Opinion
Introduction
Musgrave, Judge:
Plaintiffs Sharp Corporation and Sharp Electronics Corporation (collectively, “Sharp”) challenge the final results of the [569]*569administrative review of the antidumping finding T.D. 71-76 by the International Trade Administration (“ITA”) for the period April 1, 1980 through March 31, 1981, announced in Television Receivers Monochrome and Color From Japan; Final Results of Antidumping Duty Administrative Review and Determination Not to Revoke in Part, 54 Fed. Reg. 35,517 (August 28, 1989). Plaintiffs move for judgment on the agency record under Rule 56.1. This Court has jurisdiction under 28 U.S.C. § 1581(c) (1991).
Sharp raises three grounds for review. First, Sharp alleges that no substantial evidence supports the ITA’s rejection of the methods used in the 1983 preliminary finding1 of no dumping margins for Sharp. Second, Sharp argues that the use of best information available (“BIA”) is not supported by substantial evidence. Finally, Sharp argues that no substantial evidence supports the use of another company’s dumping rate (the highest rate for any company published in the earlier final results for the same review period2) as BIA on the ground that it was adverse to Sharp.
Background
Antidumping duties are normally assessed “in an amount equal to the amount by which the foreign market value exceeds the United States price for the merchandise.” 19 U.S.C. § 1673 (1991). Foreign market value (“FMV”) is defined in part as the price at which such or similar merchandise is sold in the principal markets of the exporting country. 19 U.S.C. § 1677b(l) (1991). In the preliminary determination of the foreign market value of the televisions covered by this review, the ITA used Sharp’s sales to its distributors in Japan. Sharp had ownership interests in all of the distributors. Supplemental Administrative Document Number 6.
Under Commerce Department regulations, sales to related parties will ordinarily not be used to calculate FMV unless the Secretary of Commerce is satisfied that they are at prices comparable to sales to unrelated persons, or colloquially, at “arms-length.” 19 C.F.R. § 353.22(b) (1981); 19 C.F.R. § 353.45(a) (1991). From the outset of the review, the ITA was concerned with whether Sharp’s sales to its related Japanese distributors were at arms-length.
In a letter dated September 25, 1981, the ITA requested that Sharp provide gross sales amounts, cost of goods sold, gross profit, selling, general and administrative expenses and net operating income for five of its related distributors, to be used in a statistical test of whether the sales to the distributors were at arms-length. Administrative Document Number 274. Sharp submitted the information, which was later verified. [570]*570Administrative Document Number 278. An ITA memorandum describes the test:
Based upon detailed information pertaining to four of Sharp’s distributors selected by the Department, we developed weighted average ratios of gross margin (net sales less cost of goods) to net sales, and operating margin (gross margin less S.G. & A. [selling, general and administrative expenses]) to net sales, with regard to the distributors’ performance in the sales of color TVs. * * * We then developed similar ratios for these selected distributors with regard to their performances in the sales of all electric appliances. * * * The same ratios [were calculated for Sharp’s remainingten distributors as a group, and for the aggregate of all fourteen distributors.] This information was considered in conjunction with data published by the Small and Medium Enterprise Agency of the Japanese Ministry of International Trade and Industry. The MITI data for Electric Appliance Wholesalers indicates the industry-wide performance with regard to sales of all products. * * * We considered the foregoing information in an effort to determine whether the [color television] experience of the Sharp distributors selected by the Department was comparable to the all-product experience of those distributors, whether the all-product experience of the selected distributors was comparable to the all-product experience of all Sharp distributors, and whether the all-product experience of all Sharp distributors was comparable to the all-product experience of the industry in Japan.
Supplemental Administrative Document Number 6.
After performing this analysis, the ITA noted that although it found that the gross profits of the examined distributors differed from the experience of the industry, it also found that the variation was reasonable because Sharp’s related distributors assume greater marketing responsibilities than would be expected of unrelated purchasers. Supplemental Administrative Document Number 6. The ITA concluded “Sharp’s sales to related distributors in the home market have been satisfactorily demonstrated to be at prices comparable to those at which such or similar merchandise is sold to persons unrelated to the seller.” Using the home market prices, the ITA preliminarily determined that Sharp had zero dumping margins for the period. Television Receiving Sets, From Japan; Preliminary Results, 48 Fed. Reg. at 37507-8 (August 18,1983).
The ITA subsequently became suspicious of the statistical method. In January, 1984, the ITA notified Sharp that it wanted additional information, Plaintiff’s Memorandum, at 10; and in a meeting to discuss the request with counsel on May 15, 1984, the ITA informed Sharp that it had “grave doubts about the type of test used” in the review. Administrative Document Number289, at 2. Although no copy of the request appears in the record, the only topic discussed at the meeting was the use of Sharp’s sales to related distributors to establish FMV. Administrative Document Number 289.
At the meeting, counsel for Sharp noted that the ITA apparently wanted to disregard past practice and examine sales at the next level of [571]*571distribution. He argued against the change, but said that Sharp would respond promptly once a decision was made. Administrative Document Number 289, at 2. The ITA, through a representative, noted that the deadline for submission of the information had passed, but indicated that an extension would be granted. Id. The ITA stated that it was looking for something to verify the validity of the test. Id.at 3.
At a follow-up meeting held on June 13,1984, the ITA provided Sharp with a copy of an ITA memorandum, dated May 22, that details the problems the ITA perceived with the statistical analysis:
1. The type of test used — we [ITA] used a confidence interval test but probably should have used a hypothesis test * * *.
2. Errors in the calculations — the formula used to arrive at the degrees of freedom was incorrect. In addition the record indicates that the calculation taking the weighted mean of the two sample variances was omitted. Consequently the figure derived for the standard error of the differences between the means was incorrect.
3. We did not use direct comparisons — instead of making direct comparisons on sales of the subject merchandise we compared profit results for loosely related products * * *.
4. Sample size was too small — our basic sample consisted of four distributors and our entire sample consisted of fourteen related distributors.
5. We obtained no information on the MITI data used — the MITI data submitted was excerpted from a volume titled “Management Indicator of Small and Medium Sized Enterprises” and the record contains one page of data showing average operating results of electric appliance distributors.
6. There was wide variation of profit experience on sales of color televisions among the four Sharp distributors examined * * *.
Administrative Document Number 286. The memo concluded that the only solution was to request that Sharp submit information on sales to unrelated dealers. Id. at 2. A representative of the ITA then stated that the test used in the preliminary determination was inadequate, and that Sharp should submit the information requested in January, 1984. Administrative Document Number 290.
Sharp responded by letter dated July 19,1984 with comments upon the memoranda provided by the ITA. Administrative Document Number 123. The letter states that the statistical analysis used by the ITA
is completely devoid of value because there is no basis in this case for inferential statistical analysis. Under a proper analysis it is absolutely clear that the use of Sharp’s selling prices to related distributors for purposes of calculating foreign market values is both reasonable and appropriate in this case.
Administrative Document Number 123, at 1. Because the ITA had information about all of Sharp’s Japanese distributors, the letter argues, the ITA should have used descriptive, rather than inferential, statistical methods. Mat 5. Therefore, the ITA analysis should be rejected outright. Id.
[572]*572Two years later, on August 1, 1986, the ITA wrote to Sharp stating that it had found certain (but unspecified) deficiencies in Sharp’s questionnaire response. Administrative Document Number 295. The letter requested that Sharp, “On computer tape, provide all home market sales data for the second review period from your related distributors to the first unrelated customers * * And, “On computer tape, provide all exporter’s sales price data from Sharp Electronics Corporation to its unrelated customers for the second review period. ”/ci.
Sharp responded on August 18, 1986, stating that, “[Tjhere are no tapes in existence containing the requested data. It will therefore be necessary to keypunch the data at very considerable expense. ’’Supplemental Administrative Document Number 7. The letter requested that the ITA reconsider its request for a computer tape. Id. By telephone, the ITA subsequently notified Sharp that it waived the requirement that the data be submitted on computer tape, and gave Sharp an extra month to comply. Administrative Document Number 296.
On September 23,1986, the ITA notified Sharp that despite the extension, the ITA had not received the information requested of Sharp, and had determined to use BIA for the second review. Administrative Document Number 297.
Sharp responded on September 29,1986, again arguing that the sales to related distributors were adequate to determine FMV, and that the data the ITA requested was unnecessary. Administrative Document Number 298. Sharp also advised the ITA, apparently for the first time, that neither Sharp nor its related distributors had the requested data. Id., at 2.
On October 8,1986, Sharp submitted the affidavit of a statistical expert, who reiterated many of the criticisms Sharp had earlier voiced in Administrative Document Number 123. The affidavit concludes that the profit experience of Sharp distributors is statistically equivalent to the profit experience of the entire industry in Japan as measured by the government data. Administrative Document Number 299.
About two years later, the ITA announced that it was going to use as BIA for Sharp the highest rate from the prior review of other firms. Preliminary Results of Antidumping Duty Administrative Review and Intent to Revoke in Part, 53 Fed. Reg. 33,164 (August 30, 1988).
At a hearing held on October 24,1988, the ITA produced a memorandum dated April 3,1985, which recommends rejecting the statistical test proposed by Sharp in its July 19,1984 letter. Administrative Document Number 291. The memorandum argues that similarity of profit experience between Sharp’s distributors and other Japanese distributors does not indicate that the sales were at arm’s length. It points out that by in effect transferring to the distributor selling, general and administrative expenses that would normally be borne by the manufacturer, the manufacturer can reduce the price to a related distributor (achieving a lower FMV) while the price to the first unrelated seller and net profit remain the same. Id.
[573]*573The April 3,1985 memorandum also points out that a chart in Sharp’s July 19,1984, submission showed aggregate data for Sharp’s distributors, not restricted to color televisions. It notes that an examination of the data for just color televisions shows one of the distributors to be outside the range of statistical equivalence. The memorandum also reiterates the ITA’s questions about the MITI data. Id.
The memorandum states that, “The statistical test doesn’t prove anything about arm’s length transactions. All it shows is that profit experience is similar.” Id., at 2. Finally, the memorandum indicates that the ITA requested that Victor of Japan and NEC provide information about sales by related distributors to unrelated dealers, and questions whether a different approach should be used only for Sharp. Id.
Sharp had not previously seen the 1985 memorandum, and submitted comments on December 7, 1988. Administrative Document Number 309. Sharp’s comments address the issues it raises here. On August 28, 1989, the ITA published the final results at issue in this action, and established a 0.86 percent antidumping duty rate for Sharp. 54 Fed. Reg. 35,517, 35,521.
Discussion
The standard of review for this case is set forth in 19 U.S.C. § 1516a(b)(l): “The court shall hold unlawful any determination, finding, or conclusion found * * * to be unsupported by substantial evidence on the record, or otherwise not in accordance with law.” Substantial evidence is “such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.” N.A.R. S.p.A. v. United States, 14 CIT 409, 412, 741 F. Supp. 936, 939 (1990). “This is something less than the weight of the evidence, and the 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 Commission, 383 U.S. 607, 619-620, 16 L. Ed. 2d 131, 141, 86 S. Ct. 1018, 1026 (1966) (citations omitted).
The decision to use BIA was based upon Sharp’s failure to provide the supplemental information requested by the ITA. Television Receivers, From Japan; Final Results, 54 Fed. Reg. at 35,521 (Comment 14). The decision to request supplemental information was in turn based upon the conclusion that the statistical information used in the preliminary determination was insufficient to demonstrate that sales between Sharp and its related distributors were at arm’s length. Id. Sharp argues that this conclusion is not supported by substantial evidence on the record.
The ITA responds only that there is no basis for requiring Commerce to demonstrate that it properly rejected its own analysis. Defendant’s Memorandum, at 33. While it is true that the ITA has discretion to change its methods or request additional information, See, e.g., Ud-[574]*574deholm Corp. v. United States, 11 CIT 969, 971, 676 F. Supp. 1234, 1237 (1987), its discretion is not without bounds.
“The grounds upon which an administrative order must be judged are those upon which the record discloses that its action is based. ’’Securities Exchange Commission v. Chenery Corp., 318 U.S. 80, 87, 87 L. Ed. 626, 633, 63 S. Ct. 454, 459 (1943). The final determination states that the request for additional information which led to the use of BIA was based upon the conclusion that “for reasons stated in the record * * * the statistical information was insufficient to demonstrate that sales between Sharp and its related distributors were at arm’s length.” Television Receivers, From Japan; Final Results, 54 Fed. Reg. at 35,521. Since the request for additional information rests squarely upon this conclusion, this Court must hold the determination unlawful unless this conclusion is supported by substantial evidence on the record. 19 U.S.C. 1516a(b)(l). The Court finds that it is.
While many of the ITA’s criticisms of its preliminary results in the record focus upon the statistical methods used, the final determination refers to statistical information. This is perhaps in response to Sharp’s unusual argument that although the ITA’s statistical test “was completely devoid of value” because it used inferential methods, another statistical test applied to the same information compels the ITA to accept its earlier conclusion as a matter of law.
Among the reasons stated in the record for the conclusion that the statistical information was inadequate, is the ITA’s dissatisfaction with the comparison of aggregate MITI sales data for electrical appliances with Sharp’s sales information. For example, one ITA memorandum states, “We obtained no information on the MITI data used — the MITI data submitted was excerpted from a volume titled ‘Management indicator of Small and Medium Sized Enterprises’ and the record contains one page of data showing average operating results of electrical appliance distributors.” Administrative Document Number 286.
Sharp responds without elaboration that the author oí Administrative Document Number 286 was evidently unfamiliar with the proceedings, because the ITA had earlier described the MITI data, in its letter of September 25, 1981, to Sharp proposing the statistical test, as
[IJndustry data published by the Small and Medium Enterprise Agency of the Japanese Ministry of International Trade and Industry (“MITI”), which concerns the financial operating ratios of Japanese electric appliance wholesalers, unrelated to manufacturers. Such wholesalers are engaged in the sales of consumer electric products, including television receivers.
Administrative Document Number274. The Court does not perceive any fatal contradiction in the two descriptions.
An apparent inconsistency in the ITA’s description of the MITI data does occur in the record. In its internal memorandum of April 3,1985, the ITA questioned whether the MITI report showed only operating results only for wholesalers unrelated to manufacturers, Administrative [575]*575Document Number 291, despite the statement in Document 274 that it did. However, this may be explained by the fact that the ITA did not possess the MITI report referred to in Documents 286 and 291 until December, 1981, several months after Document 274 was written. Administrative Document Number 277.
The record contains a paucity of information on the report. The single page photocopy of the original, in Japanese, appears in the administrative record. An accompanying letter from Sharp’s counsel, who provided the report, indicates that an English translation accompanies the letter. Administrative Document Number 277. No copy of this translation appeared in the record at the submission of this case, although it is apparent that the ITA made use of it. The Court by letter requested that the ITA produce a copy of the translation, which was done, and the Court now supplements the record therewith.
The translation, which is only partial, describes the report as “Wholesale Business for Electric Appliances (By number of employees).” The headings for the columns are translated, with headings for “Classification by number of employees,” four categories ranging from “1 to 5” to “51 & more,” average for the four, standard deviation, “Deficit-ridden Business,” and average for the four categories plus deficit-ridden business. Only one row heading, of twenty-seven, is translated, entitled, “Ratio of operating profit to net sales.”
A copy of a letter from the ITA to the Japanese Embassy requesting verification and explanation of the MITI report appears in the administrative record, Administrative Document Number 120, but no response from the Embassy appears.
The record also reflects the ITA’s concern with the usefulness of the comparison of Sharp’s television sales with the aggregated profits on all sales of small electric appliances given in the MITI report. Direct comparisons of similar merchandise, not loosely related products, might be more appropriate. See Administrative Document Number 468. In addition, the ITA noted in the record that the aggregation of data in the MITI report might mask variation in profit experience in the same way that aggregation of profit information of Sharp’s distributors had. Administrative Document Number 471.
Sharp responds as it did administratively that variation in profit experience is to be expected in a free-market economy, and that the lack of such variability, not its presence, should arouse suspicion. This criticism is misplaced, because the ITA does not argue that such variability indicates a lack of arm’s length dealing. Rather, the variability makes it difficult to use the profit experience as an indicator. As stated in a memorandum in the record, “Given this variation among related distributors, our ability to use that data to determine the standard for the industry in arm’s length transactions is questionable.” Administrative Document Number286. The problem is not that the Sharp data does not meet the standard, but that the ITA could not determine what the standard was.
[576]*576The Court finds that a reasonable mind might accept the foregoing evidence as adequate to show that the statistical information contained in the record was insufficient to determine whether Sharp’s sales to its related distributors were at arm’s length. The decision by the ITA to request further information from Sharp is therefore supported by substantial evidence.
Sharp argues that, nevertheless, the ITA was not justified in resorting to BIA. Sharp cites Atlantic Sugar, Ltd. v. United States, 2 Fed. Cir. (T) 130, 134, 744 F.2d 1556, 1560 (1984) for the proposition that “the BIA rule is to be used (i) when parties are uncooperative and recalcitrant, and (2) [sic] to help Commerce comply with its statutory deadlines.” Plaintiffs’ Memorandum, at 32. Neither situation is involved here, Sharp continues, so use of BIA is unjustified.
19 U.S.C. § 1677e(c) states,
In making their determinations under this subtitle, the administering authority and the Commission shall, whenever a party or any other person refuses or is unable to produce information requested in a timely manner and in the form required, or otherwise significantly impedes an investigation, use the best information otherwise available.
The Atlantic Sugar Court noted the statute requires the International Trade Commission (“the Commission” referred to in the statute) to use BIA in the circumstances set forth. The Court went on to say that “Noncooperation by parties or other persons may, in the absence of ITC time to pursue judicial compliance, be penalized, at least in the eyes of those parties or persons, by the ITC’s mandatory use of whatever other best information it may have available.” Atlantic Sugar, 2 Fed. Cir. (T) at 134, 744 F.2d at 1560. Sharp attempts to construe this characterization of the BIA rule as a restriction on its use.
The fact that BIA may be viewed by uncooperative parties as a penalty, or may be viewed as “an investigative tool which [an] agency may wield as an informal club over recalcitrant parties,” id, does not imply that it may only be used in such instances. The ITA’s use of BIA here was based on Sharp’s failure to respond to its supplemental questionnaire. Television Receivers, From Japan; Final Results, 54 Fed. Reg. at 35,521. Assuming that the questionnaire withstands review, nothing more is required.
Similarly, although Atlantic Sugar notes that the BIA rule is set in the context of extremely short deadlines imposed on the ITC which make cooperation essential, the case does not imply that relaxation of deadlines obviates the requirement that BIA be used. In this case, the ITA relaxed its deadlines3 with Sharp’s consent, in order to allow Sharp to comply with its request for additional information. The fact that the [577]*577ITA was willing to allow Sharp more time to comply does not waive the requirements of § 1677e(c).
The schedule for § 751 reviews (19 U.S.C. § 1675, as amended) has been held to be directory, not mandatory. Nissan Motor Corp. In U.S.A. v. United States, 10 CIT 820, 824, 651 F. Supp. 1450, 1455 (1986). Although the eight year delay between the publication of the ITA’s intent to conduct this review and the publication of the final results is deplorable, the appropriate action by plaintiff to avoid delay would have been a suit to enforce the deadlines. Id. at 825.
Sharp argues further that reversal is required under Olympic Adhesives, Inc. v. United States, 8 Fed. Cir. (T) 69, 899 F.2d 1565 (1990). In that case, an ITA questionnaire inquired whether one of the companies involved had sold certain types of glue to any customers in Sweden other than one already reported. The company replied, “No. ” The ITA characterized this answer as a refusal to provide requested data, and resorted to BIA. Id. at 1573.
The Court of Appeals for the Federal Circuit reversed. The Court noted that the record showed that the company had made no sales of the type inquired after, and held that the ITA may not “characterize a party’s failure to list and give details of sales as a ‘refusal’ or ‘inability’ to give an answer where, in fact, there are no sales. ”Olympic Adhesives, 8 Fed. Cir. (T) at 76-77,899 F.2d at 1572. The ITA may not resort to BIA when the submitter cannot produce data because such data never existed. Id. at 1573.
Sharp does not argue that the information requested by the ITA never existed. In fact, Sharp’s letter of August 18, 1986, implies that Sharp possessed the information at that time. Supplemental Administrative Document Number 7. This is not a case in which the requested information “does not and could not exist.” U.H.F.C. Co. v. United States, 9 Fed. Cir. (T) 1, 15, 916 F.2d 689, 701 (1990). The record supports the conclusion that the data did exist and could exist if Sharp had not disposed of it.
As stated in Olympic Adhesives, “the ITA may properly request additional supplemental information, if needed, to fully resolve the issue, [but] section 1677e(b) clearly requires noncompliance with an information request before resort to the best information rule is justified, whether due to refusal or mere inability. ’’Olympic Adhesives, 8 Fed. Cir. (T) at 79, 899 F.2d at 1574 (emphasis in original). Such is the situation at bar: the ITA properly requested additional information, Sharp did not comply with the request, and the ITA resorted to BIA.
Sharp next asserts that the selection of the highest rate for any company for the period as BIA was punitive, and unreasonable. Sharp argues that the unavailability of the data was due to the ITA’s delay in requesting it, and not Sharp’s fault. The most accurate information available, Sharp maintains, was the data it had submitted, and that is what the ITA should have used.
In Rhone Poulenc, Inc. v. United States, 8 Fed. Cir. (T) 61, 899 F.2d 1185 (1990), the Court of Appeals noted that for an application of the [578]*578BIA rule to be characterized as punitive, the agency would have had to reject low margin information in favor of high margin information that was demonstrably less probative of current conditions. As this Court has concluded supra that substantial evidence supported the ITA’s conclusion that the statistical information was insufficient to demonstrate that the sales to related dealers could be used to determine dumping margins, that information was not demonstrably more probative of contemporaneous conditions than the highest margin for any company for the same period.
The Rhone Poulenc Court held that a permissible interpretation of the best information statute allows the ITA to raise a rebuttable presumption that the highest prior dumping margin for the company involved was the best information of current margins. Rhone Poulenc, 8 Fed. Cir. (T) at 67,899 F.2d at 1190. The presumption, the Court held, reflects the common sense inference that if the highest prior margin were not the most probative of current margins, the importer would have produced current information showing the margin to be less. Id. at 1190. The approach fairly places the burden of production on the importer, who possesses the information necessary to rebut the presumption. Id. Such a presumption is not punitive, the Court concluded. Id.
In this case, the ITA selected for use as BIA, “information that was adverse to the firms.” Television Receivers, Monochrome and Color, From Japan; Final Results, 54 Fed. Reg. at 35,518. This consisted variously of the highest rate for any company during the same period for Sharp; the company’s last rate for Funai Electric, and the firm’s own data for NEC. Adversity was thus a criterion by which the information used for BIA was selected.
The Court holds that the ITA’s use of adverse information as BIA for Sharp is permissible, under the rationale of Rhone Poulenc. Substantial evidence in the record supports the conclusion that Sharp possessed the information necessary to rebut the presumption that the best information available was information that was adverse to Sharp.
The 0.86 percent dumping margin that the ITA selected as BIA is the highest for any company covered by Television Receiving Sets, Monochrome and Color, From Japan; Final Results of Administrative Review of Antidumping Finding, 50 Fed. Reg. 24,278 (1985). The review covered sixteen of the eighteen known Japanese manufacturers and exporters of televisions to the United States for the period April 1, 1980 through March 31, 1981. Id. This is the same period covered by the review at issue. Of the sixteen covered manufacturers and exporters, seven were found to have dumping margins of 0.86 percent. The Court finds that a reasonable mind might accept this evidence as adequate to support the conclusion that the 0.86 percent dumping margin was the best information otherwise available for Sharp. The use of the 0.86 percent rate is supported by substantial evidence in the record, and is otherwise in accordance with law.
[579]*579Conclusion
The decision by the ITA to request additional information from Sharp, the ITA’s subsequent use of BIA to determine the dumping margin for Sharp, and the ITA’s selection of information used as BIA are supported by substantial evidence in the record and otherwise in accordance with law. The final results of the antidumping administrative review of Sharp are affirmed.