclusion of Bowe’s trade show machine in calculating United States price was appropriate. The resolution of this issue does not require the court to express an opinion as to whether Commerce has discretion to exclude certain nonrepresentative sales from United States price. See IPSCO, 12 CIT 384, 395-96, 687 F.Supp. 633 28 and cf. American Permac, Inc., v. United States, 16 CIT 41, 783 F.Supp. 1421 (1992). 29 This Court reserves *1149 judgment on whether Commerce has discretion under the statute to exclude nonrepresentative sales from United States price and what the contours of that discretion should be. That issue will arise when Commerce’s decision to exclude a nonrepresentative sale is challenged before the court and the court is called upon to decide whether such an application of the statute constitutes a permissible construction. Resolution of this case only requires that Commerce’s decision to include all sales made on the United States price side of the antidumping equation is a permissible construction of the anti-dumping statute. The Court believes that it is. The Court will not, as Bowe alternatively requests, direct Commerce to make a circumstances of sale adjustment to account for the discount surrounding the sale. As noted above, the decision to grant a circumstances of sale adjustment for a particular expense rests in the sound discretion of Commerce. Moreover, the adjustment claimed is unlike those which Commerce grants pursuant to the regulation. 19 C.F.R. § 353.56. The court cannot direct Commerce to make this unusual circumstance of sale adjustment without disturbing the discretion Commerce exercises in administering this portion of the statute. See Smith-Corona, 1 Fed.Cir. (T) 130, 137, 713 F.2d 1568. 4. Assigning a Zero Margin to Sales at or Above Fair Value In setting dumping margins, Commerce first calculates the average foreign market value of a sample of sales that occur during the period of investigation. Commerce then compares the United States price of each entry during this period with the average foreign market value, a practice which has been upheld by the Federal Circuit. Koyo Seiko Co., Ltd. v. United States, 20 F.3d 1156 (1994). Transactions with lower U.S. prices are considered dumped, with the difference between the U.S. price and the average foreign market value being the margin of dumping for that entry. Commerce assigns a zero margin to entries having a higher U.S. price than the average foreign market value (as opposed to assigning a negative margin). The Court of International Trade has upheld this practice. Serampore Industries Pvt, Ltd v. U.S. Dept. of Commerce, Int’l. Trade Admin., 11 CIT 866, 675 F.Supp. 1354 (1987). Commerce calculates the dumping margin by dividing the combined unit margins of the dumped sales by the value of dumped and nondumped U.S. sales. This methodology introduces a statistical bias in the calculation of dumping margins. By zeroing negative margins Commerce will find that some dumping occurred if any U.S. sales were made below the average foreign market value. “This will be true even if the vast majority of sales made by the subject foreign producers in the United States were at prices higher than the average foreign market value.” Ronald A. Cass & Stephen J. Nakkin, Antidumping and Countervailing Duty Law: The United States and the GATT, in Down in the Dumps 200, 205 (Boltuck & Litan eds. 1991). In this case, “Commerce calculated a dumping margin where 92 percent of Bowe’s U.S. sales were made at or above fair market value. Of those U.S. sales that were compared to home market sales (as opposed to those compared to constructed value sales), 99.4 percent were made at or above fair market value. Nevertheless, Commerce, adhering to its intrinsically unfair practice of disregarding negative margins, was able to arrive at a “positive” weighted-average margin above the de minimis level.” Plaintiffs Brief at 25. Several commentator’s have noted the statistical bias built into Commerce’s methodology of zeroing positive margins. See, e.g., Tracy Murray, The Administration of the Antidumping Duty Laiv by the Department of Commerce, in Down in the Dumps, supra *1150 at 37 (1991) (“... if the estimated margin of dumping is small, the bias could be several times the magnitude of the true margin of dumping; in such cases it is more likely that no dumping exists.”); Richard Boltuck, Joseph F. Francois, and Seth Kaplan, The Economic Implications of the Administration of the U.S. Unfair Trade Laws, in Down in the Dumps, supra at 155; Ronald A. Cass & Stephen J. Narkin, Antidumping and Countervailing Duty Law: The United States and the GATT, in Down in the Dumps, supra at 205; Michael S. Knoll, United States Antidumping Law: The Case for Reconsideration, 22 Tex.Int’l L.J. 265, 278-381 (1985). In responding to these observations, another commentator has written: One of the criticisms that has been voiced concerns the treatment of U.S. sales above fair value. The complaint is basically that a foreign company is not given any credit for nondumped U.S. sales. Presumably, if one U.S. sale is 10 percent above and one is 10 percent below foreign market value, the sales should cancel themselves out. There should be no finding of dumping. Instead, according to the arguments presented here, the nondumped U.S. sale is ignored, and the exporter is found to be dumping on the basis of the single dumped sale. Technically, no one seems to have considered how nondumped U.S. sales are actually treated under long-established Commerce methodology. Nondumped sales are not simply disregarded, but are actually given some credit. That is, the value of the nondumped sale is included in the value of the dumped sales (the denominator) that will be divided into the combined unit margins of the dumped sales to determine the dumping margin. This inclusion of value has the effect of diluting the dumping margin, making the margin lower than it would otherwise be if, as implied by the discussants, these sales were totally disregarded and were not given any credit at all in the analysis. Michael Coursey, Comment, in Down in the Dumps, supra at 240-41. Additionally, as the court observed in Serampore, Commerce justifies use of this methodology on the ground that it is necessary to combat masked dumping. Serampore, 11 CIT 866, 874, 675 F.Supp. 1354 (1987); see also Michael Coursey, supra at 240-41 (“... this methodology is designed to combat a practice known as spot dumping or rifle shooting, in which a foreign competitor gains U.S. customers not by dumping all at once, but by dumping to customers one or a few at a time.”) The statute is silent on the question of zeroing negative margins. Although Commerce’s practice of zeroing negative margins does not produce an absolute “apples to apples” 30 comparison, the Court will not substitute its own judgment for Commerce’s reasonable application of the statute. The Court is aware of the statistical bias inherent in Commerce’s methodology. The Court is not prepared, however, to invalidate that methodology because the practice combats masked dumping, an apparently legitimate goal consistent with the antidumping statute. Unless and until it becomes clear that such a practice is impermissible or unreasonable (because, e.g., Commerce erroneously placed too much significance on the phenomenon of masked dumping), the Court must defer to Commerce’s chosen methodology. See Daewoo Electronics Co., Ltd. v. International Union of Electronic, Elec., Technical, Salaried and Mach. Workers, 6 F.3d 1511, 1516 (Fed.Cir.1993). Conclusion Accordingly, Commerce’s denial of the level of trade adjustments is remanded for proceedings in accord with this opinion. In all other respects the review determination is affirmed. ORDER This case having been submitted for decision and this Court, after due deliberation, having rendered a decision herein; now in accordance with said decision, it is hereby *1151 ORDERED that the Department of Commerce’s final determination in Drycleaning Machinery from Germany, 56 Fed.Reg. 66,-838 (Dep’t Comm. Dec. 26, 1991) (final results admin, review), and subsequent remand determination is sustained in part and remanded in part; and it is further ORDERED that this case is remanded to the Department of Commerce to reconsider its denial of level of trade adjustments to foreign market value for advertising, headquarters sales, order entry and control, traffic shipment, and legal and finance expenses, in accordance with this opinion; and it is further ORDERED that (1) the Department of Commerce’s denial of circumstances of,sale adjustments for advertising, headquarters sales, order entry and control, technical publications, traffic shipment, sales administration and,management, and legal and finance expenses is sustained; (2) Commerce’s utilization of constructed value methodology in calculating foreign market value for the P200 dry cleaning machine is sustained; (3) Commerce’s inclusion of Bowe’s P200 trade show machine in the calculation of United States price is sustained; and (4) Commerce’s methodology of assigning a less than fair value amount of zero to United States sales of subject merchandise which were at or above fair value is sustained; and it is further ORDERED that the remand results are due on July 9, 1996. Any comments by plaintiffs are due on July 25, 1996. Any rebuttal comments by Commerce are due on August 9, 1996; and it is further ORDERED that Commerce’s final determination and remand results are sustained in all other respects. Footnotes [1] Drycleaning Machinery from Germany, 56 Fed. Reg. 66,838 (Dep’t Comm. Dec. 26, 1991) (final results admin, review). [2] Covering the period from November 1, 1989 through October 31, 1990. [3] Drycleaning Machinery from Germany, 56 Fed. Reg. 38,112 (Dep’t Comm. Aug. 12, 1991) (prelim. results admin, review). [4] "[Biased on and limited to the specific circumstances of this case, the Court finds that the Department of Commerce enforced its regulations in an arbitrary and capricious way in refusing to accept the compendium of information that plaintiff tendered to support its expense claims.” Böwe-Passat v. United States, 17 CIT 335, 343, 1993 WL 179269 (1993). [5] , See generally 19 C.F.R. § 353.6 (1989). The antidumping duty order that gave rise to the periodic review in this case has subsequently been revoked. Dry Cleaning Machinery from Germany, 60 Fed.Reg. 65635 (Dep’t Comm. Dec. 20, 1995) (revocation antidumping finding). [6] Chevron, U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 842-43, 104 S.Ct. 2778, 2781, 81 L.Ed.2d 694 (1984) ("First, always, is the question whether Congress has directly spoken to the precise question at issue. If the intent of Congress is clear, that is the end of the matter; for the court, as well as the agency, must give effect to the unambiguously expressed intent of Congress.”). [7] Id. at 843, 104 S.Ct. at 2781-82 (“If, ..., the court determines Congress has not directly ad- .. dressed the precise question at issue, the court does not simply impose its own construction on the statute, as would be necessary in the absence of an administrative interpretation. Rather, if the statute is silent or ambiguous with respect to the specific issue, the question for the court is whether the agency's answer is based on a permissible construction of the statute.”). [8] 19 U.S.C. § 1677b (1988). The terminology has since changed under the current statute: "Normal value” has replaced the .term "foreign market value.” See 19 U.S.C. § 1677b (1994). [9] 19 U.S.C. § 1677a (1988). The terminology has since changed under the current statute: "Export price and constructed export price” have replaced the term "United States price.” See 19 U.S.C. § 1677a (1994). [10] The antidumping duty assessed is "equal to the amount by which the foreign market value of the merchandise exceeds the United States price of the merchandise, ..." 19 U.S.C. § 1673c(a)(l) (1988). [11] 19 C.F.R. § 353.58 (1989). [12] 19 U.S.C. § 1677b(a)(4)(B) (1988); 19 C.F.R. § 353.56 (1989). [13] The statute has been amended and now provides for a level of trade adjustment. 19 U.S.C. § 1677b(7)(A) (1994): Level of trade The price described in paragraph (1 )(B) shall also be increased or decreased to make due allowance for any difference (or lack thereof) between the export price or constructed export price and the price described in paragraph (1)(B) (other than a difference for which allowance is otherwise made under this section) that is shown to be wholly or partly due to a difference in level of trade between the export price or constructed export price and normal value, if the difference in level of trade— (i) involves the performance of different selling activities; and (ii) is demonstrated to affect price comparability, based on a pattern of consistent price differences between sales at different levels of trade in the country in which normal value is determined. In a case described in the preceding sentence, the amount of the adjustment shall be based on the price differences between the two levels of trade in the country in which normal value is determined. [14] Order entry and control is a translation of the German name given to the department in the home market which handles and processes customer orders. Bowe explained in its questionnaire response that the processing of orders in the home market is more complicated than in the case of export orders and therefore involves a greater burden on its office workers. Bbwe’s compendium data indicated that its order entry and control employees devoted a much larger portion of their activities to sales in the home market than to sales in the United States. [15] Such submissions would establish the price comparability at the two levels of trade and obviate reliance on Bowe’s estimations. [16] The statute provides: "In determining foreign market value, if it is established to the satisfaction of the administering authority that the amount of any difference between the United States price and the foreign market value ... is wholly or partly due to—• ... (B) other differences in circumstances of sale; ... then due allowance shall be made therefor.” 19 U.S.C. § 1677b(a)(4)(B) (1988) (emphasis added). [17] 19 C.F.R. § 353.56 states in pertinent part: (a) In general. (1) In calculating foreign market value, the Secretary will make a reasonable allowance for a bona fide difference in the circumstances of the sales compared if the Secretary is satisfied that the amount of any price differential is wholly or partly due to such difference. In general, the Secretary will limit allowances to those circumstances which bear a direct relationship to the sales compared. (2) Differences in circumstances of sale for which the Secretary will make reasonable allowances normally are those involving differences in commissions, credit terms, guarantees, warranties, technical assistance, and servicing. The Secretary also will make reasonable allowances for differences in selling costs (such as advertising) incurred by the producer or reseller but normally only to the extent that such costs are assumed by the producer or reseller on behalf of the purchaser from that producer or reseller. (b) Special Rule.... (2) In comparisons with exporter’s sales price, the Secretary will make a reasonable deduction from foreign market value for all expenses, other than those described in para *1145 graph (a)(1) or (a)(2), incurred in selling such or similar merchandise. Up to the amount of the expenses, other than those described in paragraph (a)(1) or (a)(2), incurred in selling the merchandise. 19 C.F.R. § 353.56 (1989). [18] "Direct expenses are 'expenses which vary with the quantity sold,' while indirect expenses ‘are those that do not vary with the quantity sold.’" Zenith, 77 F.3d 426, 431, quoting Koyo Seiko v. United States, 36 F.3d 1565, 1569 n. 4 (Fed.Cir. 1994). [19] "[T]he purpose underlying the Treasury Department’s 1960 revision of the antidumping regulation is telling. By limiting adjustments to direct expenses, the government was attempting to curb abuses of the circumstances of sale provision enacted just two years earlier. 'Foreign producers had claimed indirect expense deductions under the rubric of differences in circumstances of sale to the point where price disparity routinely disappeared.’ Examples of such abuses include foreign producers 'claiming deductions for salesmen's salaries, ... and for any other portions of their general expenses—salaries, ... and so forth—which could be considered part of the sales, as distinguished from the production effort.’ ” Zenith, 77 F.3d 426, 432 (quoting Consumer Prods. Div., SCM Corp. v. Silver Reed Am., Inc., 753 F.2d 1033, 1038 (Fed.Cir. 1985)). [20] Advertising, headquarters sales, order entry and control, technical publications, traffic shipment, sales administration and management, and legal and finance. [21] Bowe does not claim level of trade adjustments for technical publications or sales administration and management costs. [22] Bowe cannot produce the requisite proof for a circumstances of sale adjustment for selling costs because it has no home market sales to distributors—the same reason it could not obtain a level of trade adjustment. The Court has upheld Commerce's rejection of the claimed circumstances of sale because it was appropriate under the clear language of the regulation, which regulation was supported by a historically developed rationale and which regulation Commerce has applied with consistency. Commerce’s rejection of the level of trade adjustments, however, was apparently inconsistent, and was not based on clear regulatory language or a well-developed rationale. [23] In calculating foreign market value Commerce used home market price for the P536 model and constructed value for the P200 model. [24] Section 1677b(a)(l) (1988), provided in pertinent part, that foreign market value: shall be the price.... (A) at which such or similar merchandise is sold ... in the principal markets of the country from which exported in the usual commercial quantities and in the ordinary course of trade for home consumption, or (B) if ... the quantity sold for home consumption is so small in relation to the quantity sold ... to countries other than the United States as to form an inadequate basis for comparison. [25] Usually, the home market is deemed "not viable” when home market sales are less than five percent of sales to countries other than the United States. See 19 C.F.R. § 353.48(a) (1989). Section 773(a)(1)(C) of the Uruguay Round Agreement Amendments to the United States An *1147 tidumping laws changed the viability test to provide that the home market will, in general, not be considered "viable” if the quantity of sales by the exporter in the home market is less than five percent of the quantity of sales by the exporter to the U.S. Market. 19 U.S.C. § 1677b(a)(l)(C) (1994) (emphasis added). [26] Section 1677b(a)(2) (1988) provides: (2) Use of Constructed Value. If the administering authority determines that the foreign market value of imported merchandise cannot be determined under paragraph (1)(A) [home market sales], then notwithstanding paragraph (1)(B) [third country sales], the foreign market value of the merchandise may be the constructed value of that merchandise, as determined under subsection (e) of this section [factors considered in determining CV]. 19 U.S.C. § 1677b(a)(2) (1988). [27] The record supports the contention of the defendants that the trade show machine was sold at a reduced price because of the wear and tear it underwent in being used as a floor model, and in being shipped to and from the trade show. [28] In IPSCO, the court recognized that Commerce had excluded some U.S. sales from its calculation of United States price in the past. The Court ordered Commerce to explain its litigation posture in IPSCO, i.e., that no statutory or regulatory authority exists for excluding sales outside the ordinary course of trade in calculating United States price, when the department had in the past excluded such sales. IPSCO, 12 CIT at 395-96, 687 F.Supp. 633. In its remand determination, Commerce explained its policy as follows: The Department’s determination to include IPSCO's U.S. sales of experimental pipe not sold in the ordinary course of trade was based upon the fact that exclusion of such merchandise is not required in the definition of United States price at section 772 of the Tariff Act of 1930. The Department will, on occasion, exclude certain U.S. sales in its less than fair value calculations where those sales will not have a significant effect on the margin or where the sales are not representative of the respondent’s selling practices in the U.S. market. ... In an LTFV investigation, the Department’s objective is to set an antidumping duty deposit rate that most accurately reflects the duties that will be imposed on entries which are liable for duties.... In order to determine whether those entries are likely being dumped (and by what amount) the Department looks at sales during the period of investigation, usually a six-month period, beginning five months before petitions are filed. The sales during this period of investigation will typically be representative of the foreign seller's behavior. Thus, they usually provide the best estimate of the dumping that is occurring after liquidation is suspended. Where certain sales during that six-month period would not be representative of the seller’s behavior, or where those sales are so small that they would have an insignificant effect on the margin, the Department may decide not to include those sales in calculating the .dumping margin. Remand Determination in IPSCO, Inc., et al, v. United States. [29] The court in American Permac stated "that while U.S. sales outside the ordinary course of trade ordinarily should be included (this may be the very cause of injury), a methodology is to be applied which accounts for sales which are unrepresentative and which do not lead to a fair price comparison.” American Permac, 16 CIT at 42, 783 F.Supp. 1421. The court further stated that the determinative factor on the U.S. sales side of the equation is not whether sales are in or out of the ordinary course of trade. "Fairness, distortion, representativeness are the issues to be examined. The goal is to include the sales but to utilize whatever methodology is needed to ensure a fair comparison.” Id. at 44, 783 F.Supp. 1421. The court concluded that the question of *1149 "[w]hether total exclusion of certain U.S. sales is among the methodologies ITA has discretion to utilize at the periodic review phase need not be resolved here, but the court seriously doubts Congress intended to compel distortions if exclusion of a few sales would remedy the problem.” Id. Prior to this discussion the court had concluded that plaintiffs claim for exclusion was not timely raised before the agency. Therefore, its consideration of the excludability of nonrepresentative sales from United States price was dicta. [30] Smith-Corona Group v. United States, 1 Fed. Cir. (T) 130, 132-34, 713 F.2d 1568 (1983), cert. denied, 465 U.S. 1022, 104 S.Ct. 1274, 79 L.Ed.2d 679 (1984).