67 FR 11976, March 18, 2002 A-580-815, A-580-816 ARP: 8/1/99-12/31/99 (CR), 8/1/99-07/31/00 (CORE) Public Document IA/III/IX: MJM/MH March 11, 2002 MEMORANDUM TO: Faryar Shirzad Assistant Secretary for Import Administration FROM: Joseph A. Spetrini Deputy Assistant Secretary For Import Administration, Group III SUBJECT: Issues and Decision Memorandum for the Final Results of Antidumping Administrative Review of Cold-Rolled ("CR") and Corrosion- Resistant ("CORE") Carbon Steel Flat Products from Korea ------------------------------------------------------------------------- SUMMARY: We have analyzed the comments and rebuttal comments of interested parties in the seventh administrative review of the antidumping duty order covering cold-rolled and corrosion- resistant carbon steel products from Korea. As a result of our analysis, we have made changes, including correction of certain inadvertent programming and clerical errors, in the margin calculations. We recommend that you approve the positions we have developed in the "Discussion of the Issues" section of this memorandum. Below is the complete list of the issues in this administrative review for which we received comments in the case briefs and rebuttal briefs submitted by interested parties. BACKGROUND: The Department of Commerce ("the Department") published its notice of preliminary results of antidumping administrative review of CR and CORE from Korea on September 11, 2001. See Certain Cold-Rolled and Corrosion- Resistant Carbon Steel Flat Products From the Republic of Korea; Notice of Preliminary Results of Antidumping Duty Administrative Review ("Preliminary Results"), 66 FR 47163 (September 11, 2001). We have analyzed the case briefs and rebuttal briefs of interested parties in this review. As a result of our analysis, we have made changes from the Preliminary Results. The specific calculation changes for Pohang Coated Steel Co., Ltd. ("POCOS"), and Pohang Steel Industries Co., Ltd. ("PSI") - (collectively, "the POSCO Group"), SeAH Steel Corporation ("SeAH"), Dongbu Steel Co., Ltd ("Dongbu"), and Union Steel Manufacturing Co., Ltd. ("Union") can be found in Analysis for the Final Results in the Administrative Review of the Antidumping Duty Order on Cold Rolled and Corrosion Resistant Steel Products from Korea (March 11, 2002). GENERAL COMMENTS 1. Calculation of Indirect Selling Expenses. COMPANY-SPECIFIC COMMENTS Pohang Iron and Steel Co., Ltd. ("POSCO"), Pohang Coated Steel Co., Ltd. ("POCOS"), and Pohang Steel Industries Co., Ltd. ("PSI") (collectively, "POSCO Group") 2. Long-term Interest Income Offset 3. Calculation of Total Sales in the Indirect Selling Expense Ratio Denominator 4. Home Market Credit 5. Overruns Dongbu Steel Co., Ltd. ("Dongbu") 6. Allocation of U.S. Indirect Selling Expenses Union Steel Manufacturing Co., Ltd. ("Union") 7. Inclusion of Overruns in the Calculation of Normal Value DISCUSSION OF THE ISSUES Issues with Respect to All Companies Comment 1: Calculation of Indirect Selling Expenses All four respondents argue that the Department miscalculated their U.S. indirect selling expenses for the preliminary determination. Respondents state that the Department, which addressed the need to avoid double- counting interest expenses in the Preliminary Results, nevertheless employed a methodology that did not achieve this objective. Respondents contend that the Department's preliminary methodology distorts the calculation of the indirect selling expense ratio and the subsequent dumping margin calculation. Specifically, respondents argue that the Department mistakenly compared the sum of imputed credit expenses assigned to subject merchandise to the net interest expenses for subject and non-subject merchandise. Respondents state that the Department then applied the amount by which net interest expenses exceeded the sum of imputed credit expenses to the indirect selling expense ("ISE") pool. Because net interest expenses pertain to subject and non-subject merchandise, respondents contend that any comparison between subject merchandise-specific imputed credit expenses and total interest expenses is incorrect, an "apples-to-oranges" comparison. According to respondents, the subsequent inclusion of interest expenses in excess of the total imputed credit expenses to the ISE pool distorts the ISE ratio. Additionally, respondents argue that the Department must count imputed credit expenses as only a direct expense adjustment to gross unit price, not as an indirect expense adjustment. Respondents state that the Department's current methodology amounts to double-counting, as the interest expense attributable to subject merchandise is included both in the calculation of imputed credit and the calculation of the indirect selling expense ratio. See section 351.401 (b)(2) of the Department's regulations. Additionally, SeAH and Union state that ISEs must be reasonably attributable to sales of subject merchandise, and the inclusion of interest expenses associated with non-subject merchandise in the dumping margin calculation is also distortive. If the Department chooses not to apply respondents' submitted ISE ratios, Union additionally proposes a methodology that, it argues, remedies this distortion. Rather than using actual amounts to draw comparison between imputed credit expenses and total interest expenses, respondents state that the Department should compare the ratio of total imputed credit expenses to subject-merchandise sales value to the ratio of total interest expenses to total sales value of subject and non-subject merchandise. Noting that the ratio for imputed credit expenses is greater than the ratio of total financial expenses, Union argues that it is apparent that left-over imputed credit expenses do not exist, and therefore, any amount of imputed credit expense cannot be added to the ISE pool. Respondents alternatively propose that the Department calculate a ratio of total interest expenses to total sales of subject and non-subject merchandise and apply it to respondents' total sales value of subject merchandise, yielding an amount attributable to interest expenses for subject merchandise alone. Respondents state that, should this amount exceed the sum of calculated imputed credit expenses for subject merchandise, the difference could then be appropriately applied to the ISE pool. Finally, respondents cite the Department's methodology used in the final results of Stainless Steel Sheet and Strip in Coils from Korea, 66 FR 64950 (December 17, 2001) ("SSSS") and Stainless Steel Plate in Coils from Korea, 66 FR 64107 (December 11, 2001) ("SSPC") as a more accurate model than that contained in the Preliminary Results of the review with which to calculate ISE ratios. In SSSS, the Department determined that it should first allocate interest expense between subject and non-subject merchandise. Only then did the Department deduct imputed credit expenses from interest expense for subject merchandise. The Department then added only the remaining amount of interest into indirect selling expense to determine the applicable ratio. In SSPC, the Department agreed that its preliminary methodology, which excluded only the imputed credit expenses associated with subject merchandise from the interest expenses included in the ISE calculation, was in error. In that case the Department resolved the problem by attempting to calculate the amount of interest expense applicable to sales of the subject merchandise, thereby excluding interest expenses associated with non-subject merchandise. To do this, the Department calculated two ratios. The first was a ratio of all indirect selling expenses (excluding interest) over total sales (subject and non- subject). The second ratio calculated the amount of the U.S. affiliate's interest expenses applicable to subject sales. The Department accomplished this by first multiplying the affiliate's total interest expenses (subject and non subject) by the ratio of subject U.S. sales over total U.S. sales (all products). The Department then deducted the imputed credit expenses for subject merchandise from that figure and divided the resulting "net interest expense applicable to sale of subject merchandise" by subject U.S. sales. The ratios, added together, were applied to the gross unit price as ISEs. Petitioners argue that the Department should not deduct imputed credit expenses or inventory carrying costs from the interest expenses in determining U.S. ISEs. Specifically, petitioners contend that imputed credit adjustments, while measuring the time-value of money, are not considered "out-of-pocket" expenses, like interest expenses. Petitioners further state that any opportunity cost associated with imputed credit expenses is offset by the higher price a seller charges for sales on credit, rather than for immediate cash payment. Correspondingly, according to petitioners, the time-value of money suggests that a buyer is willing to pay a higher price for merchandise received on credit. Petitioners further argue that sales on credit are independent of any out-of-pocket interest expenses, as the selling price of merchandise inherently reflects the terms of the sale (i.e., credit or cash). Therefore, according to petitioners, interest expenses are unrelated to any opportunity cost associated with sales on credit, and the granting of credit does not increase its need to borrow funds. Petitioners additionally state that the Department itself recognizes that credit expenses are "built into" the sale price. See Import Administration Policy Bulletin 98.2 (February 23, 1998). Following this, petitioners argue that the Department's reasoning that credit expenses and seller's borrowing needs are correlated crumbles when examined in the case of other direct adjustments, such as freight, warehousing, etc. Petitioners claim that such direct adjustments also reduce "cash-flow" but obviously do not impact a seller's borrowing needs and decisions. Petitioners additionally note that credit expenses are imputed to capture the unrecorded time-value of money, as distinct from interest expenses, which reflect the recorded cost of borrowing money used in any aspect of business enterprise. In their rebuttal briefs, respondents Union and Dongbu argue that the Department must reduce net interest expenses by the amount of imputed credit expenses incurred in the United States, as affirmed by the Court of International Trade ("CIT"). See Pohang Iron and Steel Co., Ltd. v. United States, 188F. Supp.2d (CIT October 13, 2000). Additionally, Union and Dongbu cite the Department's determination in the sixth administrative review of this case, quoting "We agree with Respondents that a certain measure of double-counting will occur if we deduct both U.S. interest expenes, imputed U.S. credit costs and U.S. inventory carrying costs from the starting price." See Issues and Decision Memorandum for the Administrative Reviews of Cold-Rolled and Corrosion-Resistant Carbon Steel Flat Products from Korea, January 5, 2001. Union and Dongbu further note in the final results of SSPC that the Department recognized its obligation to eliminate double-counting of actual interest expenses and imputed credit expenses. Union further contends that interest expenses are directly related to imputed credit expenses insomuch as a firm which sells on credit does not "sit idly by waiting for money from its customers" but rather obtains loans which reduce the opportunity cost of waiting for customers' payments. See Letter from Union at 5 and 6, dated December 14, 2002. Petitioners' rebuttal briefs argue that respondents' methodological arguments for re-calculating U.S. ISEs are moot because the Department should not deduct any imputed credit expenses from the net interest expenses, and hence the ISE pool. Stating that respondents' claims of mistaken comparisons between imputed credit and net interest expenses should be disregarded, petitioners contend that the fundamental differences between the two expenses demand that they be treated differently and do not permit any consideration of possible double- counting of the same expense. By virtue of this, according to petitioners, no complaint of "apples-to-oranges" comparison is valid or relevant. Petitioners further argue that, while the issue of double-counting does not logically exist, any deduction of imputed credit expenses from net interest expenses must only consider the role of subject-merchandise, not of non-subject merchandise. Therefore, if an amount is to be deducted from net interest expenses, it should include the value of imputed credit- expenses attributable to non-subject merchandise as well. Department's Position: We agree with respondents that the calculation used in the Preliminary Results to offset interest expenses included in indirect selling expenses by the amount of subject-merchandise-related imputed expenses can be made more accurate. Following the Department's practice as set forth in SSSS and SSPC, our calculation of indirect selling expenses, which is done on a company-wide basis, takes into account all of the indirect selling expenses of the company and allocates them over all of the U.S. affiliate's sales in the United States. As a result, the offset for imputed expenses is not calculated on the same basis as all of the other expenses included in indirect selling expenses, since the offset is limited to sales of subject merchandise, whereas all of the other expenses are reported on a company-wide basis. Therefore, we have adjusted our methodology in the final determination to place the comparison between imputed expenses and net interest expenses and the consequent double- counting of imputed expenses on the same basis. To enact this refinement, we recalculated the ISE ratio as follows. Beginning with the respondent-provided indirect selling expense amount, we calculated a preliminary ISE ratio by dividing the ISE amount by the total sales amount. Additionally, we calculated an interest expense ratio. To do this, we calculated the ratio of U.S. sales of subject merchandise to total sales and applied it to the total interest expense, as reported by respondents. This yields a subject merchandise-specific interest expense amount. This allocation is appropriate to ensure that the deduction for double counting is taken from a pool of expenses at the same level as the offset, e.g., subject merchandise. This more accurately ensures that no non-subject merchandise interest or imputed expenses are applied to subject merchandise. From this amount, we then deducted the sum of imputed expenses, creating a new net interest expense amount. Following this, we divided the net interest expense amount by the total sales of subject merchandise to create the interest expense ratio introduced above. Because both the preliminary ISE ratio and the interest expense ratio are ultimately applied to gross unit price (i.e., the same variable), we added the two ratios together to create a new and final ISE ratio, inclusive of all relevant indirect selling expenses and interest expenses. We applied this final ISE ratio to the gross unit price. See Analysis Memo. Additional Issues with Respect to POSCO Comment 2: Long-Term Interest Income Offset Petitioners argue that, within POSCO's calculation of total U.S. indirect selling expenses, short-term interest expenses were offset with long-term interest income. Petitioners state that reducing short-term interest expense by long-term interest income is inconsistent with the Department's longstanding policy to allow an interest income offset to interest expense only for short-term interest income. Petitioners argue that while the Department's policy was originally applied in cases dealing with calculation of cost, the underlying principle is not cost-specific. Petitioners therefore claim that the Department should revise POSCO's submitted U.S. indirect selling expense calculations to exclude any offset for long-term interest income. POSCO contends that the Department previously considered and rejected petitioners' suggested adjustment and correctly continued to include total interest income as an offset to interest expense. POSCO argues that the Department, in its Preliminary Results, rejected petitioners argument and calculated the interest expense adjustment to POSAM's U.S. ISE; thus, petitioners' argument that POSCO inappropriately included total interest income as an offset to interest expense is incorrect. Additionally, according to POSCO, the Department, not POSCO, deducted total interest income in its calculation of U.S. ISEs, not simply short-term interest expenses. Thus, the Department itself calculated the adjustment, and did so properly with respect to the interest income offset. POSCO claims the policy petitioners referenced "was originally applied in cases dealing with calculation of cost." POSCO argues that petitioners provide no support for their conclusions that "the underlying principle is not cost-specific." POSCO argues that the Department's treatment of interest expense in a cost context is inapplicable to its treatment of these expenses in an analysis of selling activities, as these analyses have entirely different purposes. For a sales-below-cost analysis, according to POSCO, the Department examines interest expenses to calculate normal value that in turn is based on sales in the "ordinary course of trade." Under the Department's practice, significant quantities of sales that are made below the cost of production are outside the ordinary course of trade. Respondents argue that to analyze if a company has made sales below cost, the Department offsets interest expense with short-term interest income only. Respondent states that the Department has adopted this position because short-term interest income is generated by working capital and thus is directly tied to production activities, whereas long- term interest income may be generated by activities unrelated to production such as investment. See Porcelain-on-Steel Cooking Ware from Mexico, 58 Fed. Reg. 32095, 32100 (Jun. 8, 1993). Respondent claims that to calculate a cost of production reduced by income from activities unrelated to production would distort the Department's analysis of whether sales are above the cost of production, i.e., the ordinary course of trade. Respondent also argues that petitioners' logic does not apply when calculating selling expenses because the Department has consistently held that all activities of an entity that functions solely as a selling entity are related to sales. For example, in Certain Hot-Rolled Carbon Steel Flat Products from Thailand, 66 FR 49622 (Sep. 28, 2001), the Department explained that "because this affiliate... is a selling entity, it is appropriate to include G&A expenses incurred by the entity in the selling expense calculation," See also Honey from Argentina, 66 Fed. Reg. 49622 (Sep. 28, 2001) at Comment 6; Melamine Institutional Dinnerware Products from the Peoples Republic of China, 62 Fed. Reg. 1708, 1710 (Jan. 13, 1997); Sweaters Wholly or in Chief Weight of Man-Made Fiber from the Republic of Korea, 55 Fed. Reg. 32659 (Aug. 19, 1990). Thus, for example, all general and administrative expenses, as well as interest income and expense, are included in the calculation of U.S. ISEs. Accordingly, all expenses and income incurred at POSAM, including long-term interest income, are appropriately treated as direct or indirect selling expenses. Finally, POSCO argues that, contrary to petitioners' assertions, the practice of including total interest income as an offset to interest expense is in fact consistent with the Department's methodology when including interest expense as a component of ISE. For example, the Department included a full offset for long-term interest income in the previous administrative review of this case. Respondents claim that in the prior administrative review, as in the Preliminary Results of this review, the Department properly determined that all of POSAM's activities, including investment, should be considered as a component of U.S. indirect selling expenses to be deducted from U.S. price. Furthermore, the Department did not distinguish between interest income and expense, or otherwise indicate that it intended to disallow any component of interest income. In addition, the Department has in no way indicated that it would include some components and ignore others. Respondents argue that for the above reasons, the Department correctly included long-term interest income as an offset to POSAM's interest expense when calculating U.S. indirect selling expense in its Preliminary Results. Respondents argue that petitioners' suggested adjustment to U.S. indirect selling expense calculation must be rejected. Department's Position: We agree with petitioners that no offset should be made, although not for the reasons petitioners advanced. Neither petitioners' nor POSCO's arguments address the relevant facts on the record regarding this issue. Exhibit S-9 of POSCO's July 10, 2001, submission indicates that the interest income at issue earned by POSAM originated from a loan to its U.S. market affiliate. This loan is therefore an intra-company transaction, and as such, the interest income listed by POSAM must be offset by the loan-obtaining affiliate as an interest expense, reflected within POSCO's consolidated financial statements. See Exhibit S-9 of POSCO's July 10, 2001, submission. Consequently, the inclusion of this amount of interest income is inappropriate. Therefore, for the final results, we are excluding this offset from the total interest expense factor calculated in the ISE pool. See POSCO's Final Analysis Memo. Comment 3: Calculation of Total Sales in ISE Ratio Denominator Petitioners argue that POSCO used an inaccurate total sales value in the denominator of its ISE ratio, consequently generating an ISE ratio lower than would result from using an accurate total sales value. Therefore, according to petitioners, POSCO has reported significantly smaller U.S. ISEs for its U.S. sales. In addition, petitioners argue that POSCO did not reconcile the components of POSAM's U.S. ISE ratio to POSAM's financial statement in accordance with the Department's instructions. Specifically, petitioners contend that while POSAM's year 2000 consolidated financial statements reflect only the net revenue gained from inter-company transfers and are, thus, General Accepted Accounting Principles ("GAAP")-accordant, POSCO's formula for its U.S. ISE factor overstates the total sales denominator because it reflects the gross sales between affiliated companies, not merely the net revenue. This, according to petitioners, understates the per-unit U.S. ISE that is deducted from U.S. gross unit price. Petitioners cite the Financial Accounting Standards Board's Emerging Issues Task Force Abstract, Issue 99-19, dated July 20, 2000, arguing that current GAAP prohibits a company from including in its total sales those revenues it derives from sales in which it does not play a significant role. Petitioners argue therefore that POSCO's calculation of U.S. ISEs is not GAAP-accordant. Petitioners dispute POSCO's claim that it is necessary to calculate U.S. indirect selling expenses on a consistent basis with gross unit price. Petitioners state that while it is true that the Department applies the indirect selling ratio to the gross price reported in a company's dataset, that gross price is analogous to, and directly tied to, the total revenue amount reported in a company's financial statements as well as the gross unit price reported in the U.S. datasets. Petitioners argue that the Department should use the "product sales" value contained in POSCO's audited financial statements to calculate the company's ISE ratio. Petitioners argue that the total sales amount POSCO has reported for POSAM does not reconcile to the total sales amount in POSAM's consolidated financial statements. Petitioners state that the sales value POSCO has submitted inflates the value reported in POSAM's consolidated financial statements. Petitioners argue that the Department should revise POSCO's calculation of total U.S. sales used as the denominator for the calculation of U.S. ISE ratio. The Department should pro-rate the 1999 and 2000 financial statement amounts to calculate a POI sales value amount. Using this approach, the total sales value used in the denominator of the calculation would be stated on the same basis as the revenues reported in POSAM's consolidated financial statements and the sales values reported in POSCO's datasets. This would be consistent with the methodology the Department has employed in previous segments of this proceeding. Respondent argues that POSCO calculated and reported the U.S. ISE factor for the period of review based on gross sales value as recorded in its financial accounting records. Gross sales value is the accumulation of gross unit price times quantity for POSAM's sales of all products, and is the equivalent of gross unit price on the sales data files that accompany POSCO's submissions. POSCO's methodology for gross value was the same here as that employed by POSCO and accepted by the Department in every prior administrative review. The Department in turn relied on POSAM's total POI sales value when it calculated the U.S. ISE factor. Respondent argues that in 2000, POSAM changed the reporting in its consolidated financial statements to comply with changes in U.S. GAAP. Specifically, POSAM purchases the product from POSCO and resells it with a markup. In past accounting periods, for sales to all customers POSAM reported the gross sales value in its product sales and the corresponding costs of purchasing the product in the cost of sales line of its financial statements. However, as of 2000, POSAM's income statements identify revenues in two different forms. For sales to unaffiliated parties, the reporting is unchanged. That is, POSAM recognizes sales revenue for the full value of the sale and a cost of sale equal to the purchase price paid to POSCO. However, due to changes in U.S. GAAP, POSAM now records the net revenue for sales of products to UPI, an affiliated party. That is, for sales to UPI the net revenue, not the separate sale or corresponding cost of sale, is recognized as revenue on POSAM's consolidated financial statements. This change in GAAP did not affect POSAM's reporting for operation or other selling expenses. These expenses continue to be reported in full in the financial statements. Respondent claims that petitioners urged the Department to revise POSCO's U.S. ISE factor using as the denominator the POI sales revenue as restated in the 2000 consolidated financial statement. Respondent argues that under petitioners' approach, only the revenue reported in the consolidated financial statements forms a viable basis for calculating ISE, and that petitioners rely solely on a change in reporting format as support for their argument. Respondent states that POSCO has clearly demonstrated to the Department, its ISEs and sales revenue reconcile directly to its consolidated financial statement for both 1999 and 2000. According to respondent, the Department rejected petitioners' argument that POSCO's ISE calculations are not reconcilable to POSAM's financial statements. The Department, therefore, has correctly used POSCO's reported sales for the POR for its calculation of U.S ISE. Respondent argues that the Department has recognized, and petitioners acknowledge, that selling expenses are assigned to the sales values generating those expenses under the matching principle. For a sale-based allocation to reasonably and accurately allocate expenses to each transaction, sales reflected on POSAM's book must be placed on a common basis. To be consistent with how the ISE ratio will be applied, i.e., as a factor times gross selling price, it is necessary to state all revenues based on the gross value of product sales, not the net revenues earned after accounting for cost of sales. Respondent argues that the Department therefore correctly rejected Petitioners' suggested alterations to POSCO's reported ISE ratio to avoid serious distortions in the result. Respondent claims that the Department has consistently applied this matching principle when calculating ratios for its analysis. For example, in Japanese Tapered Roller Bearings, 63 FR 2558 (Jan.15, 1998) the Department explained that "if an expense is allocated on the basis of total sales value, as is the expense at issue here, the expense amount (the numerator) and the total sales value (the denominator) should reflect the same pool of sales such that the total expense amount reported by the respondent is divided by the total value of the sales for which the expense was actually incurred. Likewise, the allocation ratio should be applied to the same sales price reflected in the denominator...We would not accept the application of an allocation ratio to home market gross sales price if the denominator was calculated by totaling the value of all sales on the basis of a net price." Respondent argues that similarly, in Stainless Steel Wire Rod from Spain 66 FR 10988 (Feb. 21, 2001), the Department recognized that revenue as recognized in the financial statement might not accurately reflect total sales value for which a company may incur selling expenses. Petitioners, in that case, in addition to here, urged the Department to calculate the ISE ratio for the U.S. subsidiary as total operating expenses and net sales revenue in the audited financial statement. The Department disagreed, noting that the revenue did not include revenue from back-to-back sales, which were not recorded in the company's book. Thus, the Department concluded "the respondent plays a role in making back-to- back sales and, thus, these sales must be included in the denominator of the ratio. Respondent claims that the Department generally relies on sales value, but is not required to use it. Instead, the Department has adjusted this value where necessary to avoid distortions. In one example, for a respondent's ISE calculation, the Department allowed the respondent's ISE calculation to include sales before or after the start of the POI because the numerator applied to sales before the POI, see Large Newspaper Printing Presses and Components Thereof, Whether Assembled or Unassembled from Japan, 61 FR 38139, 38156 (July 23, 1996). The Department stated that the "numerator of the factor calculated utilizes the expenses recognized by the respondent in the normal course of business for the period in question and the denominator of that factor utilizes the sales recognized by the respondent in the normal course of business for the same period." Respondent argues that the same matching principle is reiterated in PET Film from Korea 65 FR 55003 (Sept. 12, 2000). In a similar ruling the Department correctly refused to use the net revenue reported in POSAM's financial statement because it recognized that for sales to UPI, this approach would allocate only a fraction of the appropriate expenses to the UPI sales. ISEs were incurred by POSAM in support of the sales to UPI and it would be distortive and inconsistent with the Department's prior practice to allocate those gross expenses over only the net value of the sales to UPI. Respondent argues that the Department also had adopted methodologies other than purely sales based allocations when reliance solely on sales revenue would be distortive. For example, in Structural Steel Beams from Korea 65 FR 41437 (July 5, 2000), the Department accepted an allocation based on selling division expenses. Moreover, in DRAMs from Korea 61 FR 20216 (May 6, 1996), the Department stated "it is not our policy to require allocation of indirect selling expenses based upon relative sales value in every instance." Respondent argues that the Department recognized in its Preliminary Results that POSCO reasonably applied the Department's matching principle in the present review by calculating its ratio using gross sales value for all sales. Sales reflected on POSAM's books must be placed on a common basis for a sales based allocation to reasonably and accurately allocate expenses to each transaction. POSCO's reliance on gross sales value to calculate its U.S. ISEs has been accepted in all prior reviews. Respondent states that a serious distortion would result if the Department only relies on the revenue recognized in POSAM's financial statement which include only the net revenue for sales to UPI. The gross sales value for UPI would be eliminated, yet the expenses associated with the sales would remain. Accordingly, Respondent argues that petitioners' argument should once again be rejected. Department's Position: We disagree with petitioners. Any factor (e.g., indirect selling expense ratio) applied to gross unit price normally must be calculated on the same basis as gross unit price. Petitioners have not demonstrated on the record that calculating a U.S. ISE factor based on gross sales and applying it to gross unit price amounts to an "apples-to-oranges" comparison and consequently distorts the impact of indirect selling expenses on the final U.S. price. Therefore, for the final results, we have not changed POSCO's reported total sales in the denominator of its ISE ratio. See POSCO's Final Analysis Memo. While the Department may re-examine this issue in a subsequent review, there is insufficient information currently on the record, particularly with regard to the nature of the selling expenses incurred by POSAM, to overturn the precedent cited by POSCO. As such, the Department will continue to leave unchanged the calculation of the U.S. ISE denominator as submitted by POSCO. Comment 4: Home Market Credit According to petitioners, POSCO states that to determine the rebate amount applicable to a home market customer in a given month, PSI ties the payments received from that customer in that month to the customer's total sales amount in that month. Thus, in calculating monthly rebate amounts, PSI identifies a payment date and applies it to specific sales. However, in calculating home market credit expenses, POSCO claims that because PSI has no way to tie payments to specific sales, it calculates an average credit period for each customer based on the average age of accounts receivables for that customer. Petitioners claim that POSCO's explanations of PSI's rebate and credit reporting methodologies rely on inconsistent factual assertions. For PSI's rebates, POSCO claims to tie payments to sales on a monthly basis. For calculating PSI's credit expenses, POSCO claims to be entirely unable to tie payments to sales. Petitioners argue that because POSCO has stated that it cannot tie payments to sales, the Department should reject POSCO's calculated home market credit expenses. Respondent argues that in its Preliminary Results, the Department correctly accepted POSCO's reported home market credit expenses for PSI sales; however, petitioners argue that the Department should reject these reported expenses because POSCO failed to rely on transaction-specific payment dates when calculating credit expense for PSI sales. Respondent argues that Petitioners continue to raise an issue which the Department has rejected and continue to misunderstand or deliberately mistate the concept of POSCO's open accounting system. Respondent claims that, as POSCO has described on numerous occasions, this system does not have a matching key or systematic method to tie payments to particular sales transactions. However, petitioners incorrectly assert that "PSI identifies a payment date and applies it to specific sales" for specific rebates. POSCO has explained thoroughly in this review that under PSI's open account system there are essentially two mutually exclusive events that petitioners repeatedly try to connect. First, the customer purchases product on a specific date. Second, that same customer's purchases are added together and are periodically paid. The payment PSI receives is credited to the balance of the summed purchases, not to a particular sale. Respondent argues that for specific rebates, PSI's sales department manually compares these rebates to that customer's total balance for the prior period to determine the specific rebate. Respondent claims that this calculation does not link the customer's payment with a particular sale. Instead, customers can receive rebates based on a combination of factors, as POSCO explained in its response. For example, PSI examines the total quantity purchased on a monthly basis. A higher rebate percentage may result for a customer on a specific case basis. PSI also looks to the timing of the customer's payment, and provides different rebate amounts based on time criteria. Respondent argues that home market credit expense is determined under the same open accounting system. For specific rebates under the open account system, no systematic method exists to tie payments for this calculation to specific transactions. Payments are made in lump sums and are subtracted from a customer's running balance owed. Respondent argues that the Department correctly recognized that POSCO's reporting methodology was reasonable and not distortive and requires no adjustment. Department's Position: We disagree with petitioners. Contrary to past administrative reviews in these proceedings, where POSCO has relied upon an open accounting system to calculate its home market credit expenses, in this administrative review, for certain home market customers during the POR, "POSCO has manually identified the payment date and recalculated credit expenses based on actual payment date" for POSCO, POCOS and PSI. See POSCO's Supplemental Section B submission dated March 30, 2001, at page 25 and Exhibit SB-3. Thus, for certain customers POSCO has reported its home market credit expenses based on actual payment date in accordance with the Department's instructions. Also, petitioners have offered no basis and the Department has no basis to reject these home market credit expenses which are based on actual payment date. Additionally, where POSCO was unable to identify actual payment date based on customer payment, POSCO has continued to use its open accounting system to report home market credit expenses. The Department has verified POSCO's reporting open accounting system methodology for credit expenses in prior reviews, and we continue to find it acceptable and consistent with the Department's reporting requirements. As POSCO stated in its Section B submission, "actual payment dates tied to specific transactions are not traced in POSCO's POCOS', and PSI's accounting system. As indicated in the questionnaire response, POSCO has therefore calculated the average credit period on a customer-by- customer basis across all products sold to each customer during the period of review based on average age of accounts receivable." See POSCO's Section B submission dated December 18, 2000 at page 64. Further, petitioners have not provided any new evidence to challenge the Department's Preliminary Results that home market credit expenses should be excluded from the Department's margin calculation. Therefore, for these final results, we will continue to use POSCO's reported home market credit expenses in our margin calculation. Comment 5: Overruns Petitioners argue that throughout the course of the review, POSCO consistently refused to report PSI's overruns as instructed. In answer to the Department's instructions for POSCO to identify all of its overrun sales and explain its methodology for identifying overruns, POSCO identified overrun sales for PSI as "(1) those products which have not been sold within 90 days of production, or (2) those products which were produced for export but were in fact sold to the domestic market." The Department instructed POSCO to revise its reporting methodology to properly identify the missing overrun sales; however, POSCO did not revise its response. Petitioners state that at verification, PSI officials stated -contrary to POSCO's repeated assertions that it identified products not sold within 90 days of production as overruns- that price rather than time in inventory was used as the determinative criterion in identifying which sales were reported as overruns. Petitioners argue that the Department's discovery at verification was unmistakable: whether PSI determined that it had granted a discount on a specific type of sale was the decisive factor in whether it identified products held in inventory over 90 days as overruns. POSCO's statement throughout the review regarding identification of long-term inventory sales as overruns included no mention of discounts being the decisive factor, or in any way relevant to, that identification. In additon, and contrary to the facts discovered at verification, POSCO stated in its home market questionnaire response that PSI did not grant quantity or any other discount on home market sales, and that POSCO therefore omitted from its dataset Field 19.3, which would identify "Other Discounts." Petitioners argue that the Department cannot be certain that there are no sales in which PSI granted a discount on merchandise held in inventory over 90 days and where the discount was granted on a sale initiated by the customer. Petitioners state that given POSCO's lack of compliance with the Department's instructions, and its failure to report the information requested, the Department should not rely on POSCO's identification of PSI's overruns. The Department should use adverse facts under certain conditions which are met in this case. First, petitioners assert that the general requirement for use of facts available, that "necessary information is not available," is met here. According to petitioners, POSCO used the methodology the Department repeatedly instructed POSCO to use to identify PSI's overrun sales to identify POSCO and POCOS overrun sales, but not to identify PSI's overrun sales. Second, petitioners claim that the Department repeatedly identified deficiencies in respondent's submission which respondents failed to correct. Petitioners claim that POSCO's lack of cooperation in response to the Department's request for information that would properly identify overruns warrants application of adverse facts available. Petitioners claim that despite instructions from the Department to identify PSI's overruns according to the method used to identify POSCO and POCOS overruns, POSCO made no effort to provide any information that would have enabled the Department to meaningfully assess the severity of the deficiency. According to petitioners, POSCO does not claim that it could not do so, rather it claims only that to supply a complete reporting of all of PSI's overruns according to the criteria it normally uses with POSCO and POCOS would require manual identification of overruns. Petitioners allege that in addition to making no effort to supply information responsive to the Department's instructions, at verification POSCO misled the Department in its explanation of how it identified PSI overruns. Therefore, POSCO cannot be considered to have met the statutory requirement of "acting to the best of its ability to comply with a request for information" and adverse fact available is warranted Petitioners state that because the Department does not possess the information needed to cure PSI's deficiencies in reporting long-term inventory overruns, it must adopt one of a variety of adverse facts available methodologies to ensure that POSCO does not benefit from its refusal to cooperate and its omission of discount information from its dataset. Petitioners argue that the Department should identify as PSI overruns all sales already identified as such by POSCO, plus all home market sales of any CONNUM where 1) that CONNUM is made to a foreign specification; and ii) where the home market sales volume of that CONNUM is less than half of one percent of the sales volume to the United States and all other export markets. The Petitioners argue that this methodology is reasonable and is not punitive. Petitioners also argue that the Department may reasonably infer that products produced to foreign specifications are generally intended for export. Where this is the case and the volumes of the product sold in the home market are extremely small compared to exported volumes, the Department may reasonably infer that those home market sales are of overrun merchandise. Respondent argues that petitioners suggest that PSI has improperly reported overrun sales because it "refused" to identify these sales on the same basis as POSCO and POCOS, but petitioners claims are incorrect in multiple respects. First, according to POSCO, the Department has repeatedly reviewed, verified, and accepted PSI's overrun reporting methodology in prior reviews and has recognized that PSI did not utilize the same internal coding system as either POSCO or POCOS. PSI, therefore, was permitted to develop this overrun reporting methodology, in consultations with the Department in prior reviews. Thus, PSI did not "refuse" to use the same methodology as POSCO or POCOS; rather, it could not. Second, respondents counter that petitioners also ignore the fact that the Department reviewed and verified at length PSI's overrun reporting methodology in the present proceeding as noted by the Department's statement "we did not find any discrepancies between the reporting of overruns and the company's overrun methodology." Third, according to POSCO, petitioners are incorrect when they state that PSI "refused" to identify sales on the same basis as POSCO and POCOS. Respondent argues that as clearly stated in PSI's submissions, PSI asked the Department for clarification on how to respond to the Department's request because it did not have any systematic means to identify and report overrun sales on the same basis as POSCO and POCOS. Moreover, during verification, the Department confirmed that PSI did not utilize the same internal coding system as either POSCO or POCOS and confirmed that PSI did not identify overruns in the normal course of business. Department's Position: We disagree with petitioners. First, the application of adverse facts available with regard to PSI's overruns sales is not warranted in this case. The Department has concluded that although POSCO did not comply fully with the Department's instructions in reporting PSI's home market overrun sales, verification demonstrated that PSI cannot comply fully with the Department's request because no 'overrun' designator was found among PSI's ordinary book and records. Additionally, PSI has stated that it does "not have a classification for overrun sales in its internal systems (e.g., PSI's operating system or accounting system), and therefore, does not classify any of these sales as "overrun" in the normal course of business." See POSCO's supplemental questionnaire response dated July 10, 2001, at page 3. Also, POSCO has described PSI's overrun reporting methodology on the record in this administrative review. See POSCO's supplemental questionnaire response dated March 30, 2001, at page 11. Further, the Department has determined and verified that while POSCO and POCOS have the same system (i.e., both POSCO and POCOS have a code in their respective accounting system that classify sales as either overrun or non-overrun sales) for tracking overrun merchandise, PSI does not use this system and thus is not able to track overrun merchandise in a manner similar to POSCO and POCOS. See POSCO's supplemental questionnaire response dated March 30, 2001, at pages 10 and 11. Also, we have determined that POSCO has acted to the best of its ability to comply with the Department's requested information by developing a reasonable overrun classification methodology. See POSCO's supplemental questionnaire response dated July 10, 2001, at page 3. In addition, respondent has stated on the record that "PSI has adopted this overrun classification methodology because there is no systematic means for PSI to report overrun sales on the same basis as POSCO and POCOS." See POSCO's supplemental questionnaire response dated July 10, 2001, at page 17. Second, at verification, we found no indication that PSI records overruns in the ordinary course of business. At verification, we examined PSI's methodology for reporting overruns and found its reporting to be consistent with its questionnaire responses. Further, through our home market sales traces, we found that the overrun sales were consistently reported and we found no discrepancies in any of the overrun sales. We found that sales were consistently reported as overrun and non-overrun according to the methodology constructed by the company in order to respond to the Department's questionnaires in this administrative review. Additionally, verification of the record evidence makes it clear that PSI reported as overruns all instances where PSI sold at a discounted price merchandise that had not been sold within 90 days of production, and thus, classified these sales as overruns. See POSCO's supplemental questionnaire response dated July 10, 2001, at page 4 and Verification Report of the Questionnaire Responses of the POSCO Group ("Verification Report") dated February 5, 2002, at page 5. Likewise, at verification the Department found if merchandise was sold from inventory more than 90 days after production and did not receive a price discount, PSI did not provide these sales with a special code that designated these sales as overruns. See Verification Report at page 6. Further, in the instances where a discount was applied, the discount was not applied after the sale. The verification report clearly states "the company stated that a code of [...] means that a product was offered at a discount price and the [...] is entered by salesman after the sale is made at that price." See Verification Report at page 6. Accordingly, as the price already reflects a discount, we disagree with petitioners and determine that it is not appropriate to apply a discount to these sales. Therefore, for the final results, we have accepted POSCO's reporting of overruns and have not excluded certain PSI home market sales from our dumping margin calculations. See Final Corrosion-Resistant SAS program. Additional Issue with Respect to Dongbu Comment 6: Allocation of U.S. Indirect Selling Expenses Petitioners argue that Dongbu understates its U.S. indirect selling expense amount attributable to subject merchandise. Petitioners claim that the ratio of sales of subject merchandise to total sales in the U.S. market should be consistent with the ratio of indirect selling expenses directly attributable to subject merchandise to total indirect selling expenses. According to petitioners, Dongbu's report of indirect selling expenses directly attributable to subject merchandise is inconsistent with its sales record because the vast majority of its total sales in the U.S. market were of subject merchandise, yet the amount of subject merchandise indirect selling expense to total indirect selling expenses is relatively small. Therefore, petitioners contend that the Department should re- allocate Dongbu's indirect selling expenses directly attributable to subject merchandise on the basis of sales value consistent with the Department's typical practice. Respondent argues that in calculating its U.S. indirect selling expense factor, it distinguishes its indirect expenses as either (1) expenses directly related to sales of the subject merchandise ("flat-rolled"), (2) expenses directly related to sales of non-subject merchandise ("other"), or (3) expenses that cannot be attributed to any specific product group ("common" expenses). The combined total of "flat-rolled" and "other" indirect expenses are referred to collectively as "direct" indirect selling expenses, i.e., expenses that can be tied directly to sales of a particular type of merchandise. In calculating the U.S. ISE factor, Dongbu only included the "flat-rolled" portion of "direct" indirect selling expenses-- removing the "other" expenses that are not associated with sales of the subject merchandise. This total "flat-rolled" portion of "direct" indirect selling expenses is then allocated only over sales of flat-rolled products. As for the "common" expenses which cannot be directly tied to sales of a particular type of merchandise, these expenses are allocated between "flat-rolled"and "other" on the basis of sales value. Respondent argues that the Department, in calculating the ISE factor used in the Preliminary Results, correctly followed Dongbu's allocation methodology and included only the "flat-rolled" portion of Dongbu's "direct" U.S. ISEs -- excluding the "other" portions that are not related to sales of the subject merchandise. Dongbu rebuts petitioners objection to this methodology in which petitioners claim that Dongbu's attribution basis is inconsistent with the sales data on the record. Petitioners highlight the fact that most of Dongbu USA's sales were of subject merchandise, and Dongbu allocated a minor amount of Dongbu USA's "direct" indirect selling expenses to the subject merchandise. Respondent contends that petitioners' argument fails under the weight of Department precedent in this case. In previous reviews of this case including the two most recent, the Department has accepted Dongbu's breakdown of expenses into the three categories (flat-rolled, other, and common) and allocated the indirect selling expenses tied to flat-rolled sales over total sales of flat-rolled products during the POR. In the fifth review, the only issue that the Department had with respect to Dongbu's ISE ratio was Dongbu's allocation of the "common" expenses to both subject and non-subject merchandise. Specifically, the Department determined that Dongbu had not adequately defined what expenses were included in this "common" expense category. Dongbu's split of "direct" indirect selling expenses between "flat-rolled" and "other" products, however, was accepted by the Department in the fifth review. In the sixth review, the Department again accepted Dongbu's methodology for reporting indirect selling expenses between "flat-rolled," "other," and "common" expenses. As the Department stated in the sixth review: We agree with Respondent. In the two cases cited by petitioners, the respondent was unable to demonstrate at verification that the non-financial ISEs which the respondent had allocated to a particular product were generated by selling activities of that product. In contrast, the record in this case contains no evidence that the reporting methodology used by Respondent was distortive or otherwise inappropriate. Therefore, we have accepted Dongbu's methodology with respect to its assignation of ISEs to flat-rolled sales and non-flat rolled sales division. Respondent argues that the Department should decline petitioners' request to abandon this long line of precedent and to instead allocate "direct" indirect selling expenses on the basis of sales value. Respondent states that while it is true that the expenses associated with the sale of flat- rolled products are disproportionately low compared to their sales value, the reason for this disproportionality is that while flat-rolled products purchased from Dongbu constitute a large share of the sales value, they require little of Dongbu USA's time or resources to sell and thus only account for a small portion of Dongbu USA's expenses. Respondent argues that the Department is aware that Dongbu USA is involved in a number of activities, only one of which involves the sale of flat-rolled products. Dongbu USA's primary function for Dongbu is to act as a communications link between U.S. customers and Dongbu Group companies and where appropriate, act as a processor of sales-related documentation. Respondent claims that consistent with Dongbu USA's limited role with respect to sales of flat-rolled products, the major expenses incurred by Dongbu USA are associated with activities, such as direct freight costs recorded for other products, unrelated to Dongbu USA's role as a communication link and processor of sales-related documentation. For these reasons, the level of activity related to sales of flat-rolled products is low in comparison to the activity related to sales of other products. Respondent argues that Dongbu took this same approach in the fifth and sixth reviews, and the Department accepted this part of the allocation methodology by allocating the U.S. ISEs directly associated with flat-rolled products over total sales of flat-rolled products during the POR. Therefore, the Department should continue to accept Dongbu's methodology of allocating ISEs between "flat-rolled" and "other" expenses in the Final Results. Department's position: We agree with Dongbu. In the sixth administrative review, the Department accepted Donbgu's reported categorization of indirect selling expenses, including indirect selling expenses directly associated with subject merchandise. Because the relevant facts on the record of this review are the same as in the sixth administrative review, and no evidence indicates that Dongbu's reporting methodology was distortive or otherwise inappropriate, for the final results of this review, we are accepting Dongbu's reported indirect selling expenses attributable to subject merchandise. D. Additional Issues with Respect to Union Comment 7: Inclusion of Overruns in the Calculation of Normal Value Union argues that the Department mistakenly did not eliminate overruns from the calculation of normal value. Because overrun sales are considered outside the ordinary course of trade, Union contends that such sales should be excluded from consideration in the calculation of normal value. Petitioners did not provide comments concerning this issue. Department's Position: We agree with Union. Overrun sales must be disregarded from consideration in the calculation of normal value. See SAS programs for updated language that excludes these sales from consideration. 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 reviews and the final weighted-average dumping margins for the reviewed firms in the Federal Register. AGREE ______________ DISAGREE ______________ __________________________________ Faryar Shirzad Assistant Secretary for Import Administration __________________________________ Date