65 FR 81830, December 27, 2000 A-475-828 Investigation Group III/Office 8: HMK Public document MEMORANDUM TO: Troy Cribb Assistant Secretary for Import Administration FROM: Joseph A. Spetrini Deputy Assistant Secretary Enforcement Group III SUBJECT: Issues and Decision Memorandum for the Final Determination in the Antidumping Duty Investigation of Stainless Steel Butt-Weld Pipe Fittings from Italy Summary We have analyzed the comments and rebuttals of interested parties in the antidumping duty investigation of stainless steel butt-weld pipe fittings ("pipe fittings") from Italy. As a result of our analysis, we have made changes, including correction of an inadvertent clerical error, in the margin calculation. 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 investigation for which we received comments and rebuttals by parties: 1. Cost of Production A. Combining Costs of the Affiliated Suppliers/Major Input Rule B. Facts Available C. Selling, General and Administrative Expenses D. Financial Expenses 2. Level of Trade 3. Usual Commercial Quantities and Ordinary Course of Trade 4. Circumstance-of-Sale Adjustment-Imputed Credit Expenses 5. U.S. Movement Expenses 6. Indirect Selling Expenses (ISE) 7. Ministerial Error 8. Critical Circumstances 9. Miscellaneous Issues A. Model Match B. Sample Sales and Sales to Affiliated Party C. Correction of Errors Found At Verification D. Use of Updated Cost Data Background On August 2, 2000, the Department of Commerce ("the Department") published the preliminary determination in the antidumping duty investigation of pipe fittings from Italy. See Notice of Preliminary Determination of Sales at Less Than Fair Value: Stainless Steel Butt-Weld Pipe Fittings from Italy, 65 FR 47388 (August 2, 2000) ("Preliminary Determination"). The period of investigation ("POI") is October 1, 1998, through September 30, 1999. We invited parties to comment on our preliminary determination. On November 7, 2000, we received case briefs from Coprosider S.p.A. ("Coprosider"), the sole respondent in this case, and from Alloy Piping Products, Inc., Flowline Division, Markovitz Enterprises, Inc., Gerlin, Inc., and Taylor Forge (hereinafter the "petitioners"). Both parties filed rebuttal briefs on November 15, 2000. A request by Coprosider for a public hearing was withdrawn. Scope of Investigation The merchandise covered by the scope of this investigation is certain stainless steel butt-weld pipe fittings under 14 inches in outside diameter, whether finished or unfinished. This merchandise encompasses all grades of stainless steel and "commodity" and "specialty" fittings. Specifically excluded are threaded, grooved, and bolted fittings, and fittings made from any material other than stainless steel. For a complete description of the scope see Notice of Final Determination of Sales at Less Than Fair Value: Stainless Steel Butt-Weld Pipe Fittings from Italy. Discussion of the Issues Cost of Production A. Combining the Costs of the Affiliated Suppliers Comment 1A: Coprosider argues that the Department should use the actual production costs of its wholly-owned affiliated supplier, IBF, and its partially-owned affiliated supplier, CFF, as reported, and not apply the "major input rule." Respondent states that it purchased the raw materials used in production and supplied the material to IBF and CFF to produce finished products and that neither IBF nor CFF sells inputs to Coprosider or finished pipe fittings to any party other than Coprosider. Coprosider argues that IBF and CFF provide the functions of manufacturing divisions and that Coprosider performs all marketing and selling functions. Petitioners do not comment on whether the major input rule should apply to the services provided by IBF or CFF. Department's Position: We disagree with respondent that the major input rule should not apply here and that IBF, CFF, and Coprosider should be treated as one reporting entity. The Department will treat affiliated companies as one reporting entity only if they satisfy the collapsing requirements enunciated in the Department's regulations at 19 CFR 351.401(f). In determining whether to collapse affiliated producers of the subject merchandise, the Department's regulations provide a two-prong test. According to 19 CFR 351.401(f)(1), the Department will treat two or more affiliated producers as a single entity where (1) those producers have production facilities for similar or identical products that would not require substantial retooling of either facility in order to restructure manufacturing priorities, and (2) the Department concludes that there is a significant potential for the manipulation of price or production. Section 771(28) of the Tariff Act of 1930 ("the Act") defines the term "producer" as the "producer of the subject merchandise." 19 CFR 351.401(h) further clarifies that the Department "will not consider a toller or subcontractor to be a manufacturer or producer where the toller or subcontractor does not acquire ownership, and does not control the relevant sale of, the subject merchandise or foreign like product." (Emphasis added.) In the instant case, our evaluation of the record information shows that Coprosider controls the relevant sale of the subject merchandise and dictates the production schedules followed by IBF and CFF. According to Coprosider's case brief at page 18, neither IBF nor CFF market or sell finished products to any party and both companies exists solely to produce the pipe fittings sold by Coprosider. Coprosider further explains that IBF and CFF provide the functions of manufacturing divisions, while Coprosider performs all marketing and selling functions. Furthermore, as discussed in Coprosider's July 3, 2000, section D response and September 7, 2000, supplemental section D response, Coprosider purchased the raw material and transferred it to CFF and IBF. As explained to the Department at verification, during the entire POI, Coprosider maintained ownership of all materials sent to CFF for further production, and for a portion of the POI, Coprosider maintained ownership of the materials sent to IBF for further production. Finally, at verification the Department found that neither CFF nor IBF have a sales department or sales staff. See Verification of the Cost of Production and Constructed Value Data Submitted by Coprosider S.p.A. (October 30, 2000) ("CVR"), at 7. Based upon these facts, we find that neither IBF nor CFF are "producers" of subject merchandise within the meaning of section 771(28) of the Act, and the Department's regulations. Rather, the facts indicate that both affiliates are "subcontractors" providing a tolling function, as defined by 19 CFR 351.401(h). Our determination is consistent with the Department's current policy on subcontracted or tolled operations. See, e.g., Notice of Final Determination of Sales at Less Than Fair Value: Static Random Access Memory Semiconductors from Taiwan, 63 FR 8909, 8918 (February 23, 1998) (the Department decided to exclude a foundry as a respondent because it did not control the production of wafers); Notice of Final Determination of Sales at Less Than Fair Value: Dynamic Random Access Memory Semiconductors of One Megabit and Above from Taiwan, 64 FR 56308, 56318 (October 19, 1999) (the Department determined that a company affiliated with the respondent was not a "producer," given that the terms of an agreement indicated that the company did not acquire ownership of the subject merchandise and did not control the sales). Accordingly, as affiliated subcontractors, the production services provided by IBF and CFF in the manufacture of the merchandise under consideration are subject to the major input rule and should be valued at the higher of transfer price, cost of production, or market price, in accordance with sections 773(f)(2) and (3) of the Act. B. Facts Available Comment 1B: Petitioners claim that Coprosider's reported costs are not usable and should be rejected by the Department for purposes of the final determination. Petitioners point out that, contrary to respondent's claim, Coprosider's reported costs were not reconciled with IBF's and CFF's financial statements; thus, Coprosider did not report its actual costs incurred during the POI. Petitioners contend that Coprosider's reported costs were based on IBF's and CFF's standard conversion costs and standard production minutes per-product. Petitioners point out that IBF's and CFF's average conversion cost was based on each company's entire production experience and each company's total labor hours, regardless of the specific product models and different production processes. Petitioners contend that this overall simple average methodology renders Coprosider's reported costs unusable, because they are not product-specific and do not reflect each company's actual cost of production ("COP"). Petitioners state that, given that Coprosider failed to provide appropriate information with respect to its price estimating model and transfer prices, the Department should find that Coprosider failed to cooperate to the best of its ability. Thus, the Department should use adverse facts available based on the highest antidumping margin for Italy as calculated in the petition, or at a minimum, apply the most adverse difference between Coprosider's reported costs and the actual acquisition costs obtained at verification. Coprosider argues that the COP amounts reported to the Department were IBF's and CFF's actual costs of production and were examined by the Department at verification. Coprosider points out that it does not maintain, in the normal course of business, a cost accounting system that specifically allocates the product costs, and that it had to develop an alternative method to allocate costs according to the Department's specifications. Coprosider states that, under this alternative method, the Department was able to reconcile IBF's and CFF's reported production costs to each of the affiliated company's respective financial statements. Coprosider asserts that the Department is required, to the extent possible, to rely upon verified information in the administrative record. Coprosider references section 782(e) of the Act, Krupp Thyssen Nirosta GMBH v. United States, Slip Op. 2000-89 at 18 (CIT July 31, 2000), and Ferro Union, Inc. v. United States, 74 F. Supp. 2d 1289, 1297 (1999), to emphasize that the Department should utilize the verified cost and sales data to calculate the dumping margin. Coprosider also refers to Antifriction Bearings from France, et.al., 64 FR 35590, 35596 (1999), and SKF, Inc. v. United States, Slip Op. 200-105 at 36 (CIT August 23, 2000), in arguing that the Department should use partial facts available to fill in gaps in the administrative record. Coprosider asserts that the statute permits the use of adverse facts available only in situations where the Department has affirmatively determined that an interested party has failed to cooperate by not acting to the best of its ability to comply with a request for information. Coprosider notes that there is no indication that it purposefully withheld information or impeded the Department's investigation. Coprosider contends that it offered to submit a revised cost listing that reported product-by- product raw material costs on an acquisition cost basis prior to the beginning of verification and the Department declined to accept the data for the reason that it constituted new information. Coprosider argues that the administrative record contains sufficient verified information, which would enable the Department to adjust reported production costs so as to reasonably reflect Coprosider's acquisition costs. Department's Position: We agree with petitioners, in part. As stated in comment 1.A. above, the major input rule applies to the production services provided by IBF and CFF. Pursuant to section 773(f)(2) of the Act, the Department may evaluate whether transactions between affiliated parties involving any element of value (i.e., major or minor inputs) are at prices that "fairly reflect the amount usually reflected in sales of merchandise . . . in the market under consideration . . . ." Pointing specifically to major inputs, section 773(f)(3) allows the Department to test transactions between affiliated parties for the sole purpose of examining whether major inputs are above the affiliated supplier's COP. In other words, if an understatement in the value of an input would have a significant impact on the reported cost of the subject merchandise, the Department must be satisfied that the transfer price or market price is above cost. In determining whether an input is considered "major," among other factors, the Department looks at both the percentage of the input obtained from affiliated suppliers (as opposed to unaffiliated suppliers) and the percentage the individual element represents in the product's cost of manufacture ("COM"). See Preamble to the Department's Regulations, 62 FR 27296, 27361 (May 19, 1997). In determining whether an input is considered "major" in the instant case, we considered the percentages for the production services provided by IBF and CFF. As Coprosider states in its July 3, 2000, section D response, IBF and CFF produced the merchandise under consideration from raw materials supplied by Coprosider. Based on our evaluation of the cost database, we find that these production services represent a significant portion of the product's COM. See Cost of Production and Constructed Value Calculation Adjustments for the Final Determination Memorandum to Neal Halper, dated December 15, 2000 ("Calculation Memorandum"). Given that the production services obtained from IBF and CFF are extensive, and that these services represent a significant portion in the product's COM, there is a risk of misstatement of the subject merchandise's costs to such a degree that we have determined that section 773(f)(3) of the Act (i.e., the major input rule) does apply. Once the Department determines that an input is "major," it then decides the value of major inputs purchased from affiliated parties based on the higher of: (1) the price paid by the producer to the affiliated party for the major input; (2) the amount usually reflected in sales of the major input in the market under consideration; or (3) the cost to the affiliated person of producing the major input. See 19 CFR 351.407(b). In the instant case, Coprosider did not purchase similar production services from any unaffiliated supplier, nor did IBF or CFF provide their production services to any other company. As a result, comparable market prices for these production services are not available. Therefore, in applying section 773(f)(3) of the Act, the Department can only rely upon the higher of the affiliates' COP or the transfer price between Coprosider and its affiliates. However, the Department lacks the necessary information to make this comparison, given that Coprosider failed to provide the actual transfer prices between itself and its affiliated processors. In the original questionnaire issued on March 9, 2000, at II.A.6, the Department requested that Coprosider "state whether the transfer price of the good or service reflects the market price of the item." In its July 3, 2000, response to the original questionnaire, Coprosider claimed that it was reporting the COP of its two processors, IBF and CFF. However, on page D-3 of that response, Coprosider stated that "the reported cost figures represent the actual costs incurred by Coprosider and its manufacturing affiliates, as recorded in Coprosider's normal accounting system," thus implying that the reported figures were, in fact, transfer prices. Because of the ambiguity in Coprosider's response, we met with the company's counsel on August 3, 2000. In that meeting, the counsel stated that the reported conversion costs were actually transfer prices between Coprosider and its affiliates. See Ex Parte Memorandum regarding Meeting with David R. Amerine of Manatt, Phelps & Phillips, and Geno Parisi, Consultant, International Trade Consulting, Regarding the Section D Response of Coprosider S.p.A., dated August 8, 2000. In the Department's supplemental questionnaire, dated August 10, 2000, we directed Coprosider to "[c]ontinue to report Coprosider's 'cost`. . . for each of Coprosider's CONNUMs based on the transfer prices between Coprosider and IBF and CFF." In its second supplemental response, dated September 7, 2000, at page 3, Coprosider added an additional field (CONVPRI) that contained what it then represented to be the "transfer prices" between Coprosider and its affiliates, given that the previously reported amounts were, in fact, the actual costs of production of its affiliated processors. Furthermore, in its September 7, 2000, response, Coprosider provided a worksheet that supposedly reconciled these transfer prices to its financial statements. At Coprosider's cost verification on September 18, 2000, the testing of the company's reconciliation of transfer prices revealed that "the amounts reported in the CONVPRI field were the standard conversion price per- piece, and do not reflect the actual transfer price paid to the affiliated suppliers." See CVR at 10. The verifiers found that all of the reported transfer prices between Coprosider and its affiliated suppliers were significantly lower than the actual transfer prices recorded on Coprosider's normal books. On September 19, 2000, the second day of the cost verification, Coprosider offered to provide the corrected amounts for all of the affected product control numbers (CONNUMs), but the Department declined to accept those data because it was past the deadline for accepting new information, pursuant to 19 CFR 351.301(b)(1). As a result, the major input test under section 773(f)(3) of the Act cannot be performed, given that the actual transfer prices are not on the record. Section 776(a) of the Act provides that, if the necessary information is not available on the record, or if an interested party withholds information, fails to provide such information in the form or manner requested, significantly impedes a proceeding, or provides information that cannot be verified, the Department shall use, subject to sections 782(d) and (e) of the Act, facts otherwise available in reaching the applicable determination. Pursuant to section 782(e), the Department shall not decline to consider submitted information if all of the following requirements are met: (1) the information is submitted by the established deadline; (2) the information can be verified; (3) the information is not so incomplete that it cannot serve as a reliable basis for reaching the applicable determination; (4) the interested party has demonstrated that it acted to the best of its ability; and (5) the information can be used without undue difficulties. As described above, Coprosider failed to provide, within the applicable deadline, the transfer prices between Coprosider, IBF and CFF. Despite the Department's unambiguous attempts to obtain the necessary information, pursuant to section 782(d) of the Act, Coprosider failed to follow the Department's instructions. As a result, Coprosider's incorrectly reported transfer price data are not usable in applying the major input rule and, as such, the value of the conversion costs cannot serve as a reliable basis for reaching our final determination. See section 782(e)(3) of the Act. Given that the information needed is critical in the final margin calculations, the Department must resort to facts otherwise available in reaching its final determination, pursuant to section 776(a) of the Act. In selecting from among the facts otherwise available, section 776(b) of the Act authorizes the Department to use an adverse inference, if the Department finds that an interested party failed to cooperate by not acting to the best of its ability to comply with the request for information. See, e.g., Certain Welded Carbon Steel Pipes and Tubes From Thailand: Final Results of Antidumping Duty Administrative Review, 62 FR 53808, 53819-20 (October 16, 1997). We find that the application of an adverse inference in this case is appropriate. As discussed above, Coprosider failed to provide the critical transfer price data, despite the Department's clear directions in its questionnaires and during the meeting with the respondent's counsel. It was not until the Department's testing at verification that it became known that the transfer prices submitted by Coprosider were not the actual transfer prices and that, as such, they were unusable for purposes of testing the appropriate value to apply to the conversion of pipes into pipe fittings. The omission of the data from the record is significant in that it has an impact on the vast majority of Coprosider's reported CONNUMs. Only the small number of pipe fittings purchased from unaffiliated companies remains unaffected. Moreover, the information proved to be readily available to Coprosider, as witnessed by the fact that they were able to provide it within a day of the Department's discovery of the problem. Finally, Coprosider did not demonstrate due diligence in reporting to the Department, as shown by the fact that the problem with the company's reported transfer prices could have, and should have, been discovered by their personnel when preparing the reconciliation of the reported transfer prices to their financial statements, which was submitted on September 7, 2000. This reconciliation did not, in fact, reconcile when reviewed at verification. See CVR at 10. For these reasons, we believe that Coprosider did not act to the best of its ability in responding to the Department's request for transfer prices, and, consequently, that an adverse inference is warranted under section 776(b) of the Act. See, e.g., Certain Circular Welded Carbon Steel Pipes and Tubes From Taiwan: Final Results of Antidumping Duty Administrative Review, 64 FR 69488, 69489 (December 13, 1999) (the Department applied partial adverse facts available, where respondents failed to provide necessary cost information). To ensure that an appropriate value for the conversion of pipes into pipe fittings is used in the final determination, as partial adverse facts available, we have calculated, from the sample data taken at verification, an amount to apply to the reported conversion costs of the affiliated suppliers. We note that the use of the reported conversion costs of the affiliated suppliers is appropriate, given that we were able to reconcile the costs reported for products produced by IBF and CFF to each of the affiliated company's financial statements. The amount applied to the reported conversion costs as an adjustment represents the difference between the respective affiliated supplier's cost and the transfer price between that affiliated supplier and Coprosider. For detailed information regarding the Department's methodology in this calculation, see the Department's Calculation Memorandum. C. Selling, General and Administrative Expenses Comment 1C: Coprosider argues that the inclusion of administrative costs incurred by its parent company, Coprosider International, and Coprosider International's parent, SYMI, in Coprosider's reported G&A expenses would be contrary to law and not supported by substantial evidence. Coprosider contends that Coprosider International and SYMI are merely Luxembourg holding companies and were not involved in the production or sale of the subject merchandise. Coprosider points out that, according to the Department's practice, G&A expenses are based upon home market data. Coprosider states that neither Coprosider International nor SYMI provide any corporate, administrative, personnel, or information services for Coprosider, and that the administrative expenses of the holding companies relate solely to fulfilling their legal obligations under Luxembourg law and should not be allocated to Coprosider. Petitioners argue that a portion of Coprosider International's and SYMI's G&A expenses should be allocated to Coprosider. Petitioners refer to the Department's antidumping questionnaire, which instructs the Coprosider to "include in your reported G&A expenses an amount for administrative services performed on your company's behalf by its parent company or other affiliated party." See the Department's antidumping questionnaire at D-25. Department's Position: We agree with the petitioners, and for the final determination we will allocate a portion of Coprosider International's G&A expenses and SYMI's G&A expenses to Coprosider in order to calculate its G&A expense rate. In cases where a parent company is an investment holding company, it is the Department's practice to allocate a portion of G&A expenses incurred by the parent company to the respondent, under the theory that the parent's G&A expenses are incurred on behalf of and as a result of the parent's investment holdings. See Final Determination of Sales at Less Than Fair Value: Welded Stainless Steel Pipe From Malaysia, 59 FR 4023, 4027 (January 28, 1994), and Final Determination of Sales at Less Than Fair Value: Ferrosilicon From Venezuela, 58 FR 27524 (May 10, 1993). D. Financial Expenses Comment 1D: Coprosider argues that its revised financial expense worksheet, provided at the cost verification, correctly reported all elements of its foreign exchange gains and losses. Coprosider claims that the exchange loss on the EURO parity adjustment, and the gain on exchange rate fluctuations, were correctly reported in the company's revised financial expense worksheet. Coprosider argues further that the Department's proposed reclassification of exchange gains and losses, relating to accounts receivable and accounts payable from gains, and losses relating to notes and loans payables, includes a portion relating to exchange gains and losses on forward exchange contracts, which should not be reclassified. Petitioners object that Coprosider did not include any foreign exchange losses in its financial expense rate calculation, but that it offset the financial expenses reported by the entire amount of its foreign exchange gains. Petitioners also point out that some of the foreign exchange gains and losses, identified as generated by notes and loans payables, were actually generated by accounts receivables and payables and should be reclassified. Department's Position: We agree with petitioner, in part. To calculate its reported costs, Coprosider excluded all foreign exchange losses and included all foreign exchange gains. However, it would be inconsistent to include the gains while excluding the losses. Therefore, consistent with our normal practice, we have included the current portion of exchange gains and losses. See, e.g., Notice of Final Determination of Sales at Less Than Fair Value: Steel Wire Rod from Trinidad and Tobago, 63 FR 9177, 9181 (February 24, 1998) and Notice of Final Determination of Sales at Less Than Fair Value: Stainless Steel Round Wire from Canada, 64 FR 17324,17334 (April 9, 1999). Level of Trade (LOT) Comment 2: Coprosider argues that the Department should find two identical LOTs in the United States and Italy and make fair value comparisons at the same respective LOTs. When there is no matching sale at the same LOT, Coprosider urges the Department to make an adjustment to normal value ("NV"). Coprosider claims that the first LOT consists of sales to the ultimate consumer for specific projects, i.e., sales to end users, engineering companies, equipment manufacturers and trading companies. The second claimed LOT consists of sales to distributors or stockists who maintain their own inventory for resale. Coprosider claims that the record shows that sales to LOT 1 and LOT 2 in each market are made to different points in the chain of distribution, i.e., that sales made to LOT 1 are made directly to end users, whereas sales to LOT 2 are made to distributors and stockists who maintain their own inventory and resell the subject merchandise to LOT 1 purchasers. In addition, according to Coprosider, the record establishes that Coprosider performs significantly different selling functions for LOT 1 sales than for LOT 2 sales. Coprosider states that for LOT 1 sales in Italy, it sells the subject merchandise from its finished goods inventory maintained in Cavenago di Brianza and that Coprosider's personnel conducts all sales- related activities, such as arranging for delivery, collecting receivables and providing after-sales service. Coprosider claims that customer- specific selling activities include customer education and advice on the suitability, uses and characteristics of Coprosider's products; in addition, as part of the technical advice given to customers in LOT 1, it provides significant testing and quality control assurance. See SVR at 7. Moreover, according to Coprosider, LOT 1 customers are often involved in the quality control process from the time a purchase order is received until the final inspections. The expenses associated with the additional technical advice were included in indirect selling expenses. See SVR at 8 and sales verification exhibit ("SVE") 26 at page 11005. Coprosider argues that because LOT 1 sales are made directly to the end- users for use in specific projects, sales quantities are generally small and the per-piece selling expense, and consequently the sales price, are higher. Coprosider considers sales to trading companies also as sales within LOT 1. Coprosider states that trading companies bundle various components on a project-specific basis for their customers, who pay a premium fee for these procurement services. The prices Coprosider charges trading companies are based on the specific quantities required for a project; therefore, Coprosider considers sales to trading companies, which are destined to engineering companies for specific projects, as at the same LOT as direct sales to end-users, engineering companies or equipment manufacturers. See June 20, 2000, supplemental response ("SR") at 2-3. Coprosider states that for home market LOT 2, it sells commodity products to distributors and stockists who inventory subject merchandise for resale to smaller distributors or end-users. According to Coprosider, these customers demand lower prices because they perform many selling functions normally provided by Coprosider on its sales to LOT 1 customers (such as inventory maintenance, executing the sale to the ultimate customer, arrangement of freight and delivery, and after-sale servicing), thereby incurring greater selling expenses. Coprosider explains that because LOT 2 sales are not made on a project-specific basis, the home market sales to LOT 2 customers involve large quantities of standard items. See Verification of Coprosider S.p.A. Sales in the Antidumping Investigation of Stainless Steel Butt-Weld Pipe Fittings from Italy ("SVR") at 7. Coprosider states that these sales do not require its intensive interaction with the customers in order to make the sale. Coprosider contends that the Department erred in categorizing Norca as a trading company, rather than as a distributor, in the SVR and that its conclusion that there is only one LOT in the United States is therefore incorrect. Coprosider claims it described Norca as a distributor in its Section A response at pages 4 and 7. Coprosider also claims that the sales invoices included as sales verification exhibits ("SVE") 14 and 20 clearly show that Norca is a distributor. On this basis, Coprosider argues that there are two channels of distribution in the United States. Coprosider describes Norca as a LOT 2 customer which maintains inventory and performs all selling functions necessary to complete the sales to its customers. Coprosider claims that Norca demands lower prices because it performs many selling functions, such as inventory maintenance and sales solicitation on downstream sales. According to Coprosider, LOT 1 in the United States consists of trading companies, which act as agents for end-users purchasing the subject merchandise for specific projects. Coprosider states that, on these sales, its personnel performs all selling activities. Coprosider claims that these sales, which are project-specific, require intensive technical advice and service and that sales quantities are generally smaller than for distributors. Petitioners urge the Department to reject Coprosider's claim for a LOT adjustment and to discard the information on special testing presented at the sales verification as new information, in accordance with section 351.301(b)(1) of the Department's regulations. Petitioners argue that Coprosider reported no warranty or technical service expenses in its home market and U.S. sales data, and that it was verified that it did not include in its calculation of ISE the costs of testing by IBF or personnel costs of the Quality Assurance Department. See SVR at 16. Petitioners claim that the sales verification team found that the sales process was the same for all customers examined. See SVR at 4. Petitioners cite the SVR's statement at page 13 that Coprosider coded some of its home market customers as an end-user for some sales and as a trading company for other sales, and argue that Coprosider did not provide any information to demonstrate that the selling functions it provided varied from sale to sale for these customers. Petitioners characterize Coprosider's customer distinctions as suspect and argue they cannot be relied upon for a LOT analysis. Petitioners conclude that Coprosider failed to demonstrate that selling functions varied at different LOTs both qualitatively and quantitatively. Department's Position: We agree with Coprosider, in part. Based on our evaluation of record evidence, including our findings at verification in this case, we have determined that there are two LOTs in the home and U.S. markets. However, we disagree with Coprosider's characterization of some of its customers. Two LOTs in the U.S. and Home Markets. In our preliminary determination, we denied Coprosider's request for a LOT adjustment because we found insufficient evidence on the record to establish different LOTs in either market. Coprosider claimed that it provided technical advice and after- sale services and warranties for customers in the end-user, equipment manufacturer, and engineering company categories in both the home and the U.S. markets, and also to the trading company category in the United States, but not to distributors. See June 20, 2000, Supplemental Response (Exhibit SB1). However, in its Section B and C response of May 1, 2000, it stated it incurred no warranty and technical service expenses during the POI (other than quality control expenses reported under indirect selling expenses). Thus, the only remaining differences in reported selling functions between the claimed LOTs were inventory maintenance, order solicitation and order processing. Therefore, for purposes of the preliminary determination, we did not consider these differences in selling functions sufficient to find different LOTs. See Preliminary Determination at 47390-47391. At verification, we asked Coprosider to clarify the differences in selling functions and services it offers to different classes or groups of customers in the home market and the United States. Coprosider's Sales Manager explained that Coprosider understands technical advice as including quality control and testing. See SVR at 6. The Sales Manager also stated that such quality control and testing is only performed for channel 1, but not for channel 2 customers. The verification team confirmed the accuracy of these statements by first carefully examining Coprosider's Quality Assurance Manual, which indicated that obtaining customer approvals is part of the quality control activities performed at two stages. See SVE 6. The engineering company's (channel 1 customer) purchase order requires unrestricted access to all places where the work or quality records are located for the purpose of conducting quality surveillance and audits. Id. at 10649. The sample documentation Coprosider submitted at verification to support its claim of substantial interaction with certain categories of customers during the production process, in the guise of product testing and certification, all relate to the engineering company customer in the home market, with one exception, which was a U.S. customer described in the Section A response as an end-user, to whom no sales of subject merchandise were made during the POI. Id. at 10606. However, the Department found other evidence in the sales traces relating to pre-sale assistance Coprosider provided to a U.S. trading company to obtain a contract for a project. See SVE 8 at 0529. Coprosider's Sales Manager explained that his company understands such assistance to fall under technical advice. The verification team also confirmed that the engineering company in the home market and a U.S. trading company, both of which are channel 1 customers, require strict adherence to their delivery schedules. The engineering company's purchase order contains a clause for liquidated damages for late delivery of materials. See SVE 6, page 10649, Art. 36. The trading company advised Coprosider in its purchase order that it would incur damages from their customer for a late delivery. See SVE 21 at 70276. In contrast, strict adherence to delivery schedules to meet the requirements of specific projects is not a condition for sales to distributors. We therefore find that these specific differences in delivery terms for channel 1 and channel 2 customers constitute additional evidence that there are two LOTs in the home and U.S. markets. We also note that there is a difference in inventory services provided for channel 1 and channel 2 customers. Based on Coprosider's responses to our questionnaires and our finding at verification, the record shows that, for channel 1 customers, the inventory services in both markets are not as extensive as for channel 2 customers. See Section A response at 6 and SVR at 4, 7. Based on our analysis of Coprosider's selling functions with respect to its customers, therefore, we determine that there are two LOTs in the home and U.S. markets. For LOT 1, there is considerable interaction between Coprosider and its customers, including (1) pre-sale technical advice to obtain contracts, (2) provision of technical assistance ranging from extensive to minor, (3) rigorous testing and certification procedures, including on-site inspections by customers, (4) production based on orders, and (5) requirements for adherence to strict delivery schedules. For LOT 2, in contrast, these services are not provided. However, as orders are filled from finished items held in inventory, the provision of inventory services is a more intensive selling function than for customers in LOT 1. For purposes of calculating the dumping margin, we will attempt to match sales made at the same LOT. When there is no match of identical or similar merchandise at the same LOT, we will make a LOT adjustment based on the pattern of consistent price differences between sales at the two LOTs in the home market. We note that petitioners' claim that the verifiers found that the sales process was the same for all customers examined is taken out of context. This statement referred to the document flow for sales. The sales verification report goes on to state that an additional document-a technical data sheet-is issued for sales to engineering companies. See SVR at 4. Finally, we disagree with petitioners that the LOT information presented to the Department verifiers constituted new information under section 351.301(b)(1) of the Department's regulations. As we explained in our verification report, Coprosider's original responses to the Department's questionnaire did address quality control (which included testing) and inspection in the context of indirect selling expenses. See SVR at 7. Therefore, the evidence presented at verification regarding testing as part of quality control does not constitute "new information" within the meaning of 19 CFR 351.301(b)(1) but merely corroborates and clarifies previously submitted information. The Department thus properly considered the evidence presented at verification. Coprosider's Characterization of its Customers. Throughout the entire proceeding, Coprosider's responses to the Department's questionnaires and statements made at verification regarding this issue contained several inconsistencies. For example, Coprosider initially described channel 1 in the United States as comprised of sales of commodity fittings to an unaffiliated importer, Norca Corp. Industrial Division ("Norca"), which resells the fittings to master distributors. Coprosider characterized channel 2 in the United States as comprised of sales to unaffiliated U.S. trading companies for resale to construction projects. There were no sales during the POI to the claimed channel 3, consisting of sales to end-users. See Section A response, February 17, 2000, at 4. Coprosider claimed three channels of distribution in Italy: the first comprised of sales to distributors, the second consisting of sales to end-users, and the third consisting of sales to an engineering company. Id. at 5. Subsequently, Coprosider revised its reported channels of distribution, reporting two instead of three channels. In both markets, channel 2 comprised distributors, and channel 1 comprised all others. Coprosider now included trading companies among its home market customers, contradicting its earlier statement in Section A that there were none. See Section B and C response, May 1, 2000, at B10. Coprosider identified LOT 1 as including all channel 1 customers, and LOT 2 as comprised of channel 2 customers. Id. at C17. In Exhibit C2 of this response, Coprosider coded one U.S. customer as a trading company, and Norca and other customers as distributors. However, at the ITC injury hearing in this investigation, on October 17, 2000, Coprosider's witness from Norca described the company as an importer, and stated that the large majority of their sales, consisting mostly of commodity fittings, are made to U.S. distributors, and that it does not sell directly to end-users. See Petitioners' rebuttal brief (November 15, 2000), Exhibit 1 (Excerpt from Hearing Transcript). We note that the evidence provided in petitioners' rebuttal brief establishes, in Norca's own words, that it is an importer of commodity fittings and its customers are master distributors. The record shows that the selling functions Coprosider provides to Norca are the same as those to distributors. SVR at 14. Therefore, we have concluded that it is appropriate to include Norca in channel 2 and LOT 2. At verification, Coprosider's Sales Manager characterized another customer, coded as 9216, as a trading company. See proprietary information in SVR at 4. However, this customer describes itself as a distributor of stainless and exotic alloy piping materials. See SVE 15 at 70252. We therefore find that this customer is a distributor. Customers coded as 4360 and 6030 were described as trading companies in the Section A response and at sales verification by Coprosider's Sales Director, but were reported in the sales listing (in field CUSCATU) as distributors. The Department verified that customer 4360 is a distributor by accessing that company's Web site on the Internet. We were unable to find any information about customer 6030. However, the quantities in which that customer buys individual products are typical of distributors, rather than trading companies ordering for specific projects. We have therefore accepted the coding of this company as a distributor. We agree with Coprosider that customer 11604 is a trading company, as it is buying pipe fittings for specific customers and end-uses. See SVE 21. Two home market customers were categorized as distributors for some sales, and as trading companies for others. Coprosider initially described customer 10479 as a large distributor. See Section A response at 6 and 7. Coprosider did not provide any evidence to support treating some of the sales to this customer as occurring at a different LOT. There is no information on the record regarding the other customer, customer 1951. However, on the basis of our analysis of the home market sales listing, we find that those sales to this customer which Coprosider categorized as sales to a trading company were frequently made in quantities characteristic of distributors. Therefore, for the final determination, we have treated all sales to these two customers as sales to distributors. Finally, one home market customer was coded as a trading company, although it is an end-user, because it allegedly resold the merchandise. See SVR at 13. As both trading companies and end-users are coded as at the same LOT, the discrepancy has no effect on the Department's margin calculation, and we have disregarded the discrepancy. Usual Commercial Quantities and Ordinary Course of Trade Comment 3: Coprosider claims that a large portion of the dumping margins found in the Department's preliminary determination was generated from price comparisons in which NV was based upon only one piece sold in Italy. Coprosider argues that calculation of NV based on sales of one piece during the POI does not comply with the statutory definition of sales in the "usual commercial quantities" under section 771(17) of the Act. Coprosider concludes that the Department must revise its computer program to eliminate home market sales of one piece. If the Department declines to exclude the one piece sales, Coprosider argues that it should alternatively make an adjustment to the NV to account for the differences in price between multiple-quantity sales made in the United States and single piece sales made in the home market, pursuant to section 773(a)(6)(C)(i) of the Act and section 351.409(a) of the Department's regulations. Coprosider points out that single-unit sales account for a small percentage of total U.S. sales transactions, in contrast to almost one- quarter of total home market sales transactions. Therefore, Coprosider argues that failure to adjust NV to account for the large percentage of higher-priced single piece sales when being compared to U.S. sales, which are mainly lower-priced multiple-unit sales, would distort the pricing comparison and create dumping margins based solely on the sales volumes of individual transactions. Coprosider claims that it satisfies the Department's regulatory requirements for a differences in quantity adjustment to NV, stating that the requirement that quantity discounts be granted on a least 20 percent of home market sales is met because multiple-piece transactions, which benefit from the alleged quantity discount, account for over three- quarters of Coprosider's home market sales transactions. Citing section 351.409(d) of the Department's regulations, Coprosider argues that the absence of a published discount list is not controlling. Coprosider also claims that it meets the alternative requirement for a quantity adjustment to NV, namely, that prices for multiple-unit sales reflect savings attributable to the production of different quantities. Coprosider asserts that these differentials reflect the cost saving attributable to the company in the manufacture and sale of multiple-unit items. See 19 CFR 351.409 (b)(1) and (2). Petitioners point out that Coprosider did not claim that any of its home market sales were outside the ordinary course of trade until the case brief. They argue that it is hardly unusual for Coprosider to sell single pieces in both markets. Petitioners argue that Coprosider's request is unsubstantiated because Coprosider failed to demonstrate that these sales were unique and unusual and were not made pursuant to bona fide home market demand. Petitioners point out that Coprosider specifically stated that it did not grant quantity discounts during the POI. See Section B-C response of May 1, 2000, at B16 and C15. Petitioners argue that the fact that Coprosider did not grant quantity discounts and does not have a published discount program negates Coprosider's argument for a quantity-based adjustment to NV. Petitioners argue further that Coprosider did not demonstrate a clear and direct correlation between price differences and quantities sold. Petitioners characterize Coprosider's worksheet in Exhibit 4 of the case brief as misleading and meaningless because it compares average prices regardless of the model match characteristics of the products, masking significant price differences between single pieces of different products. Petitioners urge the Department to reject Coprosider's adjustment claim. Department's Position: We disagree with Coprosider that its one-piece sales were made outside the ordinary course of trade and that any adjustments to NV are warranted in this case. The Department will exclude from its calculations sales made outside the ordinary course of trade in order to prevent dumping margins from being based on sales which are not representative of home market or third country sales. See sections 771(15) and 773(a) and (e) of the Act.; and Final Determination of Sales at Less Than Fair Value: Canned Pineapple Fruit from Thailand, 60 FR 29553, 29562 (June 5, 1995). The Department considers a number of factors when determining whether particular sales are outside the ordinary course of trade. For example, the Department may exclude from its analysis sales of merchandise produced according to unusual product specifications, merchandise sold at aberrational prices, or merchandise sold pursuant to unusual terms of sale. See Statement of Administrative Action, H.R. Doc 103-316, Vol. 1, at 834 (1994); and Static Random Access Memory Semiconductors From Taiwan; Final Results and Partial Rescission of Antidumping Duty Administrative Review, 65 FR 55005, 55006 (September 12, 2000). Generally, the ordinary course of trade is determined on a case-by-case basis by examining all the relevant facts and circumstances. See Bergerac v. United States, Slip Op. 2000-71 (CIT June 21, 2000). In the instant case, the sales that Coprosider seeks to exclude do not belong to any of the categories of sales that the Department normally excludes as outside the ordinary course of trade (e.g., high profit, sample, insignificant, below-cost, or second quality sales). We find that Coprosider did not demonstrate that, compared to multiple-piece sales, its one-piece sales have other unusual characteristics, which would place them outside the ordinary course of trade. Based on our analysis of Coprosider's responses to the Department's questionnaire, we find that the number of single-piece sales Coprosider made in the home market is too large for them to be considered unrepresentative. Moreover, these sales were made as part of orders including multiple units of similar merchandise, which were sold to the same customers that purchased multiple pieces of the same product, and to which normal shipping procedures were applied. Moreover, the prices for single-piece purchases were sometimes the same as for multi-piece purchases. Thus, there was nothing unusual about these sales, compared to multiple-piece sales. We also note that Coprosider did not raise this issue until almost the end of this investigation, namely, the submission of its case brief. Coprosider's explanation of facts and price comparisons, therefore, fails to sufficiently establish the extraordinary circumstances rendering these particular sales outside the ordinary course of trade. See, e.g., Koyo Seiko Co., Ltd. v. United States, 932 F. Supp. 1488, 1497-98 (CIT 1996); Murata MFG. Co. v. United States, 820 F. Supp. 603 (CIT 1993). Coprosider's request that, in the alternative, the Department should make adjustments to NV is similarly not supported by the record information. Section 773(a)(6)(C)(i) of the Act allows the Department to adjust NV to reflect the differences in quantities between the exporting country and the U.S. market. The Department may make a reasonable allowance for any differences in quantities to the extent it is satisfied that the amount of the price differential results from that difference in quantities. See 19 CFR 351.409(a). With respect to quantity discounts, the Department will account for them only if (1) the respondent grants quantity discounts of at least the same magnitude on 20 percent or more of its sales during the relevant period; or (2) the respondent demonstrates that the discounts reflect savings specifically attributable to the production of the different quantities. See 19 CFR 351.409(b) and Final Determination of Sales at Less Than Fair Value: Brass Sheet and Strip from the Netherlands, 53 FR 23431, 23433 (June 22, 1988) (explaining the Department's criteria in granting the quantity discounts adjustments). Finally, the Department does not grant this adjustment if it is claimed in addition to a LOT adjustment, unless the Coprosider demonstrates that the effect on price comparability due to differences in quantities is separate from that due to differences in the LOT. See 19 CFR 351.409(c). In this case, as described above, we determined that there are two LOTs in the home and U.S. markets and, accordingly, have granted Coprosider appropriate LOT adjustments. In its case brief, Coprosider did not address the issue of whether the effect on price comparability because of differences in quantities is separate from that due to differences in the LOTs. Therefore, pursuant to 19 CFR 351.409(c), we have determined that no differences in quantity adjustment is warranted. We also note that, even absent any LOT adjustments, Coprosider did not provide satisfactory evidence that its one-piece sales merit adjustments under 19 CFR 351.409(a). As pointed out by the petitioners, Coprosider reported that it granted no quantity discounts in either market. Coprosider has made no claim that it has a clear and consistent policy of charging customers more for single-piece purchases than for multiple pieces, and did not refer to or provide copies of a written policy to this effect. Coprosider also made a claim for a difference-in-quantity adjustment on the basis of savings in manufacturing costs of producing multiple-piece orders, pursuant to 19 CFR 351.409(b)(2), without offering any evidence of such savings. Cost adjustment claims must be based on direct manufacturing costs. Coprosider has not demonstrated that price differences are attributable to the production of different quantities. Accordingly, the Department has not granted Coprosider's claim for a quantity discount. See Final Determination of Sales at Less Than Fair Value: Open-End Spun Rayon Singles Yarn From Austria, 62 FR 43701, 43708 (August 15, 1997) (Comment 11); Granular Polytetrafluoroethylene Resin from Japan; Final Results of Antidumping Administrative Review, 58 FR 50343, 50345 (September 27, 1993); Final Determination of Sales at Less Than Fair Value: Steel Wire Rod From Canada, 63 FR 9182, 9186 (February 24, 1998). 4. Circumstance-of-Sale Adjustment-Imputed Credit Expenses Comment 4: Petitioners argue that the Department should deny an adjustment for Coprosider's reported home market credit expenses. They point out that Coprosider did not report dates of payment for either home market or U.S. sales, claiming that its new computer system did not permit their retrieval on an invoice-specific basis. See Section B-C response (May 1, 2000) at B12 and C12. Coprosider used terms of payment as a proxy for the actual payment dates. For the sales verification, petitioners note that Coprosider based its home market credit expense calculation on the average age of accounts receivable for all its Italian customers. See SVE 10 at 10722, 11 at 0925 and 12 at 0833. Petitioners argue that this is inappropriate and should be rejected by the Department because (1) the calculation of the average age of accounts receivable for home market sales was not based on customer-specific credit periods; (2) such a calculation included sales of subject and non-subject merchandise; and (3) the verification shows that Coprosider had the payment date information available on a sale-by-sale basis, but chose not to provide this information. Petitioners further allege that the verification report shows that Coprosider factored its receivables during the POI with its bank. See SVR at 13-14. Petitioners conclude that Coprosider withheld the information required by the Department, and used a much longer accounts receivable aging period in order to increase the home market credit period and consequently to reduce dumping margins. Petitioners urge the Department to make a finding that Coprosider did not cooperate to the best of its ability and to use adverse facts otherwise available by denying an adjustment for Coprosider's home market credit expense, and using the highest number of days outstanding for the calculation of the U.S. credit expenses for all U.S. sales. Coprosider rebuts petitioners' argument by claiming that its adjustment for credit expense was prepared in a manner consistent with the Department's longstanding practice. Coprosider originally reported customer-specific imputed credit expenses based on the terms of payment. In response to the Department's second supplemental questionnaire, Coprosider provided the average age of receivables in both markets. See Second Supplemental Response (July 3, 2000) at Exhibit SSB3. Coprosider claims that the average accounts receivable methodology has been accepted by the Department in numerous investigations and administrative reviews. To support this argument, Coprosider cites Gray Portland Cement and Clinker From Mexico: Final Results of Antidumping Duty Administrative Review (Cement), 62 FR 17148, 17163-17164 (Comment 18) (April 9, 1997), and Tapered Roller Bearings and Parts Thereof, Finished and Unfinished, From Japan, and Tapered Roller Bearings, Four Inches or Less in Outside Diameter, and Components Thereof, From Japan; Final Results of Antidumping Duty Administrative Reviews and Termination in Part ("TRB"), 63 FR 20585, 20595 (Comment 11)(April 27, 1998). Coprosider denies petitioners' allegation that it factored its receivables, pointing out that "factoring" is a term used for the sale of accounts receivable to a financial institution for collection. To demonstrate that the company did not factor its accounts receivable, Coprosider cites the statement in the SVR at 14 that, if the customer does not pay on time, Coprosider is responsible for collecting the overdue amount. Coprosider states that it reported its credit costs based on a reasonable calculation method using the information maintained in the normal course of business, and that it never claimed that the actual dates of payment were not available, but rather that the data are not readily accessible from its accounting system. Department's Position: We agree, in part, with Coprosider. The Department's preference is that respondents report credit expenses on a transaction-specific basis. Imputed credit costs are normally calculated by dividing the number of days between the date of shipment and the date of payment by 365, and then multiplying by the interest rate and price. When respondent's accounting system does not readily permit the retrieval of actual payment dates, we allow it to base the number of credit days on the average age of accounts receivables. See TRB, 63 FR at 20595 (the Department permitted the respondent to report credit expenses based on the average credit days outstanding on a customer-specific basis, where the transactions were voluminous); Color Television Receivers From the Republic of Korea; Final Results of Antidumping Duty Administrative Review (CTV), 56 FR 12701, 12708 (Comment 28) (March 27, 1991) (the Department stated that the use of an average accounts receivable turnover ratio was a sufficiently accurate measurement of the imputed credit expenses). In this case, the Department asked Coprosider in a supplemental questionnaire to report the actual dates of payment for all U.S. sales, noting that the number of days the payments were actually outstanding on a sample of sales were on average substantially greater than the number of days according to the terms of payment used by the company to calculate credit expenses. We were concerned that Coprosider's methodology would, on average, understate imputed credit expenses. In response, Coprosider reported the average age of receivables of its three largest U.S. customers, including one which did not buy the subject merchandise during the POI. On October 12, 2000, the Department requested that Coprosider report the total sales value during the fiscal year to other U.S. customers for subject merchandise, so that we could calculate the average age of receivables on a customer-specific basis. In its October 31, 2000, response, Coprosider reported the total sales value for all its U.S. customers. As the Department's preference is for customer-specific information, for the final determination we have calculated the average age of receivables for each U.S. customer and used this methodology to calculate imputed U.S. credit expenses. However, for home market sales, because Coprosider calculated the overall average age of receivables for all customers, including those who did not purchase subject merchandise during the POI, we agree with Coprosider that the use of the reported credit expenses based on terms of payment is more accurate than the use of the alternative methodology. The latter resulted in an average number of credit days which is excessive when compared to the terms of payment, and in light of Coprosider's method of collecting payments, which is proprietary information described in SVR at 13-14. We disagree with petitioners' assertion that Coprosider has been uncooperative and that we should apply adverse facts available in our calculation of imputed credit expenses. The company responded to the Department's several supplemental questionnaires, making an effort to report the requested information to the best of its ability. We note that Coprosider is a small company and, thus, requiring it to manually retrieve actual payment days would have imposed an excessive burden. The reported terms of payment, which are transaction-specific, are more accurate in this case than the average age of receivables for all home market customers. Therefore, we are accepting the reported home market credit expenses and are allowing a credit expense adjustment in the home market. 5. U.S. Movement Expenses Comment 5: Petitioners state that the Department should recalculate U.S. freight expense for all U.S. sales using the average freight expense for all merchandise (subject and non-subject) because the verification report shows that these expenses were under-reported. Coprosider counters that it reported international movement expense based on the cost associated with a shipment to Houston that consisted solely of subject merchandise. See Section B-C Response (May 1, 2000) at C20. Coprosider contends that basing movement expense on all shipments, including non-subject merchandise, is distortive. Department's Position: We agree, in part, with Coprosider. As reported in its original response, although the freight rates per kilogram are the same for subject and non-subject merchandise, the cost per kilogram varies from shipment to shipment based on the total weight of the shipping container. We find that the inclusion of the larger and heavier non-scope merchandise in a shipment would increase the container weight as well as the per-kilogram international freight expense and, as such, it would be distortive. We verified the international freight expense reported in Coprosider's supplemental response dated June 20, 2000, Exhibit SC2. See SVR at 15. However, Coprosider reported this expense in the field DINLFTPU on a per-piece basis which, when converted to a per-kilogram basis using the reported weight per piece, was less than the verified expense. For this final determination, therefore, we have used the verified per- kilogram amount for all U.S. sales. 6. Indirect Selling Expenses Comment 6: Petitioners claim that Coprosider's home market indirect selling expenses (ISE) were understated by the omission of bad debt reserves. Petitioners argue that the bad debt reserves are directly related to sales and used as an indirect reduction in accounts receivable in the period in which the sale is recorded. Coprosider contends that petitioners have misstated the Department's practice and the facts in the record. Coprosider claims that the Department has a consistent and long-standing practice of not including bad debt reserves as a direct or indirect selling expense, and that adjustments to price are made to account for bad debt reserves only in cases in which they were expensed to write off a previously reported sale. See, e.g., Porcelain-on-Steel Cookware From Mexico: Final Results of Antidumping Duty Administrative Review, 65 FR 30068, 30070 (Comment 4) (May 10, 2000)("Cookware"); Antifriction Bearings (Other Than Tapered Roller Bearings) and Parts Thereof From France, Germany, Italy, Japan, Romania, Sweden, and the United Kingdom; Final Results of Antidumping Duty Administrative Reviews, 64 FR 35590, 35607 (Comment 5)(July 1, 1999)("AFBs"). Coprosider argues that amounts recorded as bad debt reserves are not actual selling expenses incurred during the POI. Rather, they are reserve accounts set aside in case an actual expense for bad debt is incurred in the future. Coprosider concludes that, as in this case, where there has been no write-off of actual sales or expensing of the bad debt reserves during the POI, no direct or indirect deduction to NV should be made. Department's Position: We agree with petitioners and have revised ISE to include the bad debts provision. It is the Department's practice not to include bad debt reserves in the calculation of ISE. However, we do include amoounts expensed in the relevant period. See, e.g., Cookware, 65 FR at 30070; AFBs, 64 FR at 35607. In this case, we found that Coprosider's bad debts provision was included in the production operating costs for fiscal year 1999, and therefore it was expensed, in the financial statements reported in Section A, Exhibit 6B at line 12. Accordingly, for purposes of the final determination, we have allocated these expenses to the home market sales of the subject merchandise, using the methodology proposed by petitioners. See Petitioners' Case Brief at 14. 7. Ministerial Error Comment 7: Coprosider urges the Department to correct the decimal place error made in calculating U.S. credit expenses for the preliminary determination. Department's Position: We agree and have corrected this error for the final determination. 8. Critical Circumstances Comment 8: Coprosider asks the Department to make a negative final determination with respect to critical circumstances on the basis that, if the Department were to revise its margin calculation as proposed by Coprosider, the margin would be substantially reduced, if not eliminated entirely. Department's Position: In the Preliminary Determination, the Department found that critical circumstances exist with respect to imports of pipe fittings from Italy. Section 733(e)(1) of the Act provides that the Department will determine that critical circumstances exist if there is a reasonable basis to believe or suspect that: (A)(i) there is a history of dumping and material injury by reason of dumped imports in the United States or elsewhere of the subject merchandise, or (ii) the person by whom, or for whose account, the merchandise was imported knew or should have known that the exporter was selling the subject merchandise at less than its fair value and that there was likely to be material injury by reason of such sales, and (B) there have been massive imports of the subject merchandise over a relatively short period. In determining whether there is a reasonable basis to believe or suspect that an importer knew or should have known that the exporter was selling stainless steel butt-weld pipe fittings at less than fair value, the Department normally considers margins of 25 percent or more for EP sales sufficient to impute knowledge of dumping and of resultant material injury. (See, e.g., Preliminary Determination of Critical Circumstances: Certain Small Diameter Carbon and Alloy Steel Seamless Standard, Line and Pressure Pipe from the Czech Republic, 65 FR 33803, 33803 (May 25, 2000)). Coprosider has not provided any new evidence bearing on our preliminary determination of critical circumstances. As Coprosider's final dumping margin is more than 25 percent, the Department affirms its finding of critical circumstances in this case. 9. Miscellaneous Issues A. Model Match: Petitioners argue that for reducing Ts and reducers, pipe fittings that connect pipes with different outside diameters, the Department should use Coprosider's reported thickness of the smaller diameter pipe. Petitioners agree with Coprosider that different wall thicknesses at each end of these products are important physical characteristics. Coprosider urges the Department to exclude short radius products from the model match and compare these models to constructed value ("CV") because sales of these products are generally made in one or two pieces. Petitioners counter by arguing that the sales were made in the usual commercial quantities, and that the statute and regulations require comparisons to be based on sales of merchandise in the home market that are identical or most similar to the merchandise sold in the United States. Petitioners conclude that, given that Coprosider sold short radius products during the POI and has never indicated any differences in physical characteristics of these products, the Department should not compare them to CV. Department's Position: The Department finds that the interested parties' requests for changes to the model match are not timely. On January 21, 2000, we sent proposed product concordance criteria to all interested parties, asking for comments by February 4, 2000. Coprosider and petitioners submitted comments, but neither commented on the exclusion of the wall thickness of the smaller diameter pipe from the model match characteristics. Moreover, the wall thicknesses of the smaller diameter pipes have not been used in the other cases, which are also a part of the less than fair value investigation regarding pipe fittings. Therefore, we have not included this additional variable in the model match for the final determination. We agree with petitioners that there is no basis in law or regulations to exclude short radius products from the model match solely because they are sold in small quantities. These products are within the scope of this investigation. Coprosider did not raise the exclusion issue until the case brief, and did not provide any additional information on, or analysis of, the sales in question. We therefore find no rationale for comparing these products to CV. B. Sample Sales and Sales to Affiliated Party: Petitioners claim that the Department found at verification that Coprosider improperly included sample sales and sales to an affiliated company in its reported home market sales. See SVR at 10. Coprosider disagrees, stating that careful review of the verification report makes clear that Coprosider excluded the sample sales and sales to its affiliate from the database reported to the Department. Coprosider points out that the SVR states that the Department was able to confirm that these sales were excluded. Department's Position: We agree with Coprosider. Petitioners seem to have misread the SVR. Page 10 of the SVR explains the procedure whereby the Department verified the reported quantity and value and reconciled it to the financial statements. The discrepancy between the quantity and value in Coprosider's worksheet and the amounts reported to the Department was accounted for by the exclusion of sample sales and sales to its affiliate from the reported database. C. Correction of Errors Found at Verification: Petitioners argue that the Department should correct various errors found at verification, namely: (1) exclude PSUS-1; (2) add unreported bank charges and interest to PSUS- 4; (3) add bank fees and correct the credit expense for PSUS-5; (4) deduct the refunded customs duty and harbor fees from SUS-4; (5) correct the shipment date for SUS-7; and (6) correct misreported GRSUPRH and COMMH for PSHM-3. Coprosider agrees that, for the final determination, the Department should correct the sales listings, but submits that many of the petitioners' suggested corrections are themselves incorrect. Coprosider agrees that sales trace PSUS-1 should be dropped from the U.S. sales listing because it was a sale to a third country. However, in regard to bank fees, Coprosider points out that it reported all bank fees in the field INTEX, so no adjustment to credit cost reported for PSUS-4 and PSUS- 5 is required. Coprosider argues that petitioners' suggestion that the Department impute a "cost of money" for the U.S. sales reflects a misunderstanding of how Coprosider finances its home market and U.S. sales. Coprosider cites the SVR at 14 as correctly noting that the bank charges interest for each invoice and this interest is captured in the INTEX field. Coprosider concludes that the "correction" requested by petitioners would result in double-counting Coprosider's interest cost. Coprosider agrees with the petitioners' proposed adjustment to U.S. sales trace SUS-4 for the duty and harbor fees and the adjustments to GRSUPRH and COMMH for home market sales trace PSHM-3. Finally, Coprosider comments that the Department can disregard the one day difference between the reported date of shipment and the invoice date. Department's Position: We agree with Coprosider that some of the petitioners' proposed changes are the result of a misunderstanding and have made corrections for the final determination, as described below. During the sales traces, we found one sale (PSUS-1) reported as a U.S. sale that was destined for a third country, as noted on the purchase order. We have therefore deleted that sale from the database. We found that for sales trace SUS-4, Coprosider refunded the U.S. Customs duty and harbor fees to the customer. We allocated the amount of the refund to the value of the sales on that particular invoice and captured it in the OTHDISU field. The one day difference between the shipment date and the invoice date for SUS-7 has no effect on the margin calculation because the calculation of imputed credit expense is not based on the difference between the shipment date and the actual date of payment, which was not reported. We have therefore disregarded it. For sales trace PSHM-3, we found that the reported price and commission did not match the amounts on the sales documents, and have corrected GRSUPRH and COMMH accordingly. In regard to petitioners' second and third proposed changes, petitioners appear to overlook the difference between the imputed credit expense that is calculated for the fields CREDITH/U, which is a measure of the opportunity cost of capital tied up in accounts receivable, and the interest expense, which are the actual financial charges imposed by Coprosider's bank, reported in the field INTEX. Accordingly, we have not made these proposed changes for the final determination. D. Use of Updated Cost Data: Petitioners comment that the reported VCOM and TOTCOM in the September 7, 2000, cost database did not agree with the same variables reported in the July 1, 2000, Section B and C databases, and that the Department should use the VCOM values reported on September 7, 2000, for the difference in merchandise adjustment calculation. Coprosider did not comment on this issue. Department's Position: We agree with petitioners and have used the updated cost data for the final determination. Recommendation Based on our analysis of the comments received, we recommend adopting all of the above positions and adjusting all related margin calculations accordingly. If these recommendations are accepted, we will publish the final determination and the final weighted-average dumping margin for Coprosider in the Federal Register. AGREE ____________ DISAGREE _________ ______________________________ Troy H. Cribb Assistant Secretary for Import Administration ________________ (Date)