66 FR 56274, November 7, 2001 A-489-807 AR: 04/99-03/00 Public Document IA/Gr. 1/Off. 2/I. Itkin MEMORANDUM TO: Faryar Shirzad Assistant Secretary for Import Administration FROM: Richard W. Moreland Deputy Assistant Secretary, Group I Office of AD/CVD Enforcement SUBJECT: Issues and Decision Memorandum for the Antidumping Duty Administrative Review on Certain Steel Concrete Reinforcing Bars from Turkey -- April 1, 1999, through March 31, 2000 Summary We have analyzed the comments of the interested parties in the 1999-2000 administrative review of the antidumping duty order covering certain steel concrete reinforcing bars from Turkey. As a result of our analysis of the comments received from interested parties, we have made changes in the margin calculations as discussed in the "Margin Calculations" section of this memorandum. We recommend that you approve the positions we have developed in the "Discussion of the Issues" section of this memorandum. Below is the complete list of the issues in this administrative review for which we received comments from parties: 1. Use of Facts Available 2. Exchange Rates used for Currency Conversions 3. Cost Recovery Test 4. Home Market Sales Priced in U.S. Dollars 5. Kur Farki Adjustment 6. Colakoglu's Home Market Credit Expenses 7. Colakoglu's Home Market Indirect Selling Expenses 8. Different Costs for the Same Products Produced by Colakoglu 9. Unreported Costs for Colakoglu 10. Colakoglu's Production Quantities 11. Colakoglu's Depreciation Expenses 12. Colakoglu's G&A Expenses 13. Colakoglu's Financing Expenses 14. Costs for Different Grades of Rebar Produced by Diler 15. Unreported Material Costs for Diler 16. Diler's Depreciation Expenses 17. Diler's G&A Expenses 18. Diler's Financing Expenses 19. Selling Expenses for Constructed Export Price 20. Ekinciler's Home Market Freight Expenses 21. Ekinciler's U.S. Freight Expenses 22. Ekinciler's Home Market Credit Expenses 23. Ekinciler's Scrap Costs 24. Ekinciler's Depreciation Expenses 25. Ekinciler's G&A Expenses 26. Use of Consolidated Financing Expenses for Ekinciler 27. Calculation of Ekinciler's Financing Expenses 28. New Factual Information 29. ICDAS's Scrap and Labor Costs 30. ICDAS's Secondary Materials 31. ICDAS's Packing Costs 32. Treatment of ICDAS's Foreign Exchange Gains and Losses and Interest Background On May 4, 2001, the Department of Commerce (the Department) published the preliminary results of the administrative review of the antidumping duty order on certain steel concrete reinforcing bars (rebar) from Turkey. See Certain Steel Concrete Reinforcing Bars from Turkey; Preliminary Results of Antidumping Duty Administrative Review, 66 FR 22525 (May 4, 2001) (Preliminary Results). The product covered by this order is rebar. The period of review (POR) is April 1, 1999, through March 31, 2000. We invited parties to comment on our preliminary results of review. Based on our analysis of the comments received, we have changed the results from those presented in the preliminary results. Margin Calculations We calculated export price and normal value using the same methodology stated in the preliminary results, except as follows: 標e have corrected a clerical error in the exchange rates used to convert Turkish lira into U.S. dollars for all respondents. See Comment 2. 標e increased Colakoglu's depreciation expenses to account for a portion of the depreciation expenses recorded in a separate cost center, using facts available. See Comment 11. 標e included certain translation losses in the calculation of Colakoglu's financing expenses. See Comment 13. 標e increased Diler's reported costs to account for certain movement charges on imported materials. See Comment 15. 標e have adjusted Diler's cost of manufacturing (COM) to reflect the increase in depreciation expense related to the revaluation of buildings to current currency levels. In addition, we increased Diler's reported depreciation expenses to account for revaluation of buildings. See Comment 16. 標e removed packing expenses from the denominator of Diler's general and administrative (G&A) expense ratio. See Comment 17. 標e removed foreign exchange gains and losses associated with accounts payable transactions from G&A or financing expenses, as appropriate, for Diler, Ekinciler, and ICDAS. In addition, we incorporated certain auditor's adjustments reflected in Diler's financial statements in the calculation of the company's financing expenses. See Comments 18, 27 and 32. 標e increased Ekinciler's depreciation expenses to account for unrecognized depreciation on one category of assets. See Comment 24. 標e adjusted ICDAS's reported transfer prices for scrap in the months of May and September 1999 to reflect market prices. See Comment 29. 標e increased ICDAS's reported costs to account for the cost of internal packing materials. See Comment 31. Discussion of the Issues General Issues Comment 1: Use of Facts Available The petitioner argues that the Department should reject the responses of each of the companies participating in this administrative review and instead base the final dumping margins for them on adverse facts available. Specifically, the petitioner contends that the Department discovered at verification that each of the respondents failed to report certain home market sales made during the POR. For example, the petitioner notes that Ekinciler did not report sales of "non-standard" rebar, while ICDAS did not report a small quantity of mixed-length bundles. According to the petitioner, neither the fact that the volume of omitted sales was small for each company nor the fact that the average price of Ekinciler's unreported sales was lower than the average price of the reported sales is a mitigating factor in this case. The petitioner contends that even a single sale could be critical in its impact on normal value because it might be matched to many U.S. sales. Moreover, the petitioner asserts that average prices do not represent an appropriate benchmark, especially where the home market experienced high inflation during the POR. As a consequence, the petitioner argues that each of the respondents failed the test of home market completeness, and therefore failed verification. Accordingly, the petitioner argues that the Department should reject each company's response and rely on total facts available for purposes of the final results. Failing that, the petitioner argues that the Department should apply adverse facts available to all U.S. sales for which there is no identical home market match. As facts available, the petitioner contends that the Department should use the higher of the "all others" rate or the most adverse margin established in this review. The petitioner maintains that the Department has a long-standing practice of using total facts available when there are essential components of the response which are inaccurate or unreliable. As support for this assertion, the petitioner cites Steel Authority of India v. United States, 149 F. Supp. 2d 921, 926 (Court of International Trade (CIT) 2001) (Steel Authority of India). According to the petitioner, home market sales are such an essential component of the response that their absence makes it impossible to serve the basic purpose of the statute - determining the current margin as accurately as possible. Furthermore, the petitioner contends that failing to use total facts available in this situation enables the respondents to control the results of the review by providing partial information. The respondents disagree that they failed verification. Indeed, the respondents state that the petitioner's assertion that they did so borders on the frivolous, given that the quantity of the unreported sales in question were small. (1) Moreover, Colakoglu and Diler note that the differences between the home market sales listings and their accounting records were explained to the Department, and (in Diler's case) corrected, at verification. Ekinciler asserts that it reported all home market sales of rebar made to standard specifications and that the sales of non-standard length rebar were of non-prime merchandise. According to Ekinciler, the Department has a long-standing practice of comparing U.S. sales of prime merchandise only to home market sales of prime merchandise. To support this assertion, Ekinciler cites Final Determinations of Sales at Less Than Fair Value: Certain Hot-Rolled Carbon Steel Flat Products, Certain Cold-Rolled Carbon Steel Flat Products, and Certain Cut-to-Length Carbon Steel Plate from Korea, 58 FR 37176, 37183 (July 9, 1993), Final Determination of Sales at Less Than Fair Value: Certain Stainless Steel Cooking Ware from Korea, 51 FR 42873, 42877 (Nov. 26, 1986) (Korean Cookware), and Final Determination of Sales at Less Than Fair Value: Certain Fresh Cut Flowers from Colombia, 52 FR 6842, 6850 (Mar. 5, 1987) (Colombian Flowers). For example, Ekinciler asserts that, in Colombian Flowers and Korean Cookware, the Department excluded from its calculations sales of non-prime merchandise when such sales only existed in the home market, without determining whether the merchandise differed in terms of the Department's model matching methodology. Therefore, Ekinciler contends that the Department would not have used the sales of non-standard length rebar in its price comparisons, and the non-reporting of non-standard rebar had no effect on the Department's preliminary results. ICDAS contends that it could have only benefitted from reporting the sales in question, because, for each month of the POR, the weighted- average gross unit price of the unreported sales was lower than the weighted-average price of the reported sales. ICDAS, therefore, concludes that the reporting of these sales would have reduced the weighted-average price used in the preliminary results and, by extension, the dumping margin. Furthermore, ICDAS notes that the reporting of these sales would have no effect on the results of the cost test because they were of short- and mixed-length bundles. According to ICDAS, the company's accounting records do not consistently reflect the diameter of these bundles, which would prohibit ICDAS from accurately assigning them control numbers. In fact, ICDAS notes that, given the limitations of its accounting records, it cannot determine with any degree of certainty whether a significant portion of the omitted sales were of subject merchandise. In any event, ICDAS notes that these sales were made in such a small quantity that, even were the Department to assume that they were all of the same control number and find that they were all made at below-cost prices, there would still be no effect on the dumping analysis because the vast majority of sales used as the basis for normal value were above cost. Finally, ICDAS contends that the petitioner's suggestion that the omitted sales might be significant for any U.S. sale for which there is no identical match does not apply to ICDAS, given that the Department matched each of ICDAS's U.S. sales to numerous identical products sold in the home market. According to ICDAS, the Department has recognized that applying facts available to a respondent who, to its detriment, omits an insignificant number of sales serves no justifiable purpose. To support this statement, ICDAS cites Final Determination of Sales at Less Than Fair Value: Oil Country Tubular Goods from Austria, 60 FR 33553, 33554 (June 28, 1995), where the Department found that respondent's failure to report certain sales was adverse to its interests and as a consequence declined to take further adverse action. ICDAS maintains that it has already suffered sufficient adverse effects as a result of the omission of the sales in question because the Department both disallowed the offset to direct materials costs for the revenue generated by these sales and did not adjust the calculation of the per-unit cost of production (COP) to include the production quantity of the omitted products. ICDAS asserts that the petitioner's citation to Steel Authority of India is misleading. According to ICDAS, in Steel Authority of India, the Department resorted to facts available because the essential components of the respondent's submissions were so "riddled with errors and inaccuracies as to be unreliable." ICDAS asserts that, in that case, the respondent failed to: 1) report a significant number of home market sales; 2) report accurate prices; 3) reconcile production costs to its financial statements; and 4) provide constructed value data for products sold to the United States. In contrast, ICDAS asserts that the Department verified that the sales information that it reported was substantially complete and useable. Indeed, ICDAS contends that the Department recognized this fact by relying on this information for purposes of the preliminary results. Based on the foregoing, the respondents maintain that there is no legal or factual basis to support the use of facts available in this review. Department's Position: We agree with the respondents that total facts available is not warranted in this case. According to section 776(a) of the Act, the Department shall use the facts otherwise available in reaching a determination if: 1) necessary information is not available on the record, or 2) an interested party or any other person -- A) withholds information that has been requested by the administering authority or the Commission under this title, B) fails to provide such information by the deadlines for submission of the information or in the form and manner requested, subject to subsections (c)(1) and (e) of section 782, C) significantly impedes a proceeding under this title, or D) provides such information but the information cannot be verified as provided in section 782(i). In this proceeding, none of these requirements is met for any of the respondents. Specifically, we find that each company cooperated fully in this proceeding by responding to each of the Department's requests for information in a timely manner. Moreover, while it is true that each company failed to report a very small quantity of home market sales, none of the submitted responses were so incomplete that they could not be used. Regarding Diler, we found at verification that this respondent's home market sales listing: 1) did not include a small quantity of unreported home market sales of rebar which should have been reported; 2) contained certain reported sales which should not have been reported; and 3) contained a limited number of discrepancies regarding date of sale, date of shipment, and product characteristics. See the March 20, 2001, memorandum to Louis Apple from Irina Itkin and Jason Hoody, entitled "Verification of the Sales Questionnaire Responses of Diler Demir Celik Endustrisi ve Ticaret A.S. in the Antidumping Duty Administrative Review on Certain Steel Concrete Reinforcing Bars from Turkey" at page 4. However, because company officials were able to identify each of these errors at verification, we instructed Diler to submit a corrected database which we used in our preliminary margin calculations. Given that complete and accurate home market sales information exists on the record of this case for Diler, there is no reason to base Diler's final dumping margin on facts available. Regarding the remaining respondents, at verification we found that Colakoglu and Ekinciler failed to report sales of short-length rebar, while ICDAS failed to report sales of mixed-length bundles. All of these sales were unusual in some respect vis-a-vis the respondents' other home market sales. For example, Colakoglu's sales were further processed to meet the requirements of specific customers, and thus had a different cost structure. In addition, Colakoglu invoiced certain of the short-length sales by hand, while computer-generated invoices were issued for sales of standard-length rebar. Similarly, ICDAS's invoices for mixed-length bundles did not reference the diameter of the rebar (in contrast with standard-length sales), and in general, it charged lower prices for these transactions. Finally, Ekinciler considered these sales to be of second quality merchandise, and it also charged lower prices for them. Despite the petitioner's assertion that any of the sales in question could be potential matches for U.S. sales, we disagree. Based on the information currently on the record, we do not consider it likely that any of these sales would be appropriate matches for U.S. sales. Specifically, given the respondents' selling and pricing practices vis-a-vis these non- standard products, we find it inappropriate to compare U.S. sales of standard-length rebar to home market sales of non-standard-length rebar. (2) Therefore, for purposes of these final results, we have not revised our treatment of the unreported sales from the preliminary results. We disagree with the petitioner that the circumstances in Steel Authority of India are similar to the situation in this case. In Steel Authority of India, the Department applied total facts available to the respondent only after concluding that its questionnaire responses could not be verified. Specifically, in that case the Department found that the respondent "failed to report a significant number of home market sales; was unable to verify the total quantity and value of home market sales; and failed to provide reliable cost or constructed value data for the products." See Steel Authority of India, 149 F. Supp. at 925, quoting Final Determination, 64 FR at 73127. As noted above, each of the respondents in this case reported reliable and verifiable sales and cost data. Therefore, we find that the factual scenario in Steel Authority of India is not representative of the situation here. Comment 2: Exchange Rates used for Currency Conversions Colakoglu and Ekinciler contend that the exchange rate database used for the preliminary results appears to contain clerical errors. Specifically, the respondents note that the Department stated that it intended to use exchange rates for the Turkish lira published by the Dow Jones News/Retrieval Service (Dow Jones) to perform all currency conversions, but the rates actually used differed from the Dow Jones rates in a number of instances. According to the respondents, a similar problem occurred in the 1998-1999 administrative review involving carbon steel pipe and tube from Turkey. See Certain Welded Carbon Steel Pipe and Tube From Turkey; Final Results of Antidumping Duty Administrative Review, 65 FR 37116, 37117 (June 13, 2000). The respondents maintain that, in that case, the Department admitted that a clerical error had occurred in transferring the exchange rate data from Dow Jones to a spreadsheet, and, as a consequence, it used the correct rates for the final results. The petitioner asserts that there is no way for it to tell which of the sets of rates in question (i.e., the rates used for purposes of the preliminary results or the ones provided in the respondents' case briefs) is accurate. Nonetheless, the petitioner contends that, if the Department uses different rates for purposes of the final results, it should fully explain the reasons for the change. The petitioner contends that the Department should not allow the respondents to "cherry-pick" rates from various sources simply to achieve a desired result. Department's Position: The Department's preferred source for daily exchange rates is the Federal Reserve Bank. However, the Federal Reserve Bank does not track or publish exchange rates for Turkish Lira. Therefore, for purposes of the preliminary results, we intended to make currency conversions based on the daily exchange rates from Dow Jones. We agree that the exchange rates used for the preliminary results did not correspond to the Dow Jones rates in a number of instances. As a result, we have corrected the rates in our database to reflect the actual rates published by Dow Jones and have used them for purposes of the final results. Comment 3: Cost Recovery Test The Department is conducting a sales-below-cost test for each of the respondents in this proceeding. Three of these companies (i.e., Diler, Ekinciler, and ICDAS) notified us that they had U.S. sales in only certain months of the review period and requested that they be allowed to report home market sales data only for the six-month periods surrounding these sales. As part of its request, each respondent agreed to forego the cost recovery test. Accordingly, we allowed each company to report home market sales data for the six-month period requested. The petitioner argues that the Department should ignore all sales found to be below the cost of production, based on the respondents' agreement to forego the recovery of cost test. The respondents note that there are two separate tests under the Act which must be met before the Department may disregard sales made at prices below the cost of production. One is the cost recovery test cited by the petitioner, while the other is the "substantial quantities" test. Regarding this latter test, the respondents assert that section 773(b)(2)(C) of the Act requires that the volume of below-cost sales be at least 20 percent of the volume of total sales of a given product before such sales may be disregarded. The respondents contend that, even though they may have waived the cost recovery test, they did not waive the substantial quantities test. (3) Thus, they argue that, where less than 20 percent of the sales volume of a particular product is found to be made at below-cost prices, those below-cost sales should not be disregarded in accordance with section 773(b) of the Act. The respondents note that this interpretation of the Act is consistent as well with the Department's practice. In support of this assertion, the respondents cite Preliminary Results and Partial Rescission of Antidumping Duty Administrative Review; Certain Pasta from Turkey, 65 FR 48474, 48476 (Aug. 8, 2000), where the Department stated that it performed the substantial quantities test for one respondent who reported only six months of home market sales data. Department's Position: In accordance with section 773(b)(1) of the Act, the Department may disregard home market sales made at prices below the cost of production when the below-cost sales are made: 1) within an extended period of time in substantial quantities; and 2) at prices with do not permit recovery of all costs within a reasonable period. It is not sufficient for only one of these conditions to apply; both have to be met in order for the Department to disregard sales at below-cost prices. See, e.g., Certain Pasta from Turkey; Final Results of Antidumping Duty Administrative Review, 65 FR 77857 (Dec. 13, 2000). In this case, each of the three respondents in question agreed to forego the cost recovery test as a condition of its being allowed to limit the home market reporting period. However, none of these companies agreed to forego the substantial quantities test. As a result, we performed this latter test in accordance with our normal practice for purposes of the preliminary results. Specifically, we determined the volume of below-cost sales made of each product over the six-month reporting period. Where that volume equaled or exceeded 20 percent, we found that the below-cost sales of that product had been made in substantial quantities. The petitioner implies that, because the respondents did not report data for the entire POR, the Department does not have sufficient information to conduct the substantial quantities test. We disagree. Section 773(b)(1)(A) of the Act provides that the Department may disregard below-cost sales only where they are made "within an extended period of time," which is defined under section 773(b)(2)(B) of the Act as not less than six months. Because each respondent reported home market sales data over a six-month period, adequate information exists on the record of this case to allow us to perform the substantial quantities test in accordance with the Act. Therefore, we have continued to disregard below-cost sales only where such sales were made within an extended period of time in substantial quantities, pursuant to section 77(3)(b)(1) of the Act. Company-Specific Issues Colakoglu Comment 4: Home Market Sales Priced in U.S. Dollars During the POR, Colakoglu negotiated the prices for certain home market sales in U.S. dollars. Colakoglu invoiced the customer in Turkish lira for these sales. In order to determine the invoice price, Colakoglu converted the U.S.-dollar amount to Turkish lira on the date of the sale, then it added an additional sum to account for estimated inflation between the time of sale and payment. Finally, when the customer paid, Colakoglu again converted the U.S.-dollar amount shown on the invoice to Turkish lira and then either reimbursed the customer for the difference between the original estimate and the actual amount (in cases where Colakoglu overestimated the amount of inflation) or issued another invoice for the difference (in cases where Colakoglu underestimated inflation). Colakoglu termed this price adjustment kur farki. For purposes of the preliminary results, we based normal value on the U.S.-dollar price, rather than the Turkish lira price adjusted for kur farki. We reasoned that the only price agreed upon was a U.S.-dollar price, and this price remained unchanged; the buyer merely paid the Turkish lira-equivalent amount at the time of payment. See Certain Steel Concrete Reinforcing Bars From Turkey; Preliminary Results of Antidumping Duty Administrative Review, 66 FR 22525, 22528 (May 4, 2001). In addition, we also calculated credit expenses for the sales in question using the U.S.-dollar price and a U.S.-dollar-denominated interest rate. Colakoglu contends that the Department erred when it used these U.S.- dollar prices. First, Colakoglu contends that the Department's methodology is contrary to recent precedent. Specifically, Colakoglu asserts that, in two cases decided in the last two years, the Department held that the price in local currency must be used as the starting price for its dumping analysis, even where: 1) the price negotiated between the parties is in dollars; and 2) both the dollar price and the exchange rate used to convert it into local currency appear on the invoice. See Notice of Final Determination of Sales at Less Than Fair Value: Certain Cut-to-Length Carbon-Quality Steel Plate Products from Japan, 64 FR 73215, 73226 (Dec. 29, 1999) (Plate from Japan) and Notice of Final Determination of Sales at Less than Fair Value: Hot-Rolled Flat-Rolled Carbon Quality Products from Japan, 64 FR 24329, 24345 (May 6, 1999) (Hot-Rolled Steel from Japan). Colakoglu further asserts that, in those cases, as here, the invoice was tracked in the seller's books in local currency and the customer ultimately paid in local currency. Therefore, Colakoglu maintains that, consistent with its practice, the Department should use the Turkish-lira price as the starting price, plus the kur farki amount subsequently invoiced to the customer. Colakoglu further argues that the Department's methodology does not yield accurate cost-price comparisons. Specifically, Colakoglu notes that the Department compared costs to prices converted into Turkish lira on the dates of the home market sales. However, Colakoglu asserts that these comparisons are inappropriate because the Turkish lira values did not include the total revenue earned on the sales (i.e., the price plus the kur farki). According to Colakoglu, it is the amount of the anticipated devaluation which was included in the prices plus the kur farki which compensates Colakoglu for the delayed payment terms extended to the customer. Colakoglu concludes that this methodology resulted in the Department's comparing a "sight" price with a cost of production which includes financing activities (including financing accounts receivables), in direct contradiction to its practice. In support of this assertion, Colakoglu cites Final Determinations of Sales at Less than Fair Value: Certain Carbon Steel Products from Brazil, 49 FR 28298, 28301 (July 11, 1984) (Carbon Steel from Brazil) (where the Department included late payments in the price before performing the cost test) and Silicon Metal from Brazil; Amended Final Results of Antidumping Duty Administrative Review, 62 FR 47441, 47444 (Sept. 9, 1997) (Silicon Metal from Brazil) (where the Department did not deduct credit expenses from home market price before performing the cost test). Finally, Colakoglu contends that the Department's methodology leads to arbitrary distinctions between sales reported in Colakoglu's home market database. Specifically, Colakoglu asserts that it used indices other than the U.S. dollar to set its home market prices. For example, Colakoglu states that, in some cases, it merely predicted inflation between the time of sale and payment and included that amount in price without using a benchmark. According to Colakoglu, under the Department's proposed methodology, in a circumstance in which Colakoglu invoiced two separate customers on the same day for the same product at the same invoice price for the same invoice term, the price that is compared to the cost of production will be significantly higher for the sale negotiated in Turkish lira than it will be for the sale originally negotiated in U.S. dollars. Colakoglu asserts that this is due to the fact that the Turkish-lira price includes a built-in amount for inflation, whereas the U.S.-dollar price does not. Colakoglu contends that, because each sale yields the same net revenue and is treated the same in Colakoglu's accounting system, there is no reason in law or policy why the two sales should generate a different price-to-cost comparison or a different normal value. In addition, Colakoglu argues that the Department similarly erred when it calculated credit expenses on these transactions using an interest rate denominated in U.S. dollars. According to Colakoglu, the appropriate credit costs in this case are those which are calculated in the currency in which the sale is invoiced to, and paid by, the customer because the firm's opportunity cost over the payment term is the opportunity cost of not having those funds at its disposal on a sight basis. Colakoglu contends that computing credit expenses using a U.S.-dollar interest rate insufficiently adjusts the price for the opportunity cost to Colakoglu of deferred payment on those sales. Colakoglu notes that it might have done so if the Turkish lira interest rate was precisely equal to the devaluation of the Turkish lira against the dollar, plus the U.S.-dollar interest rate. In such case, Colakoglu asserts, application of the dollar rate to the dollar price should yield approximately the same result as application of a Turkish lira interest rate to the total Turkish lira price. However, Colakoglu contends that the interest rate in Turkish lira was often substantially higher than the rate of devaluation of the Turkish lira against the U.S. dollar, plus the U.S.-dollar interest rate. Accordingly, Colakoglu contends that its opportunity cost for deferred payments denominated in Turkish lira was much higher than that calculated by the Department. Finally, Colakoglu contends that it is appropriate to calculate the credit costs over the full invoice price plus kur farki because it is that price that reflects the credit terms that were actually extended to the customer. The petitioner asserts that the Department should continue to use the U.S.-dollar prices in question. According to the petitioner, Colakoglu's approach recently was rejected by a dispute panel of the World Trade Organization (WTO). Specifically, the petitioner notes that, given similar facts, the WTO found in the Report of the Panel on United States Anti- Dumping Measures on Stainless Steel Plate in Coils and Stainless Steel Sheet and Strip from Korea WT/DS179/R (Dec. 22, 2000), that the home currency price was in no way controlling. In contrast, the petitioner asserts that the facts in the cases cited by Colakoglu are not similar to those here, because the issue before the Department in both Plate from Japan and Hot-Rolled Steel from Japan was what constituted the appropriate currency of certain U.S. sales. The petitioner also disagrees that there is any distortion in the cost test as a result of using the U.S.-dollar prices. According to the petitioner, the sales in question are made in U.S. dollars and are simply accompanied by an estimate of the Turkish lira amount necessary to cover the U.S.-dollar price when payment is made. The petitioner notes that the prices stated in Turkish lira can vary significantly between the time of sale and payment. For this reason, the petitioner asserts that Colakoglu's argument ignores the effects of high inflation. Moreover, the petitioner notes that the Department defines costs by reference to the month of sale; consequently, these costs must be compared to the lira price on the date of sale. The petitioner contends that comparing costs in the month of sale to the price on the date of payment represents an apples to oranges comparison. Finally, the petitioner contends that, contrary to Colakoglu's assertions, prices set in Turkish lira and those set in U.S. dollars do not have the same economic effect on Colakoglu and the invoice prices are not the same. According to the petitioner, when Colakoglu sets its prices in U.S. dollars, it takes no risk that the Turkish lira amount that it ultimately receives will be lower than anticipated. In contrast, the petitioner asserts that prices set in Turkish lira are exposed to the risk that the relative value of the Turkish lira will decline more than expected. In addition, the petitioner asserts that the fact that Colakoglu makes a kur farki adjustment on sales negotiated in U.S. dollars demonstrates that this sale differs from a similar sale in which the price was set in Turkish lira. According to the petitioner, the Turkish lira price of the sale invoiced in U.S. dollars is not known on the date of sale or the invoice date, and thus the invoice price does not reflect the true price. Regarding the question of credit expenses, the petitioner contends that the Department should continue to base these expenses on the U.S.-dollar interest rate used in the preliminary results. The petitioner contends that, for those sales negotiated in U.S. dollars, the U.S.-dollar amount controls the price of the home market sale. Therefore, the petitioner argues that Colakoglu does not experience the opportunity cost that it would if that sale were in Turkish lira. Department's Position: We disagree with Colakoglu that we should disregard its U.S.-dollar prices in preference for Turkish-lira prices adjusted for kur farki. As noted in the preliminary results, Colakoglu and its customers negotiated the prices for the sales in question in U.S. dollars, and these prices did not change once agreement was reached; rather, the buyer merely paid the Turkish-lira equivalent amount at the time of payment. In contrast, the Turkish lira prices shown on the invoices were not the final prices between the parties, but instead represented estimates of what Colakoglu believed would be the final lira amount at the time of payment. In essence, Colakoglu is claiming that the total amount of Turkish-lira proceeds from the sale is the price it would have set if it had set the price in Turkish lira. While we agree that a Turkish-lira price actually set on the date of sale would be the appropriate price to use, for the sales at issue, Colakoglu in fact did not set a price in Turkish lira. It set the price in U.S. dollars. (4) We disagree with Colakoglu's assertion that the Department's calculations did not yield accurate price-to-cost comparisons. We find that by converting the U.S.-dollar price on the date of sale and comparing it to the cost for the product in the same month, we have made price-to-cost comparisons based on Turkish lira stated in equivalent currency units (i.e., currency units having the same value). Contrary to Colakoglu's assertion that we have not accounted for the total revenue earned on the sale, the U.S.-dollar price reflects the total revenue earned on a particular sale at any point in time; the fact that Colakoglu received more Turkish lira in nominal terms on the date of payment is irrelevant to the analysis. Furthermore, we find Colakoglu's reliance on Carbon Steel from Brazil and Silicon Metal from Brazil is misplaced. Before performing the cost test, in the former case the Department included interest revenue in the home market price, while in the latter it did not deduct imputed credit expenses. In this case, we have included all revenue items in the gross unit prices before comparing them to the COP (as in Carbon Steel from Brazil); because the U.S. dollar price is the equivalent of the Turkish lira value adjusted for kur farki, it is not necessary to add kur farki separately. (5) In addition, as in Silicon Metal from Brazil, we have made no adjustment to the gross unit prices for credit expenses before comparing them to COP. (6) We disagree with Colakoglu's claim that reliance on the U.S.-dollar prices in this case is contrary to the Department's practice. The circumstances in the cases cited by Colakoglu are only superficially similar to those here. While we agree that we declined to use the U.S. dollar prices in both Plate from Japan and Hot-Rolled Steel from Japan because the customers paid in local currency, we note that neither of those cases involved high inflation. Moreover, although the prices in those cases were initially set in U.S. dollars, the local-currency price was fixed at invoicing and the respondents received payment at a later point in local currency at that invoiced amount. Because the exchange rate differed on the date of payment from the rate in effect at the time that the price in local currency was established, these companies did not receive the local-currency equivalent of the U.S.-dollar price stated on the invoice. In contrast, although Colakoglu was paid in Turkish lira, the amount that it received corresponded exactly to the U.S.-dollar amount reflected on the invoices to its customers. Regarding Colakoglu's claim that the Department's margin calculations arbitrarily distinguish between sales which were negotiated in U.S. dollars and Turkish lira, we note that Colakoglu itself makes a distinction between these sales when it chooses to price certain sales in dollars and others in lira. Moreover, while Colakoglu may price its Turkish lira sales differently, this fact is not relevant to the issue of when the price is established for the transactions at issue. When a sale is priced in U.S. dollars, Colakoglu has assured itself that it will receive an amount of Turkish lira based upon the exchange rate in effect at the time of payment. Therefore, because: 1) the price for these transactions is fixed in U.S. dollars at the time of invoicing; and 2) this price does not change prior to payment, we have used the U.S. dollar price in our analysis. (7) Finally, we disagree with Colakoglu that it is appropriate to impute credit expenses for the sales in question using a Turkish lira interest rate. As we stated in Import Administration Policy Bulletin 98-2 (Feb. 23, 1998), for the purposes of calculating imputed credit expenses, the Department uses a short-term interest rate tied to the currency of the transaction; therefore, since we have determined that the effective currency of the home market sales in question is U.S. dollars, it is appropriate to calculate credit expenses for these transactions using a U.S.-dollar short-term interest rate. Because the credit terms extended to these home market customers were, in effect, based on a U.S. dollar amount, it is appropriate to use a U.S.-dollar borrowing rate to calculate imputed credit expenses for them, regardless of the rate of devaluation of the Turkish lira against the U.S. dollar. Consequently, we have not changed our methodology for calculating home market credit expenses for sales negotiated in U.S. dollars for purposes of the final results. Comment 5: Kur Farki Adjustment According to the petitioner, the Department failed to account for the kur farki adjustment reported in Colakoglu's home market database. The petitioner contends that the Department should add this amount to the gross unit price because it represents added funds that Colakoglu received from its home market customers. Colakoglu agrees that the Department did not make an adjustment to gross unit price for kur farki. However, Colakoglu maintains that such an upward adjustment is only appropriate when the gross unit price is denominated in Turkish lira. Colakoglu notes that the Department used gross unit prices denominated in U.S. dollars for the sales in question for purposes of the preliminary results. See Comment 4. According to Colakoglu, if the Department continues to use the U.S.-dollar value of these home market sales, it should continue to make no adjustment for kur farki. However, if the Department uses the Turkish lira prices for these sales, as argued above, Colakoglu agrees that the Department should add kur farki to the gross unit price. Department's Position: We agree with Colakoglu. As noted in our response to Comment 4, above, we are continuing to base our analysis on the U.S.-dollar prices for the sales in question. Because kur farki is an adjustment to the Turkish lira value which brings it into line with the U.S.-dollar price, using both the kur farki amount and the U.S.-dollar price would result in our double- counting the kur farki. Therefore, we have continued to make no adjustment for kur farki for purposes of the final results. Comment 6: Colakoglu's Home Market Credit Expenses The petitioner argues that Colakoglu failed to respond fully to the Department's questionnaire regarding short-term borrowings in Turkish lira. Specifically, the petitioner states that Colakoglu reported that it had no unsubsidized short-term borrowings, but failed to disclose its actual borrowing rates as noted on the company's financial statements. Additionally, the petitioner alleges that Colakoglu's reported credit period is unreliable because the Department found discrepancies in the credit period at verification. According to the petitioner, the Department should disallow an adjustment for home market credit expenses. Alternatively, the petitioner argues that the Department should adjust all of the reported credit expenses to correct for the average difference in the credit period found at verification and to use either the lowest, or the average, of the short-term interest rates listed in the notes to Colakoglu's financial statements. Colakoglu asserts that both of the petitioner's claims are baseless. First, Colakoglu notes that it disclosed all Turkish-lira short-term borrowings to the Department both in its responses and during verification. Specifically, Colakoglu notes that it disclosed in a supplemental questionnaire response that it had no Turkish-lira short-term borrowings other than certain pre-shipment loans made by the Export Credit Bank of Turkey (Eximbank), which are subsidized. According to Colakoglu, because credits issued by the Eximbank are provided at subsidized, preferential interest rates it is inappropriate to use such loans to determine the opportunity cost associated with the financing of domestic receivables. Accordingly, Colakoglu argues that the Department should continue to rely on the interest rate used in the preliminary results (i.e., the company's short-term deposit rate) for purposes of calculating home market credit expenses. Regarding the credit period, Colakoglu notes that the Department found at verification that the payment dates were reported incorrectly for only a small number of sales. Colakoglu notes that the correct payments dates for these sales was in some cases earlier and in others later than the reported date. Nonetheless, Colakoglu asserts that they were minor clerical errors which were corrected by the Department in the preliminary results and which had no effect on the dumping margin. Therefore, Colakoglu argues that no further change to its credit expenses is necessary. Department's Position: We agree with Colakoglu. The company disclosed in its supplemental questionnaire response that it borrowed from the Eximbank during the POR. See Colakoglu's October 17, 2000, submission at pages 12 through 16. In that response, Colakoglu stated that these loans were subsidized and expressly tied to exports, and it provided certain shipment documents which referenced specific Eximbank loans. In the less-than-fair-value (LTFV) segment of this proceeding, we addressed the same issue for Colakoglu. In that segment, we stated: In general, the Department's practice with regard to the interest rate used to calculate home market imputed credit expenses is to base the rate on a company's actual borrowings in the home market currency. The Department makes exceptions to this practice either when there are no loans in the home market currency or when a company is able to prove that its loans in that currency do not form an appropriate basis for the home market interest rate (e.g., when they are tied to specific export transactions). In Porcelain-on-Steel Cooking Ware, it was demonstrated to the Department's satisfaction that the loans at issue were tied directly to exports of subject merchandise. In this case, however, not only is there no evidence on the record showing that these loans are tied to U.S. sales of rebar, but there is also no evidence that they are tied to exports at all. Moreover, these loans are based on Turkish-lira-denominated borrowings and bear interest rates into which inflation has been factored. Consequently, we find that the interest rates paid on these loans are more indicative of Colakoglu's actual borrowing experience than are the interest rates published by the IMF. Accordingly, we have used them in our calculation of home market credit for purposes of the final determination. See Notice of Final Determination of Sales at Less Than Fair Value: Certain Steel Concrete Reinforcing Bars From Turkey, 62 FR 9737, 9745 (Mar. 4, 1997). Notwithstanding our use of these rates in the LTFV investigation, we have reconsidered our position on this issue. We have examined the documents provided by Colakoglu and find that they demonstrate that the Eximbank loans in question were linked to exports. In addition, we note that the Department has found in past proceedings that these types of Eximbank loans were related to export activities and were provided at subsidized rates. See, e.g., Certain Welded Carbon Steel Pipes From Turkey; Preliminary Results of Countervailing Duty Administrative Review, 65 FR 18070, 18071-72 (unchanged in the final results). Consequently, we find that these loans do not form an appropriate basis for the home market interest rate, and we have not used them for the final results. Rather, we have continued to base the interest rate for Colakoglu on the company's short-term deposit rates because these rates provide a conservative estimate of Colakoglu's actual borrowing experience. Regarding the credit period, we disagree with the petitioner that Colakoglu's data is unreliable. At verification, we found that Colakoglu reported the correct dates of shipment for all transactions examined and the correct dates of payment for the majority of these transactions. Moreover, of the four sales for which the payment dates were inaccurate, we found that: 1) the actual dates did not differ significantly from the reported dates; and 2) there was no pattern to the errors (i.e., one credit period was understated, while three were overstated slightly). Because these errors were minor in nature, we find that it is not appropriate to base the credit period for all sales on facts available nor to reject Colakoglu's credit expenses entirely. Consequently, we have continued to use Colakoglu's reported data (as adjusted for the preliminary results) in our final analysis. Comment 7: Colakoglu's Home Market Indirect Selling Expenses The petitioner claims that the Department did not account for home market indirect selling expenses when it conducted the cost test for Colakoglu. Specifically, the petitioner notes that Colakoglu claimed that its reported G&A expenses included all indirect selling expenses incurred by the company. However, the petitioner points out that the G&A expense ratio used for purposes of the preliminary results was less than the sum of the G&A and selling expense ratios reflected in Colakoglu's consolidated financial statements. Therefore, the petitioner concludes that the reported G&A expense ratio must not include selling expenses. According to the petitioner, Colakoglu's selling expenses must be taken into account in the cost test. The petitioner argues that the Department should base the amount of these expenses on facts available. As facts available, the petitioner contends that the Department should use the selling expense ratio shown on Colakoglu's consolidated financial statements. According to Colakoglu, the Department correctly included indirect selling expenses in conducting the cost test. Colakoglu points out that, because its consolidated financial statements reflect the G&A expenses of multiple companies within the Colakoglu group and not just Colakoglu itself, the petitioner's comparison is meaningless. Nonetheless, Colakoglu asserts that the expenses shown on the consolidated financial statements were mostly freight-related expenses and minor direct selling expenses. Colakoglu maintains that, because it properly included all indirect selling expenses in its G&A expense ratio, the Department correctly performed the cost test. Department's Position: At the cost verification conducted in this case, we examined the expenses included in Colakoglu's calculation of G&A expenses. We confirmed that Colakoglu reported all expenses shown on its financial statements as operating expenses, as well as all G&A expenses included in company's cost of goods sold, as part of the G&A expense ratio. In addition, we reconciled the non-G&A expenses portion of the company's cost of goods sold, on a test basis, to the manufacturing costs reported in the Colakoglu's COP database. See Colakoglu cost verification exhibits 1, 8, 9, and 14. We noted that Colakoglu had included in these G&A expenses total personnel expenses for the company, as well as certain other types of indirect selling expenses (e.g., bad debt). In addition, at the sales verification we examined the expenses recorded in Colakoglu's sales and marketing account during the POR. We found no evidence that Colakoglu recorded any expenses other than freight in this account. Consequently, based on the testing performed at verification, we find that Colakoglu's response captured all of the company's costs (including selling expenses) during the POR. We disagree with the petitioner's conclusion that Colakoglu failed to report certain expenses based on an examination of the company's consolidated financial statements. As Colakoglu correctly points out, the financial statements in question reflect the financial results of multiple companies, and not just the results of Colakoglu alone. Thus, it is not surprising that the selling, general, and administrative (SG&A) expenses shown on these financial statements exceed the expenses shown on the financial statements of a single company. Because we verified that Colakoglu completely reported all non- manufacturing costs during the POR, we have continued to accept these costs without further adjustment for purposes of the final results. Comment 8: Different Costs for the Same Products Produced by Colakoglu According to the petitioner, the Department discovered at verification that Colakoglu reported different costs for the same product rolled by different subcontractors. Specifically, the petitioner alleges that the Department found that Colakoglu used one subcontractor to produce product sold in the domestic market and another for exported product. The petitioner contends that cost differences based on the destination of the finished product are inappropriate. Moreover, the petitioner asserts that the Department followed an unsustainable approach when it calculated weighted-average costs for the same control number for purposes of the preliminary results. According to the petitioner, this approach is unjustified because only one of the two reported costs is correct for each control number. Thus, the petitioner argues that the Department should use the higher of the two costs as facts available for purposes of the final results. Colakoglu contends that the Department's methodology of calculating the costs for the products at issue was not only accurate, but it was also consistent with its long-standing practice of calculating weighted-average costs by control number, regardless of the market to which the product was sold. In support of this contention, Colakoglu cites the Department's standard questionnaire at page D-2, which states the following: "There should be a single weighted-average cost for each unique product as represented by a specific matching control number." Therefore, Colakoglu asserts that the Department should reject the petitioner's suggestion to use the higher of the two costs. Department's Position: During the POR, Colakoglu used subcontractors to roll rebar with a diameter of eight millimeters. Colakoglu also produced rebar of this same size in its own production facilities. In its questionnaire response, Colakoglu reported the costs associated with subcontracted rebar in one computer file and those associated with rebar rolled at its own factory in another. Therefore, while we agree that these costs differed, we disagree with the petitioner that this difference was "discovered" at verification. For purposes of the preliminary results, we calculated a weighted average cost for these products, by control number, in accordance with our practice. See page D-2 of the Department's standard questionnaire. We disagree with the petitioner's assertion that this approach is unjustified because only one of the two reported costs is correct for each control number. As noted in our cost verification report, we verified the accuracy of the costs reported for both the subcontracted and non-subcontracted products. As a consequence we have continued to use the weighted average of these costs in our analysis for purposes of the final results. Comment 9: Unreported Costs for Colakoglu The petitioner contends that the Department discovered at verification that Colakoglu failed to report production costs for the entire POR. According to the petitioner, Colakoglu's failure represents a significant deviation from the requirements in this review and, consequently, the Department should base Colakoglu's dumping margin on total facts available. Colakoglu asserts that the petitioner's suggestion is without merit. Colakoglu maintains that it disclosed to the Department at verification that it had inadvertently omitted cost information for a small number of products produced during certain months of the POR. According to Colakoglu, the best alternative source of information would have been the actual manufacturing costs that the company reported for those products, because there is no reason to believe that these costs were unrepresentative of the costs over the entire POR. Nonetheless, while Colakoglu disagrees with the necessity for punitive action, it contends that the Department's application of partial facts available in the preliminary results was at least measured to the degree of the minor error. Accordingly, Colakoglu contends that the use of total facts available is unwarranted. Department's Position: During the POR, Colakoglu produced rebar in coil form in the company's wire rod mill. Colakoglu reported the costs associated with producing this rebar in its cost response for the months of January through March 2000. At verification, we found that Colakoglu also produced rebar in coil form in other months of the POR, and, as a consequence, we based the dumping margin for the U.S. sales of these products on facts available. Specifically, we stated: Regarding Colakoglu, we were unable to make product comparisons for certain models which were produced and sold during 1999 because this respondent failed to report cost information for them, including both difference-in-merchandise and CV data. Consequently, for purposes of the preliminary results, we based the margin for the sales of these products on facts available pursuant to section 776(a)(2)(B) of the Act. As facts available, we used the highest non-aberrant margin calculated for any U.S. transaction for Colakoglu, in accordance with our practice. See, e.g., Static Random Access Memory Semiconductors From Taiwan; Preliminary Results and Partial Recission of Antidumping Duty Administrative Review, 65 FR 26577, 26579 (May 8, 2000) (unchanged by the final results); and Final Determination of Sales at Less Than Fair Value: Stainless Steel Sheet and Strip in Coils from Germany, 64 FR 30710, 30732 (June 8, 1999). In selecting a facts-available margin, we sought a margin that is sufficiently adverse so as to effectuate the statutory purposes of the adverse facts-available rule, which is to induce respondents to provide the Department with complete and accurate information in a timely manner. We also sought a margin that is indicative of the respondent's customary selling practices and is rationally related to the transactions to which the adverse facts available are being applied. To that end, we selected the highest margin on an individual sale which fell within the mainstream of Colakoglu's transactions (i.e., transactions that reflect sales of products that are representative of the broader range of models used to determine NV). See the Preliminary Results, 66 FR at 22527. We disagree with the petitioner that the application of further adverse facts available is warranted in this case. We note that the products at issue represented only a small portion of the company's total production and sales of rebar during the POR, and Colakoglu's failure to report complete costs for these products could have only a limited impact on the company's dumping margin. In addition, we were able to confirm the accuracy of the vast majority of Colakoglu's reported sales and cost information at verification. Under such circumstances, we find that the use of total adverse facts available is inappropriate. Consequently, as in the Preliminary Results, we have continued to apply facts available only to the U.S. sales of the products in question. Comment 10: Colakoglu's Production Quantities The petitioner alleges that at verification the Department was unable to reconcile the production quantities reported for three months of the POR to the company's records. Accordingly, the petitioner contends that the Department should take this failure into account in the final results by applying facts available in some form. However, the petitioner does not indicate what type of facts available is appropriate. Colakoglu argues that the Department reconciled the reported production quantities to production records for each month of the POR and noted only minor differences. Specifically, Colakoglu contends that the differences observed at verification amounted to significantly less than one percent of the company's total production volume. Department's Position: At verification, we found that the production quantities reported by Colakoglu reconciled without discrepancy to the company's production records for eight of the 12 months of the POR. In addition, we found that the production quantities for the remaining four months of the POR reconciled to the production records with only minor discrepancies. (8) Because we were able to confirm the accuracy of Colakoglu's reported production quantities at verification, we find no basis for rejecting the reported information. Consequently, we have continued to accept it for purposes of the final results. Comment 11: Colakoglu's Depreciation Expenses The petitioner argues that Colakoglu significantly understated its depreciation expenses because it: 1) failed to account for certain expenses associated with one of Colakoglu's cost centers; (9) and 2) did not completely report all expenses shown in the company's consolidated audited financial statements. Regarding this latter point, the petitioner asserts that the expenses shown on the consolidated financial statements, when converted into Turkish lira, exceed the reported expenses by a significant margin. For these reasons, the petitioner claims that the Department should base Colakoglu's depreciation expenses on facts available. However, the petitioner does not suggest an appropriate source for facts available. According to Colakoglu, it based its reported depreciation expenses on the amounts reflected in its unconsolidated financial statements, and thus Colakoglu contends that the Department should continue to rely on these expenses for purposes of the final results. Regarding the first point raised by the petitioner, Colakoglu notes that the expenses in question were incurred in a separate division of the company which did not produce subject merchandise. Colakoglu asserts that the petitioner provided no support for its claim that the depreciation expenses incurred in this division should be added to the cost of producing rebar because no support is available. Regarding the second point, Colakoglu notes that the petitioner made the argument elsewhere in its case brief that the Department should reject Colakoglu's consolidated financial statements because they were unreliable. (See Comment 13.) Therefore, Colakoglu implies that these statements cannot be reliable for one purpose but unacceptable for others. In any event, Colakoglu asserts that it is unsurprising that depreciation expenses stated in U.S. dollars differ from expenses in Turkish lira. Department's Position: We agree with the Colakoglu, in part. At verification, we reconciled Colakoglu's reported expenses to the amounts recorded in the company's cost accounting system. Based on our verification procedures, we concluded that Colakoglu had completely reported all depreciation expenses associated with the fixed assets in its melt shop, bar mill, and rod mill. See the Colakoglu cost verification report at pages 21 and 22. We disagree with the petitioner that it is possible to conclude that Colakoglu's depreciation expenses are understated based solely on a comparison of the reported amounts to the amounts reflected on the company's audited financial statements. We note that these financial statements include the consolidated results of Colakoglu and its subsidiaries. Moreover, these financial statements are stated in U.S. dollars, whereas the company's books and records are maintained in Turkish lira. Finally, the audited financial statements reflect certain depreciation expenses associated with cost centers within Colakoglu that are not related to the production of rebar. See Colakoglu cost verification exhibit 10. Thus, it is not surprising that the amounts shown on Colakoglu's audited financial statements differ from the amounts reported in the company's COP database. Regarding the depreciation expenses related to the separate cost center, however, we agree with the petitioner that a portion of these expenses should have been allocated to the production of rebar. At verification, Colakoglu indicated that the "majority" of the product generated by this cost center was not used in the production of rebar. See the Colakoglu cost verification report at page 21. Based on this statement, we conclude that some of this product was consumed during steel production. Because Colakoglu was unable to quantify how much of this product was sold and how much was actually consumed, as facts available we have assumed that 49 percent (i.e., a minority) of the depreciation expenses in question was consumed in steel production. We then allocated a portion of these expenses to rebar beginning with the month in which Colakoglu began production of the new product. (10) For further discussion, see the October 31, 2001, memorandum from Elizabeth Eastwood to the File, entitled "Calculations Performed for Colakoglu Metalurji A.S. for the Final Results in the 1999-2000 Antidumping Duty Administrative Review on Steel Concrete Reinforcing Bars from Turkey." Comment 12: Colakoglu's G&A Expenses The petitioner contends that Colakoglu did not include certain expenses in its calculation of the G&A expense ratio. According to the petitioner, the Department should make an adverse assumption when adding these expenses to G&A expenses because the amounts in question are apparently in unadjusted currency. As the adverse inference, the petitioner argues that the Department must assume that the full amount was incurred at the beginning of the year, thereby requiring forward indexing for each month in the POR. According to Colakoglu, the expenses in question were reported in the company's revised G&A expense worksheet, which was submitted in a supplemental questionnaire response and subsequently verified by the Department. Therefore, Colakoglu maintains that these expenses were correctly reported. Department's Position: Contrary to Colakoglu's assertions, we note that the expenses in question had not been included in the company's reported G&A expense ratio. Nonetheless, we obtained a revised worksheet at verification which incorporated this figure in the month in which the charges were recorded in the company's accounting system. Moreover, we used the revised G&A expense ratio in our calculations for purposes of the preliminary results. Therefore, no further revisions to Colakoglu's G&A expenses are necessary. For further discussion, see pages 24 and 25 of the Colakoglu cost verification report. Comment 13: Colakoglu's Financing Expenses For purposes of the preliminary results, the Department relied on data contained in Colakoglu's audited financial statements when calculating financing expenses. The petitioner argues that the Department should not have relied on these financial statements because they were prepared for the first time at the end of the 1999 fiscal year, and, as a consequence, were not historically maintained in the ordinary course of business. In addition, the petitioner implies that these financial statements are rendered unreliable because they were signed by the company's auditors after the end of the POR. According to the petitioner, because the Department's policy is to reject financial statements which are not maintained in the ordinary course of business, it should do so here. See Elemental Sulfur from Canada; Final Results of Antidumping Duty Administrative Reviews, 62 FR 37970, 37985 (July 15, 1997). In addition, the petitioner asserts that Colakoglu's consolidated financial statements do not contain any explanation of how the Turkish- lira amounts were converted to U.S. dollars. According to the petitioner, if this conversion was made at the end of the period, both the income and expense amounts could be misstated, rendering them even more unreliable. Finally, the petitioner asserts that these financial statements reflect certain translation losses which were not captured in the calculation of the financing expenses used in the preliminary results. Consequently, the petitioner argues that the Department should either: 1) add the translation losses to Colakoglu's costs; or 2) base financing expenses on the amounts shown on Colakoglu's unconsolidated financial statements. Colakoglu contends that the Department should continue to base financing expenses on the data contained in its consolidated financial statements. According to Colakoglu, these financial statements were not only prepared in the normal course of business, but they were audited by an affiliate of an international accounting firm. Colakoglu further states that these financial statements were not generated for purposes of the antidumping review, but rather to present to potential investors and banks, many of which require financial statements in U.S. dollars. Regarding the petitioner's implication that these statements were improper because they were not signed until after the POR, Colakoglu notes that it generally takes several months to prepare financial statements after the end of the fiscal year; thus, Colakoglu notes that it is not surprising that the auditors did not sign them until May 2000. Furthermore, Colakoglu asserts that it is patently incorrect to claim that the amounts on these statements could be misstated, given that they were not only audited by an international accounting firm, but also verified as accurate by the Department during the verification of Colakoglu's sales data. Colakoglu asserts that the financing expenses shown on the consolidated financial statements differed from those shown on the unconsolidated statements for several reasons. First, Colakoglu notes that the consolidated statements cover multiple entities, whereas the unconsolidated statements only cover Colakoglu. Moreover, the financing expenses shown on the consolidated statements were in U.S. dollars, whereas the expenses on the unconsolidated statements were in Turkish lira. Colakoglu notes that the Turkish lira expenses differ because they include foreign exchange losses on U.S. dollar-denominated loans. Finally, Colakoglu notes that it conducts much of its business (e.g., purchasing raw materials, selling finished goods, etc.) in U.S. dollars. Thus, Colakoglu asserts that it should not be surprising that a company with very little debt should have financial income in excess of financial expenses on a U.S. dollar basis. Finally, Colakoglu contends that the Department was correct in excluding translation losses from the financial expense calculation. According to Colakoglu, these losses are an adjustment resulting from the conversion of an entire financial statement from Turkish lira to U.S. dollars. Colakoglu notes that this adjustment would occur regardless of whether the company had any borrowings. As a consequence, Colakoglu argues that it is not part of financial expenses, and should not be treated as such for purposes of the final results. Department's Position: The Department's longstanding practice with regard to financing expenses is to base net financing expenses on the full-year net interest expense and cost of sales from the audited fiscal year financial statements at the highest level of consolidation which correspond most closely to the POR. See Notice of Final Determination of Sales at Less Than Fair Value: Certain Cut-to-Length Carbon-Quality Steel Plate Products from France, 64 FR 73143, 73152 (Dec. 29, 1999) (Carbon Steel Plate from France). This practice has been upheld by the CIT. See Gulf States Tube, 981 F. Supp. at 647-48. In its October 17, 2000, submission, Colakoglu provided consolidated financial statements for its 1999 fiscal year. These financial statements were prepared in U.S. dollars and included the results of both Colakoglu and its subsidiaries. These statements were not only audited by an international accounting firm in accordance with internationally- recognized accounting principles, but the company's auditors gave them an unqualified opinion. (11) Because these financial statements were audited in accordance with standard accounting principles, we have relied upon them for purposes of this proceeding. As a consequence, we have continued to base Colakoglu's financing expense ratio on data taken from these financial statements in accordance with our practice. We disagree with the petitioner's contention that the consolidated financial statements were not prepared in the ordinary course of business. Although these statements were prepared for the first time in 1999, they were not prepared for the purposes of this administrative review. Moreover, we note that it is common for companies to issue financial statements several months after the close of the fiscal year, because many auditors do not begin the auditing process until after the fiscal year data are complete. We also disagree with the petitioner's assertion that these financial statements are unreliable because they contain no explanation of how the Turkish lira amounts were converted into U.S. dollars. As noted above, these financial statements were audited by an internationally-recognized accounting firm and the auditors concluded that the financial results presented therein were an accurate reflection of Colakoglu's financial position. Had these statements misstated the company's income and expenses, the auditors could not have reached such a conclusion. Finally, regarding the translation losses shown on these financial statements, we agree with Colakoglu that the entire amount of these losses should not be included in the calculation of the company's financing expenses. We do consider it necessary, however, to include those translation gains and losses associated with foreign denominated debt. Because the consolidated financial statements are presented in U.S. dollars, none of the translation losses normally associated with foreign currency debt is included in the amounts reflected on the financial statements. Nonetheless, we note that the consolidated financial statements show that the Colakoglu group had loans in foreign currencies during 1999, and thus the group would presumably have incurred translation losses related to those loans. Therefore, as facts available, we estimated the amount of these losses using the exchange rates in effect on the first and last days of Colakoglu's fiscal year. Specifically, we calculated the change in the foreign currency loan principal balances using these rates and then included the entire amount in Colakoglu's financing expenses. We believe that this analysis is both reasonable and conservative. Even so, we note that Colakoglu's 1999 short-term financial income continues to exceed its total 1999 financing expenses. Thus, we have made no changes to the company's financing expense ratio for purposes of the final results. Diler Comment 14: Costs for Different Grades of Rebar Produced by Diler The petitioner notes that Diler's cost accounting system does not track costs by grade. Therefore, the petitioner argues that, in instances where Diler reported different costs for particular grades of rebar in the same month, the Department should use the higher of the reported costs. Diler agrees that it does not distinguish between different grades in the cost accounting system at either of its two rebar factories. (12) As a consequence, Diler notes that any differences in the reported monthly costs were attributable to: 1) the production of different sizes of rebar; and 2) the relative proportion of particular products produced at each factory. Regarding the latter point, Diler notes that the production efficiency differed during the POR at each factory. Therefore, Diler points out that, where it produced two different grades of the same size of rebar at both factories during the POR, the reported costs differed due to the fact that it may have produced more of one grade at one factory and less at the other. (13) According to Diler, accepting the petitioner's suggestion of using the higher of the reported costs for any grade of a particular size would produce a gross distortion in the margin calculation and would effectively escalate its costs beyond the true average replacement costs of the merchandise. Department's Position: The questionnaire issued in this case instructed the respondents to report sales and cost data for each grade, form, specification, and size of rebar. In accordance with these instructions, Diler defined the control numbers associated with each product sold during its reporting period using these product characteristics. Specifically, Diler reported separate control numbers for products in each unique grade/form/specification/size combination. Because Diler sold rebar in grades 40 and 60 during the period, it reported different control numbers associated with each grade. As noted in our response to Comment 8, above, it is the Department's practice to calculate costs for each product by control number. Moreover, when a company produces a product in more than one factory, we require the respondent to report the weighted average of the costs incurred in each location. At verification, we tested the accuracy of Diler's calculation of its weighted-average costs and noted no discrepancies. See the Diler cost verification report at page 21. Consequently, because Diler reported its costs on a control number- specific basis in accordance with the questionnaire instructions, we find no reason to use facts available for them. Therefore, we have continued to accept Diler's costs (adjusted as noted below) for purposes of the final results. Comment 15: Unreported Material Costs for Diler According to the petitioner, the Department discovered at verification that Diler failed to report certain movement charges associated with the importation of materials. Accordingly, the petitioner contends that the Department should include these movement charges in Diler's costs for purposes of the final results. Diler did not comment on this issue. Department's Position: We agree. We have included the unreported costs in our analysis for purposes of the final results. Comment 16: Diler's Depreciation Expenses The petitioner contends that the Department discovered at verification that Diler's reported depreciation expenses: 1) do not appear to capture the total depreciation expense for the month; and 2) are based on the historical cost of its assets. The petitioner maintains that International Accounting Standards (IAS) require that all fixed assets be revalued to account for the effect of inflation before depreciation is calculated. Accordingly, the petitioner argues that the Department should disallow the reported depreciation expenses in favor of depreciation that would be in accordance with IAS. The petitioner notes that this revised amount is set forth in the Department's cost verification report for Diler. Diler disagrees. According to Diler, its reported depreciation expenses accurately capture the amounts recorded in its monthly books and records, which in turn are reflected in its audited financial statements. Diler asserts that the Department confirmed at verification that Diler estimates the annual revaluation index on a monthly basis and then records depreciation expense based on the revalued asset value. Diler concedes that the Department also found that there were "jumps" in the depreciation amounts recorded at the end of each quarter which were related to corrections of the estimated revaluation rate to the actual rate based on quarterly indices published by the government of Turkey. Nonetheless, Diler contends that the existence of these "jumps" does not call into question the accuracy of its response. Specifically, Diler asserts that any lag in the depreciation expense recorded in the beginning of the quarter was picked up and included in the "jump" month. According to Diler, when the Department indexed and weight-averaged these expenses on a control-number-specific basis over the home market comparison period, in accordance with its normal practice in hyper-inflationary economies, it picked up the full depreciation expenses reported for the six-month time period and redistributed the expenses to each month on a weighted-average basis. Therefore, Diler asserts that no adjustment to the reported costs is required. Moreover, Diler disagrees that the Department should restate the company's depreciation expenses in accordance with IAS. Diler notes that it based the reported expenses on the amounts recorded in its normal books and records, which are kept in conformance with generally accepted accounting principles (GAAP) in Turkey. According to Diler, Turkish GAAP permits the revaluation of machinery and equipment for inflation but requires that depreciation on fixed assets be calculated on a historical- cost basis. Diler states that Turkish GAAP also permits accelerated depreciation, whereas IAS requires the use of a straight-line method. According to Diler, when the company's auditors compared the reported depreciation expenses (based on Turkish GAAP) with its expenses based on IAS, they found that the depreciation in Diler's books was larger for one of the company's two rebar factories and smaller for the other. Diler further notes that, on net, the two factories' depreciation expenses per their statutory books exceeded that required by IAS and were therefore conservative. Nonetheless, Diler asserts that, if the Department decides to adjust the depreciation expenses to IAS for one of the companies, it should make a similar adjustment for the other company for the sake of consistency. Department's Position: We agree in part with both the petitioner and the respondent. The petitioner asks the Department to reject Diler's reported depreciation expense, which is calculated based on its normal books and records in accordance with Turkish law, in favor of depreciation expense calculated based on IAS. However, the Department's long-standing practice, codified at section 773(f)(1)(A) of the Act, is to rely on data from a respondent's normal books and records where those records are prepared in accordance with home country accounting principles and reasonably reflect the costs of producing the merchandise. Normal GAAP accounting practices provide both respondents and the Department a reasonably objective and predictable basis by which to compute costs for the merchandise under investigation. In those instances where it is determined that a company's normal accounting practices result in a misallocation of production costs, the Department will adjust the respondent's costs or use alternative calculation methodologies that more accurately capture the actual costs incurred to produce the merchandise. In this case, evidence on the record indicated that Diler's calculation of depreciation expense in its normal books and records was consistent with both Turkish law and the Turkish Capital Market Board's regulations. As noted above, Diler revalues machinery and equipment for inflation, but requires that depreciation on buildings be calculated on a historical cost basis. As a result, because Turkey has high inflation, Diler's calculation of depreciation expense on buildings in its normal books and records does not properly reflect the cost in current currency levels. Instead, the reported depreciation expense on buildings is stated in currency levels from prior years. Therefore, we have adjusted Diler's cost of manufacturing to reflect the increase in depreciation expense related to the revaluation of buildings to current currency levels. We agree with petitioner that the depreciation expenses recorded monthly fail to capture the precise amount of inflation adjusted depreciation expense for each month. As mentioned above, Diler estimates the annual revaluation index on a monthly basis and then records depreciation expense based on the revalued asset value. At the end of each quarter, the difference between the estimated depreciation expense amounts and the actual amounts, based on quarterly revaluation rate indices published by the government of Turkey, are captured and recorded in Diler's normal books. We concur with Diler that the total depreciation expense has been captured; however, based on the Department's high inflation methodology, the reported monthly COP must capture the total cost related to that month. In this case, the depreciation expense in the last month of the quarter is overstated, and conversely the expense for the first two months of the quarter are understated. The impact of recording a portion of a particular month's inflation adjusted depreciation expense in a subsequent month is to under-inflate that month's underestimated depreciation expense when computing a single weighted-average cost of production in year-end currency levels. Therefore, we adjusted Diler's reported depreciation expense to account for the impact of the revaluation adjustment. Comment 17: Diler's G&A Expenses According to the petitioner, the Department verified that Diler included packing costs in the cost of goods sold used to calculate the G&A expense ratio. However, the petitioner claims that Diler only applied the G&A expense ratio to the COM. Therefore, the petitioner contends that the Department should apply the G&A expense ratio to Diler's manufacturing cost plus packing expenses. Diler agrees that the denominator used to calculate the G&A expense ratio included some packing costs. However, Diler points out that the two companies producing rebar, Diler Demir and Yazici Demir, record some (or most) of their packing costs in accounts other than cost of goods sold. Diler argues that multiplying the G&A expense ratio by the reported cost of manufacturing plus packing would seriously overstate the company's G&A expenses. Finally, Diler noted that the amount of packing costs included in the denominator of the G&A expense ratio is so small as to be insignificant. Therefore, Diler argues that any adjustment to G&A expenses related to packing expenses can be disregarded in accordance with section 777A(a)(2) of the Act and 19 CFR 351.413. Department's Position: At verification, we found that Diler included certain packing costs in the denominator of its G&A expense ratio. Because we calculated Diler's per-unit G&A expenses using the reported cost of goods sold (which did not include packing expenses), we have removed the amount of packing costs found at verification from the denominator of this ratio. We then recalculated G&A expenses for purposes of the final results. Comment 18: Diler's Financing Expenses According to the petitioner, the Department discovered at verification that, in calculating its reported financing expenses, Diler had not: 1) included net foreign exchange losses on materials purchases; or 2) considered any adjustments made to these expenses by the company's auditors. The petitioner contends that the Department should increase financing expenses to account for these costs. Diler argues that the Department should not adjust financing expenses for either of the items in question. Specifically, Diler notes that the foreign exchange losses at issue were generated when Diler purchased imported raw materials in a given month but did not pay for them immediately. (14) According to Diler, the foreign exchange loss generated by this delay in payment is part of the historical cost of the materials. Diler points out that, in cases involving high inflation, the Department does not rely on historical costs, but rather requires respondents to provide the associated replacement costs. Diler notes that the purpose of using a replacement cost methodology is to restate the historical cost at the current value of the currency in the month of production. For this reason, Diler contends that adding foreign exchange losses to the replacement cost would either double-count these expenses or produce a cost which is equivalent to the cost at a later time than the month of production. (15) In either case, Diler contends that the Department would end up with something other than the replacement cost in the month, and therefore it would be inappropriate to include the foreign exchange losses as part of Diler's costs. Regarding the auditor's adjustments noted by the petitioner, Diler contends that these adjustments are so small as to be insignificant. Therefore, Diler contends that the Department should disregard them in accordance with section 777A(a)(2) of the Act and 19 CFR 351.413. Department's Position: We agree with Diler in part. Regarding kur farki, Diler is assuming that changes in the exchange rate between the Turkish lira and the U.S. dollar are the result of inflation alone. Its argument does not recognize the fact that currencies have a value which fluctuates according to other factors such as the rules of supply and demand. Both are accounted for in the recognition of foreign exchange gains and losses. The Department notes, however, that information on the record shows that the change in the exchange rate between the U.S. dollar and Turkish lira during the year closely mirror the inflation rate experienced in Turkey during the same period. See the October 31, 2001, memorandum to the file from Elizabeth Eastwood entitled "Chart Comparing the Turkish Wholesale Price Index to Average Monthly Exchange Rates in the 1999-2000 Antidumping Duty Administrative Review on Certain Steel Concrete Reinforcing Bars from Turkey." Thus, we consider it reasonable not to adjust the reported costs for the kur farki as the fluctuation in the value of the U.S. dollar and Turkish lira outside of the effect of inflation is minimal. See Notice of Final Determination of Sales at Less Than Fair Value: Certain Cold-Rolled Flat-Rolled Carbon-Quality Steel Products from Turkey, 65 FR 15123 (Mar. 21, 2000) and accompanying decision memorandum at Comment 3.1 (Cold-Rolled Steel from Turkey). Moreover, because Ekinciler and ICDAS reported foreign exchange gains and losses during the POR, we have similarly removed these gains and losses from the calculation of their costs as well. See also Comments 27 and 32. Finally, we agree with the petitioner that it is appropriate to account for the auditor's adjustments reflected in the company's financial statements. Even so, we note that Diler's 1999 short-term financial income continues to exceed its 1999 financing expenses. Thus, we have made no changes to the company's financing expense ratio for purposes of the final results. Ekinciler Comment 19: Selling Expenses for Constructed Export Price According to the petitioner, certain pages in one of Ekinciler's sales verification exhibits were faxed to Turkey from the company's U.S. affiliate, Ferromin, in preparation for verification. The petitioner argues that this fact proves that Ferromin was involved in Ekinciler's U.S. sales, and, as a consequence, the Department should make a constructed export price (CEP) selling expense adjustment to account for Ferromin's selling expenses. Specifically, the petitioner contends that the Department should base the amount of this adjustment on the SG&A expenses shown on Ferromin's 1999 income statement. Ekinciler did not comment on this issue. Department's Position: According to Ekinciler's response, Ferromin had no involvement in Ekinciler's U.S. sales of rebar during the POR. See the October 20, 2000, supplemental response at page 2. At verification, Ekinciler provided certain sales documents which appeared to have been faxed from Ferromin's offices in the United States. However, these documents do not indicate that Ferromin had any actual role in making the sales in question or in negotiating their terms. Indeed, the only conclusion that can be drawn from these documents is that Ferromin transmitted certain contracts to Ekinciler by fax. Moreover, we found no evidence at verification to contradict Ekinciler's claim that all of its U.S. sales were made by the company's staff in Turkey. Specifically, we found that Ekinciler personnel signed all contracts, arranged for all shipments, and collected all payments from U.S. customers. Under these circumstances, we find that Ekinciler's sales were properly classified as export price transactions. Consequently, we have made no adjustment for Ferromin's indirect selling expenses, in accordance with our practice. Comment 20: Ekinciler's Home Market Freight Expenses According to the petitioner, the Department discovered at verification that Ekinciler failed to report freight revenue on certain transactions in the home market. The petitioner contends that the Department should increase the gross unit prices for these sales to account for this revenue. Ekinciler did not comment on this issue. Department's Position: For purposes of the preliminary results, we treated freight revenue as an offset to freight expenses. Because we have already accounted for it in our calculations, there is no need to make any further adjustment for purposes of the final results. Comment 21: Ekinciler's U.S. Freight Expenses According to the petitioner, the Department found at verification that Ekinciler reported its home market freight expenses inclusive of value added taxes (VAT) in Turkey, but reported U.S. freight expenses exclusive of VAT. The petitioner contends that the Department should increase Ekinciler's U.S. freight expenses to include VAT. Ekinciler maintains that it reported freight expenses for both markets on a VAT-exclusive basis. Therefore, Ekinciler contends that no adjustment to freight expenses is required. Department's Position: We reviewed the documents examined at verification and found that Ekinciler reported its freight expenses in the home market exclusive of VAT. Because it is the Department's policy to use charges and adjustments on a VAT-exclusive basis, we find that no adjustment is necessary for purposes of the final results. Comment 22: Ekinciler's Home Market Credit Expenses The petitioner argues that the Department should base Ekinciler's home market credit expenses on facts available because Ekinciler did not report them it its home market sales listing. According to the petitioner, the Department found at verification that Ekinciler's average credit period actually was negative, and thus its credit expenses constituted a benefit that should be added back to home market price. The petitioner accordingly requests that the Department base the amount of the credit benefit on the longest negative credit period reported by any other respondent and apply this negative credit period to all of Ekinciler's home market sales. Ekinciler contends that it correctly reported its credit period in the home market. Specifically, Ekinciler notes that the credit period reported in its questionnaire response was less than one day, and for this reason it reported that shipment and payment occurred on same date. Ekinciler notes that, although the Department discovered an error in the calculation of this credit period at verification, the corrected figure remained less than one day. Therefore, Ekinciler contends that no change to its reported credit expenses is necessary. Department's Position: In accordance with section 777A(a)(2) of the Act, the Department may decline to take into account adjustments which are insignificant in relation to the price or value of the merchandise. Under 19 CFR 351.413, the Department defines insignificant adjustments as those having an ad valorem effect of less than 0.33 percent of the export price or normal value. In this case, we find that a credit period of less than one day to be so small as to render the resulting credit expenses insignificant within the meaning of 19 CFR 351.413. Consequently, we do not find it appropriate to base the amount of these expenses on facts available, nor do we deem it necessary to adjust Ekinciler's normal value to account for the verified credit period. Comment 23: Ekinciler's Scrap Costs During the POR, Ekinciler purchased scrap used in the production process from an affiliated party. According to the petitioner, the Department confirmed at verification that the prices paid to the affiliate were higher than those paid to unaffiliated parties. The petitioner concludes that these prices were not at arms-length and, thus, asks the Department to reduce these prices to those paid to unaffiliated parties. Ekinciler asserts that the prices paid to its affiliate were at arms- length because they were higher than the prices paid to unaffiliated parties. Ekinciler notes that scrap is the primary raw material input into the production of rebar. According to Ekinciler, the petitioner appears to believe that the scrap costs in question were used as an offset to Ekinciler's production costs, rather than as part of its cost of materials. Nonetheless, Ekinciler notes that it would not object if the Department were to restate its purchases from its affiliate at the contemporaneous cost of scrap obtained from unaffiliated parties. Department's Position: We disagree with the petitioner. The costs at issue were associated with Ekinciler's purchase of raw materials, rather than the sale of scrap produced during the production cycle. Because scrap is the major input into the production of rebar, at verification we confirmed that Ekinciler purchased this scrap at prices higher than either: 1) the price paid to unaffiliated parties; or 2) the cost to the affiliate for purchasing the scrap from unaffiliated parties. See page 16 of the December 7, 2000, memorandum from Shawn Thompson and Jim Nunno to Louis Apple entitled "Verification of the Cost of Production and Constructed Value Data Submitted by Ekinciler Demir Celik A.S. and Ekinciler Holding A.S. in the Antidumping Duty Administrative Review on Certain Steel Concrete Reinforcing Bars from Turkey" (the Ekinciler cost verification report). Consequently, we find that these prices were at arm's length, in accordance with section 773(f)(3) of the Act, and we have continued to use them for purposes of the final results. Comment 24: Ekinciler's Depreciation Expenses Ekinciler based the depreciation expenses reported in its response on the expenses recorded in its accounting system for each month of its reporting period. In addition, Ekinciler also reported an additional amount of depreciation associated with the revaluation of its fixed assets, as required by the Ministry of Finance in Turkey at the end of each year. At verification, we confirmed that Ekinciler did not recognize this additional amount of depreciation in its accounting system for the current year, but rather recorded it in a separate fixed asset ledger which it updated each year. On April 13, 2001, the petitioner submitted a letter in which it alleged that Ekinciler's depreciation expenses were significantly understated because they did not correspond to the fixed asset values shown on the company's 1999 balance sheet. Based on this letter, we requested that Ekinciler reconcile the total asset value shown in its 1999 fixed asset ledger (after revaluation) to the total fixed asset value reflected on its balance sheet. We also requested that Ekinciler describe certain assets which did not appear to be depreciated in 1999 and provide a schedule of their useful lives. Ekinciler provided this information on May 11, 2001. In its case brief, the petitioner continues to assert that Ekinciler has significantly understated its reported depreciation expenses. First, the petitioner alleges that these expenses cannot be reconciled to the asset values shown on the company's balance sheet. The petitioner speculates that this is due to Ekinciler's statement that it has not recognized any additional depreciation expenses in its accounting records since 1996. The petitioner concludes from this statement that the beginning asset values used to compute the reported depreciation expenses were understated, and thus do not form an appropriate basis for the calculation. According to the petitioner, because the Department does not have Ekinciler's asset ledgers for each year since 1996 on the record, it cannot determine the correct depreciation figures. In addition, the petitioner claims that the Department similarly cannot use either Ekinciler's income statement or audited financial statements to calculate depreciation expenses because: 1) the former was prepared for purposes of the response; and 2) the latter are unreliable because they do not include inflation-adjusted depreciation expenses after 1996. According to the petitioner, there are several additional indications that Ekinciler's depreciation expenses are understated. First, when the reported depreciation expenses are compared to those of other respondents, Ekinciler's differ significantly. Second, when Ekinciler's reported depreciation expenses (assumed to be at a 1996 inflation level) are inflated to the fourth quarter of 1999 and the first quarter of 2000 (i.e., the time period for which Ekinciler reported sales and cost data), the reported expenses again are significantly lower. (16) Finally, the petitioner contends that Ekinciler admitted omitting depreciation on idle assets. The petitioner argues that, in light of these problems, the Department should apply total facts available in determining depreciation expenses for Ekinciler. As facts available, notwithstanding its arguments above regarding the unreliability of Ekinciler's financials statements, the petitioner argues that the Department should base the amount of these expenses on an amount derived by the petitioner from Ekinciler's balance sheet. Ekinciler maintains that it correctly calculated its depreciation expenses. Ekinciler notes that, contrary to the petitioner's claim, the Department verified that it included the revaluation amount for each year after 1996 in the year-end asset values that formed the basis for depreciation in the following year. According to Ekinciler, the only action that Ekinciler did not take with respect to depreciation was to not include the current-year revaluation in the depreciation taken in the same year. Ekinciler notes that including current-year asset revaluation in the asset value basis for a firm's depreciation is permissible, but not required, under Turkish law. Nonetheless, Ekinciler notes that it separately reported the amount of additional depreciation related to the revaluation of its 1999 and 2000 fixed assets in its response. Ekinciler notes that, because the 2000 revaluation rate was not published by the government of Turkey at the time that its response was prepared, it computed the amount of additional depreciation associated with the revaluation using the Turkish Wholesale Price Index (WPI). Ekinciler contends that this was a reasonable method of adjusting Ekinciler's depreciation expenses to those required by IAS. Regarding the petitioner's assertion that Ekinciler's financial statements were either prepared for purposes of this proceeding or otherwise unreliable, Ekinciler notes that it provided its official financial statements to the Department at verification. Ekinciler further notes that the Department not only tied the POR financial statements to these official statements at verification, but to the company's asset ledgers and trial balances as well. Regarding the petitioner's assertion that Ekinciler's depreciation expenses differ from that of certain other respondents, Ekinciler asserts that no conclusions can be drawn from such a comparison. Ekinciler notes that relative depreciation expenses have to do with the age of the assets in use. According to Ekinciler, because it depreciates machinery and equipment over a useful life of eight years, any equipment put into service prior to 1991 would have been fully depreciated prior to the commencement of the POR. While Ekinciler asserts that it is unable to point to specific facts on the record because the issue was not raised until the petitioner's case brief, it notes that the company was founded in 1983. Finally, regarding the petitioner's claim that Ekinciler admitted omitting depreciation on idled assets, Ekinciler asserts that this statement was taken out of context. Specifically, Ekinciler notes that it indicated that the company has some permanently idled equipment from a rolling mill that was closed during 1995. According to Ekinciler, it reported no depreciation expenses on these assets because they had already been fully depreciated or written off in prior years. Based on the foregoing, Ekinciler contends that it has accurately and completely reported depreciation expenses. Therefore, Ekinciler argues that no further adjustments are necessary for purposes of the final results. Department's Position: We agree with Ekinciler, in part. At verification, we found that Ekinciler completely reported all depreciation expenses recorded in its fixed asset ledgers covering the POR. Contrary to the petitioner's assertions, we found that these depreciation expenses were calculated using Ekinciler's accounting records, which reflected book values computed as the sum of the historical costs of the company's assets plus an accumulated amount for revaluation through the end of the previous year. See the Ekinciler cost verification report at page 19. Therefore, we disagree with the petitioner that the beginning asset values used to compute the reported depreciation expenses were understated in any respect. (17) We similarly disagree with the petitioner that Ekinciler's income statement is unreliable because it was prepared for purposes of the response. At verification, Ekinciler noted that it had created an income statement for the home market reporting period using the data shown on its fiscal year financial statements. We confirmed that this income statement accurately reflected Ekinciler's financial data by reconciling it to the data shown on company's fiscal year financial statements, as well as to the tax returns filed with the government of Turkey. See the Ekinciler cost verification report at pages 6, 10, and 11. Furthermore, we disagree with the petitioner's allegation that the depreciation expenses in Ekinciler's fixed asset ledger cannot be reconciled to the asset values shown on the company's balance sheet. In its May 11 submission, Ekinciler provided a schedule which reconciled the asset values in the fixed asset ledger to those shown on the balance sheet. According to this schedule, the difference between the asset values was related to land, construction in progress, and assets located in the company's sales office in Istanbul. We note that the former two are not subject to depreciation according to Turkish law, while the depreciation expenses associated with the latter were reported as part of G&A expenses. We find that the petitioner's comparison of Ekinciler's depreciation expenses with the expenses reported by other respondents in this proceeding is not appropriate. We note that this analysis is not valid because there are many factors which could lead to differences in depreciation expenses between companies. Foremost among these is the age of the assets in question and the time period over which they are depreciated. Because we tied Ekinciler's depreciation expenses to its accounting system, including its trial balances and financial statements, without discrepancy, we have continued to rely on the reported expenses (adjusted as noted below). Notwithstanding our verification findings, we have included certain additional depreciation expenses in the reported amounts for purposes of the final results. Specifically, we have increased the reported depreciation to account for the amortization of certain expenses related to capitalized financing expenses that Ekinciler would have recorded during the POR, had it not been prevented from doing so under Turkish tax law. (18) We find that this adjustment is appropriate because the asset values associated with them remain on Ekinciler's books. (19) To calculate the amount of the adjustment, we determined the portion of the capitalized expenses which would not have been fully depreciated during the home market reporting period (i.e., the expenses related to the last three months of 1999). We then inflated this amount to reporting period levels using the Turkish WPI published by the International Monetary Fund and allocated it over production during the same period. For further discussion, see the October 31, 2001, Memorandum from Elizabeth Eastwood to the File, entitled "Calculations Performed for the Ekinciler Group (Ekinciler) for the Final Results in the 1999-2000 Antidumping Duty Administrative Review on Steel Concrete Reinforcing Bars from Turkey." Comment 25: Ekinciler's G&A Expenses According to the petitioner, the Department found at verification that the cost of sales used to allocate G&A expenses incorrectly included packing expenses, thus understating the G&A expense ratio. The petitioner argues that the Department should recalculate G&A expenses, in order to exclude packing expenses. Ekinciler notes that the Department recalculated G&A expenses to exclude packing for purposes of the preliminary results. Therefore, Ekinciler asserts that no additional adjustment is required. Department's Position: We agree with both parties that the packing figures should be excluded from the denominator of the G&A expenses calculation. However, because we excluded these figures from the calculation for purposes of the preliminary results, no additional changes are necessary. Comment 26: Use of Consolidated Financing Expenses for Ekinciler Ekinciler asserts that the Department should base the calculation of its financing expenses on the consolidated financial experience of the entire Ekinciler Group, rather than on the experience of Ekinciler Demir Celik A.S. (Ekdemir) (i.e., the group member which produces rebar). According to Ekinciler, it is the Department's long-standing practice to base interest expenses for an affiliated group of companies on the group's consolidated financial expenses. As support for this assertion, Ekinciler cites Notice of Final Determination of Sales at Less Than Fair Value: Steel Wire Rope From India and the People's Republic of China; Notice of Final Determination of Sales at Not Less Than Fair Value: Steel Wire Rope From Malaysia, 66 FR 12759 (Feb. 28, 2001) and accompanying decision memorandum for Malaysia at Comment 1 (Steel Wire Rope from Malaysia) and Silicon Metal From Brazil; Final Results of Antidumping Duty Administrative Review and Determination Not To Revoke in Part, 66 FR 11256 (Feb. 23, 2001) and accompanying decision memorandum at Comment 6. Moreover, Ekinciler asserts that the Department ordinarily calculates a consolidated financial expense ratio even in those cases where an affiliated group of companies does not prepare a consolidated financial statement. See Final Results of Antidumping Duty Administrative Review; Silicon Metal From Brazil, 65 FR 7497, 7499 (Feb. 15, 2000) (Silicon Metal from Brazil). Finally, Ekinciler asserts that the Department's practice in this area has been repeatedly sustained by the CIT. See, e.g., Camargo Correa Metais, S.A. v. United States, 17 CIT 897, 902 (CIT 1993). In addition, Ekinciler asserts that the Department based Ekinciler's financing expenses in the most recently completed segment of this proceeding on the consolidated expenses of the Ekinciler group. Ekinciler contends that it is especially appropriate to use the group's consolidated financial expenses in this case because group companies borrow from each other interest-free. As a consequence, Ekinciler contends that allocating Ekdemir's financing expenses over only its own cost of sales would fail to take into account the fact that a good part of the financial burden assumed by Ekdemir was incurred to finance the goods and services sold by other group companies. According to Ekinciler, adequate information exists on the record to determine the group's financing expenses on a consolidated basis. Specifically, Ekinciler contends that it submitted the total group financing expenses as part of its October 25, 2000, supplemental response, which it claims were verified during the cost verification at Ekdemir. Moreover, although Ekinciler notes that the Department found during the cost verification that the cost of sales reported for the group contained intra-group transactions, it asserts that it remedied this problem by providing a revised cost of sales during the subsequent sales verification. (20) Thus, Ekinciler asserts that the Department should recalculate financing expenses using this information for purposes of the final results. The petitioner contends that the Department should not revise its calculation of financing expenses for Ekinciler. As a threshold issue, the petitioner contends that certain of the data necessary for this calculation is untimely-filed new factual information, and as such should be rejected in accordance with the Department's practice and regulations. Specifically, the petitioner contends that Ekinciler "slipped" the revised consolidated cost of sales information into a verification exhibit, (21) while it asserts that there is nothing in writing on the record to support Ekinciler's assertion that Ekdemir loaned money to other companies in the group on an interest-free basis. Nonetheless, the petitioner argues that if the Department does not reject Ekinciler's information as untimely, there are still several reasons why using the consolidated financial expenses of the Ekinciler group is inappropriate. First, the petitioner contends that the Ekinciler group's method of consolidation is suspect, because Ekinciler did not explain what factors it considered in determining which companies were included in the consolidation. The petitioner notes that Ekinciler included certain members of the group, while it excluded several others. According to the petitioner, under standard accounting principles affiliated companies are not automatically consolidated into a single entity. (22) The petitioner asserts that, because Ekinciler did not explain the basis for its consolidation, the Department cannot determine whether the companies included in it were appropriate. Similarly, the petitioner asserts that, while Ekinciler did not explain the basis for excluding certain companies, it would be inappropriate to exclude them due to product lines (e.g., rebar vs. non-subject merchandise) or geographic boundaries (e.g., operations in Turkey vs. the United States). According to the petitioner, even assuming that Ekinciler selected the appropriate companies for consolidation, there is no reliable evidence that Ekinciler accurately performed the consolidation. The petitioner notes that Ekinciler provided only summary worksheets showing final consolidation figures, without accompanying worksheets setting out details of any intra-company eliminations. The petitioner asserts that, due to this lack of data, there is no way for the Department to verify that Ekinciler's consolidation is proper. In any event, the petitioner argues that the Department's practice has been limited to situations where it could rely on a financial statement. For this reason, the petitioner asserts that Ekinciler's reliance on Silicon Metal from Brazil is misplaced. Specifically, the petitioner contends that in Silicon Metal from Brazil, the Department accepted unaudited, consolidated financial statements, which were prepared at the direction of the Department and which the Department was able to reconcile to the data reflected on the respondent's audited financial statements. In contrast, the petitioner argues that, in the instant case, Ekinciler did not provide an unaudited consolidated financial statement, nor does the Department have the ability to verify the veracity of the so-called consolidated data to audited financial statements of each company in the group. Finally, the petitioner asserts that the CIT has held that the Department must determine the "true costs of the specific exporter," which in this case is Ekdemir. According to the petitioner, in Aimcor v. United States, 69 F. Supp. 2d 1345 (CIT 1999) (Aimcor) the CIT held that the Department is statutorily mandated to use a financing ratio which will more accurately reflect actual costs incurred, especially in cases where there is no evidence of inter-company borrowings or other indicia that the parent company determined the respondent's cost of money. The petitioner asserts that the circumstances in Aimcor are the same as those here, because there is neither any evidence on the record that Ekdemir loaned money to any group companies nor any other indicia that Ekdemir's parent company determined Ekdemir's cost of borrowing. Department's Position: The Department's longstanding practice with regard to financing expenses is to base net financing expenses on the full-year net interest expense and cost of sales from the audited fiscal year financial statements at the highest level of consolidation which correspond most closely to the POR. See Carbon Steel Plate from France, 64 FR at 73152. This practice has been upheld by the CIT. See Gulf States Tube, at 647-48. In this case, no company in the Ekinciler "group" prepares consolidated financial statements. Therefore, in order to determine its financing expenses, Ekinciler attempted to consolidate the financing expenses of Ekdemir (the group rebar producer) with all other companies located in Turkey. Ekinciler also reported the financing expenses of Ekdemir but did not use these expenses in its cost database. At verification, we reviewed both of the submitted calculations. We found that the "consolidated" figure was not prepared using generally accepted methods, and consequently it did not accurately represent the real financing experience of the Ekinciler group. Specifically, we stated the following in our cost verification report: Ekdemir calculated its financing expenses using the combined expenses and cost of sales of all companies in the Ekinciler Group which operate in Turkey. According to Ekdemir officials, these companies sell goods and services to each other. At verification, we found that Ekdemir had not adjusted its cost of sales figures to eliminate these inter-group transactions. Therefore, it may not be appropriate to base financing expenses on the combined expenses of the Ekinciler Group as reported. See the Ekinciler cost verification report at page 2. Ekinciler attempted to remedy the above deficiency during the cost verification. However, given that: 1) the number of companies included in the calculation was voluminous; and 2) the changes necessary to correct the problem were complicated and extensive, the verification team determined that it would not be possible to review a revised worksheet in the time allotted for verification. As a consequence, the verifiers were unable to review the information beyond that already on the record, although they did tie this information to Ekinciler's accounting system. At the subsequent sales verification, Ekinciler again attempted to submit a revised calculation of Ekinciler's cost of sales. The sales verification team also informed the company that the Department: 1) did not intend to verify this information due to the time constraints of the verification process; and 2) as a consequence, would not accept this new information in the context of this administrative review. Although a worksheet containing this new information was provided as part of a sales verification exhibit, the verification team did not intend to accept this document as part of the official record of this case. Moreover, when the Department realized that this unverified information was on the record, it was returned to Ekinciler. See the September 25, 2001, memorandum from Irina Itkin to the File entitled "Rejection of Document contained in Sales Verification Exhibits of the Ekinciler Group (Ekinciler) in the 1999-2000 Antidumping Duty Administrative Review on Certain Steel Concrete Reinforcing Bars from Turkey." On October 9, 2001, Ekinciler filed a letter with the Department in which it requested that the Department reconsider its rejection of the document in question. We responded to this letter on October 25, 2001. See letter from Richard W. Moreland to counsel for Ekinciler, entitled "Rejection of Document in the Antidumping Duty Administrative Review on Steel Concrete Reinforcing Bars from Turkey." Regarding the calculation on the record, we find that it is inappropriate to use this information for two reasons. First, the calculation does not present a true picture of the financing experience of the Ekinciler group because Ekinciler did not include all of the group's subsidiaries in the "consolidation." More importantly, however, because Ekinciler failed to eliminate inter-company transactions from the cost of sales figure used as the denominator of the "consolidated" ratio, the denominator of the calculation is overstated (and, therefore, the financing expense ratio is understated). As a consequence, we have continued to rely on the financing expenses of Ekdemir for purposes of the final results. We recognize that we accepted Ekinciler's "consolidated" financing expenses in the most recently completed segment of this proceeding. However, this action does not change our finding that Ekinciler's calculation in this segment contains fatal flaws. Moreover, we note that the information submitted in the last administrative review was not verified, and as a result similar problems were possibly not discovered. Therefore, the factual circumstances present in the different segments of this proceeding are, in fact, different. For this reason, our rejection of the "consolidated" ratio in this case is not inconsistent with our acceptance of it in the previous unverified segment. Comment 27: Calculation of Ekinciler's Financing Expenses The petitioner asserts that the Department should revise Ekinciler's financing expense ratio, based on its findings at verification. Specifically, the petitioner contends that the Department should: 1) include various financial expense and income items that had not been taken into account in the reported figure; 2) disallow short-term interest income which the petitioner claims is related to accounts receivable, because this type of income would typically be classified as interest revenue; and 3) exclude gains related to accounts payable, in accordance with the Department's practice. Ekinciler notes that the Department included the unreported items discovered at verification in its recalculation of G&A expenses. Thus, Ekinciler maintains that no further adjustment is necessary for these items. Regarding the petitioner's argument related to the short-term interest offset, Ekinciler notes that the income in question was related to a short-term deposit of funds associated with a sale to Syria. However, Ekinciler points out that, at the time that the income was earned, the funds from the sale had been already received and deposited in Ekinciler's bank account. According to Ekinciler, once this money was deposited, the source of it was irrelevant. Ekinciler asserts that the only important fact about this transaction was that the deposit was short-term in nature, and consequently the Department should continue to grant the full offset for interest income related to it. Finally, regarding the petitioner's argument that the Department should disallow exchange gains earned on accounts payable transactions, Ekinciler disagrees that the Department routinely disallows such gains. Indeed, Ekinciler maintains that the Department's standard practice is to allow them. In support of this conclusion, Ekinciler cites Certain Preserved Mushrooms From Indonesia; Preliminary Results of Antidumping Duty Administrative Review, 66 FR 13903, 13906 (Mar. 8, 2001) and Notice of Final Results of Antidumping Duty Administrative Review and Final Determination Not To Revoke Order in Part; Canned Pineapple Fruit From Thailand, 65 FR 77851 (Dec. 13, 2000) (Thai Pineapple) and accompanying decision memorandum at Comment 10. Therefore, Ekinciler argues that the Department should continue to allow these gains. Department's Position: We disagree with the petitioner. Regarding the unreported items discovered at verification, we note that we included these items in G&A expenses. Therefore, we find that it is not necessary to make any further adjustment for these items for purposes of the final results. Regarding the short-term interest offset, we found at verification that the income in question related to interest earned on a short-term deposit of funds in a bank account. See the Ekinciler cost verification report at page 24. Therefore, we find that this income is not analogous to the type of short-term income disallowed by the Department (i.e., interest revenue received from customers for late payments of sales invoices). As a consequence, we have continued to allow an offset for this income for purposes of the final results. Finally, regarding the issue of exchange gains related to accounts payable, we note that it is the Department's general practice to allow such gains. See, e.g., Thai Pineapple at Comment 10. Nonetheless, in this case, because we are using a replacement cost methodology, we have determined that it is not appropriate to include any foreign exchange gains or losses generated from accounts payable transactions in the calculation of G&A expenses. See Comment 18, above. Therefore, we have excluded both gains and losses related to accounts payables from the calculation of financing expenses for purposes of the final results. ICDAS Comment 28: New Factual Information The petitioner contends that a submission made on May 29, 2001, by ICDAS contained untimely new factual information. Specifically, the petitioner asserts that this information relates to the physical attributes of certain unreported sales in the home market and the manner in which ICDAS treats them in the normal course of business. Accordingly, the petitioner maintains that the Department should reject this new factual information and resort to facts available for ICDAS. ICDAS disagrees that the submission in question contained new factual information. Rather, ICDAS maintains that this submission simply contained a synopsis of factual information that was submitted in its Section D response. Therefore, ICDAS asserts that the Department should reject the petitioner's allegation as unfounded. Department's Position: We agree with ICDAS. The company originally submitted its May 29, 2001, letter on April 13, 2001. We reviewed the information in the April 13 submission and determined that it contained new factual information related to various facilities owned by certain home market customers, as well as further processing performed by those companies. As a result, we requested that ICDAS redact its April 13 submission to remove the new factual information. ICDAS resubmitted this letter on May 29. We have reexamined the information contained in ICDAS's letter and have again concluded that it does not constitute additional new factual information. Specifically, we note that the issue of ICDAS's unreported sales, as well as the physical attributes of the rebar sold, was discussed in depth with company officials at both the sales and cost verifications. These attributes can be seen on the source documents contained in sales verification exhibit 6; the production process for the company's standard and non-standard products is discussed on pages 6 and 18 of the ICDAS cost verification report. (23) The production and sales process for the products in question are also discussed on page D-6 of ICDAS's September 29, 2000, response to section D of the questionnaire. Accordingly, we have continued to accept ICDAS's May 29 letter as part of the record in this proceeding. Comment 29: ICDAS's Scrap and Labor Costs The petitioner contends that the Department discovered at verification that ICDAS understated the cost of affiliated party scrap purchases in May, September, and December 1999. The petitioner asserts that the Department should correct ICDAS's cost of manufacturing for purposes of the final margin calculations to account for the understatement. The petitioner also contends that the Department discovered at verification that ICDAS's March 2000 labor costs in its rolling mill were out of proportion in comparison to its January and February costs. The petitioner implies that ICDAS recorded its labor costs in the month in which these costs were paid, rather than the month in which they were accrued. The petitioner argues that the March costs should be reduced by the liability incurred during the months of January and February, and the January and February costs should be increased accordingly. The petitioner asserts that, because this case involves high inflation, accurate monthly costs are important. ICDAS states that it based the reported transfer prices for its affiliated party scrap purchases on the price of the last domestic scrap purchase from non-affiliated parties at the end of each month. Based on this methodology, ICDAS notes that the reported transfer prices were higher than the weighted-average prices for four of the six months for which ICDAS reported sales and cost information, and lower for the remaining two months. ICDAS asserts that, when costs for the six months are averaged on a constant currency basis, these differences cancel each other out. In any event, ICDAS asserts that, given that the percentage of total scrap purchases made from affiliates is minuscule, the difference between the transfer and market prices for the two months in question is not material to the calculation of ICDAS's costs. Regarding the question of labor, ICDAS notes that the amounts in question relate to employee bonuses. ICDAS notes that these bonuses were paid in March and accounted for as March expenses in ICDAS' general ledger. Therefore, ICDAS asserts that any adjustment related to the timing of these expenses would be inappropriate. Finally, ICDAS asserts that, because these bonuses were paid outside the cost reporting period, it is irrelevant whether they are treated as January, February, or March expenses for purposes of the calculation of COM. Department's Position: The Department's regulations at 19 CFR 351.407(b) state that the Department will normally determine the value of a major input purchased from an affiliated party based on the higher of the price paid to the affiliated party (i.e., the transfer price), the amount usually reflected in sales of the major input in the market under consideration (i.e., the market price), or the cost to the affiliated party of producing the major input. Because scrap is the major input in the production of rebar, the major input rule set forth in 19 CFR 351.407(b) applies here. At verification, we found that the weighted-average transfer prices of ICDAS's scrap purchases from an affiliated party in May, September, and December of 1999 were below the weighted-average prices of its scrap purchases from unaffiliated parties in those months. (24) For the purposes of these final results, we have adjusted ICDAS's reported purchase prices for scrap for the months of May and September to reflect market prices. Because the Department granted ICDAS' request for a shortened reporting period of April 1999 through September 1999, December is not relevant to our analysis. Therefore, we have no need to adjust this month. Regarding labor, we agree with ICDAS that the timing of the lump-sum labor payment that occurred in March 1999 is irrelevant in this case because the payment was for a retroactive wage increase for the months of January and February 1999 (25) and these months, as well as the month of payment, are outside the cost reporting period of April through September 1999. Consequently, the timing of labor costs for a period which precedes the reporting period cannot affect the reported costs, and we have made no adjustment for purposes of the final results. Comment 30: ICDAS's Secondary Materials The petitioner asserts that the Department found at verification that there was a decrease in the amount of secondary material used by ICDAS during the second quarter of 1999, as compared to the first quarter of that year. The petitioner contends that, since ICDAS could neither explain nor verify this decline, the Department should base the cost of secondary materials on the company's first quarter 1999 data and index those figures to the second quarter. According to ICDAS, the company adequately explained the reason for the decline in the costs associated with secondary materials in the first quarter of 1999. Specifically, ICDAS notes that the verification report states that: 1) the amount of secondary materials used in the second and third quarters decreased; and 2) roller refurbishing prices dropped. See ICDAS cost verification report at page 13. ICDAS points out that secondary materials are only used when parts or components wear out, which may be at irregular intervals. Thus, ICDAS maintains that period-to-period comparisons for such items may not be appropriate. Finally, ICDAS notes that it based the reported costs for secondary materials on its inventory accounting records. Thus, ICDAS asserts that there is no basis for rejecting these costs. Department's Position: We disagree with the petitioner that the decrease in ICDAS's cost of secondary materials in the second quarter of 1999 could be neither explained nor verified. At verification, ICDAS indicated that the cost of secondary materials declined due to a drop in roller refurbishing prices. We tested this statement and found it to be true. Specifically, the cost verification report for ICDAS states: "We reviewed the material movement listing for the first and second quarters which showed that the roll refurbishing prices had in fact dropped significantly in the second quarter of 1999." (26) Therefore we find that it would not be appropriate to adjust the second quarter cost of secondary materials for purposes of the final results. As noted by the respondent, secondary materials costs, as shown on ICDAS' cost accounting report, also declined in the third quarter of 1999 as compared to the second quarter of 1999. However, contrary to the respondent's argument, this decline was not explained as the result of roller refurbishing prices. As noted in our cost verification report, (27) ICDAS officials were not able to explain the decrease in the cost of secondary materials in the third quarter of 1999 shown on the cost accounting report. Nonetheless, although the officials could not provide an explanation for this decrease, the testing we conducted during verification substantiated that the fiscal year 1999 cumulative total of secondary material costs for ICDAS's rolling mill was supported by ICDAS's normal books and records and that all of these costs were included in ICDAS's reported costs. (28) For this reason, we have not adjusted ICDAS's rolling mill's secondary materials costs. Comment 31: ICDAS's Packing Costs The petitioner notes that the Department discovered at verification that ICDAS failed to include the costs of packing for internal use in the company's reported costs. The petitioner asserts that the Department should add these costs to the costs reported by ICDAS because there is no reason to exclude them. ICDAS concedes that it inadvertently excluded from its reported costs an account containing costs for packing for internal use. However, ICDAS contends that the Department should consider this omission immaterial. Department's Position: During verification, we found that ICDAS did not include internal packing costs for the company's anthracite plant in its reported costs. See the ICDAS cost verification report at page 15. All parties agree that these costs should have been reported. Consequently, we have increased ICDAS's reported costs for internal packing based on our findings at verification. Comment 32: Treatment of ICDAS's Foreign Exchange Gains and Losses and Interest According to the petitioner, in its normal books and records ICDAS records foreign exchange gains and losses, as well as interest related to imported scrap, as part of its cost of materials. The petitioner notes that ICDAS reclassified these expenses as financing expenses for both external reporting purposes and its questionnaire response. The petitioner contends that this reclassification is distortive because it results in the costs in question being allocated to all products, rather than solely to the subject merchandise. Accordingly, the petitioner contends that the Department should rely on ICDAS's books and records to compute the amount of foreign exchange gains and losses and interest related to imported scrap purchases. According to ICDAS, the gains and losses in question are treated as part of its financing expenses in its normal books and records. Specifically, ICDAS notes that, while it is true that financing expenses and G&A expenses are capitalized as part of COM and inventories for purposes of the company's internal and tax accounting, this treatment is reversed by the company's auditors at year end and reclassified as period operating expenses on the audited financial statements. ICDAS further notes that this treatment is consistent with GAAP in both the United States and Turkey. ICDAS disagrees that the inclusion of these expenses in financing costs is distortive. Specifically, ICDAS notes that the only products that it produces are steel billets, rebar made from these billets, and a small amount of by-products. Thus, ICDAS notes that there are no other products to which these costs might have been allocated. Furthermore, ICDAS notes that, while the company does sell certain services (e.g., it runs a fleet of ships), the company's auditors made a similar reclassification of foreign exchange losses associated with these services. Therefore, because the amount of exchange losses related to services significantly exceeded the exchange losses on rebar, ICDAS contends that it has been conservative in its treatment of these expenses. As a consequence, ICDAS contends that, if the Department were to consider an adjustment in this area, it should remove from financing expenses the amounts reclassified from both the cost of goods sold and from the cost of services before adding the losses in question to COM. Department's Position: The Department's long-standing practice, as set forth in section 773(f)(1)(A) of the Act, is to rely on data from a respondent's normal books and records where those records are prepared in accordance with home country GAAP and reasonably reflect the costs of producing the merchandise. As noted in ICDAS's submission to the Department, (29) and the Department's cost verification report, (30) ICDAS's inclusion of its foreign exchange gains and losses and interest in its cost of materials is for tax reporting purposes only. At the end of each fiscal year, ICDAS reclassifies these costs from the cost of goods sold to operating expenses in accordance with the Turkish Capital Market Board's regulations. (31) In most instances, following the respondent's home country accounting principles provides both the respondent and the Department a reasonably objective and predictable basis by which to compute costs for the merchandise under investigation. However, in those instances where it is determined that a company's normal accounting practices result in a misallocation of production costs, the Department will adjust the respondent's costs or use alternative calculation methodologies that more accurately capture the actual costs incurred to produce the merchandise. (32) As such, it is the Department's normal practice in cases involving significant inflation (i.e., use of replacement cost methodology) not to adjust a respondent's costs for foreign exchange gains and losses related to accounts payable transactions. (33) See Cold-Rolled Steel from Turkey at Comment 3.1 and Comment 18, above. We disagree with the respondent's proposal that the Department remove from ICDAS's financial expenses the amounts reclassified from cost of services. It is evident from the exhibits obtained during the Department's cost verification (34) that the amount reclassified from the cost of services is considered to be part of financial expenses and not foreign exchange gains and losses. Therefore, we have only removed the amount of foreign exchange gains and losses on ICDAS's purchases of raw materials from ICDAS's financial expenses. Regarding the issue of interest, we disagree with the petitioner's argument that interest related to the financing of raw materials should also be included in the COM. ICDAS, under the Turkish Capital Market Board's regulations, does not include interest expenses in its cost of goods sold. Instead, it classifies interest expenses as operating expenses. It is the Department's long-standing practice to calculate interest expense, on the consolidated level, separately from COM. See Notice of Final Determination of Sales at Less than Fair Value: Certain Carbon Steel Butt-Weld Pipe Fittings from Thailand, 60 FR 10555, 10557 (Feb. 27, 1995). Interest expenses incurred in the financing of raw materials are directly related to the company's management's decision on the usage of its capital. For these reasons, the Department's practice is not to distinguish whether interest expense is related or unrelated to the merchandise under review. See e.g., Gray Portland Cement and Clinker From Mexico; Final Results of Antidumping Duty Administrative Review, 62 FR 17148, 17161 (Apr. 9, 1997) and Certain Welded Carbon Steel Pipes and Tubes From Thailand; Final Results of Antidumping Duty Administrative Review, 61 FR 56515, 56519 (Nov. 1, 1996). Consequently, for purposes of this final results, we have not reclassified interest related to ICDAS's purchases of raw materials as part of the company's COM. Recommendation Based on our analysis of the comments received, we recommend adopting all of the above positions. If these recommendations are accepted, we will publish the final results of review and the final weighted-average dumping margins for the reviewed firms in the Federal Register. Agree____ Disagree ____ Faryar Shirzad Assistant Secretary for Import Administration (Date) 1. ICDAS asserts that the Department generally finds volumes of less than five percent to be insignificant for purposes of the dumping analysis. See Gulf States Tube Division of Quanex Corporation v. United States, 981 F. Supp. 630, 642 (1997) (Gulf States Tube). 2. Because this product comparison issue was not raised by any party in this segment of the proceeding, we did not request comments from the parties on this topic. However, we intend to consider this issue further in the context of the next administrative review. 3. In fact, both Diler and Ekinciler note that they specifically requested to limit the home market reporting periods to six months in order to provide the Department sufficient information to determine whether their below-cost sales were made in substantial quantities within an extended period of time (as defined under section 773(b)(2)(C) of the Act), which is the final condition which must be met before below-cost sales may be disregarded. 4. Although the method of determining the amount of Turkish lira ultimately paid was known at the time of sale, it was based on the U.S.- dollar price set on the date of sale and merely resulted in the payment of that U.S.-dollar price, albeit in the Turkish lira equivalent. 5. We disagree with Colakoglu that we have not captured the full revenue received on the sale by using the U.S.-dollar price; indeed, the U.S.- dollar price is the full price which we have merely stated in terms of the Turkish-lira value in the month of sale. In contrast, Colakoglu would have us use the Turkish-lira equivalent based on lower-value Turkish lira in a later month. In either case, the total revenue in real terms is the same. 6. Colakoglu's argument with respect to Silicon Metal from Brazil is that the Department cannot calculate profit accurately if it does not take into account the amount of kur farki received on its sales priced in U.S. dollars. However, both Colakoglu's home market prices and its costs were stated on the same basis, and thus any comparisons of these prices and costs are valid. 7. We note that the use of this methodology allows price-to-price comparisons without introducing unnecessary conversions. Moreover, while it may result in more sales falling below cost as a result of the cost test, this methodology results in more accurate price-to-cost comparisons because, as noted above, prices and costs are stated in terms of the same currency value. 8. Because Colakoglu has claimed proprietary treatment for the exact figures, we are unable to disclose them here. For further discussion, see pages 16 and 17 of the November 30, 2000, memorandum from Shawn Thompson and Jim Nunno to Louis Apple entitled "Verification of the Cost of Production and Constructed Value Data Submitted by Colakoglu Metalurji A.S. in the Antidumping Duty Administrative Review on Certain Steel Concrete Reinforcing Bars from Turkey" (the Colakoglu cost verification report). However, we note that the discrepancies were all less than one half of one percent of the reported production quantities (and in some cases significantly less). 9. Due to the proprietary nature of these expenses, we are unable to discuss them here. For further information, see page 6 of the petitioner's case brief and pages 21 and 22 of the Colakoglu cost verification report. 10. We note that Colakoglu began recognizing depreciation on this new cost center in 1999, before it began actual production. Because none of these expenses were associated with the production of rebar during 1999, we have not allocated any of these costs to rebar during that year. 11. Specifically, the auditors' opinion stated: "In our opinion, the financial statements present fairly, in all material respects, the financial position of Colakoglu Metalurji A.S. at 31 December 1999 and the results of its operations for the year then ended in accordance with International Accounting Standards." See Exhibit S-5 of the October 17, 2000, submission. 12. Nonetheless, Diler notes that its cost accounting system tracks the production quantities of the grades exported to other countries, even through it does not assign them different costs. 13. For example, assume that Diler's costs for size 1 at factory 1 was TL 10 per ton and at factory 2 was TL 15. If Diler produced 100 tons of grade 40 at factory 1 and 500 tons at factory 2, the average cost would be TL 12.50 for grade 40. If Diler produced 300 tons of grade 60 at factory 1 and 50 tons at factory 2, the average cost would be TL 10.71 for grade 60. This difference is solely attributable to the difference in production quantities at the two factories. 14. Diler notes that these losses are called kur farki in Turkey. 15. Diler sets forth the following example in its brief. Assume that the company purchases a ton of scrap for $75 per ton in April and pays for it in May. Also assume that the price remains $75 per ton in May and June, and the exchange rates for April, May, and June are TL 100/$, TL 110/$, and TL 120/$ respectively. In constant currency terms, the material cost in April, May, and June should be the same (i.e., $75). In Turkish lira, however, these costs would be recorded in the company's books as follows: April: TL 7,500 (or $75 * 100) May: TL 8,250 (or $75 * 110) May: TL 750 exchange loss on the April purchase (or the paid amount of TL 8,250 (i.e., $75*110) less the booked amount of TL 7500) June: TL 9,000 (or $75 * 120) Diler asserts that the this example demonstrates that the replacement cost reflects the full kur farki amount between April and May, and May and June (i.e., the respondent would report replacement costs of TL 7,500 for April, TL 8,250 for May, and TL 9,000 for June). Diler asserts that adding the TL 750 of exchange losses to the April purchase price would convert this price into its May equivalent (i.e., TL 7,500 plus TL 750 = TL 8,250). 16. The petitioner also performs its own calculation of Ekinciler's 1999/2000 depreciation expenses based on estimated net book value of each asset group tracked by the company, which it offers as additional evidence that the company's depreciation expenses are understated. 17. We also disagree with the petitioner that Ekinciler failed to report depreciation on its idled assets. We found no evidence at verification that Ekinciler idled any production assets during the reporting period. Moreover, while Ekinciler may have idled certain assets in previous years, these assets were fully depreciated in Ekinciler's books and records. See the October 23, 2001, memorandum to the File from Irina Itkin entitled, "Placing Information Regarding Ekinciler Demir Celik A.S. from the 1996- 1998 Antidumping Duty Administrative Review on the Record of the 1999-2000 Antidumping Duty Administrative Review of Certain Steel Concrete Reinforcing Bars from Turkey." 18. In our April 23, 2001, supplemental questionnaire, we identified a category of assets shown in Ekinciler's fixed asset ledger on which Ekinciler had not recognized any depreciation in 1999. We requested that Ekinciler describe each of these assets and explain why it had not depreciated them. In its May 11 response, Ekinciler indicated that these assets were capitalized financing expenses related to investments made from 1989 through 1994. According to Ekinciler, under Turkish tax law, companies are allowed to recognize 100 percent of those expenses in the related year's taxable income but cannot depreciate them in subsequent years. Ekinciler asserted that it took advantage of this tax provision, with the result that it could not depreciate these assets. Nonetheless, Ekinciler maintained that, even had it not chosen to do this, the assets would have been fully depreciated by 1999 because they had a five-year depreciation schedule. 19. It is the Department's long-standing practice, codified at section 773(f)(1)(A) of the Act, to rely on data from a respondent's normal books and records where those records are prepared in accordance with home country GAAP and reasonably reflect the costs of producing the merchandise. However, in those instances where it is determined that a company's normal accounting practices result in a misallocation of production costs, the Department will adjust the respondent's costs or use alternative calculation methodologies that more accurately capture the actual costs incurred to produce the merchandise. See also Notice of Final Determination of Sales at Less Than Fair Value: Fresh Atlantic Salmon From Chile, 63 FR 31411, 31432 (June 9, 1998) (Salmon from Chile) and Comment 16, above. 20. This information was returned to Ekinciler on September 25, 2001. See below. 21. The petitioner concedes that verification exhibits may be part of the record. However, it argues that the Department should reject such exhibits if they contain new factual information. 22. For example, the petitioner notes that companies viewed as investments are not subject to consolidation. 23. See the April 17, 2001, memorandum from LaVonne Jackson and Taija Slaughter to Neal M. Halper entitled "Verification Report on the Cost of Production and Constructed Value Data Submitted by ICDAS Celik Enerji Tersane ve Ulasim Sanayi A.S." (the ICDAS cost verification report). 24. See ICDAS cost verification exhibit 10 at page 15. 25. See the ICDAS cost verification report at page 24. 26. See the ICDAS cost verification report at page 13. 27. See the ICDAS cost verification report at page 13. 28. Id. 29. See ICDAS's September 29, 2000, section D response at page D-13. 30. See the ICDAS cost verification report at page 8. 31. Company officials stated during verification that the Turkish Capital Market Board's regulations closely follow U.S. GAAP. See the ICDAS cost verification report at page 9. 32. See Salmon From Chile, 63 FR at 31432. 33. This practice applies in cases where it can be shown that the exchange rate and inflation rate during the POR were closely linked. 34. See ICDAS cost verification exhibit 6 at pages 3 and 45.