[Federal Register Volume 63, Number 4 (Wednesday, January 7, 1998)]
[Notices]
[Pages 781-805]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 98-276]


-----------------------------------------------------------------------

DEPARTMENT OF COMMERCE

International Trade Administration
[A-580-815]


Certain Cold-Rolled Carbon Steel Flat Products From Korea: Final 
Results of Antidumping Duty Administrative Review

AGENCY: Import Administration, International Trade Administration, 
Department of Commerce.

ACTION: Final results of antidumping duty administrative review.

-----------------------------------------------------------------------

SUMMARY: On December 19, 1995, the Department of Commerce (``the 
Department'') published the preliminary results of the administrative 
review of the antidumping duty order on certain cold-rolled carbon 
steel flat products from Korea. This review covers two manufacturers/
exporters of the subject merchandise to the United States and the 
period August 18, 1993, through July 31, 1994. We gave interested 
parties an opportunity to comment on our preliminary results. Based on 
our analysis of the comments received, we have changed the results from 
those presented in the preliminary results of review.

EFFECTIVE DATE: January 7, 1998.

FOR FURTHER INFORMATION CONTACT: Charles Rast (Dongbu), Alain Letort 
(Union), or Linda Ludwig, Office of Agreements Compliance, Import 
Administration, International Trade Administration, U.S. Department of 
Commerce, 14th Street and Constitution Avenue, N.W., Washington, D.C. 
20230, telephone (202) 482-3793 or fax (202) 482-1388.

SUPPLEMENTARY INFORMATION:

Background

    On July 9, 1993, the Commerce Department published in the Federal 
Register (58 FR 37176) the final affirmative antidumping duty 
determination on certain cold-rolled carbon steel flat products from 
Korea, for which we published an antidumping duty order on August 19, 
1993 (58 FR 44159). On August 3, 1994, the Department published the 
``Notice of Opportunity to Request an Administrative Review'' of this 
order for the period August 18, 1993 through July 31, 1994 (59 FR 
39543). We received a request for an administrative review from Dongbu 
Steel Co., Ltd. (``Dongbu'') and Union Steel Manufacturing Co., Ltd. 
(``Union''). We initiated the administrative review on September 8, 
1994 (59 FR 46391).
    In a letter dated February 1, 1995, petitioners formally requested 
that the

[[Page 782]]

Department consider Union and Dongkuk Industries Co., Ltd. (``DKI''), 
which was not a respondent initially, as related parties and 
``collapse'' them as a single producer of cold-rolled carbon steel flat 
products. On May 22, 1995, the Department decided to ``collapse'' Union 
and DKI for purposes of this review. (See the Department's internal 
memorandum from Joseph A. Spetrini to Susan G. Esserman, dated May 22, 
1995.) Unless otherwise indicated, all references to Union in this 
notice include DKI.
    On December 19, 1995, the Department published in the Federal 
Register the preliminary results of the first administrative review of 
the antidumping duty order on certain cold-rolled carbon steel flat 
products from Korea (60 FR 65284). The Department has now completed 
this administrative review in accordance with section 751 of the Tariff 
Act of 1930, as amended (``the Act'').

Applicable Statute and Regulations

    Unless otherwise indicated, all citations to the statute and to the 
Department's regulations are references to the provisions as they 
existed on December 31, 1994.

Scope of the Review

    These products include cold-rolled (cold-reduced) carbon steel 
flat-rolled products, of rectangular shape, neither clad, plated nor 
coated with metal, whether or not painted, varnished or coated with 
plastics or other nonmetallic substances, in coils (whether or not in 
successively superimposed layers) and of a width of 0.5 inch or 
greater, or in straight lengths which, if of a thickness less than 4.75 
millimeters, are of a width of 0.5 inch or greater and which measures 
at least 10 times the thickness or if of a thickness of 4.75 
millimeters or more are of a width which exceeds 150 millimeters and 
measures at least twice the thickness, as currently classifiable in the 
HTS under item numbers 7209.11.0000, 7209.12.0030, 7209.12.0090, 
7209.13.0030, 7209.13.0090, 7209.14.0030, 7209.14.0090, 7209.21.0000, 
7209.22.0000, 7209.23.0000, 7209.24.1000, 7209.24.5000, 7209.31.0000, 
7209.32.0000, 7209.33.0000, 7209.34.0000, 7209.41.0000, 7209.42.0000, 
7209.43.0000, 7209.44.0000, 7209.90.0000, 7210.70.3000, 7210.90.9000, 
7211.30.1030, 7211.30.1090, 7211.30.3000, 7211.30.5000, 7211.41.1000, 
7211.41.3030, 7211.41.3090, 7211.41.5000, 7211.41.7030, 7211.41.7060, 
7211.41.7090, 7211.49.1030, 7211.49.1090, 7211.49.3000, 7211.49.5030, 
7211.49.5060, 7211.49.5090, 7211.90.0000, 7212.40.1000, 7212.40.5000, 
7212.50.0000, 7217.11.1000, 7217.11.2000, 7217.11.3000, 7217.19.1000, 
7217.19.5000, 7217.21.1000, 7217.29.1000, 7217.29.5000, 7217.31.1000, 
7217.39.1000, and 7217.39.5000. Included are flat-rolled products of 
nonrectangular cross-section where such cross-section is achieved 
subsequent to the rolling process (i.e., products which have been 
``worked after rolling'')--for example, products which have been 
bevelled or rounded at the edges. Excluded is certain shadow mask 
steel, i.e., aluminum-killed, cold-rolled steel coil that is open-coil 
annealed, has a carbon content of less than 0.002 percent, is of 0.003 
to 0.012 inch in thickness, 15 to 30 inches in width, and has an ultra 
flat, isotropic surface. These HTS item numbers are provided for 
convenience and customs purposes. The written description remains 
dispositive.
    The period of review (``POR'') is August 18, 1993 through July 31, 
1994. This review covers sales of certain cold-rolled carbon steel flat 
products by Dongbu and Union.

Verification

    As provided in section 776(b) of the Act, we verified information 
provided by Dongbu and Union using standard verification procedures, 
including the examination of relevant sales and financial records, and 
selection of original source documentation containing relevant 
information.

Analysis of Comments Received

    We gave interested parties an opportunity to comment on the 
preliminary results. We received comments and rebuttal comments from 
Dongbu Steel Co., Ltd. (``Dongbu'') and Union Steel Manufacturing Co., 
Ltd. (``Union''), exporters of the subject merchandise 
(``respondents''), and from Bethlehem Steel Corporation, U.S. Steel 
Group--a Unit of USX Corporation, Inland Steel Industries, Inc., Gulf 
States Steel Inc. of Alabama, Sharon Steel Corporation, Geneva Steel, 
and Lukens Steel Company (``petitioners''). Petitioners requested a 
public hearing, but subsequently withdrew their request in a timely 
manner.

Petitioners' Comments

Comment 1
    Petitioners argue that the Department should use alternative 
information on the record to determine the market value of transaction 
handling fees that Dongbu paid to a related party for imported raw 
materials. Petitioners contend that Dongbu did not provide substantive 
evidence to support its claim that the transfer prices paid to the 
related party were at arm's-length or at least equal to the related 
party's actual costs for providing the services. Moreover, the 
petitioners argue that since the Department was unable to test the 
transfer price at verification, the possibility exists that Dongbu may 
have selectively structured these related-party transactions to 
maximize adjustments that would lower Dongbu's production costs of the 
subject merchandise. Thus, the petitioners state that the Department 
should make an adverse inference and increase the costs of raw 
materials based on the comparison of similar arm's-length transaction 
handling fees charged by unrelated parties that Dongbu's U.S. sales 
affiliate (``DBLA'') used to import subject merchandise into the United 
States.
    Dongbu contends that there is no basis for adjusting its raw 
material costs to account for transaction fees paid to a related party 
as suggested by the petitioners. Dongbu states that the services this 
related party provides to the company are not of any tangible economic 
value other than lending its internationally recognized name to the 
transaction. Dongbu additionally states that the arrangement between 
the related party and itself simply reflects an intra-company transfer 
that benefits the related party and its shareholders. Therefore, Dongbu 
believes that the Department should accept the submitted transaction 
fees that the related party charged the company.
Department's Position
    For the final results, we accepted Dongbu's submitted transaction 
fees that were paid to a related party. The transaction fees in 
question were for assistance in handling and processing the related 
paperwork created by the importation of the material. See Dongbu's 
February 21, 1995 submission at page 12. The value of the service was 
based on a constant percentage of the acquisition price of the input. 
Dongbu was unable to substantiate that the submitted transaction fees 
reflected the market value of the service provided. At verification, 
company officials stated that they did not obtain similar services for 
the importation of inputs from any other party, nor did the related 
party provide this service to any other entity. See Cost Verification 
Report of Dongbu Steel Co., Ltd. (May 19, 1995) at page

[[Page 783]]

12. However, after further review of information on the record, we have 
concluded that the transfer prices submitted by Dongbu did fairly 
represent the amount usually reflected for such services. This 
determination was made by comparing Dongbu's submitted transaction fees 
(expressed as a percentage of the purchase price) to the weighted-
average cost (also expressed as a percentage of the purchase price) of 
similar arm's-length transaction fees charged by unrelated parties that 
DBLA used to import subject merchandise into the United States. This 
comparison showed that the submitted transaction fees were above the 
weighted-average amount charged by unrelated parties. We therefore 
accepted the submitted transaction fees that were paid to a related 
party because they reasonably reflected a market value.
Comment 2
    Petitioners contend that the costs submitted by Dongbu for its 
research and development (``R&D'') department, raw material department, 
quality control department, and procurement department should be 
included in Dongbu's manufacturing costs rather than in its general 
expenses. The petitioners argue that Dongbu's submitted description of 
the functions performed by these departments sufficiently demonstrates 
that they are manufacturing costs. They add that neither the cost 
verification report nor the accompanying exhibits contained any 
indication that Dongbu attempted to provide additional explanations, 
documentation, or schedules to support its claim that the expenses were 
general in nature. Therefore, the petitioners believe that the 
Department should include all general expenses that are not 
attributable to Dongbu's sales department in the company's cost of 
manufacturing.
    Dongbu believes that its submitted classification of these 
departmental costs as general expenses is appropriate. The company 
argues that these costs were classified as general expenses on its 
audited income statement because they benefit the entire company as a 
whole. This fact was confirmed by the Department at verification. 
Furthermore, the company argues that reclassifying these as 
manufacturing costs would have an inconsequential effect, if any, on 
its cost of production (``COP'').
Department's Position
    We agree with respondent. In this specific case, we are satisfied 
that the costs in question were properly classified as general 
expenses. For the final results, we accepted Dongbu's inclusion of 
costs from its R&D department, raw material department, quality control 
department, and procurement department as general expenses. At 
verification, the Department reviewed Dongbu's source documentation and 
noted that these costs were general in nature and related to all 
merchandise sold during the POR. Furthermore, we noted that these 
expenses were reported as general expenses on the company's audited 
income statement and not as a part of its cost-of-sales. Nor were these 
costs included as part of the inventoried costs reported in Dongbu's 
finished product inventory ledgers. See Final Determination of Sales at 
Less Than Fair Value: Certain Hot-Rolled Carbon Steel Flat Products, 
Certain Corrosion-Resistant Carbon Steel Flat Products, and Certain 
Cut-to-Length Carbon Steel Plate from Korea, 58 FR 37176, 37191 (July 
9, 1993).
Comment 3
    Petitioners argue that the Department should include foreign 
exchange losses among Dongbu's manufacturing costs to ensure that the 
cost of production is calculated accurately and that the statutory 
minimum amounts for general expenses and profit are properly computed 
for constructed value (``CV''). The petitioners state that it is the 
Department's normal practice to include foreign exchange gains and 
losses related to the production of subject merchandise in the cost of 
manufacturing and not as G&A expenses.
    Dongbu believes that its net foreign exchange losses were 
appropriately submitted as general expenses and not as costs of 
manufacturing. Dongbu states that it recognizes that it is the 
Department's normal practice to include foreign exchange gains and 
losses related to material purchases in the cost of manufacturing. 
However, Dongbu states that its submitted methodology is consistent 
with the classification of those expenses on its audited income 
statement, and that such an adjustment would needlessly result in a 
deviation from the company's normal accounting records. Furthermore, 
Dongbu argues that an adjustment to reclassify the costs is needless.
Department's Position
    We agree, in part, with both petitioners and respondent. Foreign 
exchange losses arising from the purchase of raw materials normally 
should be included in material cost because this is a component of the 
cost of manufacturing. However, in this particular instance we have not 
reclassified these losses from general expenses to cost of 
manufacturing as it would have no impact on the submitted cost of 
production. See 19 CFR Sec. 353.59(a). The slight increase in 
manufacturing costs the reclassification creates is offset by 
coinciding decreases in G&A and financing costs. See Final 
Determination of Sales at Less Than Fair Value: Dynamic Random Access 
Memory Semiconductors of One Megabit and Above from the Republic of 
Korea, 54 FR 15467, 15475 (March 23, 1993).
Comment 4
    Petitioners contend that the Department should deny all of the 
claimed miscellaneous income offsets (e.g., dividends, gains on 
investments) that were applied against Dongbu's submitted G&A costs. 
The petitioners argue it is not the Department's practice to allow a 
reduction of G&A costs unless it can be substantiated that the 
offsetting income can be tied to specific expenses related to 
production. The petitioners contend that Dongbu failed to do both of 
these steps and, therefore, the Department should deny all of Dongbu's 
claimed offsetting adjustments to G&A costs.
    Dongbu contends that it properly offset G&A costs with its various 
miscellaneous income items. Dongbu states that it submitted a complete 
list of miscellaneous income items used to offset G&A costs and that 
the Department reviewed each of these items during verification. 
Therefore, the company believes that the Department should ignore the 
petitioners' request and allow the miscellaneous income offsets to G&A 
costs.
Department's Position
    For the final results, we continue to disallow certain non-
production-related income offsets to Dongbu's G&A costs. At 
verification, we reviewed source documentation and obtained 
explanations from company officials on all the income items that were 
used to offset Dongbu's G&A expense. We disallowed certain offsetting 
income from the calculation of G&A expense because Dongbu could not 
substantiate that they related to the production of subject 
merchandise. Consequently, the offsetting revenue we disallowed 
included income received from investments (e.g., dividends, gain on 
investments) because it related to investments, and not to the 
production of subject merchandise. See Final Determination of Sales at 
Not Less Than Fair Value: Saccharin from Korea, 59 FR 58826, 58828 
(November 15, 1994).

[[Page 784]]

Comment 5
    Petitioners contend that the Department should exclude Dongbu's 
duty payments from the calculation of the company's G&A and interest 
expense factors. According to the petitioners, the addition of the duty 
to the cost-of-sales figure inappropriately overstates the figure. The 
petitioners argue that Dongbu's duty drawbacks represent a refund of 
import duties incurred in the production of finished merchandise that 
is subsequently exported. Therefore, the cost-of-sales figures in 
Dongbu's audited income statements, which is net of import duties 
refunded on certain export sales, accurately represented Dongbu's final 
cost of manufacturing. Petitioners continue this argument by stating 
that duties paid on imports used to produce merchandise sold in Korea 
are not refunded, and are included in both the net cost of sales and 
Dongbu's domestic sales price. Thus, using the net cost of sales to 
allocate general expenses and interest results in an appropriate 
comparison of prices and costs that reflect import duties.
    Dongbu believes that it properly increased its cost-of-sales figure 
to include the duty in order to calculate G&A and interest expense 
factors. Dongbu contends that the increase to its cost-of-sales is 
necessary in order to ensure comparability. Dongbu notes that its 
audited income statement cost-of-sales figure is net of duty drawback, 
while its submitted costs of manufacturing figures include the duty 
because the Department requested that it be submitted in this manner. 
Therefore, the respondent states that any G&A or interest factor that 
is applied to its duty-inclusive cost of manufacturing must itself be 
determined on a duty-inclusive basis.
Department's Position
    For the final results, the Department added the import duties paid 
by Dongbu to the cost of sales, which was used as the denominator in 
calculating G&A and interest expense factors. The cost of sales in 
Dongbu's audited income statement was net of import duty drawback, 
while the Korean and U.S. cost of manufacturing submitted by Dongbu 
included the cost of import duties. Thus, the cost of sales and the 
cost of manufacturing were not reported on a consistent basis. 
Therefore, Dongbu appropriately determined the interest and G&A factor 
on a duty-inclusive basis because the submitted cost of manufacturing 
included import duties.
Comment 6
    Petitioners assert that the Department's analysis must account for 
the difference between U.S. sales by Dongbu and its U.S. sales 
affiliate, DBLA. They argue that the Department is in error in its 
treatment of DBLA's and Dongbu's sales and requests that DBLA's sales 
be treated as exporter's sales price (``ESP'') sales. Petitioners note 
that Dongbu makes sales to the United States through three separate and 
distinct channels: directly to customers in the United States; through 
related and unrelated trading companies in Korea; and through its 
affiliate in the United States, DBLA, which purchases subject 
merchandise from Dongbu and resells it to unrelated customers in the 
United States. Petitioners assert that Dongbu is incorrect in claiming 
that sales made through each of these channels are purchase-price 
sales. They state that Dongbu's contention implies that if sales 
through each of these channels are treated as such, the U.S. prices 
calculated by the Department will represent prices at the same point in 
the chain of commerce in all cases, and thus implying that the charges 
by DBLA to the first unrelated customer in the United States represent 
the arm's-length prices that Dongbu would charge for the same 
merchandise if sold directly to an unrelated U.S. customer, without the 
involvement of DBLA. Petitioners claim that Dongbu's own sales data 
indicate that there is a systematic and significant difference between 
Dongbu's and DBLA's pricing structure which is the result of the fact 
that DBLA's involvement in the sale of subject merchandise results in 
significant costs which are included in the prices it charges its U.S. 
customers.
    Petitioners also argue that because DBLA's selling prices are 
distinct from Dongbu's, the Department must analyze DBLA's sales 
differently from Dongbu's sales in order to ensure consistency with the 
fundamental purpose of the Tariff Act regarding the calculation of 
United States price. They argue that the Tariff Act identifies two 
types of U.S. sales, purchase price (``PP'') and ESP, and mandates 
different adjustments to each so that United States price is 
reconstructed at the same point in the chain of commerce regardless of 
whether a U.S. affiliate of the manufacturer or exporter is involved in 
the transaction. Citing 19 U.S.C. 1677a(b), petitioners contend that 
the Tariff Act defines purchase price as the price at which merchandise 
is purchased, or agreed to be purchased, prior to the date of 
importation, from either a reseller, manufacturer, or producer of the 
merchandise for exportation to the United States. Conversely, say 
petitioners, ESP is defined as the price at which merchandise is sold 
or agreed to be sold in the United States, prior to or after 
importation by or for the account of the exporter. See 19 U.S.C. 
1677a(c). Thus, ESP is typically used when an affiliate of the 
manufacturer or exporter imports merchandise into the United States. 
Also, petitioners cite Smith Corona Group v. United States, 713 F.2d 
1568, 1571-72 (Fed. Cir. 1983), in arguing that when a U.S. affiliate 
of a foreign respondent imports merchandise in question, all costs and 
expenses incurred by the affiliate must be deducted from the 
affiliate's resale price in order to derive a United States price 
(``USP'') that reflects the price that the merchandise would command in 
an arm's-length transaction. They further state that this is the case 
whether the sales are from the importer to an independent retailer or 
directly to the public, as if the affiliate had no role in the 
transaction. Petitioners note that DBLA's role in selling subject 
merchandise results in selling prices that are distinct from Dongbu's 
prices for the same product, and that as a result, DBLA's role in 
selling subject merchandise creates the type of bias that is addressed 
by the provisions of the Tariff Act regarding United States price.
    Petitioners also contend that Dongbu's sales through DBLA do not 
meet the statutory definition of purchase price. They argue that the 
Department utilizes a three-part test to determine whether ESP or 
purchase price should be used to determine USP when the sale is made 
prior to the date of importation; and the focus must be on the third 
factor in this test; that is, that if the related party in the United 
States only acts as a conduit between the first unrelated purchaser and 
the seller, the resulting sale is a sale for export to the United 
States. Petitioners contend, however, that before the Department can 
accurately determine that the related party is just a processor of 
documentation, there must be evidence on the record supporting that 
conclusion. They argue that documents submitted by Dongbu, which 
include DBLA's sales contracts and production order requests, do not, 
by themselves establish that Dongbu sets the essential terms of sale in 
Korea. Petitioners maintain, rather, that there is no documentary 
evidence in the record in support of Dongbu's contention. Citing to 
Creswell Trading Co., et al. v. United States, 15 F.3d 1054 (Fed. Cir. 
1994) (``Creswell''), petitioners claim that Dongbu has the burden of

[[Page 785]]

producing information that proves its point, which it has not done; and 
in the absence of such information, the Department cannot conclude that 
the indirect PP sales at issue were made in Korea by Dongbu for 
exportation to the United States. Instead, petitioners conclude that 
the Department must determine that the sales were made in the United 
States by DBLA, and that they must be treated as ESP sales.
    Petitioners further argue that the price at which DBLA sells 
subject merchandise to the unrelated purchaser is different from the 
price at which DBLA purchases it from Dongbu. They contend that these 
prices reflect the fact that DBLA performs significant selling 
activities in the United States which require the Department to treat 
the sales in question as ESP sales. Petitioners note also that DBLA 
extends credit to certain customers by permitting them to delay payment 
for subject merchandise; that DBLA identifies customers, negotiates 
prices, and provides some warranty-related services; and that DBLA is 
engaged in marketing activities that include development of downstream 
applications for subject merchandise. Petitioners contend that another 
significant selling function performed by DBLA is the posting of cash 
deposits of antidumping and countervailing duties on behalf of its U.S. 
customers. They argue that in a typical purchase price transaction, the 
U.S. customer, as the importer of record, would be required to deposit 
cash deposits with the U.S. Customs Service upon importation of the 
merchandise, resulting in additional costs. In ESP transactions, 
however, the customer is relieved of this burden and of the risks of 
uncertain future liabilities. Petitioners contend that DBLA's selling 
activities can be demonstrated in several ways. First, the activities 
performed by DBLA are significant in the context of the totality of 
activities required to sell subject merchandise. In other words, DBLA 
performs all of the functions required to sell subject merchandise in 
the United States. Second, the significance of DBLA's selling 
activities, and the economic benefit these provide to DBLA's customers, 
is reflected in DBLA's prices. Finally, petitioners cite declarations 
made by DBLA on Customs Form 7501 which indicate that it was more that 
a processor of sales related documentation.
    Respondent counters these arguments by stating that Dongbu's sales 
through DBLA meet the statutory definition of PP sales, and that 
petitioners even concede that Dongbu satisfies the first two prongs of 
the test: (1) Dongbu's sales through DBLA are shipped directly from 
Dongbu to the unrelated buyer without being introduced into DBLA's 
inventory, and (2) such shipment is customary in the industry. 
Respondent notes that the sole issue thus raised by petitioners is 
whether Dongbu USA satisfies the third prong of the test (i.e., does 
Dongbu USA act solely as a processor of sales-related documentation and 
a communication link with its unrelated U.S. buyers). Respondent 
contends, however, that verification reports and associated documents 
confirm that sales through DBLA meet the third requirement of the test, 
and that DBLA played only a limited role as a processor of sales 
related documentation and as a communications link to the customer.
    Respondent argues that all of the selling activities carried out by 
Dongbu USA in connection with these sales are within the range of 
activities determined by the Department and the Court of International 
Trade (``CIT'') to be consistent with purchase price classification. 
Respondent notes further that petitioners make the same argument here 
that they made during the original less-than-fair-value (``LTFV'') 
investigation with respect to sales of cut-to-length plate made by 
Dongkuk Steel Mill Co., Ltd. through its affiliated selling agent in 
the United States. In that case, as with Dongbu, the U.S. affiliate was 
responsible for payment of customs duties and brokerage and handling 
charges, invoicing and collecting payment, and financing accounts 
receivable. Respondent states that the Department in that case 
determined that all of the functions identified by petitioners were 
within the scope of activities consistent with a purchase price 
classification. See letter from Morrison & Foerster to the U.S. 
Department of Commerce (June 8, 1995) at 11-13; concurrence memorandum 
in Cut-to-Length Carbon Steel Plate from Korea, Inv. A-580-817 (January 
20, 1993) at 13. Respondent notes that DBLA facilitates the sales by 
processing the documents needed to ensure that the merchandise is 
delivered in accordance with the negotiated sales terms: that is, 
delivery to the customer after clearance through U.S. Customs and 
payment of brokerage and related charges. In detailing these functions, 
respondent argues that all of the selling activities carried out by 
DBLA in connection with these sales are within the range of activities 
determined by the Department to be consistent with purchase price 
classification in previous cases.
    Regarding petitioners' argument that the Department should classify 
sales through DBLA based upon comparative pricing patterns, respondent 
counters that there is no legal or factual basis for reclassifying 
these sales as ESP. Respondent contends that selling functions, not 
selling prices, are the basis for the Department's classification of 
sales as purchase price or ESP. With regard to Dongbu's sales through 
DBLA, respondent argues that the Department must consider DBLA's 
selling functions in connection with the fact that these products are 
sold to the unrelated U.S. customer on an ex-dock duty-paid basis and 
must thus be delivered to the possession of the customer after 
clearance through U.S. Customs. Respondent notes that in this case, 
Dongbu has simply transferred these routine selling functions to a 
related selling agent in the United States, and that the substance of 
the transaction is not changed, which is that they are purchase price 
rather than ESP.
Department's Position
    We have determined that purchase price is the appropriate basis for 
calculating USP. Typically, whenever sales are made prior to the date 
of importation through a related sales agent in the United States, we 
conclude that purchase price is the most appropriate determinant of the 
USP if the following factors exist: (1) the merchandise in question was 
shipped directly from the manufacturer to the unrelated buyer, without 
being introduced into the inventory of the related shipping agent; (2) 
direct shipment from the manufacturer to the unrelated buyers was the 
customary commercial channel for sales of this merchandise between the 
parties involved; and (3) the related selling agent in the United 
States acted only as a processor of sales-related documentation and a 
communication link with the unrelated U.S. buyers. See, e.g., Certain 
Stainless Steel Wire Rods from France: Final Determination of Sales at 
Less that Fair Value, 58 FR 68865, 68868-9 (December 29, 1993); 
Granular Polytetrafluoroethylene Resin from Japan: Final Results of 
Antidumping Duty Administrative Review, 58 FR 50343-4 (September 27, 
1993). This test was first developed in response to the Court of 
International Trade's decision in PQ Corporation v. United States, 652 
F. Supp. 724, 733-35 (CIT 1987). It has also been used to uphold 
indirect purchase-price transactions involving exporters and their U.S. 
affiliates. See, e.g., Zenith Electronics Corp. v. United States, 
Consol. Ct. No. 88-07-00488, Slip Op. 94-146 (CIT 1994).
    We disagree with petitioners' argument in citing to Creswell that 
Dongbu has not met the burden of

[[Page 786]]

producing information that demonstrates that the related party in the 
United States functions only as a processor of documentation. Dongbu 
has placed information on the record which we have verified describing 
the functions of its related party. Furthermore, the Department has 
recognized and classified as indirect PP sales transactions involving 
selling activities similar to those of DBLA's in other antidumping 
proceedings involving Korean manufacturers and their related U.S. 
affiliates. See, e.g., Final Determination of Sales at Less Than Fair 
Value; Circular Welded Non-Alloy Steel Pipe from the Republic of Korea, 
57 FR 42942, 42950-1 (September 17, 1992). In the present review, we 
found that: (1) Dongbu's sales though DBLA, its related sales agent in 
the United States, are shipped directly from Dongbu to the unrelated 
buyer without being introduced into DBLA's inventory; (2) such 
shipments are the customary channel of distribution for the parties 
involved; and (3) DBLA performed limited liaison functions in the 
processing of sales-related documentation and a limited role as a 
communication link in connection with these sales.
    When all three of the criteria described above are met, we consider 
that the exporter's selling functions have been relocated 
geographically from the country of exportation to the United States, 
where the sales agent performs them. We determine that DBLA's selling 
functions are of a kind that would normally be undertaken by the 
exporter in connection with these sales. Furthermore, we conclude that 
DBLA's role in the payment of cash deposits of antidumping and 
countervailing duties, extension of credit to U.S. customers, the 
processing of certain warranty claims, and project development does not 
involve the development of downstream applications for subject 
merchandise; rather, DBLA's role is not in consistent with purchase 
price classification and is a relocation of routine selling functions 
from Korea to the United States.
Comment 7
    According to petitioners, the Department is required by law to 
deduct the cost of ``actual'' antidumping and countervailing duties 
from USP when the record demonstrates that those costs are included in 
the prices paid by the first unrelated purchaser. Petitioners contend 
that these duties are costs to Dongbu and must be deducted from the 
price paid by the first unrelated purchaser in order to obtain a fair 
comparison between USP and foreign market value (``FMV'').
    Petitioners assert that the statute provides authority for 
deducting the cost of actual antidumping and countervailing duties 
incorporated in the price used to establish USP. Citing 19 U.S.C. 
Sec. 1677a(d)(2)(A), they argue that USP shall be reduced by ``the 
amount, if any, included in such price which is attributable to 
additional costs, charges, and expenses, and United States import 
duties, incident to bringing the merchandise into the United States.'' 
The costs of antidumping and countervailing duties thus fall within the 
scope of this provision as costs, charges, and expenses or as U.S. 
import duties. The former, petitioners note, is a subset of the latter, 
and as a matter of law they must be deducted from the price to the 
first unrelated purchaser. The also argue that the statute provides 
that USP shall be increased by the amount of any countervailing duty 
imposed to offset an export subsidy.
    According to petitioners, in order to prevent double-counting, the 
Department must deduct the full amount of the countervailing duties 
paid by Dongbu for those entries covered by the first and second annual 
reviews of the countervailing duty order. They claim that none of the 
arguments for not deducting the estimated antidumping duties applies in 
the case of the countervailing duty payments. First, petitioners argue 
that Dongbu has presented evidence that DBLA paid those duties and that 
they have an impact on the price. Second, they contend, there is no 
danger of double-counting since the countervailing duties are not paid 
to offset past price discrimination. In this case, the countervailing 
duties are paid to offset domestic subsidies and have nothing to do 
with Dongbu's price discrimination practices. Thus, petitioners assert 
that the countervailing duties are a cost separate from the payment of 
antidumping duties and should be treated as normal customs duties. 
Also, petitioners claim that since no party requested a review of the 
countervailing duty order at the time of the first or second 
anniversary, those duties have become final duties. They also assert 
that the Department must deduct the cost of antidumping duties equal to 
the amount of the calculated margin in this review. Petitioners note 
that the court acknowledged in Zenith Elec. Corp. v. United States, 18 
CIT __ Slip Op. 94-146 (September 19, 1994) that the deduction from USP 
of actual antidumping duties remains an open issue. Accordingly, 
contend petitioners, the court expects that the Department will 
approach the payment of actual antidumping duties differently than it 
does the payment of estimated antidumping duties.
    Respondent argues that in the absence of reimbursement, it is 
unlawful and contrary to Department practice to deduct antidumping and 
countervailing duties from USP. Respondent contends that petitioners' 
reading of the statute is contradicted by both long-standing 
administrative and judicial precedent. See, e.g., Final Results of 
Antidumping Duty Administrative Review: Certain Hot-Rolled Lead and 
Bismuth Carbon Steel Products from the United Kingdom, 60 FR 44009 
(August 24, 1995), Antifriction Bearings (Other Than Tapered Roller 
Bearings) and Parts Thereof from France, et al.; Final Results of 
Antidumping Administrative Reviews, 60 FR 10900, 10907 (February 28, 
1995), PQ Corp. v. United States, 652 Supp. 724, 735-37 (CIT 1987), 
Federal-Mogul Corp. v. United States, 813 F. Supp. 856, 872 (1993), and 
Torrington Co. v. United States, 881 F. Supp. 622 (CIT 1995. Respondent 
further argues that the Department and the courts have long since 
recognized that such deductions are not authorized under the 
antidumping laws because they are, inter alia, not ``selling expenses'' 
within the meaning of the statute. Respondent notes that making the 
required adjustment would unlawfully result in the double-counting of 
dumping duties, and would perpetuate dumping orders thereby violating 
both the letter and remedial purposes of the statute. They also state 
that Congress has refused to yield to lobbying by the U.S. steel 
industry for the enactment of legislation that would for the first time 
authorize such a deduction.
    Respondent asserts that petitioners are incorrect in their argument 
that the issue of deducting antidumping and countervailing duties 
should be considered differently in this case because the Department is 
determining ``actual'' rather than ``estimated'' antidumping duties. 
Respondent also states that petitioners are wrong in their extension of 
this argument to Dongbu's countervailing duty deposits on the theory 
that such deposits represent ``actual'' duties because the amounts 
deposited are ``conclusive'' since no party requested an administrative 
review. Respondent notes that the countervailing duty order is 
currently on appeal to the Court of International Trade and liquidation 
of these entries has been suspended pending the outcome of that appeal.

[[Page 787]]

    By assessing duties beyond the actual margins of dumping, according 
to respondent, petitioners' recommended deduction would also violate 
international law as embodied in the World Trade Organization's 
antidumping agreement. See Final Act Embodying the Results of the 
Uruguay Round of Multilateral Trade Negotiations, April 15, 1994, and 
Agreement on Implementation of Article VI of the General Agreement on 
Tariffs and Trade 1994, article 2para. 4.
    Respondent claims that petitioners are incorrect in arguing that 
their proposal will not result in a double-counting of antidumping 
duties. Rather, respondent asserts it is a ``mathematical certainty'' 
that this will be the result. Respondent argues that if petitioners' 
suggestion were followed, it would be impossible for a company engaged 
in indirect PP sales to ever eliminate its margins. Respondent 
concludes its argument by stating that petitioners have provided no 
legal support for their position either in the language of the statute, 
legislative history, court decisions, international law or the 
Department's historical interpretation of the law.
Department's Position
    We disagree with petitioners. In Final Results of Antidumping Duty 
Administrative Review: Certain Hot-Rolled Lead and Bismuth Carbon Steel 
Products from the United Kingdom (``UK Lead and Bismuth''), 60 FR 
44009, 44010 (August 24, 1995), petitioners made arguments similar to 
those presented here `` that ``actual'' antidumping duties are a 
``selling expense'' and that the Department has not previously 
considered whether to deduct ``actual'' expenses under section 772 
(d)(2)(A). In UK Lead and Bismuth, we responded that ``[a]ntidumping 
duties are intended to offset the effect of discriminatory pricing 
between the two markets. In this context, making an additional 
deduction from USP for the same antidumping duties that correct this 
price discrimination would result in double-counting. Therefore, we 
have not treated cash deposits of estimated antidumping duties as 
direct selling expenses.'' Id. at 44010. See also Color Television 
Receivers from the Republic of Korea, Final Results of Administrative 
Review, 58 FR 50333, 50337 (September 27, 1993); and Antifriction 
Bearings (Other Than Tapered Roller Bearings) and Parts Thereof from 
France, et al.; Final Results of Antidumping Administrative Reviews, 60 
FR 10900, 10906 (February 28, 1995). This same reasoning would also 
hold where ``actual'' antidumping duties are known. The fallacy of 
petitioners'' argument for treating antidumping duties as a cost is 
that antidumping duties, although paid by an importer, are not selling 
expenses, nor are they normal customs duties. Antidumping duties are 
unique in that they represent antidumping duty margins--a measure of 
price discrimination between FMV and USP. The statutory remedy for such 
unfair price discrimination is to assess antidumping duties against the 
imported merchandise in an amount equal to the amount by which the FMV 
exceeds the USP for the merchandise. 19 U.S.C. 1673. To then subtract 
this amount from USP in order to recalculate a supra-antidumping duty 
margin would be creating additional price discrimination that did not 
exist. This is the same as saying that dumping margins must be adjusted 
to account for dumping margins. Such double counting of antidumping 
duties is contrary to the Act, which is designed to comport with 
Article 8para. 2 of the Agreement on Implementation of Article VI of 
the General Agreement on Tariffs and Trade (``GATT'') 1994 in that the 
duty collected must not exceed the margin of dumping.
    We also disagree with petitioners' extension of their argument to 
Dongbu's countervailing duty deposits on the basis that the amounts 
deposited are ``conclusive'' since no party has requested an 
administrative review. Even though the countervailing duty order is 
currently on appeal to the Court of International Trade and liquidation 
of these entries has been suspended pending the outcome of that appeal, 
we still would not deduct the actual duties from USP for the reasons 
outlined above.
Comment 8
    Petitioners note that in the preliminary results of this review, 
the Department calculated Dongbu's dumping margins using Dongbu's 
reported U.S. credit expenses. However, at verification, the Department 
determined that Dongbu's short-term interest rate during the period of 
review should be revised upward.
Department's Position
    We agree with petitioners. The Department recalculated Dongbu's 
credit expenses using the revised interest rate as determined at 
verification for the final results of this review.
Comment 9
    Petitioners argue that Dongbu's freight charges for home-market 
sales should be reduced by the amount of the intra-company transfer of 
funds between Dongbu and Dongbu Express. They assert that 
transportation services for Dongbu's home-market sales are provided by 
unrelated trucking companies pursuant to contracts with Dongbu's 
wholly-owned subsidiary, Dongbu Express; and that as such, Dongbu's 
payment to Dongbu Express for those services is nothing more than ``an 
internal price constructed for bookkeeping purposes.'' Petitioners 
contend that the Department should revise these expenses to exclude 
markups charged by Dongbu Express on the grounds that such markups 
represent intra-company transfers of funds. They cite Final 
Determination, Rescission of Investigation, and Partial Dismissal of 
Petition High Information Content Flat Panel Displays and Display Glass 
Therefor from Japan, 56 FR 32376 (July 16, 1991), and Final Results of 
Antidumping Duty Administrative Review: Color Picture Tubes from Japan, 
55 FR 37915 (September 14, 1990), in arguing that the Department has 
previously disregarded the same type of markup paid to Dongbu Express 
when calculating adjustments to FMV, and that the Department attempts 
to value sales-related services at actual market rates, rather than at 
the rates established between related parties.
    Respondent counters that payment of a markup for such valuable 
services in this case is consistent with commercial considerations. 
Respondent argues that the Department has similarly acknowledged and 
accepted that an administration fee paid by a respondent to its related 
shipper reflected additional services which would have to be assumed by 
either another trucking company or the respondent itself. According to 
respondent, there is no dispute regarding the services covered by the 
markup (i.e., that Dongbu Express acts as a freight forwarder in 
arranging for and subcontracting trucking services for Dongbu). 
Respondent states that Dongbu has also demonstrated that the markup 
reasonably reflects the value of those services.
    Dongbu states that it has shown that, on average, the percentage of 
Dongbu Express' general expenses to its cost of sales is equal to the 
profit it earns. The sum of these two elements equals the markup to the 
cost from the unrelated freight company charged to Dongbu. Thus, 
according to respondent, to ensure that the reported freight amounts 
accurately reflect market rates, the Department must use the price from 
Dongbu Express to Dongbu.

[[Page 788]]

Department's Position
    We disagree with petitioners. We find that the markups charged by 
Dongbu Express to Dongbu were commercially reasonable charges for the 
services provided by Dongbu Express. Although the Department does not 
have a standard policy requiring it to deduct related-party markups in 
all cases, in Final Determination, Rescission of Investigation, and 
Partial Dismissal of Petition: High Information Content Flat Panel 
Displays and Display Glass Therefor from Japan, 56 FR 32376, 32393 
(July 16, 1991), the Department rejected the price between related 
parties not because there was a markup, but because it was determined 
that the reported amount reflected a price constructed for ``internal 
bookkeeping purposes'' rather than a market value. Also, in Final 
Results of Antidumping Administrative Review: Color Picture Tubes from 
Japan, 55 FR 37915, 32922-23 (September 14, 1990), the Department 
acknowledged and accepted the respondent's argument that an 
administrative fee paid by the respondent to its related shipper 
reflected additional services that would have been sustained by either 
another trucking company or the respondent directly. In the present 
review, we verified the arm's-length nature of Dongbu's freight charges 
and found no basis for reducing home-market inland freight charges. We 
agree with respondent that Dongbu has demonstrated that: (a) on 
average, the percentage of Dongbu Express's general expenses to cost of 
sales is equal to the profit Dongbu Express earns; (b) the sum of these 
two items equals the markup to the cost from the unrelated freight 
company to Dongbu; and (c) the prices charged to Dongbu by Dongbu 
Express accurately reflect market rates.
Comment 10
    According to petitioners, the amounts reported by Dongbu and used 
by the Department to determine the market rates for Dongbu's foreign 
brokerage and handling charges are incorrect. They reject the amounts 
used for the following reasons: (1) the evidence presented by Dongbu 
that freight charges are provided at arm's-length rates is irrelevant 
to whether the same company also provides unloading charges at arm's-
length rates, and (2) Dongbu has not demonstrated that Dongbu Express 
provides freight services at arm's-length rates. On this basis, argue 
petitioners, the Department must determine the value of unloading 
charges incurred in Korea using alternative information, specifically, 
the highest reported brokerage and handling charge for any U.S. sale as 
the adjustment for all of Dongbu's U.S. sales.
    Respondent argues that the record demonstrates that the charges 
Dongbu reported in connection with related party transactions are at 
arm's-length, and that the small amounts reported which reflect Korean 
unloading charges are for a service performed solely by Dongbu Express 
and provided solely for Dongbu. Respondent argues that Dongbu has shown 
that other, more valuable and significant services provided by Dongbu 
Express (i.e., inland freight charges, both to the United States and in 
the home market) are on an arm's-length basis. Respondent also notes 
that it is a matter of record that Dongbu Express was profitable 
throughout the review period. Accordingly, states respondent, this 
evidence provides a sufficient and reasonable basis to conclude that 
the transactions for relatively small brokerage and handling charges 
are also at arm's length.
Department's Position
    We disagree with petitioners. Although the Department generally 
prefers to demonstrate that a related-party service was provided at 
arm's length by comparing those rates with charges for similar services 
provided by unrelated companies, the Department does not automatically 
resort to best information available when that methodology is 
unavailable. Verification is the Department's means of testing 
information; it is not intended, nor is it possible, that every single 
item be examined during verification. See Monsanto Co. v. U.S., 698 F. 
Supp. 275, 281 (CIT 1988). As our verification report indicates, we 
performed an arm's-length test on Dongbu's related party, Dongbu 
Express. We reviewed invoices from an unrelated trucking company to 
Dongbu Express, and found that inland freight charged by the unrelated 
party in question was lower than that charged by Dongbu Express. On the 
basis of this verification, we have no reason to believe that Dongbu's 
brokerage and handling expenses are not also at arm's length.
Comment 11
    Petitioners contend the Department should have applied total BIA to 
Union because of the respondent's inability, at verification, to 
properly document home-market product characteristics. As a consequence 
of the flawed verification, petitioners believe that the Department 
cannot be confident that (1) it is matching U.S. sales to the proper 
home-market transactions in price-to-price comparisons; (2) it is 
matching the COP assigned to a home-market model to the proper home-
market price in the sales-below-cost test; and (3) it is properly 
resorting to CV in cases where there is no similar, contemporaneous 
home-market product or the home-market sale price is below the COP.
    Petitioners also argue that failure to verify Union's product 
characteristics taints not only Union's product comparisons, but also 
Union's COP and CV data, since those data are reported on the basis of 
specific control numbers, and each control number (``CONNUM'') is 
defined by a unique set of unverified product characteristics. To 
derive the per-ton cost of each CONNUM reported in its response, 
petitioners state that Union allocated costs on the basis of the total 
quantity produced of that CONNUM. If the home-market product 
characteristics used as a basis for defining CONNUMs are suspect, 
according to petitioners, then the production quantities and cost 
allocations based on those CONNUMs are unreliable.
    Petitioners claim that, in a number of cases where the use of 
unverified data would have rendered meaningless any calculation 
employing that data, or where the Department was unable to verify a 
respondent's home-market product characteristics, the Department has 
resorted to total, rather than partial, BIA. In addition, petitioners 
note that the Department has routinely resorted to total BIA where a 
respondent has destroyed, or has been unable to produce, documents 
supporting critical aspects of its submitted data. Petitioners point 
out that the CIT has recognized that parties who initiate unfair trade 
proceedings--as did Union by requesting this review--bear the burden of 
maintaining and retaining records relevant to the proceeding. See, 
e.g., Krupp Stahl AG v. United States, 822 F. Supp. 789 (CIT 1993) 
(``Krupp Stahl''). Indeed, petitioners note, even DKI, a company that--
unlike Union--did not anticipate being reviewed in this proceeding, 
retained production records and other customer correspondence relevant 
to home-market sales during the POR. Petitioners contend that Union's 
data deficiency, which was caused by its failure to retain relevant 
production records and customer correspondence in a review that it 
requested, is every bit as pervasive and significant as in prior cases 
where the Department has resorted to BIA. According to petitioners, 
when this data deficiency is combined with the Department's inability 
to verify the accuracy of Union's home-market date of sale and Union's 
failure to report

[[Page 789]]

accurate dates of sale for a significant percentage of its U.S. sales, 
the Department has no alternative but to resort to total BIA in its 
final results in petitioners' view.
    Petitioners cite Krupp Stahl in support of their contention that 
the choice of which information to use as BIA must not reward a 
respondent. Because it applies only to price-to-price comparisons, 
petitioners argue that the Department's BIA methodology rewards Union 
Steel by failing to account for the possibility that costs assigned to 
a particular CONNUM might not be matched to the correct home-market 
price in the sales-below-cost test, or that the use of CV as a result 
of home-market sales falling below COP or the lack of a home-market 
match would be improper. It also fails to address the possibility that 
Union's reported COP/CV amounts do not correspond to the product to 
which they are assigned. Petitioners also take issue with the 
Department's presumption that the largest possible adjustment to the 
prices of comparable products is no more than 20 percent of the cost of 
manufacturing (``COM'') of that product. Petitioners claim that the 
Department can have no idea of the extent to which improper matches may 
understate FMV because some or all home-market products may be 
improperly matched. Therefore, petitioners state, any sales of any 
product in Union's home-market database could theoretically be compared 
to U.S. price, and the record shows that price differences between U.S. 
and Korean sales are in fact far greater than the adjustment 
preliminarily used by the Department. According to petitioners, the 
Department has therefore rewarded, rather than penalized, Union for its 
improper record-keeping procedures. Should the Department fail to use 
total BIA in its final results, the Department will invite manipulation 
and circumvention of the antidumping process by respondents, 
petitioners say.
    Under the partial BIA methodology employed by the Department, 
petitioners claim a respondent could request a review and then destroy 
critical supporting documentation associated with any sale under the 
guise that such destruction is its normal business practice and assign 
to such sales the product characteristics it desires to ensure the most 
favorable price-to-price comparisons and sales-below-cost test result, 
secure in the knowledge that the Department will cap any BIA adjustment 
at a mere 20 percent of the product's COM. Similarly, petitioners 
argue, knowing that reported COP/CV amounts will not be adjusted 
despite the Department's inability to verify home-market product 
characteristics, respondents could simply assign costs to specific 
CONNUMS as they desire to ensure the most favorable outcome. The 
Department's inability to verify Union's home-market product 
characteristics taints price-to-price comparisons, the sales-below-cost 
test, and the decision to resort to CV, as well as Union's submitted 
COP/CV data.
    The Department stated that its BIA methodology was designed to 
address the possibility that (1) ``U.S. sales are not being compared to 
sales of the most similar home-market models' and (2) ``reported costs 
of home-market models may not correspond to the costs of the home-
market products that were actually shipped.'' (Department's internal 
memorandum from Joseph A. Spetrini to Susan G. Esserman, ``Treatment of 
Union Steel With Respect to Certain Corrosion-Resistant Carbon Steel 
Flat Products from Korea,'' dated August 8, 1995. Because the 
Department's partial BIA methodology assigns a 20 percent COM 
adjustment to FMV used only in price-to-price comparisons, does not 
contain any adjustment to Union's COP/CV data, or affect the sales-
below-cost test and the basis for resorting to CV, it fails to account 
for the latter. For all of the above reasons, petitioners urge the 
Department to apply total BIA to Union for the final review results.
    Respondent rejects both petitioners' claim that there are pervasive 
and significant data deficiencies sufficient to warrant total BIA and 
the Department's use of partial BIA. Union states that the Department 
verified home-market date of sale and that the Department has already 
adjusted the data with regard to U.S. date of sale. Union contends that 
there is no evidence on the record indicating that the home-market 
codes are wrong. It notes that product code questions for home-market 
sales have no implications for any of the cost data.
    Respondent states that petitioners' reliance on Cold-Rolled 
Stainless Steel Sheet from Germany and Krupp Stahl is misplaced. In 
that case, Union states, all records had been destroyed, preventing it 
from preparing a response to the Department's questionnaire and 
preventing the Department from conducting a verification. In this case, 
Union claims only two types of documents are at issue: mill 
certificates and customer correspondence. In Union's view, it had no 
reason to suspect that these documents, which it does not normally 
retain, would be deemed necessary at verification. Union concludes that 
the precedents ``underscore that the use of total BIA is appropriate 
only for a noncooperative respondent or a respondent whose submission 
is so fundamentally flawed that it cannot be used even with partial 
BIA.'' See, e.g., Antifriction Bearings (Other Than Tapered Roller 
Bearings) and Parts Thereof from France, 60 FR 10900. Thus, respondent 
states that the Department must reject petitioners' request to use 
total BIA.
    Respondent notes that the statement in the verification report that 
the Department was ``unable to verify the accuracy of the product code 
system for [Union's] home-market sales, or determine the basis behind 
Union's coding of certain model-match characteristics,'' upon which 
petitioners rest their claim for application of total BIA, is 
contradicted by factual evidence on the record. Union asserts that, as 
part of the verification, the Department: (1) Repeatedly tied the 
product codes reported on Union's tape to the product codes used on 
commercial invoices maintained in the normal course of business; (2) 
traced the reported invoice data, including the product code, from the 
commercial invoice to Union's sales ledgers, and thus to the audited 
financial accounting system; (3) compared the product codes with 
Union's product manual, and found no discrepancies; and (4) repeatedly 
checked product codes for U.S. sales (which are the same product codes 
used in the home-market) against mill certificates. Union also asserts 
that the decision memorandum forwarded to the Assistant Secretary 
failed to mention the first three of these facts. Rather, Union avers, 
the Department's memorandum gives central status to two types of 
documents--mill certificates and customer notifications--on no basis 
other than the fact that these documents were not retained. Union also 
claims that, by not notifying the company during verification of its 
concerns with regard to product characteristics, the Department 
deprived Union of an opportunity to address those concerns.
    Union, citing recent cases (see, e.g., Brass Sheet and Strip from 
Canada, and Oil Country Tubular Goods from Korea), argues that the 
Department routinely relies on commercial documentation, such as 
invoices and sales ledgers, to verify internal product codes, and does 
not normally trace product codes to production records.
    Union maintains that there exists on the record production 
information, viewed by the Department at verification, supporting its 
internal product characteristics. The Department, according to Union, 
examined post-POR mill certificates. In

[[Page 790]]

addition, Union claims that the Department's cost verifiers ascertained 
that Union used a single product coding system, which enabled them to 
test the quality and specifications of input materials to the quality 
and specifications of the finished product. It is Union's view that the 
Department's verifiers could have tied Union's product codes to its 
inventory withdrawal records and to entries into the finished goods 
inventory, which in turn could have been tied to other production 
records, but they did not do so. Alternatively, Union suspects the 
Department could have reconciled total sales to total inventory entries 
or withdrawals, thereby confirming that the amount sold of a given 
product matched the total amount produced and entered into finished 
goods inventory, but it did not.
    Respondent reiterates that there is only one internal product 
coding system used for home-market sales, U.S. sales and cost of 
manufacturing. Respondent claims it is beyond dispute that the 
Department verified both the U.S. sales data and cost data, which 
confirms the integrity of the entire internal product coding system, 
even if the Department was not fully satisfied that it could tie home-
market sales to mill certificates or customer correspondence.
    Union also asserts that its recordkeeping practices do not differ 
significantly from Dongbu's, which, like Union, did not retain home-
market mill certificates or customer correspondence. Even if Union had 
kept records in a significantly different manner from Dongbu's, Union 
cites Coated Groundwood Paper from Finland; Final Determination of 
Sales at Less Than Fair Value (56 FR 56363--November 4, 1991) as an 
example where the Department relied on very different documentation to 
verify two respondents' respective product characteristics. In that 
case, Union claims that the Department relied upon respondent Metsa-
Serla's product coding sheet to verify Metsa-Serla's product 
characteristics. It says Metsa-Serla was not penalized because it was 
unable to provide mill orders and the other respondent, UPM/Rupola, 
was. Union maintains that the context in which the Department examined 
certain documents at verification is irrelevant; the key point is that 
the Department routinely uses commercial documentation as satisfactory 
evidence of the accurate reporting of product codes and 
characteristics.
    Union disputes the Department's assertion that a majority of 
Union's reported home-market characteristics--derived from the internal 
product code--did not identify such characteristics, and therefore did 
not support respondent's conclusion with record evidence.
    Union states that the record of this review does not provide any 
explanation or reasoned basis for the Department's product hierarchy. 
Under those circumstances, it is Union's opinion that the Department 
may not lawfully use partial BIA even if Union fails to support its 
product distinctions sufficiently.
    Even assuming certain product characteristics could not be 
verified, Union argues, the Department's conclusion that the maximum 
possible adjustment for differences in physical characteristics of the 
merchandise (``difmer'') is necessary to account for the worst case is 
unwarranted. The Department could have drawn an adverse inference with 
respect to the specific product characteristics at issue.
    Union asserts that information on the record of the cost 
investigation allows the Department to limit its use of partial BIA to 
only those product characteristics that the Department erroneously 
considers not to be verified. Union suggests that the Department could 
allow those product characteristics dependent on the product code to 
vary to determine the maximum possible universe of products for each 
reported product code. The Department could then choose the highest 
home-market variable cost of manufacture (``VCOMH'') for each such 
universe and use it to calculate the difmer (subject, of course, to the 
20 percent difmer cap). Union suggests even if the Department finally 
concluded that the product codes were not verified, it could still 
calculate a margin based on submitted data.
    Union also rejects the idea that its COP and CV data are tainted by 
the alleged failed verification of home-market product characteristics. 
Union claims that the Department never expressed any concern that the 
post-verification issue of product characteristics extends to the 
calculation of Union's production costs. Indeed, Union asserts, its 
costs were developed on the basis of withdrawals from materials 
inventory and pass-through quantities, which are entirely independent 
from the quantity of product shipped. Union claims that the cost 
verification report and exhibits demonstrate that the Department could 
trace any product's characteristics back to the daily line production 
reports for the final stage of processing; that these reports indicate 
the internal product code and the nature of the final stage of 
processing; and that the product could be traced back through the 
production process based on the mill order number. The Department's 
verifiers, Union maintains, identified the input coils and determined 
the chemistry of the input coils from the suppliers' mill certificates.
    Union finally notes that, in the parallel review of Union's 
corrosion-resistant products, petitioners explicitly conceded that 
Union has a single product coding system in both the U.S. and home 
markets. Therefore, to the extent that the product coding system was 
verified in one market, it was verified generally.
    Union protests that petitioners' alleged claim that the 
Department's final determination is driven by a single sentence in the 
verification report makes a ``mockery'' of the antidumping law and 
procedures in general and of this proceeding in particular. Union 
states that in petitioners' view the final decision in this case was 
effectively made on May 16, 1995, by the authors of the verification 
report when they inserted the allegedly damning sentence into the 
record. Union further notes that petitioners portray that one sentence 
as ``handcuff[ing]'' the Department without regard to any analysis of 
all the other information on the record of this proceeding. As a matter 
of law, Union avers, the Department's preliminary and final 
determinations must be based upon a comprehensive analysis of the total 
record of the proceeding. Union contends that conclusory statements in 
internal Department memoranda are of value only if supported by the 
record.
    Union argues that the Department's preliminary determination that 
Union's home-market product characteristics were not fully verified was 
based on an incomplete and erroneous presentation of the facts on the 
record. Union claims that the decision memorandum elevates two 
potential ancillary means of verification (mill certificates and 
customer notifications) to central status on no basis other than the 
fact that these documents were not retained. Union claims the 
Department verified the accuracy of Union's home-market product 
characteristics through other means, and had many others available. 
Union also takes issue with the verifiers not having advised Union at 
verification of any outstanding concerns over product characteristics 
based on product codes. Had the Department expressed any such concerns, 
Union argues it could have suggested additional ways to verify its 
data, but was denied the opportunity.
    Petitioners protest what they term Union's ``eleventh-hour attempt 
to ``clarify,''' long after its May 23, 1995,

[[Page 791]]

submission purporting to correct certain aspects of the Department's 
sales verification report, the sentence in that report which stated 
that the Department was unable to verify Union's home-market product 
characteristics, by saying the sentence ``[was] in error or [ . . . ] 
misleading because overly broad when written.'' Petitioners argue that 
Union should not be allowed, at this late date in the proceeding, to 
assert that the report was inaccurate in this regard when it could have 
raised this concern long ago but elected not to do so. Union, 
petitioners state, simply failed to present to the Department during 
verification any documentation supporting the premise that Union's 
home-market product characteristics had accurately been verified.
    Petitioners dispute Union's suggestion that only a minority of 
product characteristic variables were derived from the internal product 
code. Petitioners point out that the verification report specifically 
says the opposite in three different places, and that Union never 
attempted to clarify or rebut these statements. Union's claim that 
certain product characteristics were derived from the product's name is 
a non sequitur in petitioners' view. They argue that while these 
physical characteristics may be associated with the product name, that 
claim in no way demonstrates that the product actually produced and 
sold possesses the physical characteristics attributable to it by 
virtue of its product name. Petitioners add that such a demonstration 
could only have been effected by providing the Department with 
production records indicating the physical characteristics of the 
products produced and sold (e.g., production orders or mill 
certificates), which Union failed to do. In any event, petitioners 
argue, even if a minority of Union's reported product characteristics 
were derived from its internal product code, it would be unreasonable 
to limit application of partial BIA to specific product 
characteristics, because Union's home-market sales, cost, and CV data 
would still be tainted. Petitioners suggest that the Department, if it 
persists in applying partial BIA to Union, could use as partial BIA the 
highest VCOMH reported in Union's database for purposes of calculating 
the difmer adjustment as well as COP and CV.
    Respondent denies that the Department's preliminary results reward 
Union and urges the Department to reject the notion that, absent any 
evidence of manipulation, a 20 percent difmer adjustment would provide 
future respondents with an incentive to manipulate the model-match 
process.
    Union argues that even if the Department justifiably determined 
that Union's product characteristics had inadequately been verified, 
its decision to resort to partial BIA was wrong, since the statute 
affords the Department broad discretion to base FMV on CV. Because 
Union's CV data was verified and reflects the cost of the products sold 
in the United States, and the Department's stated policy is to use as 
much of a respondent's data as possible, the Department had a 
responsibility to use Union's own, verified data rather than using a 
flat, across-the-board difmer of 20 percent as BIA. Respondent notes 
that a comparison of U.S. price to CV is totally unaffected by the 
perceived problems with the verification of product characteristics and 
suggests that in light of the Department's concerns, the use of CV is 
``the obvious alternative.''
    Petitioners counter that Union's CV database is just as tainted by 
the failure adequately to verify product characteristics as Union's 
sales database. Union, they claim, mistakenly believes that, because 
the product characteristics associated with the merchandise sold by 
Union in the U.S. market are not in dispute, the costs associated with 
producing that merchandise are also not in dispute. Petitioners state 
that, due to the Department's inability to verify the accuracy of Union 
Steel's reported home-market product characteristics, the physical 
characteristics of the products whose production levels Union used in 
calculating the unit cost of each given product are either unknown or 
unreliable.
    Petitioners argue further that even if the per-unit costs used to 
calculate CV are not tainted by the product-characteristic deficiency, 
CV is nevertheless unreliable because the home-market profit component 
of CV is tainted by that deficiency. The profit component of CV is 
based on the weighted-average profit made on home-market sales. Since 
the Department cannot be certain that the reported COP of home-market 
products is being compared to the proper home-market sales and prices, 
the Department cannot be certain that the profit component of Union's 
CV is accurate. This deficiency, petitioners contend, renders the 
reported CV amounts unreliable.
    Petitioners also affirm that the statute does not give the 
Department discretion to use CV as FMV when home-market sales data is 
not verified. They note the statute provides that the Department may 
use CV when home-market sales are found to be below cost in significant 
numbers and when there are no matchable numbers in the home-market 
because they exceed the 20 percent difmer test. In those situations, 
petitioners observe, the Department has before it otherwise usable and 
properly verified data which cannot be used in margin calculations. In 
this case, however, the Department did not have home-market sales data 
that was otherwise usable according to petitioners. Petitioners argue 
that when the Department is unable to verify submitted data, as it was 
in this case, the statute requires the Department to resort to BIA, 
which is always an adverse inference. In this case, they claim using 
Union's CV data is not adverse to Union and would reward Union.
    Petitioners counter that the record is unclear as to whether the 
Department ``repeatedly'' tied the product codes to sales and 
production documents, as claimed by Union. Even if the Department did 
repeatedly perform each of these tasks cited by Union, petitioners 
argue that none of these tasks (i.e., tying product codes from sales 
invoice to sales tape, tracing invoice data to sales ledgers, checking 
product codes against a product code key, checking U.S. product 
characteristics against mill test certificates) in any way confirmed 
that products sold in the home-market possessed the physical 
characteristics reported by Union.
    Petitioners claim that the statute requires the Department to 
verify the accuracy of the data submitted, not some proxy thereof. They 
note that Union has admitted on the record that its home-market 
customers are somewhat less concerned than U.S. customers with the 
accuracy of product specifications. Therefore, petitioners argue, 
verification of U.S. product characteristics cannot serve as proxy or 
surrogate for verification of home-market product characteristics. 
Petitioners allege that, to the extent that the internal product code 
was the basis for matching home-market products to U.S. products, Union 
had an incentive to ensure that the product code assigned to an 
individual home-market sale resulted in the most favorable match. 
Petitioners claim that Union does not seem to recognize that submitted 
data must be verified not to its own satisfaction, but to the 
Department's.
    Petitioners also argue that the verification reports cited by Union 
as evidence that the Department normally applies a lower standard for 
verification of product characteristics than was the case here are all 
inapposite. In those

[[Page 792]]

cases, petitioners claim, the Department was not verifying the accuracy 
of product characteristics as reflected by product codes, but rather 
whether the merchandise was in-scope versus out-of-scope, or whether 
the respondent had completely reported all sales of the subject 
merchandise. In those cases, according to petitioners, the Department 
was provided with other documentation, including documentation 
furnished by the customer, such as purchase orders and order 
confirmations.
    Petitioners contend there is a critical distinction between, on the 
one hand, verifying whether merchandise is in-scope or whether all 
sales of the subject merchandise during the POR were reported, and, on 
the other hand, whether the reported products actually possess the 
physical characteristics reported. The former is a preliminary, general 
inquiry which is designed to ascertain whether all sales have been 
reported, while the latter is a separate, detailed inquiry designed to 
ensure that the physical characteristics of comparison products were 
accurately reported. In this case, petitioners assert, the Department 
was unable to verify the accuracy of Union Steel's reported home-market 
product characteristics in the context of the latter inquiry.
    Further, as Union has conceded, the verification techniques 
employed in a given instance are dependent on the specific facts of 
each case. Petitioners state that the Department has considerable 
latitude in conducting verification and ``[t]he decision to select a 
particular method of verification rests solely within [the 
Department's] sound discretion.'' See Floral Trade Council v. United 
States, 822 F. Supp. 766 (CIT 1993). Petitioners stress that Union, as 
the requester of the review, has only itself to blame for not 
preserving vital documentation months after the review had started. In 
addition, petitioners note that Union gave the Department reason to 
distrust the company's reported product characteristics by placing on 
the record a report, prepared by a private consulting firm in Union's 
employ, which stated that the respondent was incapable of tracing its 
production records to individual shipments.
    Petitioners claim that Union's post hoc explanation of the 
production records it allegedly maintained does not demonstrate the 
accuracy of its reported home-market product codes. Petitioners allege 
that the explanation furnished by Union with regard to post-POR records 
allegedly examined by the Department's verifiers constitutes new 
factual information that should be stricken from Union's case brief. 
Petitioners argue that explanation does not exist anywhere on the 
record, nor is it clear that verification reports or exhibits support 
that purported explanation. Consequently, petitioners request that this 
explanation be stricken from the record and ignored on the grounds that 
it is untimely submitted. In any event, these materials were examined 
by petitioners for the limited purpose of ascertaining the accuracy of 
Union's reported date of sale in the home-market. Therefore, 
petitioners claim any assertion that these materials support home-
market product characteristics is post hoc and unverified.
    Petitioners also deny that the cost verification supports the 
validity of Union's internal product coding system. They claim that the 
cost verifiers did not ascertain whether the reported internal codes 
accurately reflected the characteristics of products produced and sold. 
Rather, petitioners say, the verifiers tested input costs on the basis 
of the specifications of Union's internal product code and physical 
dimensions. It is unclear, petitioners note, whether the products that 
Union reported as coming off its production line actually possessed the 
physical characteristics represented by the internal product code 
assigned to them in the accounting records maintained with respect to 
production. Finally, petitioners argue, the fact that the accuracy of 
the internal code may have verified with respect to one market (the 
United States) does not mean it verified with respect to the other 
(Korea). Even if the Department incorrectly concluded that the accuracy 
of Union's internal product code with respect to products produced for 
the home-market was verified, the accuracy of the codes appearing on 
self-generated commercial invoices for home-market sales remains 
unverified. Petitioners object to Union's suggestion that the 
Department could have employed alternative verification techniques, 
thereby trying to usurp the Department's role. They note that the 
verification outline clearly put the respondent on notice as to the 
goals of the verification and as to the type of supporting 
documentation Union would be required to produce. It was therefore 
``unconscionable'' for Union to destroy records that would have allowed 
the Department to verify the accuracy of the most critical component of 
antidumping analysis--the product characteristics assigned to each 
control number, according to petitioners. It is incumbent upon a 
respondent to volunteer to the Department's verifiers information as to 
what sort of documentation is available to permit verification. It 
would appear that by inserting the consulting firm's report on the 
record of the verification, Union was fully aware of the problem posed 
by verifying home-market product characteristics. Yet it was not until 
the case brief that Union volunteered the existence of documents which 
Union claims would have permitted such a verification. Union had 
repeatedly denied that production records could be tied to shipment 
records. Union also suggests post hoc that inventory records could have 
been used to verify product characteristics, yet the consulting firm's 
report states outright that these records are inaccurate. If the 
product code could not be verified for home-market sales, petitioners 
suggest, it is doubtful that the accuracy of the product codes in the 
inventory records could have been verified. Petitioners affirm that 
there is no requirement that the Department inform a respondent, during 
verification, of errors and deficiencies discovered during same.
    Petitioners dispute Union's contention that the Department's 
preliminary decision to use BIA was arbitrary because it was based on a 
comparison of Union's recordkeeping practices with those of Dongbu. 
Petitioners find this ``strange,'' since in its case brief, Union 
itself compared its recordkeeping practices to those of other 
respondents in non-flat-rolled-steel cases in an attempt to demonstrate 
the validity of its records. As to Union's contention that, in fact, 
its recordkeeping practices differ little from Dongbu's, petitioners 
point out that Union officials or counsel were not present at Dongbu's 
verification, that Dongbu never asserted (as Union did) that it was 
incapable of tracing production to shipment, that it was able to show 
certain production records to the Department, and that Dongbu had not 
destroyed all of its home-market production records relating to the 
POR.
Department's Position
    We disagree with petitioners that the Department should have 
applied total BIA to Union. The Department applies total BIA when a 
respondent refuses to provide the information requested in a timely 
manner or in the form required, or otherwise significantly impedes a 
proceeding. See Antifriction Bearings (Other than Tapered Roller 
Bearings) and Parts from France, et al.; Final Results of Antidumping 
Administrative Reviews, 60 FR 10900, 10908 (February 28, 1995), Allied-
Signal Aerospace Co. v. United States, 996 F.2d 1185 (Fed. Cir. 1993); 
NTN Bearing Corp. of America v. United States, Slip Op. 93-

[[Page 793]]

 129 (CIT July 13, 1993). The Department considers the errors and 
inconsistencies in Union's submission to be of such a nature that they 
do not warrant the use of BIA, as discussed below. With respect to U.S. 
date of sale discrepancies, we agree with respondent that this has 
already been addressed in the preliminary results by using date of 
shipment as date of sale.
    We agree with respondent that the cases cited by petitioners 
regarding the destruction of records are not applicable to this 
instance. In Krupp Stahl AG v. United States, 822 F. Supp. 789 (CIT 
1993), for instance, respondent purposefully destroyed all records for 
the POR, making it impossible for them to respond to our questionnaire 
or enable us to verify any submitted information. That is not the case 
with Union. Following its normal procedures, Union did not retain mill 
certificates or other documents needed to verify home-market product 
characteristics. However, all other documentation was maintained and 
there is no evidence that respondent's failure to retain certain 
records was intended to impede our ability to conduct this proceeding.
    Although we reassert our determination that applying only partial 
BIA to Union is warranted, after analyzing all comments received and 
re-evaluating the information on the record, we are modifying our 
application of partial BIA compared to the preliminary results. Because 
Union's reported home-market product characteristics were not 
verifiable, it was not possible for the Department to make reliable 
price-to-price comparisons. Such deficiencies may warrant the use of 
total BIA in many circumstances. In this particular case, however, the 
Department has concluded that the use of total BIA is unwarranted for 
the following reasons:
     Union's normal business practice at the time was not to 
retain certain production records, such as mill certificates;
     there is no evidence on the record that Union deliberately 
refrained from retaining those records with the purpose of impeding the 
Department's ability to conduct this proceeding;
     we were able to verify product characteristics of the 
merchandise sold in the U.S. market and to link specific U.S. sales to 
control numbers; and
     CV was associated with specific control numbers.
In light of the above, and because the Department is treating Union and 
DKI as a single producer of certain cold-rolled carbon steel flat 
products for purposes of this review, we determined to use DKI's home-
market sales and our usual below-cost sales test as bases for 
comparison in cases where U.S. sales by Union were matched to similar, 
contemporaneous sales by DKI in the home market. Where Union's U.S. 
sales could not be matched to similar, contemporaneous DKI transactions 
in the home market, or where such DKI transactions failed the below-
cost test, we determined that basing FMV on CV, in accordance with 
section 773(a)(2) of the Act, was warranted. While we were able to 
match all of Union's U.S. sales to similar, contemporaneous, DKI 
transactions in the home market, all of these DKI transactions were 
below cost, which caused CV to be used as the basis for FMV in all 
instances.
    Section 773(e)(1)(B)(ii) of the Act requires that, as a component 
of CV, an amount for profit shall be used that is equal to that usually 
reflected in the sales of the merchandise made by producers in the 
country of exportation, except that the amount of profit shall not be 
less than 8 percent of the sum of such general expenses and cost. In 
this instance we were unable to determine the actual amount of Union's 
profit because the profit component of Union's reported CV data is 
derived from Union's home-market COP database, which, as we explained 
above, is not usable because we could not verify Union's home-market 
sales product characteristics. Because these product characteristics 
could not be verified, we were unable to match specific sales to 
specific costs; thus, it was not possible to determine the actual 
profit for specific products based on a transaction-by-transaction 
analysis. Consequently, because of this failure of verification, the 
Department, pursuant to section 776(c) of the Act, resorted to the use 
of BIA in order to determine the profit component to be used in 
calculating CV. As partial BIA, we have used the higher of the 
weighted-average profit for all of Union's above-cost home-market sales 
or the statutory 8 percent profit.
    In order to determine which of Union's sales were made at prices 
above the COP, we calculated a simple average of all COPs reported by 
Union. We were unable to calculate a weighted-average COP because we 
could not link Union's COP database to individual home-market sales as 
Union's home-market sales product characteristics could not be 
verified. After calculating the simple average COP, we compared that 
cost to each individual home-market sale to determine which sales were 
made at prices above the average COP.
    Once we had determined which home-market transactions were made at 
prices above the simple average COP, we calculated the transaction-
specific profit on each of those sales. This was done by first 
calculating the sales value of each individual home-market transaction 
(i.e., net price times sales quantity). From each sales value we 
subtracted the value of the COP for that particular transaction to 
determine the transaction-specific profit (i.e., sales value minus 
simple average COP times sales quantity). Finally, we weight-averaged 
the transaction-specific profits for purposes of deriving an overall 
profit percentage for use in the CV calculation. We were able to 
weight-average profit because we verified the quantities and prices of 
Union's individual home-market sales transactions.
    Given Union's home-market data deficiencies, we determined that 
this approach was a reasonable means to calculate the profit component 
of CV. We used as much of Union's verified data as possible. However, 
where verified data were not available, we resorted to partial BIA, 
still using Union's data but in a more adverse manner than if the data 
in question had not failed to verify. We concluded that adopting this 
partial BIA approach, rather than using the statutory minimum profit, 
comported with the statute, the Department's practice, and with Court 
precedent. As the Department has previously noted, ``the noncomplying 
respondent cannot find itself in a better position as a result of 
failing to comply with the Department's information request than had 
the respondent provided the Department with complete, accurate and 
timely data.'' Replacement Parts for Self-Propelled Bituminous Paving 
Equipment From Canada; Final Results of Antidumping Duty Administrative 
Review, 56 FR 47451, 47453 (September 19, 1991). See also National 
Steel Corp., et al. v. United States, 870 F. Supp. 1130, 1135 (CIT 
1994) (approving use of adverse partial BIA when only part of the 
submitted information is deficient). Because the calculated weight-
averaged profit was lower than 8 percent, however, we used the 
statutory minimum profit for CV purposes.
    In any future review of this order, however, the Department expects 
Union to retain any and all records, including production records, 
necessary to permit the Department to verify Union's home-market 
product characteristics.
    Union argued that use of even partial BIA by the Department was 
inappropriate for the following reasons. Union claimed that the 
difficulty in verifying home-market product characteristics was limited 
to those defined by the internal product code,

[[Page 794]]

which is only partially correct. The internal product code did serve as 
the basis for categorizing many of the cold-rolled model-match 
variables; however, it was the basis for a majority of the variables, 
rather than just the five referenced by respondent. In fact, five of 
the six most important variables in the model-match hierarchy were 
derived from the internal product code, and Union's methodology for 
categorizing an additional variable (yield strength) on specific sales 
was not explained to the Department. Since Union did not maintain 
records of any correspondence with its home-market customers prior to 
shipment indicating the product being sought, and the description of 
products sold in the home market and appearing on the commercial 
invoices was only the internal product code, with the exception of 
thickness and width, the Department could not verify that the product 
code represented an accurate reflection of the product sold and 
shipped. The fact that Union did not preserve production records for 
its home-market sales, such as mill certificates, which would provide 
this detailed information on products produced and which would link 
these products to specific sales, prevented the Department from 
determining the accuracy of this system.
    With respect to Union's claims that the Department relies on 
commercial documentation, such as invoices and sales ledgers, to verify 
internal product codes, we note that Union's invoices--unlike those for 
many companies--do not contain a detailed product description of the 
product sold. Neither did Union maintain any customer correspondence or 
any documentation which contained such a detailed product description. 
With respect to the cases cited by Union, we note that the reference in 
Brass Sheet and Strip from Canada was not relevant to verifying product 
characteristics as it involved a volume and value trace. The reference 
to Small Diameter Circular Seamless Carbon and Alloy Steel Standard, 
Line and Pressure Pipe from Brazil and Germany was also not relevant to 
the present case, as that case involved the use of industry-wide 
product codes. No such claim was made by Union; indeed Union 
consistently referred to its codes as ``internal'' product codes.
    Union also alleges that the internal product code was the same as 
that used for U.S. sales and the Department was able to verify its 
accuracy. Products sold in the United States, however, had commercial 
invoices with detailed descriptions of the product sold, and the 
necessary mill certificates that could be used to confirm these product 
descriptions. In addition, products sold in the two markets possess 
different physical and mechanical characteristics, are made to 
different specifications, and are coded differently in the internal 
product code.
    We note that Union, in its case brief of October 2, 1995 (at 15 et 
seq.), almost seven months after the verification and five months after 
the sales verification report (``SVR'') was issued, suggests that the 
Department could have used alternative verification techniques to 
verify Union's home-market product characteristics. If that were true, 
respondent could have suggested these techniques during the 
verification itself, but did not do so. Only the respondent is in a 
position to know what documentary evidence there exists in its 
possession; it is the respondent's responsibility to determine, prior 
to the verification, what documentary evidence exists in its records 
which supports the information previously supplied to the Department, 
and to provide such documentary evidence to the Department's verifiers. 
It is not the responsibility of the Department's verifiers to guess 
what records might be in the respondent's possession and to suggest to 
the respondent how it might best document the information provided in 
the questionnaire responses. We note further that, at verification, 
Union entered as a verification exhibit a consulting report stating 
that Union's production and inventory records are inaccurate. See 
Union's SVR of May 16, 1995, at 10. This calls into question the 
possibility of successfully employing the alternative techniques Union 
is now advocating. Finally, contrary to Union's claim, the Department 
did not examine at verification post-POR mill certificates as well as 
``factory inspection cards'' for certain home-market sales within the 
POR.
    Union's assertion that its recordkeeping practices do not differ 
significantly from Dongbu's is also incorrect. Dongbu, like most other 
parties in these flat-rolled steel proceedings, did maintain mill 
certificates on at least some of its home-market sales during the POR. 
Dongbu also retained various customer correspondence containing product 
descriptions. While it is not the Department's practice to mandate that 
respondents keep their records in a particular manner, in this case all 
of this information, as well as any alternative documentation which 
could have served to verify reported product characteristics, was 
lacking for Union, or not brought to the Department's attention.
    We disagree in part with petitioners' assertion that the CV cost 
data are not viable because production quantities were used to allocate 
costs. While it is true that the quantities of each control number sold 
were used to reconcile total costs to respondent's financial 
statements, these quantities were not used to build up individual costs 
by control number. Instead, Union used average material costs based on 
withdrawals from inventory. The weighted-average costs were then 
applied to a specific control number, and therefore, the final 
production quantity of that control number was not relevant. For 
fabrication costs, Union used the pass-through quantities for each 
process to accumulate and allocate costs to a specific control number. 
Again, the final production quantity was not used to allocate costs, 
and therefore, is irrelevant. Thus, we are satisfied that Union's 
method of assigning a cost to a specific control number is reasonable 
and that total costs (i.e., materials, labor, overhead) were allocated 
to either home-market, third-country, or U.S. merchandise.
    We agree with petitioners that the explanation in Union's case 
brief with regard to post-POR records examined by the Department's 
verifiers does not exist anywhere on the record and that the 
verification reports or exhibits do not support that explanation. In 
fact, we had already requested that the parties delete this information 
from their briefs, on the grounds that it was untimely submitted. This 
information, therefore, is no longer on the record.
    As we are not using total BIA, comments regarding the choice of a 
total BIA margin are moot.
Comment 12
    Petitioners contend that Union Steel's submitted COP and CV data 
must be revised to reflect product-specific costs. According to 
petitioners, Union improperly assigned the same cost of manufacturing 
to multiple products in its COP and CV databases when these products' 
physical characteristics differed in yield strength, width, temper-
rolling, annealing, and/or surface finish in its home-market sales 
listing, and differed in thickness tolerance in its U.S. sales listing. 
The petitioners argue that products with such differences in physical 
characteristics are not identical and have distinct production costs. 
For example, producing a product to a smaller tolerance, temper-rolling 
or annealing a product, or adding various surface finishes all require 
further processing and, consequently, entail additional costs. Union, 
therefore,

[[Page 795]]

should not have reported these products as having the same COM. Even 
more troubling, according to petitioners, is the fact that Union 
reported different COMs for certain products possessing identical 
physical characteristics with the exception of width. Thus, to avoid 
any manipulation of cost, the petitioners request that the Department 
adjust Union's cost data to eliminate the distortion caused by 
inappropriate cost allocations.
    Union contends that its cost data were reported to an appropriate 
degree of specificity. Union states that the petitioners claim is made 
without any substantial support because the Department's hierarchy is 
not based on physical characteristics alone, and that there are no 
reasons to expect any given company to track possible small differences 
in costs that may be associated with different classifications in the 
hierarchy. Additionally, the Department's hierarchy classification 
chose to conform to commercial practices rather than production 
characteristics which cause some products to have similar costs of 
manufacturing. Furthermore, Union states the Department thoroughly 
verified product costs by control number and found no discrepancies.
Department's Position
    For the final results, we accepted Union's CONNUM-specific costs. 
We found that Union's cost data were allocated to a sufficient level of 
product detail following the Department's section D questionnaire 
instructions. Following these instructions, it is possible for some of 
Union's control numbers to have identical COMs for products that varied 
only in yield strength and width. Specifically, a product's yield 
strength is based mainly on the carbon content and, to some extent, 
micro alloying elements of the raw-material input. A raw material input 
with a higher carbon level will produce a product with a higher yield 
strength. However, even though raw-material inputs may vary in carbon 
content, their acquisition cost can be identical. Additionally, Union 
weight-averaged its raw materials based on characteristics of the 
material other than the carbon content (i.e., commercial quality, 
drawing quality, and ASTM grade). Hence, it is possible for some of 
Union's products that are in different strength bands to have no cost 
differential. As for petitioners' concern that the cost of 
manufacturing should differ for products with different width, we are 
satisfied that the respondent reasonably allocated costs associated 
with width differentials. For certain types of cost, Union used 
processing times to allocate fabrication costs by deriving an average 
cost. This average cost was then applied to specific control numbers. 
Therefore, due to this averaging it is possible for identical products, 
with the exception of width, to have the same cost of manufacturing.
Comment 13
    Petitioners contend that the conversion factor used by Union to 
convert home-market sales of sheet reported in theoretical-weight terms 
to actual-weight terms was flawed, because Union was unable to document 
the basis for its formula at verification and because the formula, by 
Union's own admission, was based on incomplete data covering only a 
portion of the POR. Petitioners suggest instead that the Department 
apply a conversion factor derived from the lowest ratio experienced by 
Union on the basis of information on the record.
    Respondent counters that the Department was able to verify the 
theoretical-to-actual weight conversion factor. Union states that the 
sales verification report was inaccurate on this point, and that it 
explained the nature of the discrepancy immediately following the 
issuance of the report.
Department's Position
    Because none of Union's home-market sales were used in our FMV 
calculations, and all of DKI's sales were in coil (rather than sheet) 
form, this comment is moot.
Comment 14
    Petitioners argue the Department should deny Union's claimed 
circumstance-of-sale adjustment for inventory carrying costs, since 
during verification Union prevented the Department's staff from 
actually examining the area in the mill where the physical inventory is 
stored. Petitioners claim that allowing the claimed adjustment would 
only reward Union's obstructiveness.
    Respondent retorts that these costs were fully verified. Union 
notes that it does not have a distinct warehouse for finished goods, 
and the verification team did examine inventory areas at the mill.
Department's Position
    We disagree with petitioners. During the sales verification, the 
Department's verifiers were given to understand that there was a 
separate area in Union's mill dedicated to storing inventory, but did 
not in fact see this area, despite their request to do so. The cost 
verifiers, however, ascertained that steel coils were being stored on 
the mill floor. The Department also verified Union's calculation of 
inventory carrying costs and traced the figures to Union's accounting 
records. Accordingly, there is sufficient information on the record in 
support of this adjustment.
Comment 15
    Petitioners contend that in calculating Union's USP, the Department 
must deduct actual countervailing and antidumping duties when they are 
paid by the respondent or related parties because (1) the plain 
language of the statute requires this conclusion; (2) court decisions 
are also consistent with this conclusion; and (3) the record evidence 
demonstrates that Union America (``UA'') is paying for countervailing 
and antidumping duties on behalf of Union's U.S. sales and that those 
costs are included in the price to the first unrelated party.
    With respect to the first point, petitioners cite section 772(d)(2) 
of the Act, which provides in relevant part that ``the purchase price 
and the exporter's sales price shall be * * * reduced by--except as 
provided in paragraph (1)(D), * * * United States import duties, 
incident to bringing the merchandise from the place of shipment in the 
country of exportation to the place of delivery in the United States'' 
(19 U.S.C. Sec. 1677a(d)). Antidumping and countervailing duties are 
plainly import duties ``incident to bringing the merchandise from the 
place of shipment in the country of exportation to the place of 
delivery in the United States.'' The language of the statute does not 
indicate that antidumping and countervailing duties are to be excluded 
from the phrase ``import duties.'' Moreover, petitioners say, when this 
provision is read in conjunction with section 772(d)(1)(D) of the Act, 
the conclusion that antidumping and countervailing duties constitute 
``import duties'' under section 772(d)(2)(A) is inescapable. Section 
772(d)(1)(D) provides that USP shall be increased by the amount of any 
countervailing duty imposed to offset an export subsidy. By including 
the phrase ``except as provided in paragraph (1)(D)'' in section 
772(d)(2)(A), the drafters clearly understood the subsection's 
reference to ``import duties'' as including countervailing duties 
imposed to offset an export subsidy. This exception was necessary to 
ensure that the statute was consistent with Article VIpara. 5 of the 
GATT, which prohibits the assessment of both antidumping and 
countervailing duties to compensate for the same cause of unfairly low-
priced imports, whether by dumping or as a result of an export

[[Page 796]]

subsidy. Had the exception not been inserted, an amount would be added 
to USP by section 772(d)(1)(D) and deducted by section 772(d)(2)(A). 
Therefore, petitioners believe, Congress contemplated that antidumping 
and countervailing duties were to be treated as ``import duties'' and 
deducted from USP.
    With respect to the second point, petitioners argue that the 
Department must also deduct the cost of antidumping duties equal to the 
amount of the calculated margin for the period being reviewed. In 
Federal-Mogul Corp. v. United States, 813 F. Supp. 856, 872 (CIT 1993), 
according to petitioners, the court recognized that section 
772(d)(2)(A) of the Act requires the Department to deduct any import 
duties that can accurately be determined at the time the Department is 
calculating the current dumping margins. In this case, once the final 
results are issued, Union's antidumping duties will actually be 
determined. Therefore, petitioners urge the Department, in its final 
results, to deduct the difference between FMV and USP (i.e., the actual 
duty amount) from USP before the final margin is calculated.
    With respect to the third point, petitioners cite the verification 
report as evidence that UA is incurring the cost of antidumping and 
countervailing duties on behalf of Union, and that those costs are 
passed on to the first unrelated purchaser in the United States.
    Petitioners state that the Department must deduct the full amount 
of the countervailing duties paid by UA for those entries covered by 
the first administrative review of the countervailing duty order on the 
subject merchandise. Since no party requested a review of this order, 
those duties have become final and they represent a calculable cost to 
Union apart from the payment of the estimated antidumping duty deposit. 
Therefore, petitioners claim, the payment of countervailing duties must 
be treated as actual import duties for purposes of calculating Union's 
dumping margin.
    Union replies that the Department has repeatedly rejected the 
notion of treating AD/CVD duties as expenses to be deducted from U.S. 
price. Union adds that, in Antifriction Bearings (Other Than Tapered 
Roller Bearings) and Parts Thereof from France, et al.; Final Results 
of Antidumping Duty Administrative Reviews, Partial Termination of 
Administrative Reviews, and Revocation in Part of Antidumping Duty 
Orders, 60 FR 10900 (February 28, 1995), the Department stated as 
follows:

    We agree with respondents that making an additional deduction 
from USP for the same antidumping duties that correct for price 
discrimination between comparable goods in the U.S. and foreign 
markets would result in double-counting. Thus, we have not deducted 
antidumping duties or antidumping duty-related expenses from ESP in 
this case.

Union states that the Department disagreed with petitioners' claim that 
antidumping duties constitute a selling expense, and notes that the 
Department's practice has been upheld by the courts. Finally, Union 
denies that the intent of Congress has been that AD/CVD duties be 
deducted from USP, citing the Statement of Administrative Action that 
accompanied the URAA that the law ``is not intended to provide for the 
treatment of antidumping duties as a cost.''
Department's Position
    We agree with respondent. See DOC Position to Petitioners' Comment 
7 supra.
Comment 16
    Because on three separate occasions the Department requested 
information from Union regarding its early-payment discount policies 
for U.S. customers, and Union failed to provide the requested 
information, petitioners argue that the Department should adopt BIA 
with respect to those discounts. Petitioners suggest, as a reasonable 
adverse inference, that the Department assume that Union granted an 
early-payment discount on any transaction where payment was received 
before the due date.
    Union claims that it was fully responsive to the Department with 
regard to information about this discount and that it was fully 
verified. Union states that its discount ``policy'' does not matter; 
all that matters is that it did extend early-payment discounts, that it 
did report them, and that they were verified.
Department's Position
    We agree with respondent. Although Union did not explain its policy 
with respect to early-payment discounts in the U.S. market, the 
Department was able to ascertain that Union in fact extended certain 
early-payment discounts, and to verify to its satisfaction the amount 
of such discounts. See Union's SVR of May 16, 1995, at 33.
Comment 17
    Petitioners argue that the Department must revise Union's reported 
G&A expenses to account for expenses incurred by the Dongkuk Steel Mill 
(``DSM'') group as a whole. As part of its decision to collapse Union 
and DKI, the Department determined that neither Union nor DKI operates 
as a single independent entity, but rather as interrelated entities 
both under the control of the Chang family through its ownership in 
DSM. In prior cases, the Department has adjusted a respondent's 
submitted data to include an allocated portion of the parent company's 
expenses. The record in this case, petitioners assert, clearly 
indicates that expenses were incurred at the headquarters or DSM group 
level (e.g., chairman's salary, group product brochures, group training 
center, and personnel welfare center, office costs, security expenses, 
entertainment expenses, etc.).
    Since Union failed to furnish complete information regarding these 
expenses, petitioners argue that the Department should, as BIA, 
increase Union's calculated G&A expense by the ratio of all G&A 
expenses incurred at DSM over the consolidated DSM group's cost-of-
sales.
    Union contends that the Department should reject the petitioners 
proposed combination of DSM's and Union's G&A expenses. Union argues 
that there is no parent-subsidiary relationship between the two 
entities and that there are no DSM general expenses to attribute to 
Union's activities. Union also counters that Dongkuk Steel Mill was a 
respondent in the 1993 antidumping investigation of Certain Cut-to-
Length Carbon Steel Plate from the Republic of Korea, and in that case 
the Department concluded that Dongkuk Steel Mill's G&A expenses were 
appropriately allocated to Dongkuk Steel Mill's activities and not to a 
group.
Department's Position
    We disagree with petitioners. For the final results, we did not 
combine Dongkuk Steel Mill and Union's general and administrative 
costs. It is the Department's normal practice to include a portion of 
the G&A expense incurred by affiliated companies on the reporting 
entity's behalf in total G&A expenses for COP and CV purposes. However, 
in this specific case, we did not identify any allocable parent company 
costs after reviewing the information on the record. See, e.g., Final 
Determination of Sales at Less Than Fair Value: Small Diameter Circular 
Seamless Carbon and Alloy Steel, Standard, Line and Pressure Pipe from 
Italy, 60 FR 31981, 31992 (June 19, 1995); Final Determination of Sales 
at Less Than Fair Value: Welded Stainless Steel Pipe from Malaysia, 59 
FR 4023, 4027 (January 28, 1994).

[[Page 797]]

Comment 18
    Petitioners contend that, in contrast with the preliminary results 
in the parallel administrative review of certain corrosion-resistant 
carbon steel products from Korea, the Department's ``model-match'' 
computer program accidentally eliminated the fixed 20 percent BIA 
difmer adjustment with respect to Union's price-to-price sales 
comparisons. Petitioners request that, if the Department does not 
revise its BIA methodology as discussed above, the Department at the 
very least make the cold-rolled model-match program conform with the 
corrosion-resistant model-match program in order to ensure that the BIA 
difmer methodology is correctly applied.
    Union counters that petitioners are themselves in error when they 
claim the Department's model-match program contains an error. Union 
believes that the lines questioned by petitioners set the limits on 
permissible matches in the home market. Without them, any given U.S. 
sale could be matched to any home-market sale, which was clearly not 
the Department's intention in the preliminary results. Union states 
that the Department's preliminary methodology was to set Union's difmer 
at 20 percent for the margin calculation, but only after a proper model 
match had been conducted to exclude comparisons resulting in a difmer 
of more than 20 percent. The model-match program exactly reflects that 
intention, according to Union.
Department's Position
    This comment is rendered moot as the Department is applying a 
different partial BIA methodology, which does not comprise a flat 20 
percent difmer adjustment, for purposes of these final results. Where 
DKI sales are used as a basis for comparison, the Department is using 
the difmers reported by DKI, capped at 20 percent of the cost of 
manufacture of the U.S. product, which is the Department's usual 
practice. Because none of DKI's above-cost home-market sales were 
similar to any of Union's U.S. sales, the Department based FMV on CV 
(see the Department's response to Petitioners' Comment 11 supra).
Comment 19
    Petitioners assert, and Union concurs, that, although the 
Department correctly created a new, ``other'' thickness tolerance 
category to account for home-market sales by DKI, it failed to adjust 
the numerical weighting factors associated with Union's U.S. sales to 
conform with the weighting factors associated with DKI's home-market 
sales, thereby making it impossible for any home-market sale to be 
considered an identical match to a U.S. sale, even though the home-
market product may in fact be identical. This error also allegedly 
undermines the accuracy of the selection of most similar home-market 
matches. Petitioners and Union request that the Department correct this 
ministerial error in the model-match program for purposes of the final 
results.
    Union adds that code 16, created for the new, ``other'' thickness 
tolerance category, should be corrected to one of the other codes, if 
necessary on a sale-by-sale basis. Otherwise, the problem identified by 
petitioners remains, in that ``identical'' products are not compared. 
Union presumes that the Department did not intend for all DKI sales to 
be within a single hierarchy category differentiated from those already 
defined. If the Department were to modify its model-match hierarchy to 
make DKI sales a category unto themselves, Union argues, the Department 
would need to explain its reasons for such a change.
Department's Position
    We agree in part with both petitioners and respondent. For purposes 
of these final results, we have harmonized the format of the numerical 
weighting factors for thickness tolerance in the model-match programs, 
thereby insuring that the program will function as intended. In 
addition, we have coded DKI material so that it most closely 
approximates half-mill-tolerance material produced and sold by Union in 
the U.S. market. The necessity for this additional thickness-tolerance 
category (``16'') arises from differences in thickness tolerance 
between Union's ``standard'' and ``half-mill'' material and that of 
DKI.
    Because the only price-to-price comparisons we are making for 
purposes of these final results are those involving home-market sales 
by DKI, none of which are identical in physical characteristics to any 
U.S. sale by Union, petitioners' comment regarding the impossibility 
for any home-market sale to be considered an identical match to a U.S. 
sale is moot. By harmonizing the format of the weighting factors, DKI 
sales of similar, contemporaneous merchandise will now be matched to 
U.S. sales by Union, as the Department originally intended.
    Union's presumption that the Department did not intend for all DKI 
sales to be within a single hierarchy category differentiated from 
those already defined is, in fact, incorrect. DKI reported only one 
thickness tolerance, which it categorized as ``standard,'' but provided 
no record evidence of any thickness-tolerance differences that may have 
existed during the review period. It was, and still is, the 
Department's intention to modify its model-match hierarchy to make DKI 
sales a category unto themselves. As the Department stated in its 
preliminary sales analysis memo dated September 21, 1995,

    We disagree, however, with DKI's categorization of its thickness 
tolerances as ``standard.'' Based on the Department's model-matching 
criteria, we have concluded that, DKI's thickness tolerances are 
much closer to U.S. ``half-mill'' tolerances than to Union's 
``standard'' tolerances. We have therefore created a new category of 
thickness tolerance--called ``other''--for DKI, permitting the 
comparison of Union's U.S. sales of ``half-mill'' to DKI's home-
market sales.

Since the verification, Union has not submitted any record evidence 
that would lead the Department to change its analysis. Therefore, we 
have maintained the new, ``other'' thickness tolerance category (coded 
``16'') in the model-match program.
Comment 20
    Petitioners allege that section 2 of the Department's margin 
calculation program regarding Union accidentally created additional 
U.S. observations, or ``clones,'' which were inadvertently included in 
the Department's analysis. The problem arises when two products are 
sold in the home market that are equally similar to the comparison U.S. 
product. In such cases, the program weight-averages the prices of the 
two home-market products and calculates a single transaction specific 
margin (``UMARGIN'') by comparing that weighted-average home-market 
price to the U.S. price. However, where one of the equally similar 
home-market products fails the cost test, but the other does not, the 
program inadvertently calculated two-transaction specific margins using 
the same U.S. sale. Specifically, for the same U.S. transaction, the 
program calculated one price-to-price margin using the weighted average 
home-market price of the equally similar product that does not fail the 
cost test, and another price-to-CV margin to account for the equally 
similar product that failed the cost test. The net effect of this 
inadvertent programming error is to reduce Union Steel's calculated 
margin. Petitioners therefore request that the Department correct this 
ministerial error and eliminate the second transaction specific price-
to-CV margin for purposes of the final results.

[[Page 798]]

    Union agrees with petitioners with regard to the problem but not to 
the solution. According to Union, the Department's rule in cases in 
which there are two equally similar products is to use an average of 
both in the calculation of FMV, regardless of the basis of computation 
for FMV. If the Department incorrectly calculated separate margins with 
respect to each of the home-market products where one of the products 
was below cost, Union argues, then to remedy this error the Department 
should average the two FMVs.
    Petitioners, according to Union, would have the Department change 
its policy and base its margin calculation only on the price-based FMV, 
without providing any compelling reasons to do so. Indeed, Union 
asserts, the Department has a well established policy of using the most 
similar product comparisons, regardless of whether the basis for FMV is 
price or CV. Ironically, Union avers, for years respondents have argued 
that the Department not rely on CV when a similar home-market product 
would permit a price comparison--but U.S. producers have steadfastly 
opposed such a notion, and the Department has consistently sided with 
the latter. In this instant case, Union contends, the Department's 
policy leads to two equally similar comparison products, and consistent 
with its policy, the Department should average the two FMVs.
Department's Position
    We agree with petitioners and have fixed this programming error for 
the final results. It is the statutory preference to calculate FMV 
based on home-market sales rather than CV. As noted in the Department's 
position on Comment 11, it is our preference based on the facts of this 
case to match U.S. sales to DKI's home-market sales whenever there are 
appropriate matches. Accordingly, in any instances in which there are 
equally similar comparison products, and certain of these comparisons 
would result in using FMV based on a DKI price-to-price comparison and 
others would result in FMV based on CV, we have chosen the match or 
matches based on price-to-price comparisons.
Comment 21
    Petitioners claim that the Department should treat Union's U.S. 
sales through UA as ESP transactions for purposes of the final results. 
Petitioners base this claim on three broad reasons: (1) Union's U.S. 
sales through UA do not meet the statutory definition of purchase-price 
transactions; (2) the limited factual information on the record only 
supports a conclusion that the subject sales are ESP transactions; and 
(3) declarations made on Customs form 7501 clearly indicate that UA is 
the purchaser of the imported merchandise.
    In determining whether a U.S. sales transaction meets the statutory 
definition of purchase price, the Department looks at whether (a) the 
merchandise was shipped directly from the manufacturer to the first 
unrelated purchaser in the United States, without being introduced into 
the inventory of the related shipping agent; (b) direct shipment from 
the manufacturer to the unrelated parties was the customary commercial 
channel for sales of the merchandise between the parties involved; and 
(c) the related selling agent in the United States acted only as a 
processor of sales-related documentation and a communications link with 
the unrelated U.S. buyers. Petitioners claim that the first two factors 
may be indicia pointing to the conclusion that sales took place in a 
foreign country for exportation to the United States, but are not 
dispositive of the issue. In the steel industry, petitioners contend, 
these factors are not informative because most international shipments 
are shipped directly to the customer and not carried in inventory. 
Therefore, even if the merchandise is shipped directly to the customer 
and not placed in inventory in the United States, more evidence is 
needed to conclude that a sale is a purchase-price transaction, 
according to petitioners. Under the circumstances, they argue, the 
focus must be on the third factor of the Department's test.
    Petitioners contend that the record evidence demonstrates that UA 
acts as more than a mere processor of sales-related documentation on 
behalf of Union's U.S. purchasers. They state that UA is involved in 
the following activities: the arrangement and payment for warehousing 
expenses on U.S. sales; the financing of U.S. sales; and the hiring of 
commission agents and entrance into commission arrangements with same. 
Petitioners state that UA reported substantial inventories of steel 
products in 1993, and that UA will, for certain warranties, 
independently authorize a compensatory cash discount without contacting 
Union. Petitioners further emphasize that:
     UA has the authority to grant rebates;
     UA is engaged in advertising on behalf of Union;
     UA assumes the seller's risk pursuant to the terms of the 
invoices issued to U.S. customers;
     UA is the carrier of Union's marine insurance policy and 
pays the premium for that insurance;
     UA is the importer of record and pays U.S. duties, 
brokerage, and handling on U.S. sales;
     UA pays Union the transfer price for the merchandise and 
in turn is paid by the U.S. customer, thereby bearing the risk of non-
payment by U.S. customers; and
     UA takes title to the merchandise at the time it is loaded 
in Korea.
Petitioners assert that UA repeatedly declared on Customs form 7501 
(``Entry Summary'') that it purchased the merchandise. Therefore, the 
transaction between Union and UA is a purchase ``for export to the 
United States,'' so that the transactions between UA and its unrelated 
purchasers are necessarily sales ``in the United States'' meeting the 
definition of ESP transactions, in petitioners' view. They add that UA 
entered the merchandise in question for appraisement at its 
``transaction value,'' which is defined as ``the price actually paid or 
payable for the merchandise when sold for exportation to the United 
States.'' If the importer of record (UA) has entered the merchandise at 
the price established between the related parties as the transaction 
value, then by definition the sale was for export to the United States 
and the sale between UA and the first unrelated U.S. purchaser cannot 
also be the sale for export to the United States. It follows, say 
petitioners, that the latter sale must be an ESP transaction.
    Respondent answers that the Department properly treated the vast 
majority of Union's U.S. sales through UA as PP sales. The terms of 
sales are set prior to importation. Union claims that petitioners 
concede that the merchandise in question was shipped directly from the 
manufacturer to the unrelated buyer, without being introduced into 
inventory of the related shipping agent, and direct shipment was the 
customary channel of distribution.
    With regard to whether UA acted only as a processor of sales-
related documentation and a communications link, Union points outs the 
following:
     UA does not warehouse the imported merchandise;
     UA does not sell from inventory;
     UA does not finance U.S. sales;
     UA does not have the authority to authorize a cash 
discount for warranty claims;
     Union sets guidelines for hiring of any commission agents;
     UA does not enter into rebate agreements;

[[Page 799]]

     UA does not engage in any significant advertising on 
behalf of Union;
     Union ultimately assumes the seller's risk pursuant to the 
terms of the invoices issued to U.S. customers;
     UA's procurement of marine insurance is a normal function 
of a related selling agent; and
     UA's role as the importer of record and payment of U.S. 
duties, brokerage, and handling on U.S. sales is a normal function of a 
related selling agent.

Union further states that although UA issues commercial invoices as 
Union's proxy, it merely processes sales-related documentation, Union 
Steel bearing the final responsibility for the transaction. Union notes 
that whether or not UA takes title to the merchandise at the time of 
loading in Korea is irrelevant, since it must take title of the 
merchandise in order to resell it to an unrelated customer in the 
United States. Thus, in respondent's view, Union has strictly limited 
the role of UA to that of a conduit for Union's sales and processors of 
sales-related documentation and these sales should be treated as 
purchase price.
Department's Position
    We agree with respondents. We determined that purchase price was 
the appropriate basis for calculating USP. Typically, whenever sales 
are made prior to the date of importation through a related sales agent 
in the United States, we conclude that purchase price is the most 
appropriate determinant of the USP based upon the following factors: 
(1) the merchandise in question was shipped directly from the 
manufacturer to the unrelated buyer, without being introduced into the 
inventory of the related shipping agent; (2) direct shipment from the 
manufacturer to the unrelated buyers was the customary commercial 
channel for sales of this merchandise between the parties involved; and 
(3) the related selling agent in the United States acted only as a 
processor of sales-related documentation and a communication link with 
the unrelated U.S. buyers. See, e.g., Certain Stainless Steel Wire Rods 
from France: Final Determination of Sales at Less than Fair Value, 58 
FR 68865, 68868-9 (December 29, 1993); Granular Polytetrafluoroethylene 
Resin from Japan: Final Results of Antidumping Duty Administrative 
Review, 58 FR 50343-4 (September 27, 1993). These criteria were first 
developed in response to the Court of International Trade's decision in 
PQ Corporation v. United States, 652 F. Supp. 724, 733-35 (CIT 1987). 
These criteria have also been considered in cases with indirect 
purchase-price transactions involving exporters and their U.S. 
affiliates. See, e.g., Zenith Electronics Corp. v. United States, 
Consol. Ct. No. 88-07-00488, Slip Op. 94-146 (CIT 1994).
    Furthermore, the Department has recognized and classified as 
indirect PP sales transactions involving selling activities similar to 
those of UA's in other antidumping proceedings involving foreign 
manufacturers and their related U.S. affiliates. See, e.g., Final 
Determination of Sales at Less Than Fair Value; Circular Welded Non-
Alloy Steel Pipe from the Republic of Korea, 57 FR 42942, 42950-1 
(September 17, 1992). In the present review, for sales considered to be 
purchase price in the preliminary results we found that: (1) Union's 
sales through UA, its related sales agent in the United States, are 
almost always shipped directly from Union to the unrelated buyer and 
only rarely are introduced into UA's inventory; (2) Union's customary 
channel of distribution is direct shipment, although certain limited 
sales are normally introduced into UA's inventory; (3) UA performed 
limited liaison functions in the processing of sales-related 
documentation and a limited role as a communication link in connection 
with these sales. UA's role, for example, in extending credit to U.S. 
customers, processing of certain warranty claims, limited advertising, 
processing of import documents, and payment of cash deposits on 
antidumping and countervailing duties, is consistent with a purchase-
price classification. These selling services as an agent on behalf of 
the foreign producer are thus a relocation of routine selling functions 
from Korea to the United States. In other words, we determined that 
UA's selling functions are of a kind that would normally be undertaken 
by the exporter in connection with these sales. More specifically, we 
regard selling functions, rather than selling prices, as the basis for 
classifying sales as purchase price or ESP. While in some cases certain 
merchandise sold by Union was entered into UA's inventory, this 
merchandise was sold prior to the importation of the merchandise, but 
not from UA's inventory. When all three of the factors already 
described for sales made prior to the date of importation through a 
related sales agent in the United States are met, we regard those 
selling functions of the exporter as having been relocated 
geographically from the country of exportation to the United States, 
where the sales agent performs them on behalf of the exporter. The 
substance of the transaction or the functions do not change whether 
these functions are performed in the United States or abroad. In this 
case, Union has transferred these routine selling functions to its 
related selling agent in the United States and the substance of the 
transaction is unchanged.

Respondents' Comments

Dongbu: Comment 1
    According to respondent, the Department is required to make an 
additional upward adjustment to USP to account for export subsidies 
subject to countervailing duties. Citing Article VIpara. 5 of the 
General Agreement on Tariffs and Trade (Uruguay Round Agreements Act, 
Pub. L. 103-465, Th. Sec. 101 (approving the Final Act Embodying the 
Results of the Uruguay Round of Multilateral Trade Negotiations, Annex 
1A 1(a)), respondent states that it provides that ``[n]o product * * * 
shall be subject to both antidumping and countervailing duties to 
compensate for the same situation for dumping or export 
subsidization.'' This provision was implemented into U.S. law by 
section 772(d)(1)(D) of the Tariff Act of 1930, amended, 19 U.S.C. 
Sec. 1677a(d)(1)(d). Thus, argues respondent, purchase price and 
exporter's sales price shall be increased by the amount of any 
countervailing duty imposed on the merchandise to offset the export 
subsidy. Respondent also asserts that, during the original LTFV 
investigation of flat-rolled carbon steel products from Korea, the 
Department made upward adjustments to USP of this type. See Final 
Determinations of Sales at Less Than Fair Value; Certain Hot-Rolled 
Carbon Steel Flat Products, Certain Cold-Rolled Carbon Steel Flat 
Products, Certain Corrosion-Resistant Carbon Steel Flat Products, and 
Certain Cut-to-Length Carbon Steel Plate from Korea, 58 FR 37176, 37191 
(1993). Dongbu states that such an adjustment is required both for 
assessment purposes and for purposes of determining the cash deposit 
rate applicable to future entries. As reported in the Final 
Determinations, the level of export subsidies determined in the final 
countervailing duty determination for cold-rolled products was 0.05 
percent ad valorem. Because Dongbu has made deposits reflecting these 
amounts in conjunction with the entries of cold-rolled flat products 
under review in this proceeding, Dongbu claims it is therefore entitled 
to a further adjustment of USP in this amount.
    Petitioners agree with respondent provided that the Department 
fully

[[Page 800]]

implements the statute, which they assert also requires under section 
772(d)(2)(A) of the Act that USP also be reduced by ``(A) except as 
provided in paragraph (1)(D), the amount if any, included in such 
price, attributable to any additional costs, charges and expenses, and 
United States import duties, incident to bringing the merchandise from 
the place of shipment in the country of exportation to the place of 
delivery in the United States' (19 U.S.C. Sec. 1677a(d)). Thus, 
petitioners argue that if the Department adds the amount of the export 
subsidy to USP, it should also treat the remaining part of the 
countervailing duties paid on those shipments as costs, charges and 
expenses, and United States import duties in accordance with the 
statute. Thus, petitioners agree with respondent that the amount of the 
export subsidy should be added to USP, but only if the Department also 
treats the countervailing duties paid on those shipments as costs, 
charges and expenses, and U.S. import duties, as defined by the 
statute. Petitioners conclude by stating that for Dongbu's direct PP 
sales, any export subsidy adjustment should be calculated against the 
reported gross unit price net of any movement charges incurred outside 
Korea, and exclusive of any duty drawback and value-added (``VAT'') 
adjustments. For indirect PP sales, petitioners state that the 
appropriate base for calculating the export subsidy adjustment is the 
entered value of the subject merchandise, which reflects the f.o.b. 
(freight-on-board) foreign port price of the merchandise.
Department's Position
    We agree with petitioners and respondent in their arguments that 
Dongbu is entitled to a 0.05 percent ad valorem adjustment to the USP. 
However, we disagree with petitioners regarding their contention that, 
if the portion of the countervailing duties attributable to export 
subsidies is added to USP, any remaining countervailing duties paid on 
those shipments must also be treated as costs, charges and expenses, 
and United States import duties. As noted earlier in our comments, we 
determined in Certain Hot-Rolled Lead and Bismuth Carbon Steel Products 
from the United Kingdom; Final Results of Antidumping Duty 
Administrative Review (60 FR 44009, 44010--August 24, 1995) that making 
an additional adjustment to USP for the same antidumping duties that 
correct the price discrimination between the U.S. and home markets 
would result in double-counting, and inconsistency with administrative 
and judicial precedent. The same principle applies with regard to 
countervailing duties. Article VIpara. 5 of the General Agreements on 
Tariffs and Trade (``GATT'') provides that ``[n]o product * * * shall 
be subject to both antidumping and countervailing duties to compensate 
for the same situation of dumping or export subsidization.'' Section 
772(d)(1)(D) of the Act implements this provision.
Comment 2
    Respondent argues that the Department erred by including imputed 
inventory carrying expenses in the selling expenses used to calculate 
CV for the preliminary results of this review. Respondent notes that 
the Department included imputed credit expenses and inventory carrying 
expenses in CV, and that while this methodology might be acceptable if 
the comparison were being made to ESP, the inclusion of imputed 
inventory carrying expenses in CV is contrary to long-standing practice 
at the Department when the comparison is being made to purchase price 
rather than ESP sales. Specifically, respondent notes that despite its 
inclusion of inventory carrying expenses in CV for the preliminary 
results of this review, the Department did not make an additional 
adjustment to the interest rate factor for CV to account for interest 
expenses associated with the carrying of inventory. They contend that 
this is contrary to long-standing precedent and leads to double-
counting of inventory carrying expenses. Respondent asserts that should 
the Department improperly include an amount for inventory carrying 
expenses in CV, it must also make an additional adjustment to the 
interest rate factor to account for inventory carrying expenses. The 
proper approach to these errors is to simply exclude imputed inventory 
carrying expenses from the CV calculations consistent with long-
standing practice.
    Petitioners counter that the Department appropriately included in 
CV the sale-specific inventory carrying charges reported by Dongbu, 
whether Dongbu's sales are classified as either PP sales or ESP sales. 
They state that during this review, Dongbu incurs inventory carrying 
costs for home-market sales of subject merchandise, and that it 
reported sales-specific inventory carrying costs in its February 15, 
1995 response. See Letter from Morrison & Foerster to the U.S. 
Department of Commerce, Case No. A-580-814 at 15 and Exh. B-40 (Feb. 
15, 1995). Thus, according to petitioners, the Department included the 
sale-specific inventory carrying costs in CV in the preliminary results 
of this review, and given that the sale-specific amounts reported by 
Dongbu provide the most accurate measure of Dongbu's costs of holding 
subject merchandise in inventory, the Department should continue to use 
the sale-specific inventory carrying charges reported by Dongbu in 
calculating CV for the final results of this review. Petitioners 
further argue that since the Department's practice has been to reduce 
the respondent's reported financing costs by an amount that reflects 
the interest costs associated with holding inventory, the Department 
should revise its calculation of Dongbu's financing costs to eliminate 
the double-counting of inventory carrying charges in CV for the final 
results.
Department's Position
    We agree with respondent. For the final results, we have excluded 
imputed inventory carrying costs from Dongbu's CV calculation, because 
Dongbu reported only PP sales. The Department normally includes 
inventory carrying costs as an indirect expense in cases involving ESP 
transactions. In ESP transactions, the imputed inventory carrying costs 
consist of the cost of financing the inventory from the time the 
merchandise leaves the production line at the factory to the time the 
goods are shipped to the first unrelated customer. To avoid the double 
counting of interest expense, we allow the respondent to offset its CV 
interest expense by the imputed inventory carrying costs. However, the 
Department does not normally include this cost or the related offset in 
PP sales.
Comment 3
    Respondent contends that for purposes of the preliminary results of 
this review, the Department erred by excluding certain adjustments from 
the gross unit price used to determine VAT on home-market sales. 
Respondent argues that although the Department followed the newly 
adopted ``Zenith footnote 4'' methodology for determining adjustments 
to USP for home-market consumption taxes (see Federal Mogul Corp. v. 
United States, 63 F. 3d 1572 (Fed. Cir. 1995)), the Department made an 
error in determining the absolute amount of VAT paid on home-market 
sales where the customer was subsequently granted this adjustment. 
Specifically, respondent notes, the Department improperly calculated 
the amount of VAT paid on home-market sales by applying the statutory 
10 percent rate to a gross unit price net of applicable adjustments 
when, in fact, according to

[[Page 801]]

the Korean law and practice, VAT must be paid on the full gross unit 
price.
    Petitioners argue that in calculating VAT taxes for the preliminary 
results of this review, the Department has appropriately deducted 
certain adjustments from the gross unit price used to determine the tax 
base. According to petitioners, at the time Dongbu's sales transactions 
occur, these adjustments are not known and should therefore not be 
deducted from the tax base at the time of the transaction. They contend 
that although these adjustments may not be deducted from the VAT base 
at the time of sale, it is not clear whether the VAT paid by Dongbu's 
customers ultimately is net of the same adjustments. Petitioners argue 
that if the VAT paid on the amount of the adjustment were not refunded 
to the customer, the effective tax rate incurred by the customer would 
be in excess of the statutory rate of 10 percent; and that the payment 
of these adjustments therefore would be accompanied by a refund of the 
VAT amounts associated with the adjustment.
Department's Position
    We agree with respondent. Dongbu has provided information 
indicating that under Korean law, VAT taxes associated with home-market 
sales are assessed based on the price of goods and services at the time 
of delivery, and that certain adjustments made to the price after the 
goods and services have already been delivered do not result in 
adjustments to VAT taxes already paid.
Union: Comment 1
    Union contends that the Department's decision to collapse Union and 
DKI in the instant review is contrary to the Department's practice. Not 
only does a strong possibility of price manipulation not exist, 
according to Union, but the Department's standard of a strong 
possibility of price manipulation per se violates respondents' right to 
due process. In determining whether two companies should be collapsed, 
the Department should look to evidence of actual manipulation, rather 
than to suspicion or speculation of possible manipulation at an 
unspecified future time. If the Department is concerned about the 
possibility of price manipulation in the future, it should consider any 
evidence of such manipulation in future reviews.
    Petitioners assert that the Department's decision to collapse the 
two entities is entirely consistent with record evidence. Petitioners 
object to Union's statement, for the first time on the record of this 
proceeding, in its case brief that ``the Department's standard of 
`strong possibility of price manipulation' violates respondent's right 
of due process.'' In not one of its four submissions contesting 
petitioners' collapsing request did Union ever claim that the 
Department's standard for collapsing related parties violates 
respondents' due process rights. If anything, petitioners assert, Union 
explicitly endorsed the Department's standard by not contesting it 
directly and addressing each of the four criteria used to ascertain 
whether the standard has been met. Petitioners strongly protest Union's 
eleventh-hour raising of this due process argument nine months after 
the collapsing decision was made and request that the Department 
dismiss it outright. In any event, petitioners maintain, the 
Department's four-point standard is entirely reasonable and has been 
applied by the CIT. See, e.g., Nihon Cement Co. v. United States, Slip 
Op. 93-80 at 48-54 (CIT May 25, 1993). To require parties to 
demonstrate actual price or production manipulation would impose a 
quasi-insurmountable burden on petitioners.
Department's Position
    The Department's practice of collapsing affiliated parties if the 
record evidence indicates a strong possibility of price manipulation is 
longstanding and was upheld by the CIT in Nihon Cement Co. v. United 
States, Slip Op. 93-80 at 48-54 (CIT May 25, 1993). Therefore, Union's 
argument that the Department's test is legally deficient is unfounded. 
Moreover, Union has in no way been denied due process in this 
determination. Throughout the course of this proceeding, Union had 
ample opportunity to submit evidence and arguments with regard to this 
issue. We note that at no time during the period between the 
Department's decision to collapse (May 22, 1995) and the preliminary 
review results (December 15, 1995) did Union ever challenge the 
Department's collapsing test.
Comment 2
    Union claims that the Department erred in (1) concluding that Union 
had understated its U.S. credit expenses by not including bank charges 
therein, and (2) increasing Union's U.S. credit expenses by the amount 
of those charges. In fact, Union maintains, it included its U.S. bank 
charges in U.S. brokerage and handling expenses, so that they were 
double-counted by the Department. In addition, Union claims, the 
Department compounded its error by mistakenly dividing two years' worth 
of interest expenses by 18 months' worth of short-term borrowings.
    Union urges the Department, for purposes of the final results, to 
follow its own practice and treat bank charges as selling expenses. 
Union claims to have reported its bank charges on a sale-by-sale basis, 
which is the most accurate form of reporting. Also, respondent asserts, 
including bank charges in an interest-rate calculation is illogical, 
since a bank charge need not be connected to the time value of money, 
but can simply consist of a flat fee for services rendered.
    Petitioners reply that Union's claims regarding double-counting are 
unsubstantiated. Petitioners note that Union's claims that it included 
transaction-specific bank charges in its reported U.S. brokerage and 
handling expenses is not supported by any sample calculations or 
documents. Petitioners state that it is the Department's practice to 
include bank charges in credit expenses when they are not elsewhere 
reported. Because of the absence of specific data pertaining to bank 
charges alone, petitioners agree that the Department had no alternative 
but to use Union's combined interest and bank charge data for the two 
fiscal years.
Department's Position
    We agree in part with both petitioners and respondent. As there is 
no evidence on the record supporting Union's claims that it included 
bank charges in its reported brokerage and handling expenses, we have 
increased Union's reported credit expenses to account for these bank 
charges. We acknowledge our error, however, in dividing two years' 
worth of interest expenses by 18 months' worth of short-term 
borrowings, and have corrected this error for purposes of these final 
results by prorating the short-term borrowings used in the denominator 
to 24 months.
Comment 3
    Union disagrees with the Department's treatment of its home-market 
warehousing expenses as indirect selling expenses, and contradicts the 
Department's statement that these expenses were evenly allocated 
across-the-board to all home-market sales. In fact, Union affirms that 
all warehousing expenses other than labor were traced to the particular 
areas devoted to subject and non-subject merchandise, because Union 
separates warehouses subject and non-subject merchandise, and thus can 
determine the proportion of warehousing expenses attributable to each. 
Union also maintains that a selling expense is not indirect simply 
because it occurs prior to sale. For these reasons, and because

[[Page 802]]

the warehousing expenses in question are attributable to a later sale 
of the subject merchandise, Union requests that the Department treat 
these warehousing expenses as direct for purposes of the final results.
    Petitioners respond that Union stores three broad, distinct types 
of merchandise in the same warehouse--cold-rolled, corrosion-resistant, 
and pipe products. Petitioners state that Union did not link specific 
warehousing charges to specific sales, but rather allocated costs based 
on the square footage dedicated to each product type and on the total 
quantity of each product type warehoused. Petitioners believe that the 
Department's preliminary results correctly denied Union's claim that 
these expenses be classified as direct.
Department's Position
    We agree with petitioners. Union did not tie warehousing expenses 
to specific sales, but merely allocated them. The amount reported by 
Union on its computer tape for this expense in Korean won is identical 
for all sales transactions where a warehousing expense was claimed, 
regardless of the length of time the merchandise was actually 
warehoused. Therefore, we do not consider these expenses to be direct.
Comment 4
    Union disagrees with the Department's treatment of pre-sale inland 
freight expenses in the home market as indirect. Union argues that the 
Department must examine the facts of each case to determine whether 
warehousing and pre-sale freight are so linked that they must 
necessarily be treated in the same fashion. In the final results of 
redetermination on remand (January 5, 1995) pursuant to The Ad Hoc 
Committee of AZ-NM-TX-FL Producers of Gray Portland Cement v. United 
States, Slip Op. 94-151 (1994), the Department noted that ``warehousing 
and movement expenses are, for analytical purposes, inextricably 
linked'' and ``if pre-sale warehousing is an indirect expense, then, in 
the absence of contrary evidence, pre-sale movement expenses should 
also be treated as an indirect expense.'' Earlier in the case, the 
Court had stated that ``if the pre-sale warehousing expense in this 
case is not shown to be a direct expense, then it follows that the cost 
of transporting the cement to the warehouse is also not shown to be a 
direct expense.''
    Union argues that in this case, pre-sale freight and warehousing 
are not inextricably linked. Union claims that pre-sale freight was 
constant, since the merchandise was moved over the same route for all 
sales. Therefore, each ton sold from the warehouse led to an exactly 
identified increment to costs--the amount of the pre-sale freight--and 
the expense was incurred on a one-on-one basis with each unit of 
subject merchandise sold. Therefore, Union maintains the expense in 
question is clearly direct.
    Petitioners respond that the Department correctly determined that 
Union's pre-sale freight expenses were indirect. Petitioners state that 
the Department's standard is clear: pre-sale warehousing and freight 
expenses are inextricably linked; thus, in the absence of contrary 
evidence, if pre-sale warehousing is an indirect expense, so too must 
be pre-sale freight. Petitioners note that it is always true that each 
ton shipped leads to an additional charge for freight, but this does 
not mean that pre-sale freight is always a direct selling expense.
Department's Position
    In the preliminary review results, the Department stated that it 
``considers pre-sale movement expenses as direct selling expenses only 
if the movement expenses in question are directly related to the home-
market sales under consideration. In order to determine whether pre-
sale movement expenses are direct under the facts of a particular case, 
the Department examines the respondent's pre-sale warehousing expenses, 
since the pre-sale movement charges incurred in positioning the 
merchandise at the warehouse are, for analytical purposes, inextricably 
linked to pre-sale warehousing expenses. If the pre-sale warehousing 
constitutes an indirect expense, the expense involved in getting the 
merchandise to the warehouse must also be indirect. Conversely, a 
direct pre-sale warehousing expense necessarily implies a direct pre-
sale movement expense. We note that, although pre-sale warehousing 
expenses in most cases have been found to be indirect selling expenses, 
these expenses may be deducted from FMV as a circumstance-of-sale 
adjustment in a particular case if the respondent is able to 
demonstrate that the expenses are directly related to the sales under 
consideration.'' See Preliminary Results of Review; Certain Cold-Rolled 
Carbon Steel Flat Products from Korea (60 FR 65284, 65287--December 19, 
1995). The Department is continuing to treat Union's pre-sale home-
market inland freight expenses as indirect, because Union did not 
distinguish between pre- and post-sale warehousing expenses or 
demonstrate that these expenses were directly related to the sales 
under consideration.
Comment 5
    Union argues that the Department should differentiate Union's 
painted products according to specific paint types, because (1) there 
are significant cost, price, and commercial differences among Union's 
painted products; (2) these differences demonstrate that Union's 
customers perceive significantly different applications for such 
products; and (3) if the Department compares different paint types, it 
must make an appropriate difmer adjustment.
    Petitioners state the Department was correct not to revise the 
existing paint categories for the preliminary results of this review 
and should also reject this argument for the final results. Petitioners 
note that Union's arguments do not address the criteria used by the 
Department to establish categories of products and determine whether 
certain products may be compared and are not supported by the record 
evidence. Petitioners state that Union ignores that the primary basis 
for creating product categories is physical characteristics. Thus, 
according to petitioners, the Department can accept Union's proposed 
paint categories only if Union demonstrates that the physical 
characteristics of the various paint types are so dissimilar that the 
paint types cannot be compared--which Union has not done. Petitioners 
cite Koyo Seiko Co. v. United States, Slip Op. 94-1363 at 15 (Fed. Cir. 
Sept. 20, 1995) which states that products with significant physical 
similarities need not be ``technically substitutable, purchased by the 
same types of customers, or applied to the same end use'' in order to 
be compared. Petitioners add that the record does not support Union's 
contention that its different paint types exhibit significant 
differences in cost or price.
    Petitioners reject the notion of making a difmer adjustment for 
differences in paint types. Petitioners state that it is the 
Department's position in these flat-rolled proceedings that it will not 
make adjustments to account for differences between physical 
characteristics of U.S. and home-market products when the products are 
identified by the same control number. If products have the same 
control number, according to petitioners, they are in effect identical 
for purposes of this review and no difmer adjustment should be granted.
Department's Position
    We agree with petitioners. As stated in our memorandum to the file 
of August 10, 1995, discussing our preliminary results of review, Union

[[Page 803]]

provided insufficient and non-compelling information to support the 
necessity for differentiating additional types of painted products. 
Union did not demonstrate how each of the proposed additional paint 
types possesses physical characteristics that are significantly 
different from those of the other proposed paint types, and how each 
paint type is intended for significantly different applications and 
uses. Therefore, we did not create additional paint categories for 
purposes of these final results.
Comment 6
    Union argues that the Department should not combine the financing 
expenses of Union and DKI with those of other member companies of the 
Dongkuk group because this collapsing of interest expense is entirely 
at odds with the Department's practice. Union states that it is the 
Department's established policy to calculate interest expense from the 
costs of borrowing incurred by the respondent and its related parties 
only when the companies are consolidated in the normal course of 
business. Union states that there are two fundamental reasons for this. 
First, the accounting practicality of consolidating different 
companies, particularly with respect to cost of goods sold, demands 
that an audited consolidated statement be generated in the normal 
course of business. Second, the parent into which the subsidiary is 
consolidated is assumed to control the financing decisions of the 
subsidiary. See Final Determination of Sales at Less Than Fair Value; 
Small Diameter Circular Seamless Carbon Alloy Steel, Standard, Line and 
Pressure Pipe from Italy (60 FR 31981, 31990 June 19, 1995).
    In the instant review, according to Union, neither of the two 
standards articulated above applies to Union or DKI. Union states that 
the financial statements of Union and DKI are not consolidated into 
those of DSM. Union also states that no evidence on the record suggests 
that the financial decisions of Union and DKI are controlled by DSM. 
Just because two entities have been collapsed, Union claims, is not 
necessarily a reason to calculate circumstance-of-sale adjustments or 
cost adjustments on a collapsed basis. For example, selling expenses 
are not calculated for the group as a whole, but specifically; the COM 
is calculated on a company-specific basis, unless the collapsed 
entities have identical control numbers, which they do not; general 
expenses reasonably associated with the COM remain company specific. 
Likewise, Union argues, there is no reason to combine interest 
expenses, which are properly allocated on a company-specific basis. 
Union cites to our Final Determination of Sales at Less Than Fair 
Value: Dynamic Random Access Semiconductors of One Megabit and Above 
from the Republic of Korea (58 FR 15467, 15475--March 23, 1993) 
(``Korean DRAMS''), where the financial statements of two companies 
that were members of the same chaebol were not consolidated in the 
normal course of business, and where the Department did not require the 
respondent to submit a combined interest rate. Indeed, Union asserts, 
when the respondent sought to persuade the Department to use the 
interest expense for the group as a whole, the Department rejected the 
idea on the grounds that ``[t]he Department does not perform an audit 
at verification; rather, verification relies on audited records.'' 
Similarly, Union points out that in its Final Determination of Sales at 
Less Than Fair Value: Polyethylene Terephthalate Film, Sheet, and Strip 
from the Republic of Korea (56 FR 16305, 16313 `` April 22, 1991) 
(``Korean PET Film'') the Department held that, absent evidence of 
inter-company production financing arrangements, a respondent's own 
financial statements provide the most accurate picture of its financing 
activities for the production of subject merchandise.
    Additionally, Union states that the Department's calculation of its 
financing factor was incorrect because it failed to offset DKI and 
DSM's financing costs with short-term interest income. The respondent 
argues that the Department's calculation only offset Union's financing 
costs with short-term interest income. Therefore, the Department's 
calculation did not make an appropriate ``apples-to-apples'' 
comparison.
    Petitioners contend that the Department properly combined Union's 
interest expense with the interest expense of other members of the 
Dongkuk group. Petitioners state that this decision is consistent with 
the Department's normal practice because the companies are under common 
control and produce similar subject merchandise. Petitioners contend 
that capital acquisition costs are fungible and that any borrowing by 
Union, DKI, or DSM may be used for a variety of beneficial purposes for 
the group as a whole. Therefore, petitioners believe that the 
Department should continue to use the combined interest expenses of 
Union, DKI and DSM in its calculation for the final results of this 
instant review.
    In fact, petitioners claim, contrary to Union's statements, the 
Department did reduce both DKI's and DSM's reported interest expense by 
each company's respective short-term interest income. Accordingly, the 
Department should simply ignore Union Steel's comments with respect to 
this issue.
    Petitioners also state that the Department deducted an appropriate 
short-term interest income figure in its net financing factor 
calculation. Furthermore, they state that the respondent's argument of 
requiring an apples-to-apples comparison is inappropriate in this 
circumstance because symmetrical results are not necessary in this step 
of the net financing calculation.
Department's Position
    For the final results, we calculated a combined net interest factor 
using Union's, DSM, and DKI's audited financial figures obtained from 
verification exhibits, respondent's submissions and public records. 
This methodology of calculating a single net interest factor is 
consistent with our longstanding practice for computing interest 
expense in cases involving parent-subsidiary corporate relationships. 
DSM's ownership interest in Union and DKI places the parent in a 
position to influence Union's financial borrowing and overall capital 
structure. We note that, contrary to Union's assertions that Union is 
an independent company and not controlled by DSM, the two companies 
share common directors and related stockholders. Based on this 
information, it is difficult to see how Union's operations are 
independent of its parent to such an extent that we should ignore our 
normal practice of computing interest. See Final Determination of Sales 
at Less Than Fair Value; Certain Carbon Steel Butt-Weld Pipe Fittings 
from Thailand (60 FR 10552, 10557--February 27, 1995). Additionally, we 
find it appropriate to collapse the financing costs of these three 
companies in this instant review because we consider that the financing 
costs of the parent and its subsidiaries to be fungible.
    Furthermore, the facts of this instant review differ from both the 
Korean DRAMS and Korean PET Film with regard to combining interest 
expense factors. In Korean DRAMS and Korean PET Film, the respondents 
requested that the Department combine limited brother-sister companies 
to derive a consolidated group-level interest expense factor. In those 
cases, however, the Department determined that a consolidated group-
level interest factor was inappropriate because, while the respondents' 
own financial statements were audited, those of the sister

[[Page 804]]

companies and the group-level financial statements were unaudited. As 
the Department stated in Korean DRAMS, ``[a]bsent detailed testing 
usually associated with an audit, the Department cannot rely on the 
statements as submitted.'' See DOC Position on Comment 24, 58 FR 15475. 
Not only, therefore, would consolidating the entities in question have 
placed an undue burden on the Department to review unaudited 
information, but the respondents' own audited financial statements 
provided the most accurate reflection of the cost of financing the 
production of subject merchandise. In the instant review, by contrast, 
each of the entities in question--Union, DSM, and DKI--prepared 
separate audited financial statements, which we could therefore combine 
to calculate a group-level interest expense factor.
    Additionally, we agree with the respondent in that it is the 
Department's practice to allow a respondent to offset financial 
expenses with interest earned from the general operations of the 
company. See, e.g., Timken v. United States, 852 F. Supp. 1040, 1048 
(CIT 1994). The Department does not, however, offset interest expense 
with interest income earned on long-term investments. See Final 
Determination of Sales at Less Than Fair Value; Small Diameter Circular 
Seamless Carbon and Alloy Steel, Standard, Line and Pressure Pipe from 
Italy (60 FR 31981, 31991--June 19, 1995). Therefore, for the final 
results we offset the combined financing costs by the respective short-
term interest income of the three entities.
Comment 7
    Union argues that the Department should not include the company's 
``special depreciation'' that was reported as an extraordinary item on 
its audited financial statement in the cost of production of subject 
merchandise. Union contends that the Department's established policy 
with respect to this kind of expense is to exclude the cost because it 
relates solely to tax law and represents no real additional cost to the 
company. See Final Determination of Sales at Less than Fair Value; 
Stainless Steel Angles from Japan (60 FR 16608, 16617--March 31, 1995) 
(``Angles''). Therefore, Union believes that the Department should 
follow the precedent established in that determination and remove the 
special depreciation from Union's production costs.
    Petitioners argue that the Department should continue to include 
Union Steel's accelerated depreciation costs in its calculation of the 
company's COP and CV. Petitioners contend the Department does not have 
an established policy of excluding accelerated depreciation as a cost 
of production. To support their argument, petitioners state that in 
recent determination the Department rejected a similar contention made 
by the respondent and included the company's accelerated depreciation 
charges in the calculation of COP and CV. See Final Determination of 
Sales at Less Than Fair Value; Canned Pineapple Fruit from Thailand (60 
FR 29553, 29560--June 5, 1995). Furthermore, petitioners contend that 
the cost should be included in COP and CV because it is reported on 
Union's financial statements that are in accordance with generally 
accepted accounting principles (``GAAP'') in Korea.
Department's Position
    We disagree with the respondent and have included Union's entire 
special depreciation as a production cost for these final results. 
Unlike the situation in Angles, where the respondent company used 
special financial accounting treatment to reflect only its regular 
depreciation (i.e., non-tax depreciation) as a cost in its audited 
income statements for that year, Union recorded the full special 
depreciation charge as a cost in its audited income statement in 
accordance with Korean GAAP. We note that it is the Department's normal 
practice to use costs recorded in normal books and records of the 
respondent unless it can be shown that such costs do not reasonably 
reflect the amounts incurred to produce the subject merchandise. See, 
e.g., Final Determination of Sales at Less Than Fair Value; Oil Country 
Tubular Goods from Argentina (60 FR 33539, 33548--June 28, 1995); High-
Tenacity Rayon Filament Yarn from Germany; Final Results of Antidumping 
Duty Administrative Review (59 FR 15897, 15898--March 28, 1995).

Final Results of Review

    As a result of this review, we have determined that the following 
margins exist for the period August 18, 1993, through July 31, 1994:

             Certain Cold-Rolled Carbon Steel Flat Products             
------------------------------------------------------------------------
                                                              Weighted- 
               Producer/manufacturer/exporter                  average  
                                                                margin  
------------------------------------------------------------------------
Dongbu.....................................................       6.07 %
Union......................................................       1.08 %
------------------------------------------------------------------------

    The Department shall determine, and the U. S. Customs Service shall 
assess, antidumping duties on all appropriate entries. The Department 
shall issue appraisement instructions directly to the Customs Service.
    Furthermore, the following deposit requirements shall be effective 
upon publication of this notice of final results of review for all 
shipments of certain cold-rolled carbon steel flat products from Korea 
entered, or withdrawn from warehouse, for consumption on or after the 
publication date, as provided for by section 751(a)(1) of the Tariff 
Act: (1) the cash deposit rates for the reviewed companies named above 
which have separate rates will be the rates for those firms as stated 
above; (2) for previously investigated companies not listed above, the 
cash deposit rate will continue to be the company-specific rate 
published for the most recent period; (3) if the exporter is not a firm 
covered in this review, or the original LTFV investigation, but the 
manufacturer is, the cash deposit rate will be the rate established for 
the most recent period for the manufacturer of the merchandise; and (4) 
if neither the exporter nor the manufacturer is a firm covered in this 
review, the cash deposit rate for this case will continue to be 14.44 
percent, which is the ``all others'' rate in the LTFV investigation. 
See Final Determination of Sales at Less Than Fair Value: Certain Cold-
Rolled Carbon Steel Flat Products from Korea, 58 FR 37176 (July 9, 
1993).
    Article VIpara.5 of the GATT (cited earlier) provides that ``[n]o 
product * * * shall be subject to both antidumping and countervailing 
duties to compensate for the same situation of dumping or export 
subsidization.'' This provision is implemented by section 772(d)(1)(D) 
of the Act. Since antidumping duties cannot be assessed on the portion 
of the margin attributable to export subsidies, there is no reason to 
require a cash deposit or bond for that amount. Accordingly, the level 
of export subsidies as determined in Final Affirmative Countervailing 
Duty Determinations and Final Negative Critical Circumstances 
Determinations; Certain Steel Products from Korea (58 FR 37328--July 9, 
1993), which is 0.05 percent ad valorem, will be subtracted from the 
cash deposit rate for deposit or bonding purposes.
    The deposit requirements, when imposed, shall remain in effect 
until publication of the final results of the next administrative 
review.
    This notice serves as a final reminder to importers of their 
responsibility under 19 CFR Sec. 353.26 to file a certificate regarding 
the reimbursement of antidumping duties prior to liquidation of the 
relevant entries during this review period. Failure to comply with this 
requirement could

[[Page 805]]

result in the Secretary's presumption that reimbursement of antidumping 
duties occurred and the subsequent assessment of double antidumping 
duties.
    This notice also serves as a reminder to parties subject to 
administrative protective order (``APO'') of their responsibility 
concerning the disposition of proprietary information disclosed under 
APO in accordance with section 353.34(d) of the Department's 
regulations. Timely notification of return/destruction of APO materials 
or conversion to judicial protective order is hereby requested. Failure 
to comply with the regulations and the terms of an APO is a 
sanctionable violation.
    These administrative reviews and notice are in accordance with 
section 751(a)(1) of the Act (19 U.S.C. Sec. 1675(a)(1)) and section 
353.22 of the Department's regulations.

    Dated: December 29, 1997.
Robert S. LaRussa,
Assistant Secretary for Import Administration.
[FR Doc. 98-276 Filed 1-6-98; 8:45 am]
BILLING CODE 3510-DS-P