This case is before the Court to review the remand results ordered in Serampore Indus, v. United States Dep’t of Commerce, 11 CIT -, 675 F.Supp. 1354 (1987), which held that the International Trade Administration of the United States Department of Commerce (Commerce) was not required to offset less than fair value sales with fair value sales and found that no adjustment was necessary for deposits of estimated countervailing duties, but remanded for Commerce to (1) ascertain whether castings produced by Serampore Industries Pvt. Ltd. (Serampore) incurred the Indian Central Sales Tax; (2) eliminate the addition of excise duty drawback to Serampore’s United States price; and (3) explain apparently inconsistent methodologies in determining the “all others” dumping margin.
BACKGROUND
Commerce investigated four exporters of iron construction castings from India and determined that one exporter was dumping merchandise, that one exporter was not dumping, and that two other exporters were dumping at only de minimis levels. Certain Iron Construction Castings From India; Final Determination of Sales at Less Than Fair Value, 51 Fed. Reg. 9,486 (Mar. 19, 1986). The domestic industry challenged the findings of zero and de minimis margins in Alhambra Foundry Co. v. United States, 12 CIT -, 685 F.Supp. 1252 (1988). This action concerns the Indian producers’ challenges to the affirmative rate for Serampore and the “all others” dumping margin.
DISCUSSION
1. “All Others” Dumping Margin
The Court remanded to Commerce to explain its apparently inconsistent methodologies used to calculate the “all others” dumping margin. The “all others” rate applies to firms that Commerce does not include in its antidumping duty investigation and to any new exporters whose first shipment occurs prior to the liquidation of entries covered by the antidumping order. See 19 U.S.C. § 1673e(b)(l).
In calculating the “all others” dumping margin in its final determination, Commerce disregarded three of the four Indian companies investigated that were found to have zero or de minimis dumping margins and based the “all other” rate on Seram-pore’s dumping margin alone. The Court found that Commerce’s methodology was apparently inconsistent with a prior determination where Commerce included companies with de minimis margins in calculating the “all other” rate. The Court also noted the domestic industry’s observation that in even earlier determinations, Commerce set the “all other” rates at the highest dumping margin found for any firm investigated. Serampore, 11 CIT at-, 675 F.Supp. at 1361.
In Commerce’s clarification of its calculation of the “all other” rate, Commerce indicates it is a long standing practice to exclude firms that receive zero or de minim-is margins. See e.g., Carbon Steel Wire Rod From Spain; Amendment to the Final Determination of Sales at Less Than Fair Value, 49 Fed.Reg. 42,969, 42,970 (Oct. 25, 1984); Red Raspberries From Canada; Preliminary Determination of Sales at Less Than Fair Value, 49 Fed. Reg. 49,129, 49,131 (Dec. 18, 1984); Stainless Steel Woven Wire Cloth from Japan; Final Determination of Sales at Less Than Fair Value, 50 Fed.Reg. 10,520, 10,-522 (Mar. 15,1985); Egg Filler Flats From Canada; Final Determination of Sales at Less Than Fair Value, 50 Fed.Reg. 48,238, 48,241 (Nov. 22, 1985). Commerce states that several important considerations provide the basis for its exclusion of firms with zero or de minimis margins.
First, Commerce allows all companies to participate in the antidumping investigation by submitting voluntary responses. If these responses are timely and in good order, Commerce will calculate individual antidumping margins for the submitting *668companies. Consequently, any company that submits a satisfactory response cannot claim that Commerce has refused to include it in its investigation. Commerce assumes that companies which are not dumping will submit voluntary responses. Under this assumption, Commerce states that including companies that do not sell at less than fair value and that do submit a voluntary response in calculating an “all others” rate including these firms would generally be skewed to reflect the pricing practices of nondumping firms rather than those firms that decided not to respond.
Second, any shipper who requests an administrative review pursuant to section 751 of the Tariff Act of 1930, as amended, 19 U.S.C. § 1675 (1982 & Supp. IV 1986), will receive an individual margin and, at least for purposes of assessment, will not be covered by the all others rate. Thus, even companies that never submit responses in the original investigation are given a second chance in the administrative review.
Third, by establishing the all others rate at the weighted average margin level, and by applying the all others rate to new shippers, Commerce discourages existing firms from creating new corporate entities to avoid paying antidumping duties at their company-specific rate and thus circumvent an antidumping duty order.
Fourth, Commerce assumes that new shippers attempting to gain a foothold in a competitive market will be forced to price aggressively. When competing with firms that are dumping, new firms will most likely be driven to follow the same pricing strategy. Although it is possible that a new entrant will not dump, Commerce infers that as long as other firms are dumping in the market, the new entrant is also likely to dump.
For these reasons, Commerce bases the all others rate on the weighted-average rate of all the firms found to be dumping. Commerce stresses that it does not make the worst case assumption, but rather computes the rate for new companies on an average of all affirmative rates rather than the rate of the company with the highest dumping margin. Commerce acknowledges that throughout most of 1982 and during the first several months of 1983, Commerce experimented with an all others rate that was equivalent to the highest margin. See Serampore Indus., 12 CIT at-, 675 at 1361. Commerce ceased using the highest rate for an investigated company because it was believed to be unfair and punitive. Before and after this experimental period, Commerce employed an all others rate equivalent to the weighted-average of all affirmative margins. Relative to the highest margin, a weighted-average all others rate more accurately estimates the pricing practices of uninvestigated companies and new shippers.
Commerce attributed to “inadvertent error” the inclusion of de minimis margins in the all others dumping margin in Certain Welded Carbon Steel Pipe and Tube Products from Turkey; Final Determination of Sales at Less Than Fair Value, 51 Fed.Reg. 13,044 (Apr. 17, 1986). Commerce also confesses that three other administrative determinations inadvertently included a de minimis or zero dumping margin in calculating the all others rate. Lightweight Polyester Filament Fabric From the Republic of Korea; Final Determination of Sales at Less Than Fair Value, 48 Fed.Reg. 49,679 (Oct. 27, 1983); Bicycles from Taiwan; Final Determination of Sales at Less Than Fair Value, 48 Fed.Reg. 31,688 (July 11, 1983); Portland Hydraulic Cement from Japan; Preliminary Determination of Sales at Less Than Fair Value, 48 Fed.Reg. 19,445 (Apr. 29, 1985). Thus, in the absence of error, Commerce will generally not include negative or de minimis margins in calculating the all others rate.
The only instance where Commerce will intentionally include zero or de minim-is margins in an all others rate is when the number of producers and exporters is so great that Commerce is forced to investigate only a sampling of companies and not 60 percent or more. See, e.g., Certain Fresh Cut Flowers From Columbia; Final Determination of Sales at Less Than Fair Value, 52 Fed.Reg. 6842 (Mar. 5, *6691987). In cases when Commerce has conducted an investigation where the universe of producers and exporters is in the thousands, Commerce has employed a less traditional methodology such as sampling. See Southwest Fla. Winter Vegetables Growers Ass’n v. United States, 7 CIT 99, 584 F.Supp. 10 (1984). When Commerce relies on a sample, the pricing practices of a relatively small number of companies are used to represent the experience of the whole. Because Commerce does not obtain responses from companies accounting for 60 percent of all shipments to the United States, in these cases Commerce finds it appropriate to include in the all others rate firms that have zero or de minimis margins. Under these circumstances, there is no basis in law or fact to assume that a non-participating company is dumping. Even in a sample, however, Commerce will still accept voluntary responses. Thus, firms that feel margins based on Commerce’s sample will not reflect their own experience may still submit a voluntary response and receive a company-specific dumping margin if that response is complete.
In the underlying litigation defendant-in-tervenors stated that it was Commerce’s general practice to ignore individual company rates that are based upon the “best information available” within the meaning of 19 U.S.C. § 1677e(a). Commerce states that when it calculates the all others rate, it generally includes all investigated firms that receive affirmative margins, including any firm whose margin is based upon the best information available. See, e.g., Cellular Mobile Telephones and Subassem-blies From Japan; Final Determination of Sales at Less Than Fair Value, 50 Fed.Reg. 45, 447 (Oct. 31, 1985). Similar to its assumption with respect to firms that do not submit voluntary responses, Commerce assumes that investigated firms that fail to cooperate in the investigation are “more probably than not dumping.” Remand Results at 16. An all others rate that excludes margins based on the best information available would therefore, with one exception, be skewed to reflect the pricing practices of non-dumping firms rather than those firms that choose not to cooperate with Commerce’s investigation.
The exception that Commerce recognizes arises when an investigated company is unable to cooperate with the investigation for reasons largely beyond the company’s control. In an investigation on Flowers from Ecuador, for example, the company “virtually collapsed” and many of the records disappeared when the original organizers left the company. Commerce stated:
We consider it inappropriate, however, to conclude that Eden Flowers’ best information dumping margin is representative of the experience of other non-responding Ecuadorian producers/exporters. Because of the highly unusual circumstances involved in this instance, the best information available rate used for Eden Flowers has not been included in the calculation of the “all others” rate.
Certain Fresh Cut Flowers from Ecuador; Final Determination of Sales at Less Than Fair Value, 52 Fed.Reg. 2128, 2132 (Jan. 20, 1987). See also Fresh Cut Roses from Columbia; Final Determination of Sales at Less Than Fair Value, 49 Fed.Reg. 30,765 (Aug. 1, 1984). In these instances, the assumptions that compel Commerce to include a truly uncooperative company in calculating the all others rate are lacking. Commerce thus believes it is more reasonable to exclude the handful of companies that satisfy this standard, than to include their best information available margins in the all others rate.
Plaintiffs dispute that Commerce will always accept voluntary responses, and point to a determination where Commerce rejected a voluntary response. Certain Heavy-Walled Rectangular Welded Carbon Steel Pipes and Tubes from Canada; Final Determination of Sales at Less Than Fair Value, 50 Fed.Reg. 48,238, 48,239 (Nov. 22, 1985) (voluntary response of Capeo Tubular Products).
An analysis of that rejection shows the voluntary response was based on a single sale of the merchandise under investigation and that Commerce found the sale *670to be too insignificant to warrant inclusion in its analysis. Commerce acknowledges that in rare circumstances, however, one or two sales might provide Commerce with an adequate basis upon which to calculate a company-specific dumping rate. See, e.g., Final Determination of Sales at Less Than Fair Value; Offshore Platform Jackets and Piles from Japan, 51 Fed. Reg. 11,788 (Apr. 7, 1986). The Court finds the single rejection of a voluntary response based on a single sale to be insufficient, in the absence of other facts, to displace Commerce’s practice of accepting of voluntary responses that are timely filed and in good order.
The Court thus turns to plaintiffs’ argument that Commerce’s calculation of the all others rate violates the GATT Anti-dumping Code, which provides that “[t]he amount of antidumping duty must not exceed the margin of dumping as established under Article 2.” Agreement on Implementation of Article VI of the General Agreement on Tariffs and Trade Relating to Antidumping Measures, art. 8, ¶ 3, Jan. 1, 1979, 31 U.S.T. 4949, T.I.A.S. 9650. Plaintiffs argue that Commerce’s calculation of the all others rate clearly exceeds the average rate of dumping found. Because article 8 of the GATT Antidumping Code was incorporated into United States law in sections 1(c) and 2 of the Trade Agreements Act of 1979, 19 U.S.C. §§ 2502(1) and 2503, plaintiffs argue that Commerce should use a “true average” for the all others rate.
The GATT Antidumping Code does not prohibit the collection of antidumping duties in excess of the actual dumping margin. Article 8, H 3, provides in its entirety that:
The amount of the anti-dumping duty must not exceed the margin of dumping as established under Article 2. Therefore, if subsequent to the application of the anti-dumping duty it is found that the duty so collected exceeds the actual dumping margin, the amount in excess of the margin shall be reimbursed as quickly as possible.
Agreement on Implementation of Article VI of the General Agreement on Tariffs and Trade Relating to Antidumping Measures, art. 8, ¶ 3, Jan. 1, 1979, 31 U.S.T. 4949, T.I.A.S. 9650.
An administrative review under 19 U.S.C. § 1675(a) may result in a company-specific margin that is higher or lower than the all others deposit rate applied to a particular firm’s entries. If the margin is lower, any excess estimated antidumping duties will be refunded with interest as provided for in 19 U.S.C. §§ 1673f and 1677g. Accordingly, there is no conflict between Commerce’s all others methodology and Article 8 of the GATT Antidumping Code.
Finding that Commerce’s methodology does not violate the GATT Antidumping Code, the Court holds that Commerce’s practice of excluding firms with zero or de minimis dumping margins in calculating an all others rate is reasonable and is entitled to deference. See Smith-Corona Group v. United States, 1 Fed.Cir. (T) 130, 145, 713 F.2d 1568, 1571 (1983), cert. denied, 465 U.S. 1022, 104 S.Ct. 1274, 79 L.Ed.2d 679 (1984). This part of the remand is affirmed.
2. Excise Duty Drawback
In its final determination Commerce stated that it was adding “duty drawback” to Serampore’s United States price pursuant to 19 U.S.C. § 1677a(d)(l)(B) (1982), which provides for an upward adjustment to United States price for customs duties that are rebated or uncollected by reason of exportation. Certain Iron Construction Castings from India; Final Determination of Sales at Less Than Fair Value, 51 Fed.Reg. 9486 (Mar. 19, 1986). After reviewing the underlying antidumping administrative record, Commerce found the excise duty drawback should only have been deducted from Serampore’s constructed value calculations, not added to United States price. In the remand results, Commerce has eliminated the double counting that occurred with respect to the excise duty drawback. This part of the remand is affirmed.
*6713. Indian Central Sales Tax
Throughout this litigation the central question with respect to the Central Sales Tax has been the location of Serampore’s suppliers. If pig iron and scrap were sourced exclusively inside the State of West Bengal, Serampore would not have incurred the Central Sales Tax and no adjustment to the final dumping margin would be appropriate. If the pig iron and scrap that Serampore used originated from a source outside West Bengal, then the Central Sales Tax would be incurred either on transactions between Serampore and its supplier or on transactions between Serampore’s dealers and a supplier. If Serampore did pay the Central Sales Tax, Serampore’s net raw material costs would be lowered to the extent that the Central Sales Tax was incurred and subsequently rebated upon exportation by the Cash Compensatory Support program. See 19 U.S.C. § 1677b(e)(l)(A) (1982).
To determine whether Serampore incurred the Central Sales Tax on its raw material purchases, Commerce reviewed documents that identified where Seram-pore’s suppliers of pig iron and scrap were located. Commerce determined that the administrative record in the underlying an-tidumping investigation did not contain substantial evidence to answer this question. For purposes of this remand, Commerce obtained the consent of the appropriate parties to examine various documents in the public and business proprietary administrative records for the 1984 countervailing duty administrative review of Certain Iron-Metal Construction Castings From India; Final Results of Countervailing Duty Administrative Review, 51 Fed.Reg. 45,788 (Dec. 22, 1986). Commerce justifies its consideration of this information from the 1984 review because the amount of taxes that the Indian government imposed on iron-metal castings was an important issue in the countervailing duty determination, and because pig iron and scrap are the principal inputs for both iron-metal castings and iron construction castings. Commerce also determined there was a sufficient overlap in the time periods of the investigations in both periods to permit Commerce to reach a reasonable conclusion about the Central Sales Tax.
Upon its review of the documents from the countervailing duty administrative review, and upon re-examination of the documents in the administrative record for the antidumping action, Commerce determined that (i) Serampore is located within the State of West Bengal; (ii) Serampore purchased all of its pig iron from sources located outside West Bengal and the total tonnage involved incurred the Central Sales Tax; (iii) Serampore purchased all of its scrap iron from dealers located within West Bengal and the total tonnage incurred the West Bengal Sales Tax; (iv) the price Ser-ampore paid for scrap included the Central Sales Tax, because Serampore’s dealers purchased the scrap from suppliers located outside West Bengal; and (v) through the Cash Compensatory Support program the Indian government rebated the total amount of Central Sales Tax that Seram-pore paid.
Finding that Serampore paid the four percent central sales tax on all of its pig iron purchases, Commerce eliminated the adjustment for the West Bengal Sales Tax and deducted only the Central Sales Tax in its constructed value calculation to arrive at a foreign market value for Serampore. Commerce calculated the Central Sales Tax as four percent of the average unit price per pound paid to suppliers outside West Bengal.
To calculate the Central Sales Tax on scrap, Commerce relied upon the 1984 administrative review verification report which states that the dealers usually add a fifteen percent mark-up on scrap. Because the antidumping calculations are based on the price Serampore paid to the dealer, Commerce calculated the Central Sales Tax based on this price net of the fifteen percent dealer mark-up.
Serampore concurs with Commerce’s corrected calculations only insofar as the remand results relate to correction of .the Central Sales Tax. Serampore also urges that Commerce must still account for other *672taxes under 19 U.S.C. § 1677b(e)(l)(A) (1982). Serampore argues that the verification report for the 1984 countervailing duty administrative review that Commerce used to determine whether Serampore paid the Central Sales Tax indicates that Seram-pore also paid a “freight equalization fund levy” and a “turnover tax.” Serampore accordingly moves for a second remand.
Commerce states in the remand results that
Serampore did not claim during the underlying antidumping duty investigation that its castings incurred the freight equalization fund levy or the so-called turnover tax. Nor did it claim that these taxes were rebated under the [Cash Compensatory Support] program. Since the remand ordered by the Court only applies to the [Central Sales Tax], the “all others” rate, and the double-counting of the excise duty drawback, we have not examined the 1984 CVD findings for the purpose of allowing Serampore to advance new tax claims during this remand proceeding.
Remand Results at 26.
Defendant-Intervenors argue that Ser-ampore cannot now argue that Commerce should have accounted for either the freight equalization fund or the turnover tax because Serampore only argued the Central Sales Tax in the briefs and oral arguments before the Court.
A review of the myriad of papers filed in this action discloses that Serampore never abandoned its arguments concerning other taxes. For example, one document declares:
Defendant-Intervenors also state that “the only tax in dispute insofar as its calculation is concerned appears to be the Central Sales Tax.” This statement is incorrect. Plaintiffs have argued from the outset that [Commerce] failed to take into account several taxes on several different items physically incorporated into the finished product. In requesting that the Court allow remand expeditiously for calculation of the Central Sales Tax incidence, Plaintiffs have not relinquished, and do not intend to relinquish, any argument regarding the full tax incidence.
Plaintiffs’ Reply to Defendant-Inter-venors’ Motion for Leave to File Reply and Plaintiffs’ Reply to Defendant-Inter-venors’ Reply to Plaintiffs’ Reply to Defendant’s Motion for Partial Remand, at 2.
While Commerce denies that Serampore claimed an adjustment for the Freight Equalization Fund Levy, the record shows that Serampore did identify the Freight Equalization Fund Levy as an indirect tax in its dumping questionnaire response. Conf.R. 429. The countervailing duty documents upon which Commerce relied to verify the Central Sales Tax also demonstrate that the Freight Equalization Fund Levy was verified.
While the Court still encourages specific and clear objections in both administrative and judicial proceedings, the Court finds that the Freight Equalization Fund Levy was identified in the questionnaire response and verified in documents upon which Commerce relied to verify the Central Sales Tax. Plaintiffs also allege that taking the Freight Equalization Fund Levy into account will reduce Serampore’s margins to a de minimis level and would thus avoid an antidumping duty order and obviate the need to expend administrative and resources on the administrative reviews that would be necessary to revoke that order, and conserve judicial resources in reviewing objections to those administrative reviews. Judicial authority supports granting a request for a remand if it will foster and promote fundamental fairness. Alhambra Foundry Co. v. United States, 12 CIT -, 685 F.Supp. 1252, 1262 (1988); ILWU Local 142 v. Donovan, 12 CIT -, 678 F.Supp. 307, 310 (1988). In this action it is possible that after a second remand, Commerce will find that Serampore is not dumping its merchandise in the United States. This is thus a question of fundamental fair treatment of a foreign producer. The Court grants, as a matter of discretion, Serampore’s motion for a remand to account for the Freight Equalization Fund Levy.
*673The Court also finds that the information upon which Commerce relied to verify the incidence of the Central Sales Tax discloses that Serampore incurred the turnover tax. For the same concerns of fundamental fairness, and because Commerce verified the information and found it reliable, the Court also remands for Commerce to account for the turnover tax.
4. Computer Input Error
Serampore also moves for a second remand to correct an alleged computer input error at observation number 7, index document number 9, at page 2 of the data. Serampore states that the data relating to that observation indicates that the sale was f.o.b., but in the ocean freight column, there is an entry of 1.1 rather than 0.0. Serampore recognizes this as an error from the fact that observation number 6 was the same sale as observation number 7, and observation number 6 shows 0.0 ocean freight.
Commerce and defendant intervenors state that because the remand proceeding was limited to the Central Sales Tax, the excise duty drawback, and the “all others” rate, the alleged error did not fall within the scope of the remand, and argues that Serampore cannot now raise an issue that could have been raised before this Court's decision in November of 1987. See Alberta Pork Producers’ Marketing Board v. United States, 12 CIT - at -, 683 F.Supp. 1398 at 1403 (1988).
Serampore responds that a clerical error should be corrected whenever one is discovered.
The Court is loathe to affirm a determination that might be based on a questionable record. Because this action is being remanded to Commerce for calculation of the full tax incidence, this part of the action is remanded, as a matter of the Court’s discretion, for Commerce to determine whether there is an error in the computer input calculation. If Commerce agrees there is an error, Commerce is directed to make the appropriate corrections.
CONCLUSION
Commerce’s methodology in calculating the all others rate is affirmed. Commerce’s corrected calculations for the Indian Central Sales Tax and the Excise Duty Drawback are affirmed. Serampore’s motion for remand is granted as to the Freight Equalization Fund Levy, turnover tax, and the alleged computer input error. Commerce shall make the appropriate calculations within 30 days. Any party contesting the results of this second remand shall file comments with the Court within 15 days of the remand results, after which Commerce will have 10 days in which to file a reply.