65 FR 81825, December 27, 2000 A-557-809 Investigation Public Document DAS III / 9 / JHC December 15, 2000 MEMORANDUM TO: Troy H. Cribb Assistant Secretary for Import Administration FROM: Joseph A. Spetrini Deputy Assistant Secretary AD/CVD Enforcement Group III SUBJECT: Issues and Decision Memorandum for the Less Than Fair Value Investigation of Stainless Steel Butt-Weld Pipe Fittings from Malaysia: October 1, 1998 through September 30, 1999 SUMMARY: We have analyzed the comments and briefs of interested parties in the less than fair value ("LTFV") investigation of stainless steel butt-weld pipe fittings from Malaysia. As a result of our analysis, we have made changes from the Notice of Preliminary Determination of Sales at Not Less Than Fair Value and Postponement of Final Determination: Stainless Steel Butt-Weld Pipe Fittings from Malaysia, 65 FR 47398 (August 2, 2000) ("Preliminary Determination"). The specific programming changes can be found in our Analysis Memorandum for Kanzen Tetsu Sdn. Bhd.: Final Determination in the Less Than Fair Value Investigation of Stainless Steel Butt-Weld Pipe Fittings from Malaysia (December 15, 2000) ("Final Analysis Memo"). We recommend that you approve the positions we have developed in the "Discussion of the Issues" sections of this Issues and Decision Memorandum. Below is the complete list of the issues in this investigation: 1. General Issues Comment 1: Ministerial Errors From the Preliminary Determination 2. General Sales Issues Comment 2: Date of Sale/Market Viability Comment 3: Bank Charges 3. U.K. Sales Issues Comment 4: Domestic Inland Freight Comment 5: Credit Period Comment 6: FOB v. CIF Comment 7: Early Payment Discount 4. U.S. Sales Issues Comment 8: Marine Insurance Expense Comment 9: Marine Insurance Expense Discount and Denomination Comment 10: Returns Comment 11: Miscellaneous Bank Charges Comment 12: Unreported U.S. Sales Comment 13: Unshipped Sale Comment 14: Inland Freight 5. Cost issues Comment 15: Total Adverse Facts Available Comment 16: Allocation of Cost Variances Comment 17: Standard Cost Reduction Factor for Pipes Used for Fittings Comment 18: Cost of Fittings Made of Finished Pipes Comment 19: G&A Expense Ratio DISCUSSION OF THE ISSUES: 1. General Issues Comment 1: Ministerial Errors From The Preliminary Determination After the Preliminary Determination, the Department issued a ministerial error memorandum discussing four issues which petitioners alleged as requiring corrections. See Ministerial Error Memorandum for the Preliminary Determination of Sales at Not Less Than Fair Value (August 17, 2000) ("Ministerial Error Memo"). We agreed with petitioners on three of the four issues, that we should: 1) convert the indirect selling expenses from Malaysian ringgits to U.S. dollars to be consistent with all other expense elements; 2) reassign certain weighting codes to properly match products; and 3) correct the foreign unit price expressed in dollars ("FUPDOL") calculation to properly add the difference in merchandise adjustment ("DIFMER") and U.S. packing expenses to normal value ("NV"). However, as correcting the ministerial errors would not result in a change of at least five absolute percentage points in the weighted-average dumping margin, nor a weighted-average dumping margin of greater than de minimis, we chose not to make corrections at that time, but stated that we would do so, if it continued to be appropriate, for the Final Determination. Petitioners urge that the Department correct the ministerial errors for purposes of the Final Determination. See Petitioners' Case Brief at 59 (November 13, 2000) ("Petitioners' Brief"). Kanzen Tetsu Sdn. Bhd. ("Kanzen" or "respondent") did not submit comment on this issue in its rebuttal brief. See Rebuttal Brief of Kanzen Tetsu Sdn. Bhd. (November 20, 2000) ("Kanzen Rebuttal"). Department Position: We agree with petitioners that we should make corrections to these ministerial errors, as discussed in the Ministerial Error Memo. Without these corrections, we cannot properly match sales nor can we properly calculate the margin. Pursuant to 19 C.F.R. 351.224(e), the Department will, "if appropriate, correct any significant ministerial error by amending the preliminary determination, or correct any ministerial error by amending the final determination . . .{and} publish notice of such corrections in the Federal Register." In accordance with our regulations, we have edited the model match and margin calculation programs to reflect these corrections. 2. General Sales Issues Comment 2: Date of Sale/Market Viability Petitioners argue that the U.K. is not the proper market on which the Department should base NV. Instead, petitioners assert that the Department should base NV on the Malaysian home market. See Petitioners' Brief at 6- 15. Petitioners argue that the U.K. is not the proper comparison market for four reasons: 1) improper choice for date of sale; 2) anomalous sales in the U.K. market; 3) questionable reporting of Malaysian sales; and 4) questionable reporting of U.S. sales. First, petitioners argue that while invoice date is the customary choice for date of sale, Kanzen pointed to two other dates, purchase order date, and order confirmation date, in arguing that the U.K. is a viable market for sales comparisons to the U.S. during the period of investigation ("POI"). Petitioners argue Kanzen initially used purchase order date as date of sale, but after it was noted that the purchase order date of Kanzen's largest U.K. sale fell outside the POI, Kanzen instead claimed it used order confirmation date (hereinafter also referred to as "contract date") as date of sale. Petitioners argue that if the Department accepts a date other than invoice date, purchase order date is more appropriate. Petitioners claim that, according to Kanzen's description of its sales process, the material terms of sale are established in the purchase order; the order confirmation/contract merely confirms established terms. Petitioners point out that considerable negotiation goes on between Kanzen and its customers before issuance of the purchase order. Petitioners assert that a comparison of material changes in the terms of sale between purchase order and order confirmation/contract, against material changes after order confirmation further supports their position. Additionally, Kanzen has stated that it rarely changes sales terms after the order confirmation/contract, other than delivery dates and, in virtually all cases, does not revise price and quantity after the order confirmation/contract (petitioners did question some unusual post-sale adjustments to the U.K. sales, as further discussed, infra). Petitioners cite to 19 C.F.R. 351.401(i), which states that "{i}n identifying the date of sale of the subject merchandise . . . the Secretary normally will use the date of invoice, as recorded in the exporter or producer's records kept in the ordinary course of business. . . . {h}owever, the Secretary may use a date other than the date of invoice if the Secretary is satisfied that a different date better reflects the date on which the exporter or producer establishes the material terms of sale." Accordingly, petitioners argue that if the Department uses a date other than invoice date, it should use the purchase order date, as it is the date on which the material terms of sale were established, and not the contract date, the date on which sales terms were recorded and confirmed. Furthermore, if the Department uses purchase order date, petitioners argue that Kanzen's U.K. sales would fall below five percent of the U.S. sales in quantity, rendering the U.K. market unsuitable as the comparison market. See Petitioners' Brief at 6-10. Second, petitioners question Kanzen's sales to the U.K. market. Petitioners point out that the U.K. sales during the POI were made to a single customer. Petitioners also assert that regardless of the Department's choice for date of sale, in reaching the threshold for a viable market, the U.K. sales quantity is problematic. Petitioners argue that Kanzen's reported gross unit prices for its U.K. sales differ greatly from the average import value compiled in the publicly available official U.K. import statistics. Additionally, while Kanzen has stated that it rarely changes sales terms after order confirmation/contract, the evidence suggests that this was not so for sales to the U.K. customer during the POI. Petitioners point out that Kanzen offered unusual post-sale price adjustments for the customer, including offering a price reduction against increased freight charges borne by the customer even though the sale term was FOB Malaysian port, and accepting a claim by the customer involving a rare issue. Petitioners also argue that payment terms for U.K. sales were more favorable than the same terms for U.S. sales, mentioning U.K. customers' ability to view the merchandise before payment while U.S. customers purchase the merchandise sight unseen, and also mentioning the longer credit period for U.K. customers, as examples. Petitioners also note that the sales term for one of the U.K. sales changed from FOB to CIF, although FOB is the customary sales term. Additionally, while Kanzen offered an early payment discount on a U.K. sale, petitioners note that the invoice had been paid under the terms agreed upon by the parties and was not paid particularly early. Petitioners argue that Kanzen may have offered these unusual post-sale price adjustments to lower the selling prices for its U.K. market. See Petitioners' Brief at 10-12. Third, petitioners question Kanzen's reporting of Malaysian sales for the POI, which Kanzen claimed are below five percent of the U.S. sales in quantity. Petitioners question why Kanzen did not report its sales by invoice date, when its financial data are based on invoice date pursuant to generally accepted accounting principles ("GAAP"), thus creating more work for itself in reconciling its sales to its financial data. Petitioners also question why Kanzen changed its Malaysian invoice numbering system, noting that it made the same changes to its Singapore invoice numbering system and questioning how Kanzen could consider the Singapore sales as "local" in its invoicing system. Petitioners also note that Kanzen stated that it was unlikely that material sold to Singapore would be purchased back into Malaysia due to a 30% duty, yet Kanzen reported no import duties for input materials during the POI. Additionally, petitioners argue that Kanzen's preparedness for verification, in binding its documents by POI when such documents are usually bound by fiscal year, implies Kanzen selectively chose information for the Department. See Petitioners' Brief at 13-14. Fourth, petitioners question Kanzen's reporting of U.S. sales for the POI. Petitioners argue that certain documents from the sales verification indicate Kanzen may have included sales transshipped to a third country via the United States in its U.S. sales database. Petitioners argue that such sales should be excluded from the U.S. database, and such an exclusion could change the viability of the home market. See Petitioners' Brief at 14-15. Kanzen asserts that the U.K. market is a proper, viable market for calculating normal value. Kanzen disagrees with petitioners' arguments regarding date of sale and the reliability of Kanzen's U.K. sales. First, Kanzen argues that contract date is the correct date to use as date of sale, in accordance with Department practice and the Department's instructions. Kanzen points out that the Department's choice of date of sale is to increase predictability and accuracy, while remaining consistent with industry norms. Kanzen explains that it used contract date as the date of sale because: 1) it is verifiable, as Kanzen uses contract date as an integral part of its sales system; 2) the contract establishes the terms of sale; 3) the contract is legally binding and material revisions require written amendments accepted by both parties; and 4) the contract is a normal part of the sales and production system in the industry, as butt-weld pipe fittings are a time-consuming product that must be produced to order in almost all cases. Kanzen asserts that purchase order date is inappropriate because purchase orders are merely proposals, and that Kanzen will not agree to produce for a particular sale until it issues a contract. Kanzen notes the Department has recognized purchase order confirmation date as the appropriate date of sale in other cases, citing Certain Corrosion-Resistant Carbon Steel Flat Products from Japan: Final Results of Antidumping Duty Administrative Review, 64 FR 12951, 12957-58 (March 16, 1999) and Notice of Final Determination of Sales at Less Than Fair Value: Polyvinyl Alcohol from Taiwan, 61 FR 14064, 14067 (March 29, 1996). Kanzen continues that, as confirmed at the sales verification, it applied the same sales process for both the U.S. and U.K. markets. Kanzen believes the use of order confirmation date as date of sale is supported in both markets, noting that many purchase orders contain no prices or request material not available at Kanzen, and noting that because Kanzen can reject the purchase order, it cannot be considered as establishing the material terms of sale. Kanzen argues that, despite petitioners' protestations, using order confirmation/contract date as the date of sale was not difficult, as Kanzen records and uses contract date in its sales information recording system in the ordinary course of business, as well as for production planning and purchasing. Finally, Kanzen argues there are virtually no material revisions to the contract once the contract is issued and, while minor deviations are acceptable, any major changes after contract date usually resulted in the issuance of a revised contract or other documentation evidencing mutual agreement, supporting the contract as establishing the terms of sale. See Kanzen Rebuttal at 3-7. Second, regarding the viability and reliability of the U.K. market as a comparison market, Kanzen argues that petitioners have offered no support for their argument against using the U.K. market as a comparison market. Kanzen asserts that: 1) the U.K. sales were valid export sales made to an unaffiliated customer in the ordinary course of business, as confirmed in the sales verification; 2) these sales account for more than five percent of U.S. sales, regardless of whether contract date or invoice date is used as the date of sale; 3) it sold the same type, grades, and ranges of merchandise in both the U.S. and U.K. markets; 4) its sales processes were identical for both markets; 5) the customer type (distributors) was the same for both markets; 6) the sales patterns of its U.S. and U.K. customers were the same; and 7) the volume of sales in the two markets, in terms of each individual shipment and overall purchases during the POI, were comparable. Kanzen also dismisses petitioners' allegations of transshipments to a third country via the United States as unfounded. Kanzen argues that the customer clearly indicated shipment destination in the order, and the bill of lading and other documentation supported the destination city, as reviewed by the Department during verification. If any merchandise was resold to a third country, Kanzen stated it would have no knowledge of this at the time of sale. Kanzen also maintains that its differences in U.K. sales price levels, as compared to the prices indicated by U.K. import statistics, are understandable because government import statistics are reported on a landed basis, whereas Kanzen's U.K. sales were made on an FOB basis. As for other supposedly "unique" factors pointed to by petitioners, Kanzen asserts that such factors existed in both markets during the POI, were otherwise in accordance with reasonable practices in the butt-weld pipe fitting industry or were reflective of norms in the European market, and do not indicate any special or unique preferential treatment provided to the U.K. customer. Finally, Kanzen argues that two specific circumstances referenced by petitioners (i.e., payment for ocean freight on one shipment, and an early payment discount) were both verified by the Department and are not reasons to reject the use of the U.K. market. See Kanzen Rebuttal at 7-9. Department Position: We agree with petitioners, in part and with respondent, in part. We agree with petitioners that invoice date is the proper date to use in this investigation as date of sale for the comparison market. As pointed out by petitioners, it is Department practice "to use the date of invoice as the date of sale unless there is a compelling reason to do otherwise." See Final Determination of Sales at Less Than Fair Value: Stainless Steel Sheet and Strip in Coil from Germany, 64 FR 30710, 30719-721 (June 8, 1999). Although Kanzen emphasizes that order confirmation/contract date is the date on which the material terms of sale are established, the Department ascertained at verification that changes and adjustments were made to a significant portion of U.K. sales after the order confirmation/contract date, including changes in price and terms. See Sales Verification Report at 2-3. Such changes and adjustments suggest that the material terms of sale for sales to the U.K. market are not necessarily established at confirmation/contract date, but rather remain alterable by the parties. Therefore, the evidence on the record supports a finding that, although the contract nominally appears to establish the material terms of sale, in fact, the material terms of sale in the U.K. market were subject to change up to the date of issuance of the invoice. Accordingly, the information on the record suggests that for the U.K. market, the invoice is the best indication of the final terms of sale established by the parties. Thus, the Department is using invoice date as the date of sale for the comparison market, but maintaining contract date as the date of sale for the U.S. market, in the final determination. In this regard, we note that, in contrast to the U.K. market sales, the circumstances surrounding Kanzen's sales to the U.S. during the POI do not indicate that Kanzen's contracts for the U.S. market sales were subject to changes in their material terms of sale after the contract date. As to choice of comparison market, however, we agree with Kanzen that the U.K. market is a viable market and proper to use as the comparison market in the investigation. We conducted full quantity and value reconciliations of Kanzen's U.K., U.S., and home market sales at the sales verification and noted no discrepancies which would alter Kanzen's assertions regarding market viability. See Sales Verification Report at 12-17. For instance, in our verification of Kanzen's U.S. sales, we turned up no evidence of any transshipments, as alleged by petitioners, nor of any evidence that Kanzen had any knowledge that any transshipments occurred. Neither did we find any evidence that Kanzen had any knowledge that merchandise it sold to Singapore was resold to Malaysian customers (in this regard, we note that Kanzen explained at the sales verification that Singapore sales were not labeled "export" sales due to the structure of Kanzen's sales division). See Sales Verification Report at 14. As for the substantial changes which petitioners claim render the U.K. market an improper comparison market, while such changes serve to justify the choice of invoice date as the date of sale, they are not so problematical or anomalous as to place the viability of the U.K. market in question. In fact, Kanzen provided at verification full documentation regarding these changes. This documentation supports a conclusion that these circumstantial changes were made in the ordinary course of business, and do not support petitioners' assertion that these constituted "unique" post-sale adjustments specifically designed "to lower the selling price for it's {sic} U.K. market." See Petitioners' Brief at 12. As a result, the Department shall continue to use the U.K. market as the comparison market as we did in our preliminary determination for this investigation. See Notice of Preliminary Determination of Sales at Not Less Than Fair Value and Postponement of Final Determination: Stainless Steel Butt-Weld Pipe Fittings from Malaysia, 65 FR 47398, 47399 (August 2, 2000) ("Preliminary Determination"). Comment 3: Bank Charges During verification of the foreign exchange loss used in Kanzen's direct selling expense calculation for U.S. and U.K. sales, we discovered that Kanzen tied the expenses to specific invoices, affecting the direct selling expense for both U.S. and U.K. sales. See Sales Verification Report at 2. Kanzen submitted a correction to its U.K. bank charges at the start of the sales verification, for failing to report a bank charge representing the exchange loss or gain associated with the time involved between booking the sale in the company's accounting system and the bank's receipt of payment for the sale. See Sales Verification Report at 4. Petitioners argue that Kanzen's new allocation methodology is unsubstantiated and its U.K. bank charge calculations remain grossly inaccurate. Petitioners assert that a comparison of the invoices examined at verification demonstrate their point, and argue that because bank charges go into direct selling expense, Kanzen's reported U.K. direct selling expenses are affected. See Petitioners' Brief at 15-17. Petitioners point out that Kanzen acknowledged that its methodology for allocating "hedging costs" (Kanzen's terminology for foreign currency exchange costs) was flawed, after the sales verification revealed that the bank's converted U.S. dollar amount did not correspond to the invoices' U.S. dollar amounts, even though the "hedging costs" should have accounted for the difference between the invoice and bank payment values. Petitioners argue the discrepancies may have been caused by other reporting errors and omissions, such as an extended credit period for certain sales. Petitioners argue that because "hedging costs" account for the exchange losses between the sales record date and the payment date, use of an incorrect payment date resulted in the use of an incorrect exchange rate to convert the payment amount. Petitioners also question the original denomination of Kanzen's U.K. sales. Kanzen reported that its U.K. sales were denominated in U.S. dollars, but petitioners note references in the sales verification exhibits that make it appear otherwise, and argue that it is likely that a currency other than U.S. dollars or Malaysian ringgits was involved during the U.K. sales process. Petitioners argue that an additional currency conversion would generate additional exchange losses and/or gains, contributing to the difference between the sales value booked and the actual payment received. Accordingly, petitioners argue that the Department cannot rely on Kanzen's reported bank charges, and should not allow any deduction of U.K. bank charges in the price-to-price margin analysis but should use Kanzen's revised U.K. bank charges for purposes of the cost test. See Petitioners' Brief at 17-19. Finally, petitioners note that Kanzen failed to correct or even identify the allocation methodology used to calculate U.S. bank charges. Petitioners further note that the exact impact of Kanzen's flawed calculation methodology on reported U.S. bank charges is not specified in the Sales Verification Report. Accordingly, petitioners argue that there is insufficient information on the record for the Department to adjust Kanzen's reported bank charges to reflect actual "hedging costs." As a result, petitioners argue that applying adverse facts available is warranted and the Department should apply the highest unit bank charges reported by Kanzen for its U.S. sales. See Petitioners' Brief at 19-20. Kanzen argues the Department confirmed at the sales verification that Kanzen reported bank charges in the same way for both markets. Therefore, Kanzen maintains that the Department should treat these claimed adjustments in the same manner for both markets. Kanzen argues that these charges were reported "out of a superabundance of caution"; however, Kanzen asserts that the Department's policy is to exclude foreign exchange gains and losses related to sales as they are not a cost. Because the Department applies its own exchange rates to sales in its antidumping analysis, Kanzen maintains that any exchange gain or loss incurred by the company on its sales is irrelevant. Thus, Kanzen argues that the Department should disregard these expenses for both U.K. and U.S. sales, for the purpose of calculating NV and EP, and for the purpose of the sales- below-cost analysis. See Kanzen's Rebuttal at 10-11. Department Position: We agree with respondent, in part. Kanzen attempted to submit a correction to its reported bank charge at the outset of the sales verification. However, when reviewing Kanzen's correction for exchange gain or loss, we determined that Kanzen did not properly allocate the foreign currency exchange costs according to their respective invoices. Rather, Kanzen inappropriately and selectively matched exchange rates to separate invoices. See Sales Verification Report at 4. As a result, we are disallowing these bank charges on direct selling expenses for Kanzen's U.K. sales; however, we have determined that we should use facts available for Kanzen's U.S. sales, as further discussed, infra. For the U.K. sales, not only did Kanzen submit a correction that involved improper allocation of the foreign currency exchange costs, Kanzen also, as discussed, supra, reported at the sales verification that it originally failed to report a bank charge for exchange gain or loss on its U.K. sales. Id. Therefore, because Kanzen had not previously attempted to include these charges in its reported direct selling expenses prior to verification, and because we found at the sales verification that Kanzen's allocation methodology was flawed, we are disallowing these bank charges for the final determination with respect to its U.K. sales. As for Kanzen's U.S. sales, while petitioners advocate the application of adverse facts available and the use of the highest unit bank charges reported by Kanzen for its U.S. sales, we feel this is unwarranted. There is no evidence to suggest that Kanzen did not cooperate to the best of its ability. First, we note that, in contrast to Kanzen's U.K. sales market, Kanzen did report these charges, as it had indicated in its questionnaire response. See Kanzen's May 1, 2000, Section C Questionnaire Response, at C- 36. Furthermore, as demonstrated through its effort to offer a correction, Kanzen attempted to revise and clarify its allocation at the outset of the sales verification. We cannot use the correction submitted by Kanzen at the outset of the sales verification, due to the improper allocation of the foreign currency exchange costs. However, section 776(a) of the Tariff Act of 1930, as amended ("Act"), states that when necessary information is not available on the record, the Department will, subject to section 782(d) of the Act, use facts otherwise available. Here, there is other information on the record which was verified (see section 782(e) of the Act) and that can be used by the Department to calculate an average unit bank charge per ton for Kanzen's U.S. sales. Accordingly, we are applying facts available and calculating an average unit bank charge per ton for Kanzen's U.S. sales for the final determination. The calculation of this bank charge is described in detail in the Analysis Memo. 3. U.K. Sales Issues Comment 4: Domestic Inland Freight At the sales verification, we discovered that for domestic inland freight, Kanzen mistakenly calculated U.K. sales by value, when it should have calculated the sales by weight. See Sales Verification Report at 2, 18-19. Petitioners argue this miscalculation is significant and the Department should not rely on Kanzen's reported domestic inland freight for U.K. sales. Petitioners question how Kanzen could have applied an incorrect allocation methodology, arguing that the Department's questionnaire clearly sets forth the correct methodology for inland freight, and noting that Kanzen's own response indicates that it used a weight-based allocation methodology, dividing total freight charges by total sales quantity to calculate per unit inland freight, citing Kanzen's May 1, 2000 Questionnaire Response at Exhibit 2. Petitioners point out that the Department normally deducts inland freight from gross unit price for comparison market sales to arrive at NV. Petitioners suggest that Kanzen's error could result in a more favorable weighted average dumping margin for Kanzen. Petitioners argue that Kanzen failed to cooperate to the best of its ability, by knowingly misrepresenting its reporting methodology and waiting until the sales verification to disclose its incorrect methodology. Accordingly, petitioners argue that the Department should apply adverse facts available, by denying the deduction of inland freight and using the lowest verified per unit inland freight amount instead to calculate Kanzen's comparison market prices, and by using the highest reported per unit inland freight or, alternatively, the per unit inland freight for each invoice as reported by Kanzen, in the final cost analysis. See Petitioners' Brief at 20-23. Kanzen argues that the Department should not reject Kanzen's reported domestic inland freight for simple clerical errors in its calculation. While Kanzen acknowledges its erroneous calculation, it notes that at the sales verification, the Department reviewed all of the shipments reported in the U.K. sales database, and therefore has the information available to determine the correct domestic inland freight adjustments and, in fact, already made such recalculations at the sales verification (see Sales Verification Report at 19-20). Kanzen maintains that, in this regard, no adverse inference is warranted as substantial record evidence shows that Kanzen has been a cooperative participant. See Kanzen's Rebuttal at 11-12. Department Position: We agree with respondent. We reviewed all of Kanzen's U.K. invoices at the sales verification and, in the process, determined the correct domestic inland freight that should have been reported. Because we have the proper information already available to us, and because the discrepancy is relatively minor (see Sales Verification Report at 18-20), application of adverse facts available is not warranted in this instance. See Final Analysis Memo. Accordingly, we are adjusting our programming to include the corrected domestic inland freight amounts, to replace the figures reported by Kanzen. Comment 5: Credit Period During verification of Kanzen's U.K. sales, the Department discovered that Kanzen reported May 6, 1999 as payment date for certain sales observations, when actual documentation indicated payment date as March 22, 1999. See Sales Verification Report at 2, 19. Petitioners suggest that Kanzen intentionally misreported payment date in order to inflate credit expense for comparison market sales. Petitioners note that the Department uses credit period to calculate imputed credit expenses: the longer the credit period, the larger the imputed credit expenses, which are then deducted from NV. Petitioners argue the Department can draw an adverse inference from the fact that Kanzen could not explain its error in reporting, and that the error was frequent in its U.K. sales. Accordingly, petitioners assert the Department should apply adverse facts available and reduce Kanzen's reported credit expenses for all sales with the March 6, 1999 payment date. See Petitioners' Brief at 23-24. Kanzen argues that petitioners have engaged in unnecessary speculation and extrapolation with regard to payment date. While Kanzen agrees that payment date was misreported for certain observations, it maintains this error does not warrant a revision of all payment dates. Specifically, Kanzen notes that, at the sales verification, the Department in fact reviewed the payment dates for all sales made to the U.K. during the POI and found no other discrepancies in date; therefore, no additional adjustment is necessary. See Kanzen's Rebuttal at 12. Department Position: We agree with respondent. We reviewed all of Kanzen's U.K. invoices at the sales verification and, in the process, found no other such error in reporting payment date. Because the payment date has been fully verified, and because the discrepancy is relatively minor, application of adverse facts available is not warranted. Accordingly, we are adjusting our programming to correct the misreported date of May 6, 1999 and replacing the date to use March 22, 1999 as payment date for the appropriate sales observations. Comment 6: FOB v. CIF Kanzen submitted a correction to one of its U.K. invoices at the start of the sales verification, stating that although it had not reported freight payment for that particular sale, it had changed the sale term from FOB to CIF after the contract date. See Sales Verification Report at 2-3. Petitioners note at the outset that the Department would not accept information at verification regarding the precise amount paid for movement as this constituted significant new information. See Sales Verification Report at 3. More importantly, petitioners argue there was no actual change in the sale term. Rather, petitioners assert that the "freight payment" on the U.K. invoice was actually a payment agreed upon by the parties to compensate for a previous shipment, wherein two containers of non-subject merchandise were underutilized. In support, petitioners note that despite Kanzen allegedly paying freight on the invoice, the sale term on the invoice remained FOB. Petitioners also note that such a change in the sales term is inconsistent with Kanzen's stated practice of rarely changing terms of sale after order confirmation. Petitioners argue that the clearest indication that both Kanzen and the customer considered the sale term as FOB is the fact that the customer later requested a post-sale price adjustment to compensate for an unexpected increase in freight costs, a cost which was borne by the U.K. customer. With the record evidence, petitioners argue that it is obvious that Kanzen's "freight payment" was actually compensation on a previous shipment, and not payment on the invoice Kanzen offered for correction purposes. Accordingly, petitioners argue that the Department should reject Kanzen's change in sale term. See Petitioners' Brief at 25-26. Kanzen did not submit comment on this issue in its rebuttal brief. Department Position: We disagree with petitioners. Regardless of how petitioners characterize the payment on the invoice, be it freight payment or compensatory payment, it is clear that the payment necessitated a change in a sale term on the invoice, which effectively lowered the selling price for the merchandise at issue. The terms of sale were originally FOB and were later changed to CIF after the contract date. This further supports our basis for using invoice date as the date of sale in the comment section on "Market Viability," supra, and our determination that the material terms of sale for the U.K. market during the POI were not necessarily established at confirmation/contract date, but rather remained alterable by the parties after that time. While Kanzen did not physically change the terms of sale on the invoice, verification of its records tied the payment to the invoice, confirming that there was a change in the terms of sale for the invoice. Accordingly, we are making no changes or adjustments to our programming regarding Kanzen's change in the terms of sale. Comment 7: Early Payment Discount Kanzen submitted a correction on one of its U.K. invoices at the start of the sales verification, stating that while it reported no discounts for the POI, it requested early payment on that particular sale because the customer's account was overdrawn and gave the customer an early payment discount. See Sales Verification Report at 2-3. Petitioners argue that because Kanzen maintains customer-specific accounts and has no stated early payment policy for sales of subject merchandise, the Department should not treat the discount as solely associated with sales of subject merchandise under the identified invoice. Petitioners argue that the Department should instead treat the discount as an indirect selling expense and allocate it to Kanzen's total sales of subject and non-subject merchandise. However, petitioners contend that the record is not sufficiently developed on the total early discounts offered to the U.K. customer during the POI due to the customer's overdrawn account. Furthermore, petitioners argue, the record contains no information on the total POI sales to the same customer, so calculating the early payment discount applicable to Kanzen's POI sales of subject merchandise is not possible. Accordingly, petitioners argue that the Department should not accept Kanzen's correction to the invoice concerning the early payment discount. See Petitioners' Brief at 27-28. Kanzen argues the Department should accept and use the early payment discount information Kanzen reported for its U.K. sales, noting that the Department accepted this correction offered at sales verification, in accordance with the Department's established practice of accepting minor clerical corrections at the beginning of verification. Furthermore, Kanzen argues that this discount was granted on and associated with a specific invoice for a U.K. sale during the POI, and was calculated based on the amount and days outstanding for that sale; hence, Kanzen claims the discount is a price adjustment to be reflected in the Department's calculation of normal value, and is not an indirect selling expense. See Kanzen's Rebuttal at 13. Department Position: We agree with respondent. Contrary to petitioners' contention that the record is not sufficiently developed on the total early discounts offered to the U.K. customer during the POI due to the customer's overdrawn account, we noted at the sales verification that "Kanzen stated that this situation only arose once in the POI for subject merchandise and only in relation to that customer." See Sales Verification Report at 2. The record is clearly developed from the sales verification as to establish that the early payment discount was applied only to a specific invoice for a U.K. sale, and was specifically calculated according to the details and circumstances of that sale. Furthermore, we reviewed at the sales verification Kanzen's accounts "to confirm that Kanzen incurred no expenses pertaining to . . . discounts . . . (other than those already reported) on U.S. and U.K. sales of subject merchandise during the POI" and found no other early payment discounts during the POI. See Sales Verification Report at 24. We are properly treating the discount as a post- sale price adjustment (see 19 C.F.R. 351.401(c)) addressed in our normal value calculation. See Final Analysis Memo. 4. U.S. Sales Issues Comment 8: Marine Insurance Expense Petitioners argue that while Kanzen stated that it pays freight and insurance on its U.S. sales because they are on a CIF basis, petitioners observe that Kanzen fails to report freight and insurance amounts for a number of its U.S. sales. Petitioners argue that for those sales, the Department should use adverse facts available, and apply the highest amount for marine insurance reported by Kanzen for products with identical product characteristics as those U.S. sales. Petitioners argue such an application is reasonable because marine insurance as other types of insurance, is calculated based on the value of the goods insured, and identical products tend to have comparable values. See Petitioners' Brief at 28-29. Kanzen argues that the Department should not deduct marine insurance expenses where none were incurred, noting that, for a number of the contested sales, Kanzen properly reported the sales as CNF (cost and freight) sales for which no marine insurance costs were incurred. Kanzen also points to the fact that the Department noted no discrepancies with regard to marine insurance in its sales verification. As such, Kanzen argues that the Department should continue to apply no marine insurance costs for those CNF sales. See Kanzen's Rebuttal at 14-15. Department Position: We agree with respondent, in part. For those sales reported as CNF sales, Kanzen incurred no marine insurance costs and properly reported no marine insurance costs. The Department, therefore, will not apply marine insurance costs for those sales. However, not all sales without marine insurance expenses were CNF sales. For these remaining sales, which were reported as CIF (cost, insurance and freight) sales, Kanzen should have reported a marine insurance amount but did not. In reviewing Kanzen's database, it appears that some of these omissions were inadvertent, as Kanzen did in fact report marine insurance for other sales with the same invoice number. Due to these omissions, the Department must resort to the use of facts available, pursuant to section 776(a) of the Act, as Kanzen failed to provide the requested information which would permit the Department to calculate the marine insurance amount Kanzen should have reported. However, in this instance, because we find the omissions to be inadvertent, we determine that an adverse inference is not warranted. Here, there is other information on the record which was verified (see section 782(e) of the Act), that can be used by the Department to make these calculations. By reviewing other sales in the U.S. database with the same invoice number, we divided marine insurance by gross unit price to determine the applicable marine insurance rate for the invoice. We then multiplied the gross unit price on those CIF sales with unreported marine insurance by the applicable marine insurance rate to determine the unreported marine insurance on the sale. For the remaining CIF sales, where no marine insurance was reported for any of the sales on the invoice, there are no sales on the invoice from which the Department can determine an applicable percentage of marine insurance for the invoice. Therefore, we are instead applying the weighted average per unit marine insurance from Kanzen's U.S. sales as facts available, for calculating marine insurance on those sales. Comment 9: Marine Insurance Expense Discount and Denomination Kanzen submitted a correction to its marine insurance at the start of the sales verification, stating that while it reported marine insurance on U.S. sales on an average basis, it gave discounts on a number of invoices so the customer could purchase marine insurance itself. See Sales Verification Report at 3. Petitioners note that Kanzen's reported discount and total invoice value were denominated in U.S. dollars, and suggest that applying the discount ratio to related U.S. sales necessarily results in a marine insurance amount denominated in U.S. dollars. Petitioners note that Kanzen's initial questionnaire response stated that marine insurance is denominated in U.S. dollars (and petitioners imply the Department erred in taking the attached computer file layout, which reported the denomination as Malaysian ringgits, at face value). Petitioners argue that if the Department accepts Kanzen's changes, denominated in U.S. dollars, to its reported discount ratio for marine insurance, the Department must recognize that Kanzen's reported marine insurance expenses are denominated in U.S. dollars. Accordingly, petitioners maintain that the Department should revise its programming to reverse its currency conversion made in the Preliminary Determination and correct the currency denomination for the final margin analysis. See Petitioners' Brief at 29-30. Kanzen did not submit comment on this issue in its rebuttal brief. Department Position: We agree with petitioners. In order to properly calculate the margin, our programming must reflect the proper denominations in the calculations. Accordingly, in addition to making corrections to those U.S. sales invoices on which Kanzen provided discounts so the customer could purchase marine insurance itself, we have edited the margin calculation program to correctly reflect the reported denomination of marine insurance in U.S. dollars. Comment 10: Returns Kanzen reported no warranty expenses or costs for returned or under- shipped merchandise and, at the sales verification, reported one post-POI return on a pre-POI sale. However, petitioners argue that Kanzen had returns during the POI, pointing out that returns associated with POI sales could occur long after the POI, and pointing to verification exhibits which indicate U.S. sales returns during the POI. In further support, petitioners repeat Kanzen's explanation that its reported negative production quantities were due to returns of items that it did not produce during the POI. Additionally, petitioners question Kanzen's inability to provide the original U.S. sales order for which Kanzen eventually shipped replacement pieces. At the sales verification, Kanzen stated it could not trace the documentation back to the original order, but did provide correspondence on the request for replacement. See Sales Verification Report at 22. Petitioners also argue that because a customer's warranty claim must occur after receipt of the merchandise, whether the credit notes applied to sales prior to the POI does not mean that Kanzen did not incur any warranty expenses for its POI sales. Petitioners assert that Kanzen had to incur additional costs for returned and replaced goods, but such costs do not appear in Kanzen's U.S. sales database. Accordingly, petitioners contend that Kanzen did not completely report its U.S. expenses, and requests that the Department calculate an estimated warranty expense to account for the costs incurred for the returned and replaced merchandise. Petitioners suggest using the percentage of returns from the sales verification to calculate a warranty factor, or that the Department calculate a monthly warranty factor for each of the months during the POI in which Kanzen's U.S. customers made returns. See Petitioners' Brief at 31-33. Kanzen argues that it accurately reported warranty expenses associated with POI sales to the U.S. and the U.K. and, therefore, any additional deduction of warranty expenses based on data from other periods would result in double-counting. Kanzen declares that it is customary in the butt-weld pipe fittings industry for warranty-type claims to be resolved through credit notes and returns. Kanzen contends it was able to identify such credit notes and returns related to its POI sales to the U.S. and the U.K., and therefore reported them on a sales-specific basis in its databases. Kanzen notes that, in such circumstances, the Department will accept such a reporting methodology, citing Final Determination of Sales at Less Than Fair Value: Stainless Steel Sheet and Strip in Coils From Taiwan, 64 FR 30592, 30611 (June 8, 1999) ("SSSS From Taiwan Final"). As Kanzen's reporting is consistent with Department's methodology, Kanzen asks that the Department reject petitioners' request to use an allocation ratio. See Kanzen's Rebuttal at 15-16. Department Position: We agree with petitioners. Record evidence shows that warranty returns were made during the POI, which indicates that certain expenses were incurred. In this case, recognizing only warranty expenses incurred in the POI on sales in the POI would mean that certain warranty expenses would never be accounted for. Therefore, we must consider all expenses incurred in the POI. Because Kanzen did not report any warranty expenses, we must use the facts available for these expenses. Accordingly, as facts available, we multiplied the average charge for the transportation variables (international freight and marine insurance) to the quantity of returns identified at the sales verification (see Sales Verification Exhibit 6e), then divided the total by the total U.S. sales quantity, to calculate a warranty factor to apply to Kanzen's U.S. sales for the POI. Comment 11: Miscellaneous Bank Charges Petitioners take issue with Kanzen's unreported miscellaneous bank charges found under a separate account number. Kanzen stated at the sales verification that it did not include a number of these charges in its direct selling expenses or G&A. See Sales Verification Report at 24. Petitioners argue the Department must draw an adverse inference from Kanzen's failure to report these charges, and adjust Kanzen's reported unit price for U.S. sales and revise the calculations for U.S. total direct selling expense to reflect this omission. In so doing, petitioners argue it is reasonable to assume that all expenses recorded in the account were for U.S. sales, given the volume of Kanzen's U.S. sales compared to its total sales of subject merchandise during the POI, and given the extra arrangements that Kanzen took for its U.S. sales. Such "extra" arrangements include the fact that Kanzen's U.K. sales terms were FOB whereas its U.S. sales terms were CIF, and that Kanzen uses freight forwarders to move the merchandise from the U.S. ports to its U.S. customers, which petitioners suggest resulted in the "telex charges" recorded in the account. See Petitioners' Brief at 33-34. Kanzen argues that petitioners' proposal for allocating the miscellaneous bank charges to inflate Kanzen's direct selling expenses is inaccurate. Kanzen notes that the Department collected information on this account for the entire POI; the Department confirmed that the expenses in this account are associated with export sales; and the total amount of these charges for the POI is extremely small in relation to the total export sales. Notwithstanding this, however, Kanzen asserts that if the Department wants to account for these costs, it can simply allocate the total charges for the POI over the total export sales during the POI. See Kanzen's Rebuttal at 16. Department Position: We agree with respondent. Contrary to petitioners' argument, it is not reasonable to simply assume that all the unreported miscellaneous expenses in the account were for U.S. sales, merely on the basis of a greater volume of sales in the U.S. Kanzen has many markets for its merchandise; even with any purportedly "extra" arrangements for Kanzen's U.S. sales, this is not sufficient justification to assume that these unreported bank charges must apply solely to U.S. sales rather than to any other sales. We also note that the overall amount of bank charges incurred during the POI accounts for an extremely minor percentage of the sales value. See Sales Verification Report at Exhibit 31. Accordingly, using the information we have available, the Department is calculating the percentage, by value, of market share for Kanzen's U.S. and U.K. sales for the POI, and allocating the total value of the unreported miscellaneous bank charges to the U.S. and U.K. markets according to that percentage. See Final Analysis Memo. Comment 12: Unreported U.S. Sales During the sales verification, the Department requested a printout of sales after the POI and discovered that contract date was misreported for one of the U.S. sales, which actually fell within the POI. See Sales Verification Report at 15. Petitioners point out that if the Department maintains contract date as the date of sale for the U.S. market, we must include all U.S. sales with a contract date falling within the POI in the final margin analysis. Petitioners argue that because Kanzen did not report this sale in its U.S. sales listing, the Department should apply the highest calculated individual margin for Kanzen's reported U.S. sales to the unreported U.S. sale as partial adverse facts available. See Petitioners' Brief at 34-35. Kanzen argues that the Department should not apply adverse facts available for a single unreported U.S. invoice, contending that Kanzen inadvertently omitted this sale from its U.S. database due to a computer programming error in the collection of the sales information, and asserting that the Department has complete information for the two items contained in this invoice to perform a margin calculation on these sales. See Kanzen's Rebuttal at 14. Department Position: We agree with petitioners. The Department requested in Section C of the antidumping duty questionnaire that Kanzen report all its U.S. sales made during the POI. The unreported sale which we discovered at the sales verification involved two unreported items of an invoice, yet the other four items on the same invoice were reported by Kanzen. In our questionnaire, we state that "{f}or sales of merchandise that has been shipped to the customer and invoiced by the time this response is prepared, each 'record' in the computer data file should correspond to an invoice line item (i.e., each unique product included on the invoice). See Antidumping Duty Questionnaire, Section C (Sales to the United States), at C-1. Kanzen's reporting of other line items on the invoice indicates that it had the opportunity to discover its omission. Likewise, Kanzen submitted four supplemental responses, and three subsequent databases, pertaining to its U.S. sales, all opportunities in which Kanzen could have discovered its omission. Moreover, the Department's questionnaire requires, prior to verification, that respondent "provide a complete package of documents and worksheets demonstrating how you identified the sales you reported to the Department and reconciling the reported sales to the total sales listed in your general ledger." See Antidumping Duty Questionnaire, Section A (Organization, Accounting Practices, Markets and Merchandise), at A-3. However, despite being required to submit documentation which reconciled its financial statements with its reported Section C database, Kanzen failed to identify the transactions in question. In response, Kanzen provided sales reconciliation worksheets that theoretically reconciled, yet did not identify the omission. Due to this omission, the Department must resort to the use of facts available, pursuant to section 776(b) of the Act, as Kanzen failed to provide the requested information for the Department in the form of an accurate U.S. sales database. In this instance, because the omission was not discovered until the sales verification, the Department had no opportunity to note this deficiency and provide Kanzen with an opportunity to correct it. See section 782(d) of the Act. As discussed, supra, Kanzen was well aware of the Department's reporting requirements regarding all U.S. sales, and had the opportunity to discover the omission in compiling its sales reconciliation worksheets. Given its understanding of the Department's reporting requirements for U.S. sales, its numerous opportunities to submit supplemental information, and the fact that the Department was only able to discover this deficiency at the sales verification, we find, pursuant to section 776(b) of the Act, that Kanzen failed to cooperate by not acting to the best of its ability to comply with a request for information from the Department. Therefore, we are applying the highest calculated unit margin to this unreported U.S. sale. Comment 13: Unshipped Sale Kanzen previously explained that for sales which had not been shipped as of the date of the submitted database, it had reported the shipment quantity as zero, but revised the data with each successive database. See Kanzen's August 15, 2000 Supplemental Questionnaire Response at 2. Petitioners argue the Department should include an unshipped sale made during the POI, for which Kanzen reported a zero shipment quantity, in the final margin analysis. Petitioners state that it is Department practice to include unshipped merchandise in its margin analysis, as long as the sale is made within the POI, based on the chosen date of sale methodology. Petitioners argue that because Kanzen gave no indication whether the unshipped sale was cancelled after the purchase order/contract date, it is thus appropriate to include the sale in the final margin analysis. See Petitioners' Brief at 35. Kanzen did not submit comment on this issue in its rebuttal brief. Department Position: We agree with petitioners. As mentioned in our "Date of Sale" section, supra, we are retaining order confirmation/contract date as date of sale for U.S. sales. As such, the date of sale for the unshipped sale falls within the POI. Accordingly, we are using the reported sales quantity and including the sale, reported by Kanzen as unshipped, in our final margin analysis. Because Kanzen has not reported actual numbers for DINLFTPU, INTNFRU, MARNINU, CREDITU, DIRSELU1, and DIRSELU2, we are applying facts available for these variables, using the weighted average amounts for other sales of that CONNUM, in our calculation. Comment 14: Inland Freight During the Department's verification of the per unit inland freight on a pre-selected U.S. sales observation, the Department calculated a slightly different freight forwarder charge than reported by Kanzen, but noted the difference was negligible. See Sales Verification Report at 21. Petitioners disagree. While the calculated unit expenses for DINLFTPU and MARNINU were not affected, petitioners believe the overall quantity of sales were affected, noting that other sales fell under the same invoice number as the observation examined by the Department, many of which expenses could be affected by the difference in forwarder charge. Accordingly, petitioners assert that the Department should adjust Kanzen's reported DINLFTPU and MARNINU for the pertinent invoice in order to completely capture the expenses incurred by Kanzen for its U.S. sales. See Petitioners' Brief at 35-36. Kanzen argues that petitioners' request for adjusting all of Kanzen's Malaysian inland freight and marine insurance expenses is based on a single calculation error on a single sales trace. Kanzen argues the error made no difference between the reported and corrected value, and noted this was the only sale with discrepancies, out of a total of ten sales traces reviewed by the Department. Kanzen asserts that a single error of this size does not justify a complete revision. See Kanzen's Rebuttal at 14-15. Department Position: We agree with respondent. This was the only sale out of the ten sales we reviewed at verification where we found any discrepancy regarding inland freight, and that discrepancy amounted to such a slight difference in the reported forwarder charge that it made a negligible difference in Kanzen's reported per unit inland freight. See Sales Verification Report at 21. Accordingly, we find no reason to make the extensive revisions to Kanzen's reported Malaysian inland freight and marine insurance expense requested by petitioners. Therefore, we are making no changes in relation to this issue. 5. Cost Issues Comment 15: Total Adverse Facts Available Petitioners assert that the Department, in its final margin calculations, cannot rely on Kanzen's submitted cost data and should, instead, resort to total facts otherwise available due to the following problems: 1) Kanzen, while claiming that it attempted to allocate its manufacturing variances on a more product-specific basis than Kanzen maintains in the normal course of business, in fact, tried to redistribute its production costs away from the merchandise under consideration (see Petitioners' Brief at 37-38); 2) petitioners dispute the amount the standard cost of finished pipe is reduced by in order to obtain the standard cost of semi-finished pipes used for butt-weld fittings, and question whether the cost of fittings made of purchased pipes should include any allocation of the cost variances and scrap offsets related to self produced pipe (see Petitioners' Brief at 38-40); 3) due to the late disclosure of the two additional affiliated companies it is impossible to estimate the degree of their involvement and the arm's length nature of the services provided by these affiliates (see Petitioners' Brief at 40-41); 4) because certain CONNUMs were reported with negative production quantities the total reported costs allocated to individual products are understated (see Petitioners' Brief a 42-43); and 5) Kanzen grossly understated its calculated G&A expenses by including reversals of prior period provisions (see Petitioners' Brief at 43-44). Petitioners claim that these items demonstrate that Kanzen failed to cooperate with the administering authority to the best of its ability by failing to provide accurate, credible, and verifiable data in a timely manner. Petitioners argue that the Department offered Kanzen numerous opportunities to submit accurate and verifiable data in a timely manner. In addition, petitioners argue that Kanzen first delayed any revisions to its reported cost data before the Department's preliminary determination in order to secure a de minimis preliminary margin, and then submitted two sets of revised COP and CV data which continued to include glaring flaws. See Petitioners' Brief at 36-37 and 44-46. Kanzen contends that petitioners' arguments for the application of adverse facts available are inconsistent with the Department's findings at verification. Therefore, Kanzen argues that the Department should rely on Kanzen's submitted data for the final determination. Kanzen maintains that it was fully cooperative throughout this investigation and has made every effort to provide information requested on a timely basis. Kanzen points out that the variance allocation methodology in question, which it does not use in the normal course of business, was developed in an attempt to calculate more accurate costs. According to Kanzen, its normal accounting practice does not apply the variance back to the products, but rather applies total variances to the total standard costs of each section to determine the profitability of each section. Kanzen claims that these total variances were verified by the Department and can be easily applied to Kanzen's reported costs. See Kanzen's Rebuttal at 17-19. Kanzen contends that petitioners' argument to apply adverse facts available is contrary to law and the Department's practice. Kanzen argues that the Department's practice is not to apply adverse facts available to a respondent that is otherwise cooperative. Kanzen refers to the Notice of Final Determination of Sales at Less Than Fair Value: Certain Cut-to Length Carbon Quality Steel Plate Products from Indonesia, 64 FR 73164, 73167 (December 29, 1999) ("because the company was otherwise cooperative, the Department has not drawn the most adverse reference"), as well as other cases as having similar circumstances and outcomes: Mannesmannrohren- Werke AG, et. al. v. United States, 77 F. Supp. 2d 1302, 1315 (CIT 1999), Notice of Final Determination of Sales at Less Than Fair Value: Steel Wire Rod from Germany, 63 FR 8953, 8955 (February 23, 1998), and Krupp Stahl AG v. United States, 822 F. Supp. 789, 793 (Ct. Int'l Trade 1993). See Kanzen's Rebuttal at 18. Addressing the purchased and self-produced finished pipe issues, Kanzen claims that it appropriately incorporated all of the butt-weld material costs into the per-unit cost data submitted to the Department. Kanzen argues that petitioners selectively examined only one month of the POI and, therefore, their estimated amount by which to reduce standard pipe costs are not reflective of the entire POI. See Kanzen's Rebuttal at 19-20. As to the issue of affiliated suppliers identified at the cost verification, Kanzen argues that the Department easily calculated the degree of each one's involvement and relationship to Kanzen during the verification. Kanzen argues that the charges paid to the two affiliates in question were insignificant to Kanzen's total costs and were reported consistent with the Department's practice of including in costs charges paid to affiliated parties for outside services, and cites Rubber Thread From Malaysia; Final Results of Antidumping Duty Administrative Review, 64 FR 12967, 12975 (March 16, 1999) as having similar circumstances and outcome. See Kanzen's Rebuttal at 20-21. Kanzen argues that petitioners' allegations regarding negative production quantities are based on a misunderstanding and should be rejected. Kanzen maintains that it followed its normal accounting system to report production quantities in its submitted cost data. Kanzen argues that it has consistently explained that the negative production quantities are the result of internal transfers and reclassification of merchandise from finished goods inventory to WIP inventory for reworking (i.e., negative production quantity occurs when total rework quantity exceeds total actual production quantity). See Kanzen's Rebuttal at 21-22. Finally, Kanzen points out that the Department verified Kanzen's reported G&A calculation, and, according to the company, if the Department decides not to include in Kanzen's G&A calculations the reversals of prior period provisions, it does not affect the accuracy and credibility of the other G&A expenses reported by Kanzen and verified by the Department. See Kanzen's Rebuttal at 22. Department Position: We agree with respondent. While some of the issues listed by petitioners represent valid concerns, and will require certain adjustments in order to properly reflect Kanzen's cost of production during the POI, we disagree with petitioners that the application of total adverse facts available in this case is warranted. For a discussion on each of the issues raised by petitioners see each individual issue discussed below. We agree with Kanzen that the minor services provided by the two affiliates were reported properly at the transfer price. Finally, we agree with Kanzen that the insignificant negative production quantities referred to by petitioners were recorded in accordance with Kanzen's normal records, verified by the Department, and found not to distort the reported costs. Comment 16: Allocation of Cost Variances Petitioners point out that Kanzen revised its variance allocation calculations several times in the course of the investigation. They contend that, in each of these revisions, Kanzen attempted to shift costs away from the merchandise under consideration. According to petitioners, the Department should not accept such material changes, which constitute new factual information after the deadline for such information. See Petitioners' Brief at 46-49. In addition, petitioners argue that Kanzen's variance allocation methodology contains other serious flaws. In particular, they argue that it is inaccurate to allocate some of the pipe making cost variances to all butt-weld fittings, because some of those fittings were produced from pipe purchased from outside vendors. Petitioners also attempt to rebut Kanzen's statement that the transfers to the fitting section figure used as the numerator in the calculation of this allocation does not include the cost of the post-sizing processes, thus understating the allocation. They refer to Kanzen's statements showing that all the pipes are transferred to the fitting section at the standard cost of finished pipes which includes the cost of the post-sizing processes. Furthermore, petitioners contend that the pipe making conversion cost variance, as well as piping unrealized stock revaluations, should be allocated to butt-weld pipe fittings along with the raw material portion of the total pipe making cost variance. See Petitioners' Brief at 49-50. Finally, petitioners claim that in its attempts to assign the variances on a more product-specific basis, Kanzen departed from its normal books and records, and that this allocation methodology has never been fully disclosed in its responses. This, according to petitioners, constitutes Kanzen's failure to provide credible, accurate and verifiable information and failure to cooperate to the best of its ability. Therefore, petitioners contend that, for its final determination, the Department should apply partial adverse facts available. Petitioners cite Static Random Access Memory Semiconductors From Taiwan; Final Results and Partial Rescission of Antidumping Duty Administrative Review, 65 FR 55005 (September 12, 2000) and the accompanying Issues and Decision Memorandum at Comment 1 ("SRAMs" from Taiwan"), as having similar circumstances and outcome. See Petitioners' Brief at 50-53. Kanzen argues that the Department did verify Kanzen's total actual costs, standard costs, and the total variances of each production section, and found no discrepancies. Therefore, according to Kanzen, if the Department decides in the final determination not to use the detailed breakdowns of variance ratios, this does not affect the credibility of calculating variances on a production section basis. Kanzen maintains that these production section specific variances are the most credible and reliable evidence on record available to the Department. See Kanzen's Rebuttal at 23-24. Department Position: Section 773(f)(1)(A) of the Act states that costs shall normally be calculated based on the records of the exporter or producer of the merchandise, if those records are kept in accordance with generally accepted accounting principles of the exporting country and reasonably reflect the costs associated with the production of the merchandise under consideration. Section 773(f)(1)(A) further directs the Department to rely on cost allocations when they have historically been used by the producer. The Department found that in the normal course of business Kanzen only calculates total variances for each production section and applies them at year-end to the corresponding total standard costs in order to monitor the profitability of each section. However, for the reported costs, Kanzen attempted to allocate the general ledger total variances on a more specific basis (e.g., material cost by grade, conversion cost, etc.) using the information which, according to Kanzen, is maintained in its cost accounting system. The effect of the more detailed computation was to reduce the costs assigned to the merchandise under consideration. Even though it made several attempts, Kanzen was unable to produce reliable calculations allocating the total variances. Kanzen revised several times its variance allocation calculations, and submitted four different versions of the same schedule, each time claiming it found new mistakes and inaccuracies in the breakdown of the total variances. Furthermore, Kanzen's fourth attempt to submit a new version of its variance allocation calculations was made late at the verification. Because this revision was submitted well after the deadline for submitting new factual information, we declined to accept the new revision, pursuant to section 351.301(b)(1) of the Department's regulations. As a result, we did not verify Kanzen's computation of variances on a more detailed basis and, thus, only the broader production section variances were verified. We agree with respondent that the Department's decision not to accept Kanzen's product-specific variance allocation calculations does not affect the credibility of production section total variances that Kanzen calculates in the normal course of business. Furthermore, we disagree with petitioners that SRAMs from Taiwan is applicable to this case. In SRAMs from Taiwan, the respondent abandoned its normal cost accounting system and created a completely new system for reporting purposes. Kanzen reported its costs based on the standard costs normally used in its cost accounting system, but departed from that system only for the purpose of attempting to calculate a variance that was more specific to the merchandise under consideration as requested by the Department. We agree with petitioners that there is an inconsistency between the calculations related to the allocation of piping section variances to the fitting section costs. Specifically, Kanzen used the actual costs for the piping section as the denominator, but standard costs for internal transfers from the piping to the fitting section as the numerator. We also found that, consistent with Kanzen's description of its cost accounting system, the transfers to the fitting section figure used as the numerator in the calculation of this ratio does include the cost of the post-sizing processes. We also agree with petitioners that the piping conversion cost variance as well as the piping unrealized stock revaluations should be allocated to fittings along with the raw material portion of the total piping variance. However, we disagree with petitioners that the information on the record does not allow the Department to measure and verify the exact impact of these items on the reported costs. Consistent with sections 776(a) and 782(e) of the Act, there is enough verified information on the record that can be used as facts available to correct each of the inaccuracies. In this instance, we calculated two variance allocation ratios: one for fittings made of purchased pipes that is based only on the total fitting variance, and another one for all other fittings which is based on the total fitting variance plus an allocated portion of piping variance. As facts available, we used in our calculations the verified total variances for each production section as determined by Kanzen in the normal course of business. We incorporated in this calculation the corrections to each of the inaccuracies discussed above. Comment 17: Standard Cost Reduction Factor for Pipes Used for Fittings Petitioners claim that the extent of the reduction Kanzen applies to the standard costs of finished pipes to bring them down to the cost of semi- finished pipes used for fittings is not accurate. Petitioners argue that according to their calculations the average percentage of post-sizing processing costs in the per-unit standard cost of finished pipes is different from the one applied by Kanzen in the normal course of business, and, therefore, should be adjusted. See Petitioners' Brief at 53-54. Kanzen claims that the methodology used by petitioners is not consistent with Kanzen's normal accounting practices. Kanzen maintains that its normal accounting practice provides an accounting treatment that reasonably reflects the cost difference between different pipes going through different production processes, which, according to Kanzen, was verified by the Department. Therefore, Kanzen argues that, in the final determination, the Department should use the reduction factor that is used in Kanzen's normal accounting practices. See Kanzen's Rebuttal at 24-25. Department Position: We agree with respondent. The methodology petitioners used to test the accuracy of the reduction ratio applied by Kanzen to the standard costs of finished pipes to bring them down to the cost of semi-finished pipes used for fittings is not consistent with Kanzen's normal accounting practices. Petitioners' calculations are based on the standard cost build-ups for a limited sample of pipes that was not intended to and does not represent the entire population of the pipes produced by Kanzen. Furthermore, it includes pipes that Kanzen does not use for fittings because fitting production specifications require only pipes of certain standard length. In addition, the Department found that this reduction ratio was the result of an estimate on the part of Kanzen intended only to approximate the percentage of post-sizing costs in the total cost of pipes, and the accuracy of this estimate was tested by the Department verifiers to their satisfaction. Moreover, conceptually, under a standard cost system, any adjustment to standard costs has to be followed by an offsetting adjustment to the variances, which eliminates the effect of such an adjustment on total costs. This kind of adjustment basically reallocates the costs which may affect product-specific costs. However, if the adjustment is small (which is the case in this instance), the effect on product-specific costs will be negligible. Therefore, in our final determination, we did not adjust the reported reduction factor historically used by Kanzen in its standard cost calculations in the normal course of business. Comment 18: Cost of Fittings Made of Finished Pipes Pipe intended for production of the butt-weld fittings pass through fewer finishing processes than pipe produced for sale. To calculate the standard cost of pipe entering the butt-weld fittings section, Kanzen reduced the standard cost of finished pipe (i.e., the standard cost of pipe sold to outside parties) by an estimated ratio representing the finishing processes which these pipes did not receive. However, petitioners argue that the reduction ratio Kanzen applied to the standard cost of finished pipes does not address the situation when internally produced finished pipes, rather than semi-finished pipes, are used to produce fittings. Petitioners contend that the Department should adjust Kanzen's reported standard cost of materials to correct this deduction. See Petitioners' Brief at 54-55. Petitioners also raise issues related to the cost of fittings made of purchased pipes. Petitioners assert that according to their calculations, Kanzen's statement that it does not apply the reduction ratio to the standard cost build-ups of large diameter fittings because they are made of purchased (i.e., finished) pipes appears to be untrue. Accordingly, petitioners suggest the reduction should be reversed (see Petitioners' Brief at 55-57). Petitioners also contend that no piping section variances and scrap revenue offset should be applied to the cost of the fittings made of purchased pipes (see Petitioners' Brief at 57). They point out that, since these pipes were not made by Kanzen in its production facility, no variances and no scrap generated by Kanzen's piping section could be attributed to the fittings made of these purchased pipes (see Petitioners' Brief at 57). Kanzen contends that, in the final determination, the Department should not adjust the costs of fittings made from finished pipes. However, if the Department were to adjust the costs of fittings made from finished pipes, this adjustment, according to Kanzen, should not include the pipes purchased from outside suppliers because the standard cost of such purchased pipes is not reduced by the average post-sizing cost percentage. See Kanzen's Rebuttal at 25. Kanzen argues that a portion of the piping variance should be allocated to all fittings, including the ones made of purchased pipes, because, according to Kanzen, the allocation ratio calculated by Kanzen does not include the portion of pipes transferred from finished goods inventory to the fitting section. Therefore, Kanzen claims that it did not overstate the allocation ratio and the Department should apply this ratio to all fittings. See Kanzen's Rebuttal at 25. Kanzen claims that the Department should apply the full amount of scrap offset reflecting both piping and fitting production related scrap revenues to all fittings, including the ones made of purchased pipes. According to Kanzen, regardless of whether the scrap is generated from pipe manufactured by Kanzen or pipe purchased by Kanzen, the scrap revenue is recovered from the production process at Kanzen. See Kanzen's Rebuttal at 26. Department Position: We agree with petitioners in part that there are certain inaccuracies related to Kanzen's methodology that need to be corrected. However, we disagree with the adjustment to the standard cost of fittings made from internally produced finished pipes as proposed by petitioners. Even if we were to make this adjustment as petitioners suggested, we would then have to adjust the variance correspondingly, which would have made the net amount of this adjustment negligible even at the CONNUM-specific level. We also disagree with petitioners' assertion that Kanzen does, in fact, apply the reduction ratio to the standard cost build-ups of the fittings made of purchased pipes. Petitioners based their assertion on the analysis of the standard cost build-up of a pipe that Kanzen does not use to produce fittings, and therefore, can not be traced to a fitting standard cost build-up. In addition, the Department's verifiers discussed this issue with Kanzen officials and did not note any discrepancies. See Cost Verification Exhibit at 15. However, we do agree with petitioners that none of the piping section variances should be allocated to the fittings made of purchased pipes. We disagree with Kanzen that simply because the ratio allocating the piping variance to fittings calculated by Kanzen is not overstated, the allocated portion of piping variance should be applied to all fittings. The ratio in question was calculated by dividing the total internal transfers of pipes to the fitting section by total production of pipes, i.e., neither the numerator nor denominator includes purchased pipes because, according to Kanzen, these pipes are never recorded in the piping section accounts but rather are entered as raw materials for the fitting section. Therefore, it would not be appropriate to allocate any portion of the piping variance to the fittings made of purchased pipes. We also agree with petitioners that no piping scrap revenue offset should be applied to the cost of fittings made of purchased pipes. We agree that because none of these pipes were made by Kanzen in its piping facility, no scrap generated by Kanzen's piping section could be attributed to the fittings made of these purchased pipes. We disagree with Kanzen's position which seems to suggest that there does not have to be a connection between the scrap revenues and the production of merchandise under consideration in order to claim a scrap offset. The scrap revenues reduce direct material cost of the merchandise under consideration and should be as product-specific as Kanzen's reporting system would allow. We have continued to allow an offset to purchased and internally produced pipe for fitting scrap revenue consistent with this approach. Comment 19: G&A Expense Ratio Petitioners contend that the Department should disallow the reversals of prior year provisions that Kanzen used to offset its G&A expenses, and instead add the actual amounts recognized by Kanzen during FY1999. Petitioners also argue that the Department should include the foreign exchange losses classified as "Other" by Kanzen as those losses were generated by the foreign exchange contracts related to purchases. Petitioners also claim that because the Department's normal practice is to include in G&A calculations foreign exchange gains and losses on accounts payable at the non-consolidated level, the Department should use these foreign exchange gains recorded at Kanzen's level instead of the parent company's level. In addition, petitioners argue that the Department should include waste treatment expenses that Kanzen incurred in FY1999. See Petitioners' Brief at 58-59. Kanzen did not comment on the specific adjustments to G&A expense ratio calculations suggested by petitioners. Department Position: We agree with petitioners on the items they claim require adjustment in order to properly calculate Kanzen's G&A expense ratio. We agree that current period expenses should be included in the reported costs, while reversals of prior period provisions do not relate to the current period. We also agree that foreign currency exchange gains and losses generated by accounts payable should be included in the reported costs as they relate to purchases of materials and other services. Therefore, we have made the appropriate adjustments for the final determination. RECOMMENDATION: Based on our analysis of the comments received, we recommend adopting all of the above changes and positions, and adjusting the model match and margin calculation programs accordingly. If accepted, we will publish the final results of the investigation and the final weighted-average dumping margin for the investigated firm in the Federal Register. AGREE___________ DISAGREE___________ __________________________________________ Troy H. Cribb Assistant Secretary for Import Administration __________________________________________ Date