Opinion
Tsoucalas, Judge:
Plaintiff, The Torrington Company (“Torring-ton”), moves pursuant to Rule 56.1 of the Rules of this Court for judg[673]*673ment on the agency record, challenging certain aspects of the Department of Commerce, International Trade Administration’s (“ITA”) final results in the first administrative review of imports of antifriction bearings from Italy. Antifriction Bearings (Other Than Tapered Roller Bearings) and Parts Thereof From Italy; Final Results of Antidumping Duty Administrative Reviews {‘Final Results”), 56 Fed. Reg. 31,751 (1991). Substantive issues raised by the parties in the underlying administrative proceeding were addressed by the ITA in the issues appendix to An-tifriction Bearings (Other Than Tapered Roller Bearings) and Parts Thereof From the Federal Republic of Germany; Final Results of An-tidumping Duty Administrative Review {“Issues Appendix”), 56 Fed. Reg. 31,692 (1991).
Background
On June 11, 1990, the ITA initiated an administrative review of imports of ball bearings, cylindrical roller bearings and parts thereof from Italy. Antifriction Bearings (Other Than Tapered Roller Bearings) and Parts Thereof From the Federal Republic of Germany, France, Italy, Japan, Romania, Singapore, Sweden, Thailand and the United Kingdom Initiation of Antidumping Administrative Reviews, 55 Fed. Reg. 23,575 (1990).
On March 15, 1991, the ITA published its preliminary determination in the administrative review. Antifriction Bearings (Other Than Tapered Roller Bearings) and Parts thereof from Italy: Preliminary Results of Antidumping Duty Administrative Reviews and Partial Termination of Administrative Reviews, 56 Fed. Reg. 11,181 (1991).
On July 11,1991, the ITA published its Final Results in this proceeding. Final Results, 56 Fed. Reg. 31,751.
Torrington moves pursuant to Rule 56.1 of the Rules of this Court for summary judgment on the agency record alleging that the following actions by the ITA were unsupported by substantial evidence on the administrative record and not in accordance with law: the ITA’s (1) use of a methodology for adjusting United States price (“USP”) and Foreign Market Value (“FMV”) for Italy’s value added tax (“VAT”) that granted a circumstance of sale (“COS”) adjustment to FMV to achieve tax neutrality; (2) method of calculating cash deposit rates for estimated duties; (3) in regard to exporter’s sales price (“ESP”) transactions, allowance of an adjustment to FMV for inventory carrying costs; (4) failure to verify FAG Cuscinetti S.p.A.’s (“FAG”) cost response; (5) adjustment to FMV for FAG’s pre-sale inland freight; (6) treatment of SKF Industrie, S.p.A. and SKF Cuscinetti’s (“SKF”) home market discounts; and (7) treatment of FAG’s U.S. market discounts. Memorandum in Support of Plaintiff’s Motion for Judgment Upon the Agency Record {“Torrington’s Memorandum”) at 5-40.
Discussion
This Court’s jurisdiction over this matter is derived from 19 U.S.C. § 1516a(a)(2) (1988) and 28 U.S.C. § 1581(c) (1988).
[674]*674A final determination by the ITA in an administrative proceeding will be sustained unless that determination is “unsupported by substantial evidence on the record, or otherwise not in accordance with law.” 19 U.S.C. § 1516a(b)(1)(B) (1988). Substantial evidence is “relevant evidence as a reasonable mind might accept as adequate to support a conclusion.” Consolidated Edison Co. v. NLRB, 305 U.S. 197, 229 (1938); Alhambra Foundry Co. v. United States, 12 CIT 343, 345, 685 F. Supp. 1252, 1255 (1988).
1. Circumstance of Sale Adjustment to FMV for Value Added Tax:
Torrington challenges the ITA’s use of a methodology for adjusting USP and FMV for Italy’s VAT that granted a COS adjustment to FMV to achieve tax neutrality. Torrington’s Memorandum at 10-14.
Defendant argues that its actions were supported by substantial evidence on the administrative record and otherwise in accordance with law. Defendant’s Memorandum in Opposition to Plaintiff’s Motion for Judgment Upon the Agency Record (“Defendant’s Memorandum”) at 13-39.
For a more detailed discussion of Torrington and defendant’s arguments on this issue, see this Court’s decision in Torrington Co. v. United States, 17 CIT 199, 202, 818 F. Supp. 1563 (1993).
SKF and FAG essentially agree with the defendant’s arguments on this issue. Opposition of SKF USAInc. and SKF Industrie. S.P.A. to Tor-rington’s Motion for Judgment on the Agency Record (“SKF’s Opposition”) at 8-12; Memorandum of Defendant-Intervenor FAG Cuscinetti S.p.A. ("FAG”) in Opposition to Plaintiffs Motion for Judgment Upon the Agency Record (“FAG’s Memorandum”) at 24-29.
This Court has fully addressed these arguments and adheres to its decision on this issue in Federal-Mogul Corp. v. United States, 17 CIT 88, 96-99, 813 F. Supp. 856 (1993). This Court remands this issue to the ITA to allow the ITA to add the full amount of VAT paid on home market sales to FMV without adjustment in conformity with this Court’s opinion in Federal-Mogul, 17 CIT at 108-9, 813 F. Supp. 856 (1993).
2. Calculation of Cash Deposit Rates:
In this administrative review, the ITA used two different methodologies for the actual calculation of dumping margins in cases where ESP sales were used: one for assessing duties on entries covered by the review, and the other for setting the cash deposit rate on future entries of the subject merchandise. Final Results, 56 Fed. Reg. at 31,753-54; Issues Appendix, 56 Fed. Reg. at 31,698-702. To calculate the assessment rate for ESP sales, the ITA “divide[d] the total PUDD [potential uncollected dumping duties — calculated as the total difference between foreign market value and U.S. price for an exporter] for the reviewed sales by the total entered value of those reviewed sales * * *. ” Issues Appendix, 56 Fed. Reg. at 31,698-99 (emphasis added). To calculate the estimated cash deposit rate for ESP sales, the ITA “divided the total PUDD for [675]*675each exporter by the total net U.S. price for that exporter’s sales * * *.” Id. at 31,699 (emphasis added).
Torrington argues that the ITA’s use of a methodology which results in an estimated cash deposit rate different from the assessment duty rate was unsupported by substantial evidence on the record and not in accordance with law. Torrington’s Memorandum at 5-10.
Defendant argues that its actions were supported by substantial evidence on the administrative record and otherwise in accordance with law. Defendant’s Memorandum at 6-13. In addition, defendant argues that this issue is moot because of the publication of superseding cash deposit rates in Antifriction Bearings (Other Than Tapered Roller Bearings) and Parts Thereof From France; et al.; Final Results of Antidumping Duty Administrative Reviews, 57 Fed. Reg. 28,360 (1992). Defendant’s Memorandum at 3-6.
For a more detailed discussion of Torrington and defendant’s arguments on this issue, see this Court’s decision in Torrington Co., 17 CIT at 204-05, 818 F. Supp. 1563 (1993).
SKF and FAG essentially agree with the defendant’s arguments on this issue. SKF’s Opposition at 4-8; FAG’s Memorandum at 11-23.
The Court agrees with the defendant that this issue is now moot. However, the Court directs the defendant to this Court’s decision on this issue in Federal-Mogul, 17 CIT at 102-03, 813 F. Supp. 856 (1993).
3. Inventory Carrying Costs:
In the Final Results of this administrative review the ITA correctly adjusted ESP for imputed inventory carrying costs pursuant to 19 U.S.C. § 1677a(e)(2) (1988). Torrington does not challenge this adjustment.
Pursuant to its new administrative practice, the ITA also made a corresponding adjustment to FMV for imputed inventory carrying costs when comparing ESP sales to FMV sales.
Torrington objects to this adjustment by the ITA to FMV for imputed inventory carrying costs. Torrington’s Memorandum at 20-26.
For a detailed discussion of Torrington and defendant’s arguments on this issue, see this Court’s decision in Torrington Co., 17 CIT at 212-14, 818 F. Supp. 1563 (1993).
SKF and FAG essentially agree with the defendant’s arguments on this issue. SKF’s Opposition at 12-16; FAG’s Memorandum at 48-60.
This Court adheres to its decision on this issue in Torrington Co., 17 CIT at 214-16, 818 F. Supp. 1563 (1993), and finds that the ITA’s adjustment to FMV for imputed inventory carrying costs pursuant to 19 C.F.R. § 353.56(b)(2) (1991) was a reasonable exercise ofthe ITA’s discretionin implementing the antidumping duty statute and is affirmed.
4. Failure to Verify FAG’s Cost Response:
On October 4, 1990, Torrington requested the ITA to conduct verification of all responses submitted in this administrative review, including FAG’s cost response, pursuant to 19 U.S.C. § 1677e(b) (1988) and 19 [676]*676C.F.R. § 353.36 (1991).1 Administrative Record Italy Public Document No. (“AR Italy Pub. Doc. No.”) 123. FAG had not yet submitted its cost response to the ITA. AR Italy Pub. Doc. Nos. 128, 134.
After FAG submitted its cost response, Torrington renewed its request for verification alleging that FAG’s cost response was missing important information and that some of the information reported was inconsistent. AR Italy Pub. Doc. No. 213.
Inaletter dated January 25, 1991, Torrington refers to the ITA’s decision to cancel verification of FAG’s cost response, alleging that it was due in part to the outbreak of the Persian Gulf War, and goes on to request the ITA to reschedule the verification or conduct limited verification in Washington, D.C. AR Italy Pub. Doc. No. 216. However, evidence on the administrative record shows that the ITA had never scheduled verification of FAG’s cost response in Italy, but that the ITA planned to conduct limited verification of FAG’s cost response at verification of FAG-Germany. AR Italy Pub. Doc. No. 218. The ITA refused Tor-rington’s request to reschedule verification or conduct limited verification in Washington, D.C.
Torrington argues that 19 U.S.C. § 1677e(b)(3)(B) requires the ITA to verify all information used in an administrative review “if good cause for verification is shown.” Torrington cites to the House of Representatives Ways and Means Committee’s report on the Trade Remedies Reform Act of 1984, which was later incorporated into the Trade and Tariff Act of 1984, for a definition of “good cause.” The Report states that “[g]ood cause could be such factors as a significant issue of law or fact, changed or special circumstances, discrepancies found in previous verifications, or the likelihood of a significant impact on the result.” H.R. Rep. No. 725, 98th Cong., 2d Sess. 43 (1984). Torrington argues that anytime [677]*677these or comparable circumstances are shown to exist verification is required. Torrington’s Memorandum at 16-19.
Torrington argues that the analysis of FAG’s cost response that it presented to the ITA showed that discrepancies in the response proved that good cause existed for verification of FAG’s cost response. Id. at 14-16. Torrington points out that FAG used a new cost accounting system to report costs for this administrative review and that the ITA has conducted verification in situations where a respondent used a new cost accounting system. Id. at 18 (citing Final Determination of Sales at Less Than Fair Value: New Steel Rail, Except Light Rail, From Canada, 54 Fed. Reg. 31,984 (1989)). Torrington argues that at a minimum the ITA should have conducted a modified form of verification of FAG’s cost response in Washington, D.C. Id. at 19-20.
Defendant argues that Torrington failed to show that good cause for verification of FAG’s cost response existed. Defendant argues that the ITA correctly requires more than mere speculation by a party requesting verification in order for the ITA to find that good cause for verification exists. Defendant argues that ITA will not find good cause to verify if a request to verify is made prior to the submission of the data to be verified. Defendant also argues that the ITA will not find good cause to verify if the ITA finds that a respondent has addressed any perceived deficiencies in its responses and the party requesting verification has not shown that any remaining deficiencies will have an impact on dumping margins. Defendant’s Memorandum at 41-42, 45-46.
Specifically, defendant argues that the ITA evaluated FAG’s cost response, deficiency responses, its other responses and verified FAG’s sales data including FAG’s accounting systems and found the information submitted to be credible. AR Italy Pub. Doc. No. 225. Therefore, the ITA determined that there was no need to verify FAG’s cost response. Defendant’s Memorandum at 44.
Defendant also points out that Torrington failed to substantiate that any remaining problems with FAG’s cost response would have a significant effect on FAG’s dumping margin. Id. at 45-46.
Finally, defendant points out that the ITA did consider conducting limited verifications in Washington, D.C., but rejected the idea for logistical and procedural reasons. Defendant states that the biggest problem with conducting verifications in Washington was that the ITA could not rely on the veracity of documents submitted under such a procedure. Id. at 46-47.
FAG agrees with the defendant’s arguments on this issue. FAG’s Memorandum at 30-47.
In response to defendant, Torrington argues that the fact that the ITA originally scheduled verification of FAG’s cost response, but canceled it at least in part due to the outbreak of the Persian Gulf War, supports Torrington’s contention that good cause for verification existed. Reply of The Torrington Company, Plaintiff, to Responses of Defendant and [678]*678Defendant-Intervenors to Torrington’s Motion for Judgment Pursuant to Rule 56.1 (“Torrington’s Reply”) at 25-26.
Torrington points out that the defendant’s argument that Tor-rington’s request for verification was invalid because it was filed prior to the submission of FAG’s response data is misplaced because Torrington was required by 19 C.F.R. § 353.36(a)(v)(A) to request verification within 120 days of the initiation of review, which means the request had to be filed no later than October 9, 1990. FAG’s cost response was filed on October 18, 1990. Torrington’s Reply at 26; see AR Italy Pub. Doc. No. 128.
Further, Torrington argues that the defendant’s argument that a request for verification for good cause must substantiate the degree of impact of any alleged deficiencies in a response on a respondent’s dumping margin imposes an impossible standard for showing good cause. Tor-rington argues that given the complexity of responses to be analyzed and the time constraints involved, it would be impossible for a domestic interested party to meet this burden. Torrington’s Reply at 27.
Finally, Torrington argues that the defendant’s argument that verification in Washington, D.C. was rejected for logistical and procedural reasons is weak. Id. at 28-29.
As an initial matter, this Court finds that there is no support for Tor-rington’s contention that the ITA should have conducted limited verification in Washington, D.C.
After examining the administrative record, this Court finds that the ITA never scheduled verification of FAG’s cost response in Italy. See AR Italy Pub. Doc. Nos. 149, 185, 218. In addition, the ITA sent FAG a deficiency letter requesting FAG to supplement its cost response in order to deal with many of the problems identified by Torrington. AR Italy Pub. Doc. No. 154. FAG provided all the information requested by the ITA. AR Italy Pub. Doc. No. 172, 182. ITA deemed FAG’s cost response and supplemental submissions adequate and used FAG’s cost data for the Final Results.
Torrington argues that if good cause, defined “as a significant issue of law or fact, changed or special circumstances, discrepancies found in previous verifications, or the likelihood of a significant impact on the result” or comparable circumstances, exist, the ITA is required to conduct verification. See Torrington’s Memorandum at 16 (quoting H.R. Rep. No. 725 at 43). The above quotation represents examples of what Congress thought could be considered good cause. Nothing leads this Court to the conclusion that the presence of any one of these circumstances requires the ITA to conduct verification.
Torrington argues that there were significant issues of fact in dispute in regard to FAG’s cost response. However, the ITA was aware of these issues and dealt with, and resolved them, to its satisfaction. AR Italy Pub. Doc. Nos. 154, 172, 182.
However, this Court believes the ITA goes too far in arguing that a party requesting verification for good cause must be able to substantiate [679]*679the degree of impact alleged problems will have on the respondent’s dumping margin. All that is required is that the party requesting verification present a reasonable argument that there will be a significant impact. This does not mean that the party must exactly quantify the impact on the dumping margin.2
But even a showing of a potentially significant impact on dumping margins is not enough. The statute and regulations clearly leave to the ITA’s discretion the determination of whether good cause for verification exists. 19 U.S.C. § 1677e(b); 19 C.F.R. § 353.36(a)(iv). If the ITA is satisfied with a respondent’s data and determines that good cause to verify does not exist, and the ITA’s determination is supported by substantial evidence on the administrative record, this Court will uphold the ITA’s determination. 19 U.S.C. § 1516a(b)(1)(B).
As discussed above, the ITA closely examined FAG’s cost response, requested additional information, and decided that good cause for verification did not exist. This Court finds that in this case the ITA’s determination was supported by substantial evidence on the administrative record and is affirmed.
5. Adjustment to FMV for FAG’s Pre-sale Inland Freight:
In the Final Results, the ITA deducted all of FAG’s pre-sale movement expenses from its calculation of FMV stating:
[W]e have determined that pre-sale inland freight should be treated as a movement expense and deducted from foreign market value. Because we do not treat pre-sale and post-sale movement charges differently in calculating an ex-factory U.S. price, we must treat these expenses in a similar manner in the home market, to ensure an equitable price-to-price comparison.
Issues Appendix, 56 Fed. Reg. at 31,715.
Torrington argues that the ITA’s decision to deduct all presale movement expenses from FMV is contrary to past ITA practice and the decision of this court in Silver Reed America. Inc. v. United States, 7 CIT 23, 34-35, 581 F. Supp. 1290, 1298-99 (1984), rev’d on other grounds sub nom. Consumer Prods. Div., SMC Corp. v. Silver Reed America, Inc., 753 F.2d 1033 (Fed. Cir. 1985). Torrington’s Memorandum at 26-27. Torrington argues that the ITA is making this adjustment pursuant to its authority to adjust FMV for differences in circumstances of sales. See 19 U.S.C. § 1677b(a)(4)(B) (1988); 19 C.F.R. § 353.56(a) (1991).
Specifically, Torrington argues that in order for the ITA to make an adjustment to FMV for a difference in circumstance of sale, the expense at issue must be directly related to the sales being investigated. Torrington points out that the court in Silver Reed, 7 CIT at 34-35, 581 F. Supp. at 1298-99, upheld ITA’s past practice of denying an adjustment to FMV for pre-sale expenses which could not be directly tied to sales. Torrington’s Memorandum at 28-30. Torrington argues that Silver [680]*680Reed is directly on point because here FAG was unable to tie its pre-sale movement charges directly to the sales under investigation. Id. at 29-30. Torrington argues that pre-sale home market movement expenses are general overhead expenses which cannot be directly tied to sales. Id.
Torrington admits that this court recently upheld the ITA’s new practice of deducting all pre-sale movement expenses from FMV in Ad Hoc Comm. of AZ-NM-TX-FL Producers of Cray Portland Cement v. United States, 16 CIT 115, 119, 787 F. Supp. 208, 211-13 (1992), but argues that this case was wrongly decided. Torrington’s Memorandum at 30-32.
Defendant and FAG rely on this court’s decision in Ad Hoc Comm., 16 CIT at 119, 787 F. Supp. at 211-13. Defendant’s Memorandum at 56-61; FAG’s Memorandum at 63-65.
Specifically, defendant and FAG argue that the ITA is required to deduct pre-sale movement expenses from U.S. price pursuant to 19 U.S.C. § 1677a(d)(2)(A) (1988) and 19 C.F.R. § 353.41(d)(2)(i) (1991), in order to arrive at an ex-factory price for U.S. sales. The statute and regulations do not address the deduction of presale expenses from FMV Defendant points out that by not deducting pre-sale expenses from FMV which was the ITA’s past practice, the ITA unfairly compared an ex-factory U.S. price with an ex-warehouse home market price. Defendant argues that the ITA’s adjustment to FMV for pre-sale movement expenses is not dependant on its authority to adjust FMV for differences in circumstances of sale, but rather upon the ITA’s inherent authority to make an “apples with apples” comparison. Defendant’s Memorandum at 57-59. Defendant points out that the ITA’s rationale was recently sustained in Ad Hoc Comm., 16 CIT at 119, 787 F. Supp. at 211-13.
The reasoning of this Court in upholding the ITA’s treatment of pre-sale inventory carrying costs in this case is equally applicable to the ITA’s treatment of pre-sale movement expenses. The ITA’s decision to compare U.S. price to home market price at a contemporaneous point in the chain of commerce is reasonable. Torrington Co., 17 CIT at 215, 818 F. Supp. 1563 (1993). In this case, the ITAhas chosen an ex-factory price as the contemporaneous point in the chain of commerce. In order to make this comparison certain expenses need to be removed from both U.S. and home market prices. This Court finds nothing unreasonable in the ITA’s removal of pre-sale movement expenses from both U.S. and home market prices as measured from the same point in the chain of commerce, in this case ex-factory. Id. at 33-34; Ad Hoc Comm., 16 CIT at 118, 787 F. Supp. at 211-13. This method of treating pre-sale home market movement expenses has also been specifically upheld by this court in a well reasoned opinion in Nihon Cement Co. v. United States, 17 CIT 400, 414-417, Slip Op. 93-80 at 30-34 (May 25, 1993).
Therefore, this Court affirms the ITA’s deduction of FAG’s pre-sale movement expenses from FMV
[681]*6816. SKF’s Home Market Discounts:
In the Final Results of this administrative review, the ITA stated that in regard to discounts:
The Department generally allows adjustments to home market price and USP for discounts and rebates where respondents have granted and paid them on sales of subject merchandise to unrelated parties during the period of review. Such discounts or rebates should be part of a respondent’s standard business practice and not intended to avoid potential antidumping duty liability. The Department generally makes an adjustment if discounts and rebates, granted pursuant to accurately and adequately described programs, are properly reported on a sale or customer-specific basis and are directly associated with the products or sales under consideration.
}{c ******
SKF: SKF granted rebates on a customer-specific basis. The Department verified that rebates per customer were accrued for sales which occurred during an agreed-upon time period or up to a certain agreed-upon amount, and then paid to the customer. We traced from payments of rebates to documentation justifying why the payments were made. SKF demonstrated that its rebates were legitimate and based on agreements. For the purpose of allocating rebates for this review, SKF divided the total rebates given to each customer during a given time period by the total sales to that customer. We found the allocation of discounts and rebates by SKF-FRG, SKF-France, and SKF-Italy to be consistent and reasonable; therefore, we have not changed our calculations from the preliminary determination.
Issues Appendix, 56 Fed. Reg. at 31,717-18.
Torrington argues that discounts relate directly to price and therefore an adjustment to FMY as the ITA made here, should not be allowed unless SKF demonstrates a direct relationship between a specific discount and a specific sale. Torrington alleges that SKF allocated discounts over all sales without showing that each individual sale was subject to a discount. Torrington argues that allocation is allowed only when a respondent shows on the record that all sales of the subject merchandise were eligible for, and received, a discount. Torrington’s Memorandum at 32-38.
Defendant concedes that it incorrectly treated SKF’s home market discounts as directly related expenses to be deducted from FMV Defendant states that this Court’s decision in Koyo Seiko Co. v. United States, 16 CIT 539, 544, 796 F. Supp. 1526, 1529-30 (1992), requires that direct selling expenses be shown to have a reasonably direct relationship to the sales under consideration. Defendant alleges that this direct relationship was not demonstrated if discounts were incurred on a transaction-specific basis but were allocated over all sales on a customer-specific basis as SKF did here. Therefore, defendant requests this [682]*682Court to remand this issue to the ITA “for reconsideration of those home market cash discounts that were not reported on a transaction-specific or model-specific basis and recalculation of foreign market values where warranted * * Defendant’s Memorandum at 61.
SKF states that it used two different methodologies to report home market discounts: one for SKF Cuscinetti and one for SKF Industrie, S.p.A. For SKF Cuscinetti, SKF states that discounts were available to all customers. SKF states:
“[t]otal sales under these three payment terms were divided by the cash discounts granted for the sales to yield a cash discount rate.” AR (Pub.) Doc. 222 at 7. For these rates to match those set forth in the payment terms, as they do, all customers eligible for the discounts indeed must have been granted such discounts. Accordingly, the discounts granted and reported by Cuscinetti are directly related to the reported sales for which cash discounts were claimed. * * * For Industrie, as explained by SKF, total cash discounts granted in the home market were divided by total home market sales for: a) the second half of 1988, b) full-year 1989, and c) the first four months of 1990. The resultant factors were then multiplied by the price to derive the per unit cash discount value reported by Industrie in its sales response.
SKF’s Opposition at 17-18. SKF admits that it allocated all home market discounts for SKF Industrie, S.p.A. over all home market sales due to the manner in which SKF Industrie, S.p.A.’s records were kept. Id.
SKF argues that it was reasonable for the ITA to accept SKF’s methods of reporting discounts and requests this Court to sustain the ITA’s actions in regard to this issue.
In addition, SKF argues that it has been caught by surprise by the defendant’s change of position on this issue. SKF argues that there is no statutory basis for the defendant to change its position in regard to its final administrative determination after judicial review of that determination has commenced. Reply Brief of SKF USA Inc. and SKF Industrie, S.p.A. (“SKF’s Reply”) at 3-4. SKF argues that the ITA’s unilateral decision to reverse its position in regard to SKF’s discounts denied SKF due process by not allowing SKF the opportunity to comment on the ITA’s new position and possibly correct problems with its information submission. Id. at 5-7.
SKF argues that the ITA’s articulation of a new standard in regard to granting adjustments for discounts which will be given retroactive effect was unfair and unlawful. SKF points out that retroactive application of the law is frowned upon, citing Bowen v. Georgetown Univ. Hosp., 488 U.S. 204 (1988). SKF’s Reply at 7-10.
SKF also argues that it has relied on the ITA’s previous position on this issue in its efforts to try to eliminate dumping by the company. The change in ITA’s position may result in the creation of dumping margins during the second, third and part of the fourth administrative reviews [683]*683which would not have existed under the ITA’s old methodology. SKF’s Reply at 10-16.
SKF argues that Koyo Seiko, 16 CIT at 542, 796 F. Supp. at 1529-30, does not invalidate the reporting methodology used by SKF for discounts. SKF’s Reply at 16-18.
SKF argues that the ITA’s new transaction-specific standard should be rejected. Id. at 20-22. However, if this Court finds the ITA’s new standard to be reasonable and in accordance with law, SKF requests the Court to determine that SKF has met the new standard. Id. at 22-24. At a minimum, SKF requests this Court to direct the ITA to allow an indirect adjustment to FMV for SKF’s discounts. Id. at 24-25.
Torrington argues that it is well settled that the ITA may request a remand to correct errors in its application of the law as well as other types of clerical errors. Surreply of The Torrington Company to the Reply Brief of the SKF Companies.
As an initial matter, this Court finds that SKF’s argument that the defendant is precluded from requesting a remand when it believes that it has incorrectly applied the law is completely without foundation. As this court has stated:
The law is clear that remand is appropriate where an agency has followed an improper method in making a determination or where there has been a defect in the agency’s finding. See, e.g., Ford Motor Co. v. NLRB, 305 U.S. 364, 374-75 (1939); Greene County Planning Bd. v. Federal Power Comm’n, 559 F.2d 1227 (2d Cir. 1977) (en banc), cert. denied, 434 U.S. 1086 (1978).
Timken Co. v. United States, 7 CIT 319, 320 (1984).
As to the merits of this issue, the Court of Appeals for the Federal Circuit has stated that in order for a discount or rebate to qualify as a direct cost to be subtracted from FMV the discount or rebate must have been actually paid on all of the sales under consideration and allocated on the basis of actual cost and sales figures. Smith-Corona Group v. United States, 713 F.2d 1568, 1580 (Fed. Cir. 1983), cert. denied, 465 U.S. 1022 (1984). In Koyo Seiko, 16 CIT at 542, 796 F. Supp. at 1530, this Court upheld the ITA’s treatment of Koyo Seiko Co., Ltd.’s (“Koyo Seiko”) post sale price adjustments as indirect selling expenses because these post sale price adjustments could not be directly correlated with sales of the subject merchandise using verified cost and sales information. In addition, this Court has found that rebates paid on out of scope merchandise may not be used in the calculation of deductions for expenses from FMV for in-scope merchandise. Torrington Co., 17 CIT at 218, 818 F. Supp. 1563 (1993).
The statute and the ITA have a preference for respondents to provide actual expense information as opposed to allocated expense information. As a result, the ITA generally gives respondents an incentive to provide the ITA with actual expense information. The ITA does this by classifying actual expense information in a way which gives greater benefit to the respondent and classifying allocated information in a way [684]*684which gives a respondent less benefit. This can lead to differing treatment of the same kind of expenses in the calculation of USP and FMV
A respondent benefits by having home market expenses characterized as direct because generally FMV will be adjusted only for direct expenses. 19 U.S.C. § 1677b(a)(4)(B); Consumer Prods. Div., 753 F.2d at 1037-38. If the respondent fails to meet the standard for receiving a direct adjustment to price for its home market expenses, the expense will be treated as an indirect expense because this treatment is adverse to the respondent. 19 C.F.R. § 353.56(b)(2) (1991) (indirect selling expenses deducted from FMV only in relation to ESP transactions and only up to amount of indirect selling expenses deducted from USP). Allocated expenses in the U.S. market are treated as direct expenses because direct expenses will be deducted from all USP transactions which will, therefore, reduce USP and potentially increase dumping margins. 19 U.S.C. § 1677a(d)(2)(A) (1988). If these expenses were treated as indirect expenses, they would only be deducted from USP in regard to ESP transactions and will, therefore, reduce USP and potentially increase the dumping margin only for ESP transactions. 19 U.S.C. § 1677a(e)(2) (1988). Therefore, treatment of these expenses as indirect expenses would destroy any incentive a respondent has to provide the ITA with actual expense information. Defendant’s Memorandum at 62-64. This court has affirmed this method of providing an incentive for respondents to provide actual expense information. Timken Co. v. United States, 11 CIT 786, 804, 673 F. Supp. 495, 512-13 (1987).
It is clear that discounts can be deducted from FMV if the actual expense information is reported to the ITA on a transaction-specific or product-specific basis. 19 U.S.C. § 1677b(a)(4)(B). It is also clear that discounts paid on the subject merchandise can be allocated over all sales of the subject merchandise as long as discounts paid only on the subject merchandise are used to calculate the per-unit amount of discount to be deducted and the discounts “can be directly correlated with specific merchandise using verified cost and sales information.” Smith-Corona, 713 F.2d at 1580. If a respondent is unable to provide the ITA with transaction-specific or product-specific discount amounts, the ITA must look to the information provided by the respondent to determine if the reported allocated discounts were only made on and allocated to sales of the subject merchandise and can be tied to verified cost and sales data. In order for the ITA to accept discounts or rebates allocated on a customer-specific basis as direct costs, the percentage amount of each discount or rebate paid must be the same for each type of the subject merchandise sold and the total amount of discounts or rebates paid to each customer must be allocated over all sales of the subject merchandise made to that customer. If this relationship is not shown to the satisfaction of the ITA, but the aggregate amounts of discounts paid on the subject merchandise have been verified, the discounts are to be treated as indirect selling expenses. Koyo Seiko, 16 CIT at 542-43, 796 F. Supp. at 1530.
[685]*685Therefore, this Court remands this issue to the ITA to determine if SKF’s two methods of reporting discounts in the home market for SKF Cuscinetti and SKF Industrie, S.p.A. meet the standard required for those discounts to be treated as direct selling expenses and subtracted from FMV If the information on the administrative record does not support treatment of these discounts as direct expenses, the ITA will treat these discounts as indirect selling expenses since the ITA has already verified the total amounts in question. AR Italy Pub. Doc. 222. In addition, this Court cannot tell on the basis of the administrative record if discounts paid on SKF’s sales of out of scope merchandise were used in the calculation of the adjustment to FMV for SKF’s home market sales discounts. This Court cannot allow the ITA to use a methodology which allows for the inclusion of discounts paid on out of scope merchandise in calculating adjustments to FMV and ultimately the dumping margins. Torrington Co., 17 CIT at 218, 818 F. Supp. 1563 (1993). Therefore, upon remand the ITA is directed to develop a methodology which removes discounts paid on sales of out of scope merchandise from any adjustments made to FMV for discounts or, if no viable method can be developed, to deny such an adjustment in its calculation of FMV
7. FAG’s U.S. Market Discounts:
In regard to FAG’s U.S. market discounts, the ITA accepted FAG’s allocation of discounts on a customer-specific basis but treated them as direct costs to be fully deducted from U.S. price as this treatment was adverse to FAG. Issues Appendix, 56 Fed. Reg. at 31,717.
Torrington argues that, for the reasons stated above in regard to SKF’s home market discounts, since FAG’s U.S. market discounts were not reported on a transaction-specific basis, the ITA should have rejected FAG’s U.S. market discount information and should instead have used the highest actual discount paid on any U.S. sale as best information available. Torrington argues that the use of the ITA’s current methodology may under-allocate discounts on some sales of the subject merchandise. Torrington’s Memorandum at 38-40.
Defendant repeats its argument that when dealing with an expense such as a discount which is allocated on a customer-specific basis, the ITA’s treatment of that expense depends on whether the expense is a home market or U.S. market expense. As stated above, allocated expenses in the U.S. market are treated as direct expenses because direct expenses will be deducted from all USP transactions which will, therefore, reduce USP and potentially increase dumping margins. 19 U.S.C. § 1677a(d)(2)(A).
Defendant argues that this method of treating discounts is actually the drawing of an adverse inference through the use of the best information available rule and is reasonable and in accordance with law. 19 U.S.C. § 1677e (1988); Defendant’s Memorandum at 63-64.
FAG argues that the ITA verified its U.S. market discount information and that the ITA’s acceptance of FAG’s allocation methodology and [686]*686treatment of these expenses as direct costs was supported by substantial evidence on the administrative record and in accordance with law. Furthermore, FAG argues that its allocation methodology leads to the same results as the use of the allocation methodology approved in Smith-Corona, 713 F.2d at 1580. FAG’s Memorandum at 69-76.
As discussed above, if a respondent allocates expenses on a customer-specific basis and does not meet the requirements of Smith-Corona, 713 F.2d at 1580, the ITA is allowed to make an adverse inference and treat the expenses in such a way as to encourage the respondent to submit actual expense information in the future. In this case, FAG has allocated its payments of discounts in the U.S. market on a customer-specific basis. AR Italy Pub. Doc. No. 106. There is no contention that FAG has used the same allocation methodology as the one approved in Smith-Corona even though FAG alleges that the end result of the two methods is the same. Therefore, the ITA was correct to treat FAG’s discounts in the U.S. market as direct selling expenses. However, once again this Court cannot tell from the administrative record whether discounts paid on out of scope merchandise were used to calculate the adjustment to USP for FAG’s discounts. Therefore, this issue is remanded to the ITA to develop a methodology which removes discounts paid on FAG’s sales of out of scope merchandise from any adjustments made to USP for discounts or, if no viable method can be developed, to deny such an adjustment in its calculation of USP Torrington Co., 17 CIT at 221, 818 F. Supp. 1563 (1993).
Conclusion
In accordance with the foregoing opinion, this case is remanded to the ITA to add the full amount of VAT paid on each sale in the home market to FMV without adjustment; to determine if SKF’s two methods of reporting discounts in the home market for SKF Cuscinetti and SKF In-dustrie, S.p.A. meet the standard required for those discounts to be treated as direct selling expenses and subtracted from FMV or if information on the administrative record does not support deduction as direct expenses, to treat these discounts as indirect selling expenses; to develop a methodology which removes discounts paid on sales of out of scope merchandise from any adjustments made to FMV for SKF’s discounts or, if no viable method can be developed, to deny such an adjustment in its calculation of FMV; and to develop a methodology which removes discounts paid on FAG’s sales of out of scope merchandise from any adjustments made to USP for discounts or, if no viable method can be developed, to deny such an adjustment in its calculation of USP ITA’s determination is affirmed in all other respects. Remand results are due within sixty (60) days of the date this opinion is entered. Any comments or responses by the parties to the remand results are due within thirty (30) days thereafter. Any rebuttal comments are due within fifteen (15) days of the date responses or comments are due.