NOTICES

                        DEPARTMENT OF COMMERCE

                               [C-357-005]

         Cold-Rolled Carbon Steel Flat-Rolled Products From Argentina: Final
   Affirmative Countervailing Duty Determination and Countervailing Duty Order

                          Thursday, April 26, 1984

 *18006

 AGENCY: International Trade Administration, Commerce.

 ACTION: Notice.

 SUMMARY: We determine that certain benefits that constitute bounties or grants within
 the meaning of the countervailing duty law are being provided to manufacturers,
 producers, or exporters in Argentina of cold-rolled carbon steel flat-rolled products.
 The net bounty or grant for each company is identified in the "Suspension of Liquidation"
 section of this notice. We are directing the U.S. Customs Service to continue to suspend
 liquidation of all entries of cold-rolled carbon steel flat-rolled products from Argentina
 that are entered, or withdrawn from warehouse, for consumption, on or after the date of
 publication of this notice and to require a cash deposit on this product in the amount
 equal to the net bounty or grant.

 EFFECTIVE DATE: April 26, 1984.

 FOR FURTHER INFORMATION CONTACT: Barbara Tillman or Kenneth Haldenstein; Office
 of Investigations, Import Administration, International Trade Administration, U.S.
 Department of Commerce, 14th Street and Constitution Avenue, NW., Washington, D.C.
 20230; telephone: (202) 377-0192 or 377-4136.

 SUPPLEMENTARY INFORMATION:

 Final Determination and Order

 Based upon our investigation, we determine that certain benefits constituting bounties or
 grants within the meaning of section 303 of the Tariff Act of 1930, as amended (the Act),
 are being provided to manufacturers, producers, or exporters in Argentina of
 cold-rolled carbon steel flat-rolled products. For purposes of this investigation, the
 following programs are found to confer bounties or grants:
 1/8 Post-financing of exports under Circular OPRAC 1-9
 1/8 Government equity infusions
 1/8 Loans and loan guarantees during the period we consider Somisa uncreditworthy
 1/8 Somisa's purchases of "oil residue coal"
 1/8 Capital tax exemption for Somisa
 1/8 Import duty exemptions.
 We determine the net bounty or grant to be the rate specified for each company in the
 "Suspension of Liquidation" section of this notice.

 Case History

 On November 10, 1983, we received petitions from the United States Steel Corporation
 (U.S. Steel), Pittsburgh, Pennsylvania, filed on behalf of the hot- rolled and cold-rolled
 carbon steel sheet industries. The petition relating specifically to hot-rolled carbon steel
 sheet was subsequently withdrawn. In compliance with the filing requirements of §
 355.26 of our regulations (19 CFR 355.26), the remaining petition alleges that
 manufacturers, producers or exporters in Argentina of cold-rolled carbon steel sheet
 receive, directly or indirectly, benefits constituting bounties or grants within the meaning
 of section 303 of the Act.
 We found the petition to contain sufficient grounds upon which to initiate a
 countervailing duty investigation, and on November 22, 1983, we initiated such an
 investigation (48 FR 55012). We stated that we expected to issue a preliminary
 determination by February 3, 1984.
 Because Argentina is not a "country under the Agreement" within the meaning of section
 701(b) of the Act, section 303 of the Act applies to this investigation. Because the
 merchandise being investigated is dutiable, the domestic industry is not required to
 allege that, and the U.S. International Trade Commission is not required to determine
 whether, imports of this product cause or threaten material injury to a U.S. industry.
 We presented a questionnaire concerning the allegations to the government of
 Argentina in Washington, D.C., on December 16, 1983. On January 16, 1984, we
 received responses to the questionnaire.
 On February 3, 1984, we preliminarily determined that benefits constituting bounties or
 grants within the meaning of the countervailing duty law are being provided to
 manufacturers, producers, or exporters in Argentina of cold-rolled carbon steel
 flat-rolled products (49 FR 5151). Although a hearing had been requested, the request
 was subsequently withdrawn. We received briefs from the parties to the proceeding on
 February 12 and 21, March 4, 9, 15, and 20, and April 2 and 9.

 *18007

 Scope of Investigation

 The products covered by this investigation are cold-rolled carbon steel flat- rolled
 products. In the preliminary determination, we used the term cold- rolled carbon steel
 sheet in describing the products included in this investigation. This is the generic term
 used by the U.S. industry to describe the products under investigation. The more correct
 term is cold-rolled carbon steel flat-rolled products. The complete technical description
 of these products, which has not changed from the preliminary determination, is
 contained in the Product Description Appendix attached to this notice.
 There are two known producers and exporters in Argentina of cold-rolled carbon steel
 flat-rolled products to the United States. We received information from the government
 of Argentina regarding Somisa-Sociedad Mixta Siderurgica Argentina (Somisa) and
 Propulsora Siderurgica Saic (Propulsora), which represented over 85 percent of exports
 of this product to the United States during the period for which we are measuring
 bounties or grants, July 1982 through June 1983.

 Analysis of Programs

 Throughout this notice, we refer to general principles applied to the facts of the current
 investigation. These general principles are described in detail in the Subsidies Appendix
 attached to this notice. For purposes of this final determination, we are calculating
 company-specific ad valorem rates for the bounties or grants received by each company.
 We are calculating campany- specific rates because the two respondents have received
 materially different benefits and there are significant differences in the number and type
 of programs used by each respondent. The period for which we are measuring bounties or
 grants is July 1982 through June 1983, which corresponds to Somisa's and Propulsora's
 most recently completed fiscal year. To calculate ad valorem rates we used
 inflation-adjusted sales figures as of June 30, 1983; therefore, we adjusted the benefits for
 inflation as of the same date.
 As described in the Subsidies Appendix, several programs alleged by the
 petitioner--government provision of equity capital, long-term loans and loan
 guarantees--require an assessment of the producer's "equityworthiness" and
 "creditworthiness" before we can determine if, and in what magnitude, a countervailable
 benefit has been conferred.
 We have consistently held that government provision of, or assistance in obtaining,
 capital or debt does not per se confer a subsidy. Government equity purchases or
 financial backing bestow a countervailable benefit only when they occur on terms
 inconsistent with commercial considerations. To determine if such action is
 commercially unsound, we review and assess financial data for the company in question.
 For this final determination we conducted a comprehensive review, and considered fully
 the factors relevant to a determination of inconsistency with commercial considerations.
 For loans and loan guarantees, we determine whether the company was "creditworthy" in
 the years in which the loans were provided. In making this determination we focus on
 cash flow and other measures of the ability of each company to meet its long-term debt
 obligations.
 With regard to whether a company is a reasonable equity investment (a condition we
 have termed "equityworthiness"), we examined the financial ratios, operating profits or
 losses and other relevant data (e.g. effects of inflation and current market demand) to
 evaluate the company's current and future ability to earn a reasonable rate of return on
 equity investments.
 For purposes of this final determination, we assessed the creditworthiness only of Somisa.
 We did not assess Propulsora's creditworthiness because uncreditworthiness was not
 alleged by petitioners, and because at the time of the government's equity infusion into
 Propulsora, there were also purchases of stock by private subscribers. In assessing
 Somisa's creditworthiness and equityworthiness, we reviewed financial ratios and other
 measurements derived from Somisa's financial statements and accounting records for the
 fiscal years 1968/1969 through 1982/1983. We found the company to be uncreditworthy
 from fiscal year 1978/79 through 1981/82 and to be unequityworthy from 1977/78
 through 1982/83.
 In its response, the government of Argentina provided data for the applicable period
 including financial statements and debt information for both Somisa and Propulsora. In
 addition, Propulsora submitted a supplemental response containing information on
 Propulsora's use of the alleged subsidy programs. After the verification, the government
 provided amended responses on March 20 and 26 and April 3. Based upon our analysis of
 the petition, the responses to our questionnaire and our verification, we determine the
 following:

 I. Programs Determined To Confer Bounties or Grants

 We determine that bounties or grants are being provided to manufacturers, producers, or
 exporters in Argentina of cold-rolled carbon steel flat-rolled products under the
 following programs.

 A. Post-Financing of Exports Under Circular OPRAC 1-9. On September 24, 1982, the
 Central Bank of Argentina established a post-financing program for exports under
 Circular OPRAC 1-9. OPRAC 1-9 loans are granted for up to 30 percent of the peso
 equivalent of the foreign currency in which the export transaction was paid. The term of
 the loan is 180 days. The interest rate charged on OPRAC 1-9 loans is the regulated rate
 used by commercial banks, as required under Central Bank Regulations. Both Somisa and
 Propulsora received OPRAC 1-9 loans.
 The Subsidies Appendix states that the benchmark for short-term loans is the national
 average commercial rate. In the preliminary determination, we used as the benchmark
 the unregulated interest rates for short-term commercial bank loans published in the
 "Indicadores de Coyuntura" (Current Economic Indicators) by the Fundacion de
 Investigaciones Economicas Latino Americanos (FIEL). We used the unregulated rates
 because at that time, based on best information available, we considered them a better
 reflection of average commercial rates than the regulated commercial bank rates.
 However, during verification, we found that from June 1982 through September 1983, the
 interest rates on all loans were, in effect, regulated. Either the interest rate itself was
 regulated or the amount of credit available at "unregulated" rates was controlled.
 Therefore, for purposes of this final determination, we consider that a weighted average of
 the regulated and "unregulated" interest rates in effect during that period best represents
 the national average commercial rate.
 Using this weighted-average rate as the benchmark, we calculate a bounty or grant on
 exports of 0.49 percent ad valorem for Propulsora and 0.01 percent ad valorem for
 Somisa.

 B. Government Equity Infusions. Petitioner alleged that equity infusions into Somisa by
 the government of Argentina (GOA) were on terms inconsistent with commercial
 considerations. Somisa received capital contributions from the GOA between 1979 and
 1983 to support its planned expansion from 2.5 to 4.5 million tons of annual production
 capacity. In the preliminary determination, we stated 

*18008

 that the agreement on
 these equity infusions was reached in 1976; we found at verification that the agreement
 was implemented by Decrees 2.887/78 of 1978 and 1.832/81 of 1981. The agreement
 provided, in effect, a capital infusion of U.S. $80 million. The terms of the agreement
 specified that Somisa would obtain long-term loans totalling U.S. $80 million, and the
 GOA would pay the debt in exchange for stock issued at par value.
 The GOA also provided equity infusions into Somisa from 1971 through 1976 under
 Decree 2.664/70 of 1970, which authorized the GOA to provide infusions for Somisa's
 expansion to 2.5 million tons of annual production capacity. Cash was provided directly
 for shares at par value.
 In determining whether the government's equity participation in Somisa was on terms
 inconsistent with commercial considerations, we followed the guidelines outlined in the
 "Analysis of Programs" section of this notice. We determine that Somisa was not a
 reasonable commercial investment (was unequityworthy) as of fiscal year 1977/1978.
 Therefore, we find the infusions under the 1978 agreement to confer countervailable
 benefits. Following the methodology outlined in the Subsidies Appendix for calculating
 the countervailable benefit from equity infusions made on terms inconsistent with
 commercial considerations when a firm's shares are not publicly traded, we calculated a
 net bounty or grant of 0.41 percent ad valorem.
 The GOA also participated in the equity of Propulsora. The Banco National de Desarrollo
 (Banade) an entity controlled by the GOA, participated in the original issuance of
 Propulsora's share capital in 1969. Propulsora bought out the bank's shares in 1980. We
 verified that private purchases of Propulsora' stock made at the same time as Banade's
 purchase were at the same price and terms. Therefore, we determine that the government
 of Argentina's equity participation in Propulsora did not confer a countervailable
 benefit.

 C. Loans and Loan Guarantees During the Period We Consider Somisa Uncreditworthy.
 Petitioner alleged that Somisa was uncreditworthly during the period when it received
 loans and loan guarantees from government institutions. As described in the "Analysis of
 Programs" section, supra, we find Somisa to have been uncreditworthy between fiscal
 years 1978/1979 and 1981/1982. Somisa recived some loans from
 government-controlled banks during those years and many private foreign currency
 loans that were quaranteed by government institutions. Government guarantees are not
 countervailable, even in an uncreditworthy period, if they are provided on equal terms to
 a wide variety of industries. There is no evidence on the record, however, that
 government guarantees of foreign currency loans during the period when the loan
 recipient can be considered uncreditworthy were available to a wide range of industries.
 Therefore, we determine that the government loan guarantees provided to Somisa during
 the period we consider it to be uncreditworthy confer a bounty or grant.
 We treated both the loans and the guaranteed loans under the methodology outlined in
 the Subsidies Appendix for loans to uncreditworthy companies. For those loans with
 variable interest rates, however, we could not perform present value calculations.
 Instead, we compared the interest rate paid by the company to the sum of the highest
 interest rate commonly available in Argentina and the risk premium during the period
 for which we are measuring bounties or grants. Some of these variable rate loans were
 denominated in U.S. dollars, and the interest rates are quoted as a percentage above the
 London Interbank Offer Rate (Libor). Because these were non-peso loans with
 non-Argentine interest rates, it would have been inappropriate to apply Argentine
 interest rates in calculating the benefits. Therefore, we used as the benchmark the Libor
 rate in effect during the period for which we are measuring bounties or grants, plus the
 spread that prevailed in Argentina for medium-term loans denominated in U.S. dollars
 during that time, plus the risk premium. Finally, one of these guaranteed loans was
 denominated in Swiss francs at a fixed rate of interest. For this loan we followed the
 standard methodology outlined in the Subsidies Appendix for loans to uncreditworthly
 companies. We used a Swiss franc rate, published in Morgan Guaranty's World Financial
 Markets to calculate both the benchmark and discount rate. We calculated a net bounty
 or grant of 0.63 percent ad valorem.

 D. Somisa's Purchases of "Oil Residue Coal". During the verification at Somisa of alleged
 benefits under Decree 619, we examined the purchasing process and prices paid for for
 several inputs. One of these inputs was "oil residue coal." The prices of all oil products,
 including oil residue coal, are set by the GOA. Somisa provided us with the government's
 official price list for oil products. On the list, a price is quoted for oil residue coal used as a
 fuel, and a separate, lower price is quoted for oil residue coal used in the steel industry.
 Neither Somisa not the government has been able to explain or document the basis for
 the price differential. Therefore, we determine that the lower preferential price for oil
 residue coal used by the steel industry provides a benefit to a specific industry or group
 of industries. To calculate the benefit, we estimated the quantity of oil residue coal
 purchased by Somisa from July 1982 through June 1983 and multiplied it by the
 difference between the two prices for oil residue coal. We divided the result by Somisa's
 total sales to calculate a net bounty or grant of 0.09 percent ad valorem.

 E. Capital Tax Exemption for Somisa. During the verification at Somisa of allegaed tax
 incentives provided under the Industrial Promotion Law, Decree 619, and other laws and
 decrees, we found that Somisa receives a 100 percent exemption of its capital tax
 liability.
 This exemption was originally authorized by Decree 5038/61, and is currently authorized
 by Decree 548/81. Decree 548/81, which applies specifically to Somisa, provides a 100
 percent capital tax exemption and a 33 percent income tax exemption on fixed assets and
 real estate. These exemptions are countervailable because they are limited to a specific
 enterprise. Because Somisa had no taxable income in fiscal year 1981/82 even without
 the 33 percent exemption, we find that the 33 percent income tax exemption does not
 provide a countervailable benefit to Somisa. We treated the capital tax exemption as a
 grant expensed in the year of receipt. Since the value of a tax benefit is not known until
 after the tax returns have been filed, we used the tax return prepared for fiscal year
 1981/82 (July 1981 through June 1982) to estimate the benefit received in fiscal year
 1982/83, the year for which we are measuring bounties or grants. The benefit was
 calculated by multiplying Somisa's taxable capital in fiscal year 1981/82 by the tax rate of
 1.5 percent. This benefit was divided by the value of Somisa's total sales to calculate a net
 bounty or grant of 2.66 percent ad valorem.

 F. Import Duty Exemptions. Argentine tariff law authorizes import duty exemptions on
 raw materials and capital goods when there is no domestic production or insufficient
 domestic production to meet domestic demand, and when importation will not interfere
 with the market for domestic production. Because nominal general availability is not
 necessarily sufficient to prevent a program from being considered a domestic subsidy, we
 requested 

*18009

 documentation during verification to determine whether the
 exemptions are limited to a specific industry or group of industries. Regarding import
 duty exemptions for capital goods, we verified that a number of firms in a wide variety of
 industries were exempted from import duties. The evidence shows that those who apply
 for the import duty exemption on capital goods receive it. Therefore, we determine that
 import duty exemptions on capital goods are not limited to a specific industry or group of
 industries.
 Although we requested similar information concerning the duty exemptions on raw
 materials, the government did not provide sufficient documentation to demonstrate that
 import duty exemptions on raw materials are not limited to a specific industry or group
 of industries. In the absence of such evidence, we determine that the import duty
 exemption for raw materials confers a bounty or grant on the product under
 investigation.
 To calculate the benefit of the exemption to both Somisa and Propulsora, we used, as best
 information available, the verified information on f.o.b. prices paid by Propulsora to
 foreign suppliers for hot-rolled coil during fiscal year 1982/1983. These prices were
 multiplied by the import duty rate of 19 percent listed in the Argentine tariff schedules for
 most of the different types of "chapas laminadas en caliente," or hot-rolled sheet and coil.
 To find the total amount of import duties exempted, we multiplied the amount of the
 exemption by the total tonnage imported. Because we are using data verified at
 Propulsora as best information available, and because Somisa produces significantly
 more steel than Propulsora, we multiplied our calculation of the total amount of import
 duties exempted by the ratio of Somisa's production to Propulsora's production in order
 to derive a better estimate of import duties from which Somisa may have been exempted.
 Because any import duties that would have been paid would be eligible for a rebate upon
 exportation under the reembolso program, we had to factor out the import duties
 exempted on each firm's export sales from our calculation of the total amount of import
 duties exempted. The remainder constitutes the benefit received by each company. We
 divided the remainder by the total value of each company's sales to calculate a net bounty
 or grant of 1.85 percent ad valorem for Propulsora and 2.62 percent ad valorem for
 Somisa.

 II. Programs Determined Not To Confer Bounties or Grants

 We determine bounties or grants are not being provided to manufacturers, producers, or
 exporters in Argentina of cold-rolled carbon steel flat-rolled products under the
 following programs.

 A. Reembolso--Tax Rebate on Exports. The reembolso program was established in 1971.
 It authorized a refund by cash payment on export of taxes "that bear directly or
 indirectly" on exported products and/or their component raw materials for the purpose
 of promoting exports. The amount of the reimbursement is equal to a fixed percentage of
 the f.o.b. value of the exported merchandise. This percentage varies by product. Both
 Propulsora and Somisa participate in the reembolso program.
 Under the Act, the non-excessive rebate of indirect taxes levied at the final stage, and of
 prior stage cumulative indirect taxes borne by inputs that are physically incorporated
 into the fianl product, is not considered a subsidy. With respect to such non-VAT rebates,
 in order to determine whether a cash payment on export is a bona fide rebate of indirect
 taxes, we examine whether: (1) The program involved operates for the purpose of
 rebating indirect taxes; (2) there is a clear link between eligibility for payments on
 exports and indirect taxes paid; and (3) the government has reasonably calculated and
 documented the actual tax incidence borne by the product concerned and has
 demonstrated a clear link between such tax incidence and the rebate amount paid on
 export.
 The reembolso program is designed to refund taxes that "bear directly or indirectly on
 export products." We view taxes borne by a product as indirect, and taxes on, for
 example, income and labor as direct.
 Based on our review of the total tax incidence that the reembolso is designed to rebate,
 we are satisfied that the reembolso operates "for the purpose of rebating indirect taxes,"
 and that it meets our first test.
 Following a general reorganization of the reembolso program in 1976, the rate of
 reembolso on cold-rolled carbon steel sheet was set at five percent. At the time, the GOA
 analyzed the steps of production and value-added at each stage for each major sector of
 Argentine industry. The reembolso rate for each sector was then based on the estimated
 tax incidence derived from the analysis. This procedure provided the GOA only a general
 model upon which the tax incidence for specific sectors could be estimated. Without
 more precise evidence of the indirect taxes levied on cold-rolled carbon steel sheet, we
 would not find the requisite link between indirect taxes actually paid and the reembolso
 payments (the second prerequisite for considering the reembolso not to be a bounty or
 grant).
 In 1980, the Value Added Tax was established (Law 22.294/80) and in 1981, certain
 minor taxes were suspended (Law 22.374/81). As a result of these modifications to the
 Argentine tax system, the government reviewed the studies on the fiscal incidence of
 taxes in order to reevaluate the levels of the reembolso. This detailed review of specific
 taxes levied on cold-rolled carbon steel sheet, when considered in conjunction with the
 more general 1976 study, provides a sufficient basis for our determining that the
 reembolso program meets the second test of a clear link between eligibility for the
 reembolso and indirect taxes paid.
 In reviewing the studies on the fiscal incidence of taxes, the government selected certain
 firms as representative of their industries. Propulsora was selected as representative of
 the cold-rolled carbon steel sheet industry.
 In both 1976 and 1981, Propulsora, using the guidelines and methodology provided by
 the government, conducted the review of the studies on the fiscal incidence of taxes on
 the cold-rolled carbon steel sheet industry. In addition, Propulsora conducted the studies
 of the fiscal incidence of taxes at the prior stages of production.
 In the questionnaire response, the GOA provided us with the most recent analysis of the
 tax incidence on cold-rolled carbon steel sheet. This analysis shows that the taxes levied
 on cold-rolled carbon steel sheet, which the reembolso is designed to rebate, total 16.14
 percent of the f.o.b. value of the exports. Seven categories are included in the analysis:
 domestic raw material inputs, imported raw material inputs, variable costs, labor,
 indirect expenses, taxes paid directly, and export taxes.
 Of the 16.14 percent, the taxes on labor and indirect expenses, which total 0.78 perent,
 either are direct taxes or do not meet our standard for physical incorporation into the
 final product. In the preliminary determination, we also found that the taxes on variable
 costs totaling 1.01 percent did not meet our physical incorporation standard. During
 verification, we received and itemization of the variable costs included in the category
 and the taxes incident to each of these costs. We established that a number of these cost
 items, such as packing materials, are physically incorporated into the final 

*18010


 product, while a number of others, such as electricity and fuel, are not. The physically
 incorporated variable cost items account for 0.21 percent of the fiscal incidence of taxes
 on cold-rolled carbon steel sheet. The items that do not meet our standard for physical
 incorporation account for 0.80 percent.
 In its questionnaire responses, the GOA itemized only some of the taxes on domestically
 produced inputs. The remainder, which accounted for the majority of the tax incidence
 claimed, was not itemized. In the preliminary determination, we determined that because
 these taxes and their fiscal incidence on inputs were not itemized, our third test was not
 met; that is, the government had not reasonably calculated and documented the actual
 tax incidence borne by the product concerned. During verification, we found that the
 non-itemized taxes represented the total tax incidence at the prior stages of production.
 We reviewed the studies prepared by Propulsora on the total tax incidence at the
 hot-rolled coil stage of production and for production of primary raw materials, such as
 pellets and limestone. These studies are organized in the same way as the cold-rolled
 sheet study and include taxes on labor and indirect expenses, which either are direct
 taxes or do not meet our standard for physical incorporation into the final product.
 In calculating the allowable tax incidence in the domestic raw material category, we only
 included those indirect taxes at prior stages of production that meet our standard for
 physical incorporation. Applying this standard, we found that 6.12 percent of the tax
 incidence claimed is allowable and 2.84 percent is not.
 The taxes in each of the three remaining categories--imported raw material inputs, taxes
 paid directly, such as municipal taxes, and export taxes--meet our third test because they
 are itemized, and the rate of each tax and its incidence are calculated. These taxes--which
 include the stamp tax, the tax on transfer of foreign currency, insurance taxes, municipal
 taxes and the export contract stamp tax--are all indirect taxes. The total incidence of the
 taxes in these three categories is 5.39 percent.
 Therefore, of the total 16.14 percent tax incidence claimed by the government of
 Argentina for cold-rolled carbon steel sheet, we have allowed 11.72 percent and
 disallowed 4.42 percent.
 In order to determine whether the reembolso confers a bounty or grant on cold- rolled
 carbon steel sheet, we must compare what we have determined as allowable to the
 reembolso rate currently in effect for cold-rolled carbon steel sheet. Since July 5, 1982,
 the reembolso for cold-rolled carbon steel sheet has been 10 percent (Resolution M.E.
 8/82). In addition to the normal reembolso, petitioner alleged the existence of bounties
 or grants through two additional reembolsos. Decree No. 1691 temporarily established an
 additional reembolso of 15 percent. This reembolso was abolished in April 1982 and,
 thus, was not in effect during the period for which we are measuring bounties or grants,
 July 1982 through June 1983. Decree No. 2863/72 granted an additional reembolso of
 five percent for exports to new markets. This reembolso, which was abolisehd in April
 1983, was not received by the respondent companies on exports of cold- rolled carbon
 steel sheet to the United States.
 We verified that only the 10 percent reembolso is currently received by Somisa and
 Propulsora. To determine if the reembolso confers a bounty or grant, we compared the 10
 percent reembolso received to the total allowable indirect taxes of 11.72 percent.
 Because the reembolso does not exceed the total allowable indirect taxes of 11.72
 percent, we determine that the reembolso does not confer a bounty or grant on
 cold-rolled carbon steel flat- rolled products.

 B. Propulsora's Purchases of Inputs from Somisa. Petitioner alleged that Somisa, a
 state-owned company, offers semi-finished products at special discounts to other
 Argentine steel producers, particularly Propulsora, who use them to manufacture
 finished steel products for export.
 Our information indicates that Propulsora purchases approximately 50 percent of its
 hot-rolled coil requirements from Somisa. We verified that these purchases were at prices
 comparable to those paid by Propulsora to its foreign suppliers. Furthermore, we found
 no indication that Propulsora receives special discounts on purchases of inputs intended
 for export from Somisa. Consequently, we determine that Propulsora receives no
 countervailable benefits through its purchases of inputs from Somisa.

 C. Pre-Financing of Exports through Dollar-Indexed Peso Loans. Under this program,
 which was authorized under Central Bank Circular RF-153, exporters receive pre-export
 financing through peso loans equal to 60 percent of the export sale's f.o.b. value. The
 loans are given in pesos but denominated in U.S. dollars at the exchange rate prevailing at
 the time of the loan. Repayment must also be in pesos, but the peso amount is established
 by the exchange rate prevailing at the time of repayment. In addition to repaying the
 peso amount of the loan at the exchange rate prevailing at the time of repayment, the
 borrower also must pay a one percent interest rate. The funds are drawn from the Central
 Bank of Argentina and then loaned through private commercial banks to individual
 corporate borrowers. The maximum length of the loan is 120 days, but repayment must
 take place no later than 60 days from the effective export date.
 We verified that Somisa had one loan outstanding under this program during the period of
 review, while Propulsora had none. When compared the cost to Somisa for this loan with
 what Somisa would have paid commercially for the same loan, using as the benchmark the
 weighted average of the regulated and unregulated rates described, supra, we found no
 benefit. Therefore, we determine that this program does not confer a bounty or grant.

 D. Government Credit Allocations. Petitioner alleged that betwen 1973 and 1977, when
 Argentine banks "acted merely as collection agents for the central bank," the steel
 industry received a disproportionate share of available credit. This credit was alleged to
 have been available at interest rates below the rate of inflation.
 We discussed the credit allocations with GOA officials and examined a report from the
 Central Bank concerning the Argentine financing system between 1973- 1977. We found
 that during this period, the Central Bank could control all Argentine banks as part of a
 government-managed system of allocating credit. The government's design for credit
 allocation, however, consisted of general interest rate manipulations and guidelines for
 the total amount of credit to be allocated in the country. It did not call for specific rates
 or amounts of financing to specific industries. Therefore, we determine that these credit
 allocations did not confer countervailable benefits because they were not limited to a
 specific industry, group of industries, or to companies in specific regions.

 E. Loan Guarantees. Petitioner alleged the GOA guaranteed loans to the steel industry.
 During the period of review, Propulsora had one outstanding supplier credit guaranteed
 by the Banco Ciudad de Buenos Aires, a government- controlled institution. Propulsora
 paid a guarantee fee to Banco Diudad de Buenos Aires that was higher than the guarantee
 fees it paid to private banks for comparable supplier credits. Therefore, we determine that
 the loan 

*18011

 guarantee to Propulsora does not confer a bounty or grant.
 Somisa had a number of loans and supplier credits guaranteed by the GOA or official
 banks at no guarantee fee. For a discussion of the guaranteed loans during the period we
 found Somisa to be uncreditworthy, see the section entitled "Loans and Loan Guarantees
 During the Period We Consider Somisa Uncreditworthy." Concerning the guaranteed loans
 to Somisa during the period we found it to be creditworthy, we have found that standard
 practice for comparable transactions in Argentina does not involve a charge to the
 borrower by the shareholder guarantor for a guarantee. Since guarantees at no fee by a
 shareholder guarantor are the normal commercial practice, we determine that these
 guarantees do not confer a countervailable benefit on Somisa.

 F. Export Credit Insurance. During the verification of the insurance program authorized
 under Law 22.593, described infra, we found that the GOA provides export credit
 insurance against political risk. This insurance program is administered on behalf of the
 GOA by a syndicate of private insurance companies that offers commercial export
 insurance. We examined annual reports providing information on the political risk
 insurance program for the fiscal years 1971 through 1983. The annual reports include the
 political risk insurance premiums received for the years 1971 through 1981. Also
 included are indemnities paid under the political risk insurance program, net of
 recuperations, for the years 1971 through 1983. All the data in the annual reports are
 presented in dollars. Indemnities paid from 1971 through 1983 account for 11.6 percent
 of premiums received from 1971 through 1981. The government has stated that the
 private syndicate charges 10 percent of premiums received to administer the political
 risk insurance program on the government's behalf. Based on these data, we determine
 that the preimums charged for political risk insurance are not manifestly inadequate to
 cover the long-term operating costs and losses of the program. Therefore, we find that the
 export credit insurance program for political risk does not confer a bounty or grant.

 G. Multiple Exchange Rate System. Petitioner asked us to investigate whether the
 Argentine system of multiple exchange rates operates to provide countervailable
 benefits. In July 1981, the Central Bank of Argentina established a commercial rate of
 exchange which applied equally to all import and export transactions, and a financial rate
 of exchange that applied to all financial transactions. At the same time, the Central Bank
 allowed 10 percent of the f.o.b. value of "promoted exports" to be exchanged at the higher
 financial rate. This promotional aid for exports was abolished in December 1981 through
 Central Bank Circular "A" 84, which unified the exchange market again.
 In July 1982, the market was split again. Under this split, exporters of "traditional goods"
 were eligible for the higher rate of exchange. Steel is not classified in Argentina as a
 traditional good. Then in September 1982, through Circular "A" 219, the Central Bank
 abolished the provision for exporters of traditional goods and authorized a split rate for
 all commercial operations. As specified in Circular "A" 219, the exchange rate which
 applied to 85 percent of the f.o.b. value of imports and exports was the commercial rate.
 The financial rate was applied to the remaining 15 percent of the f.o.b. value. Circular "A"
 232 modified the 85/15 percent split for commercial operations in October 1982. The
 modification authorized 80 percent of the f.o.b. value of both imports and exports to be
 exchanged at the commercial rate. The remaining 20 percent could be exchanged at the
 financial rate. Circular "A" 241 unified the financial and commercial exchange markets
 again as of November 1. 1982.
 During the period for which we are measuring bounties or grants, July 1982 through June
 1983, we find that the Argentine system of multiple exchange rates did not confer a
 bounty or grant on the products under investigation. First, in July 1982, when the market
 was split and exporters of traditional goods received the higher financial rate, we find no
 bounty or grant because neither Propulsora nor Somisa export traditional goods and thus
 were not eligible for the higher financial rate of exchange. For the period September
 through October 1982, when the rate for commercial operations was split, we find no
 bounty or grant because the split generated, in effect, a uniform average rate that applied
 equally to both imports and exports (see Energetic Worsted Corp. v. United States, 53
 CCPA 36 (1966)). Third, after November 1, 1982, there was no system of multiple
 exchange rates.

 G. Discounts of Foreign Currency Accounts Receivable under Circular RF-21. During the
 verification at Somisa of pre-financing of exports through dollar- indexed peso loans, we
 found that Somisa paid off its pre-financing loan with the proceeds from discounting the
 accounts receivable for the shipment covered by the pre-financing. The discounting of
 foreign currency accounts receivable is authorized by Central Bank Circular RF-21.
 The maximum discount rate charged by the intervening bank is 7.5 percent per year, and
 the minimum is 5 percent. Somisa and Propulsora received loans under circular RF-21 at
 6 and 6.5 percent per year, respectively. The proceeds of the discounting are received in
 pesos. If any pre-financing loans are outstanding when the accounts receivable are
 discounted, the proceeds from the discounting must be used to pay off the pre-financing
 loan.
 In Argentina, firms may not hold dollar accounts receivable for more than 180 days.
 After 180 days, Circular RF-21 requires firms to discount. In addition, Argentina law
 does not permit firms to prepay their dollar liabilities earlier than 180 days from
 purchase.
 Under these circumstances, most firms would hold their dollar receivables until the
 180th day when they are required to discount them. The information we received from
 Propulsora indicates that dollar receivables are held as long as legally possible.
 We determine that this program does not confer countervailable benefits because the
 discount rates offered are set by the intervening banks between 5 percent and 7.5
 percent. We have no evidence that the government has directed discount rates lower than
 the maximum and conclude such rates represent commercial bank practice in obtaining
 the highest rate the market will bear. We will examine the operation of this program in
 greater detail during our administrative review.

 III. Programs Determined Not To Be Used

 We determine that the following programs, listed in the notice of "Initiation of
 Countervailing Duty Investigation," were not used by the manufacturers producers,
 or exporters in Argentina of cold-rolled carbon steel flat-rolled products.

 A. Forgiveness or Assumption of Debt. Petitioner alleged there are indications that the
 Argentine government may have undertaken to pay the interest on at least a portion of
 debt for state-owned companies. We have found that Somisa is a "mixed company" under
 Argentine law, not a state-owned company, and that the GOA has never forgiven or
 mandated, directly or indirectly, the forgiveness of any debt or interest on debt
 contracted by Somisa or Propulsora, except for the 1978 agreement with Somisa
 discussed in the section on "Government Equity Infusions," supra. Therefore, we
 determine that except for the 1978 

*18012

 agreement with Somisa, neither Propulsora
 nor Somisa has received countervailable benefits as a result of direct or indirect
 government action to assume or forgive its debt.

 B. Medium- and Long-Term Loans. Petitioner alleged that the Argentine steel industry,
 particularly Somisa, has benefited from government programs providing medium- and
 long-term loans on terms inconsistent with commercial considerations. We verified that
 Propulsora has not received medium- or long- term loans from any official institution. We
 found that Somisa has received loans from official banks during the period we considered
 it to have been uncreditworthy. These loans are discussed in the section entitled "Loans
 and Loan Guarantees During the Period We Consider Somisa Uncreditworthy," supra.
 Petitioner cited Argentine Law 22.510, and Decrees 989/81 and 1894/83 as having
 provided loans to steel firms. Law 1894/83 was enacted after the period for which we are
 measuring bounties or grants. We received information from the GOA indicating that
 neither Law 989/81 nor Law 22.510 authorized loans to the respondent companies.
 Further, we verified all loans outstanding to the respondents during the period for which
 we are measuring bounties or grants and identified all loans from official entities, as
 discussed above.

 C. Trade Promotion Programs. Petitioner alleged the Argentine steel industry receives
 countervailable benefits through trade promotion programs conductd by the
 government. The GOA stated that no trade promotion programs exist, and we found no
 evidence at verification that Propulsora or Somisa has used trade promotion programs.

 D. Tax Incentives. Petitioner alleged the Argentine steel industry benefits from a variety
 of special tax incentives, including those provided under the Industrial Promotion Law of
 1977, special tax treatment for state- owned enterprises, and certain exemptions
 available to firms locating in specified areas of Buenos Aires Province. We verified that
 neither Somisa nor Propulsora received tax benefits under the Industrial Promotion Law
 of 1977, nor under laws providing special exemptions for firms locating in specified areas
 of Buenos Aires Province. Regarding special tax treatment for state- owned firms, we
 verified that Somisa is not considered a state-owned company under Argentine law.

 E. Pre-Financing of Exports Under Circular OPRAC 1-1. Circular OPRAC 1-1 instituted a
 pre-financing program for Argentine exports as an alternative to the Circular RF-153
 program for pre-financing of exports through dollar-indexed pesos described supra. This
 program was initiated on August 21, 1981, and terminated on March 31, 1982. Under
 Circular OPRAC 1-1, loans could not exceed one year, and firms receiving OPRAC 1-1
 loans could not also receive Circular RF-153 loans. We verified that Propulsora used this
 program, but that no loans under circular OPRAC 1-1 were outstanding during the period
 for which we are measuring bounties or grants. Somisa did not use this program.

 F. Labor Wage Freeze. Petitioner alleged that recent wage freezes for public employees
 imposed by the GOA permit state-owned firms, including Somisa, to benefit from labor at
 subsidized wages.
 Somisa has demonstrated that, because it is governed by the laws and regulations
 applicable to corporations of mixed private and public ownership, it falls outside the
 coverage of the wage freeze.

 G. Insurance Authorized by Law 22.593. Subsequent to our preliminary determination,
 counsel to a party to the proceeding alleged that the GOA, through the National
 Reinsurance Institute, offered export insurance during the Malvinas/Falklands crisis
 under the terms of Law 22.593. We verified that under the terms of Law 22.593, insurance
 coverage is provided to all Argentine commercial traffic, whether domestic or export,
 affected by the "risks of war." The Malvinas/Falklands crisis ended on June 24, 1982.
 Accordingly, the insurance coverage under Law 22.593 was suspended because the "risks
 of war" no longer existed. We determined that insurance under Law 22.593 was not used
 by Propulsora or Somisa because the program was not in force during the period for
 which we are measuring bounties or grants.

 H. Subsidized Inputs of Raw Materials and Capital Equipment. Petitioner alleged the
 Argentine steel industry receives indirect benefits as a result of subsidies provided by the
 government to suppliers of raw materials and capital equipment used by producers of
 cold-rolled carbon steel flat-rolled products. It specifically alleged that Decree 619
 operates to direct benefits such as equity capital, loans, loan guarantees and tax deferrals
 to steel industry suppliers.
 We verified that neither Somisa nor Propulsora received benefits under Decree 619.
 Somisa is Propulsora's primary domestic supplier and accounts for approximately 85
 percent of domestic raw material inputs. Because Somisa did not receive any assistance
 under Decree 619, we determine that no benefits under Decree 619 were passed through
 to Propulsora.
 Regarding Somisa, we found that one of its suppliers received assistance under Decree
 619. In general, the assistance provided under Decree 619 is targeted to specific plant
 locations. The assistance provided to Somisa's supplier was specifically targeted to a plant
 that does not produce the input purchased by Somisa. Furthermore, the input provided
 by this supplier is relatively insignificant in terms of the final production cost of
 cold-rolled carbon steel flat-rolled products. During verification, we examined on a
 random basis Somisa's purchases of inputs. We found that these purchases (except for the
 purchase of oil residue coal discussed supra) were on commercial terms. In addition, we
 found no evidence that Somisa was related to any of its suppliers. Therefore, we
 determine that any benefits provided to Somisa's supplier under Decree 619 were not
 passed through to Somisa in the form of lower prices.

 I. Foreign Exchange Insurance. Subsequent to the preliminary determination, counsel to
 a party to the proceeding alleged that the foreign exchange insurance program
 established by the government conferred a countervailable benefit on the product under
 investigation. We found at verification that this program is available to all industries and
 to all private borrowers in Argentina. Because this program is not limited to a specific
 industry or group of industries or to companies in specific regions, we determine that it
 did not confer a countervailable benefit on the product under investigation.

 J. Additional Programs Not Alleged. In the notice announcing the initiation of this
 investigation, we stated our intent to investigate whether certain programs not
 specifically alleged by petitioner are providing bounties or grants to manufacturers,
 producers or exporters in Argentina of cold-rolled carbon steel flat-rolled products.
 The programs in question are:
 1/8 Grants from the government of Argentina
 1/8 Additional Reembolso for exports from southern ports
 Our decision to investigate these programs was prompted by experience 

*18013



 developed in prior Argentine investigations and our view that they might be relevant to
 this investigation. Based on our analysis and verification of these programs, we determine
 that these programs were not used by repondents.

 Petitioner's Comments

 Comments raised by petitioner and parties to the proceeding that pertain to the Subsidies
 Appendix are addressed at the end of the Subsidies Appendix attached to this notice.
 Comment 1. Petitioner contends that the benchmark for loans provided under the
 program for pre-export financing through dollar-indexed pesos should be a dollar
 interest rate.
 DOC Position. We do not consider that a dollar interest rate is the appropriate benchmark
 for pre-export financing through dollar-indexed peso loans. These loans are provided in
 pesos with the principal indexed to the exchange rate. Because the pre-export financing
 loans are provided in pesos we believe that the benchmark should be a peso benchmark.
 As described supra, we used a weighted average of the regulated and unregulated rates,
 which we consider is representative of the national average commercial rate, as the
 benchmark for this program.
 Comment 2. Petitioner argues that the exemptions of import duties are countervailable
 because the respondents did not provide any evidence that they are generally available.
 DOC Position. Our reasons for finding import duty exemptions on raw materials
 countervailable are explained in the section in "Import Duty Exemptions" under
 "Programs Determined to Confer Bounties or Grants." In the same section, we also discuss
 our reasons for finding import duty exemptions on capital good not countervailable.
 Comment 3. Petitioner contends that the Department must continue to find a
 country-wide rate because company-specific rates are used only when programs are
 limited specifically to certain companies and when a certification process is applied.
 DOC Position. As specified in 19 CFR 355.28(a)(3), "if separate enterprises have received
 materially different benefits, such differences shall also be estimated and stated."
 Propulsora participated in only two of the six countervailable programs, while Somisa
 participated in all six. Furthermore, the companies received materially different benefits
 under the countervailable programs. Therefore, we think that company-specific rates are
 appropriate in this case.
 Comment 4. Petitioner presented evidence that the Argentine Industrial Promotion Law
 (Law 21.608) concerning import duty exemptions has been amended by Law 22.876,
 which adds new "incentives for export trade."
 DOC Position. Because this information on Law 22.876 was not presented to us until April
 9, 1984, we are unable to consider it in this investigation. We will review this law in our
 administrative review. We verified that neither Somisa nor Propulsora was eligible for the
 incentives provided under the Industrial Promotion Law.
 Comment 5. In its comments of April 9, 1984, petitioner made four new allegations. These
 concern an exemption of stamp taxes, capital incentives, government purchase of steel at
 a premium, and funds for the development of the steel industry under Law 20.086.
 DOC Position. Because these allegations were not presented to us until April 9, 1984, we
 are unable to consider them in this investigation. We will consider them, however, in our
 administrative review.
 Comments by Parties to the Proceeding. Counsel for Bethlehem Steel Corporation and for
 Republic Steel Corporation, Inland Steel Company, Jones and Laughlin Steel
 Incorporated and Cyclops Corporation (hereinafter "the Four"), who are parties to the
 proceeding, submitted comments.
 Comment 1. Counsel for Bethlehem provided data on the corporate tax rate, national
 average cost of debt, value of outstanding debt and equity by company and the prime rate
 in Argentina for our use in calculating the weighted cost of capital. Further, they
 provided data for our use in calculating the national average rate of return on equity in
 the weighted cost of capital formula.
 DOC Position. We did not use Bethlehem's data to calculate the national average rate of
 return on equity. Instead, we are using a national average rate of return calculated by
 FIEL. FIEL's calculation is based on data reported in Prensa Economica. The data in
 Prensa Economica are comparable to the data provided by Bethlehem except that Prensa
 Economica reports equity statistics as well as sales and profits. We did not use
 Bethlehem's corporate tax rate data because we no longer include this figure in our
 weighted cost of capital formula. We did not use the prime rates provided by Bethlehem
 because we consider the ones provided by FIEL, which are higher, to be more
 representative. We did use the other figures provided by Bethlehem.
 Comment 2. Counsel for Bethlehem recommended that we use the inter-company rate for
 second line checks as the highest average spread above the average interest rate in
 Argentina for short-term foreign currency and domestic currency loans.
 DOC Position. We received information indicating that the inter-company rate for
 second-line checks was not widely used in Argentina unitl August, 1983. Because this
 falls outside the period for which we are measuring bounties or grants, we did not
 consider using it as a spread above the benchmark for calculations in this investigation.
 Comment 3. Counsel for Bethlehem argues that because there is virtually no long-term
 financing in Argentina, we should use the short-term rate it recommended for the
 national average cost of debt as the benchmark for long- term loans. This rate is the
 "effective rate for financial entities" between January 1982 and August 1983, and the
 inter-company rate for second line checks after August, 1983.
 DOC Position. We did use the "effective rate for financial entities" as a benchmark for
 domestic loans to Somisa during the period we consider it to be uncreditworthy. We did
 not consider using the inter-company rate for second line checks because we received
 information that it was not widely used in Argentina unitl August 1983, which is after
 the period for which we are measuring bounties or grants.
 Comments 4. Counsel for the Four contends that the Department erroneously used a peso
 interest rate in analyzing the pre-financing of exports through dollar- indexed pesos.
 DOC Position. See the DOC position for petitioner's Comment 1, supra.
 Comment 5. Counsel for the Four contends that the Department should have calculated a
 benchmark on the government guaranteed loan to Propulsora regardless of whether
 guarantee or other fees are paid. Counsel argues that the Department must look at what a
 company would pay absent the government guarantee.
 DOC Position. We verified that Propulsora received other loans similar to the loan
 guaranteed by the official bank that were guaranteed by private banks. The guarantee fees
 paid to the private banks were comparable and in some cases less than the fees paid to the
 official bank. Therefore, we did not find the loan guarantee from the official bank to be
 inconsistent with commercial considerations.
 Comment 6. Counsel for the Four contends that the Department 

*18014

 mistakenly
 treated as short-term loans what were, in effect, variable rate long-term loans.
 DOC Position. Almost all domestic currency loans in Argentina, whether short- or
 long-term, have variable interest rates. These interest rates change on a monthly basis. A
 borrower only knows at the beginning of each month what the interest expense will be for
 that month. If the principal of the loan is indexed, a borrower may not know until two or
 three months after the relevant month what the real interest expense was for that month.
 Under these circumstances, we believe that it is reasonable to calculate variable rate
 long-term loans like short-term loans. However, we did not use the national average
 commercial rate as the benchmark. Instead, we used the "effective rate for financial
 entities" as published in Business Trends, plus the risk premium.
 Regarding the foreign currency loans found to confer bounties or grants, the majority are
 charged an interest rate of Libira plus a spread. The benchmark for these loans is the
 Libira rate plus the average prevailing spread available in Argentina plus the risk
 premium. We consider that these loans should be calculated as short-term loans because
 the risk premium and the spread could vary substantially from year to year.
 For those foreign currency loans with constant interest rates, we followed the standard
 methodology in the Subsidies Appendix to calculate the benefits.
 Comment 7. Counsel for the Four contends that we should use the "unregulated rates"
 published by FIEL as the national average commercial rate for short-term loans because
 the weighted-average rates provided during verification include the regulated rate
 market, which is essentially a closed market except for OPRAC 1-9 loans.
 DOC Position. In June 1982, the Argentine financial system was restructured and all
 outstanding loans, both short- and long-term, were restructured under a regulated rate
 established at the beginning of every month by the Central Bank. The size of the credit
 pool under the regulated rate increased every month. In August 1982, the Central Bank
 authorized the creation of two other credit pools, one with rates tied to the wholesale
 price index and the other with "unregulated" rates. The size of both these credit pools was
 restricted to a percentage of the borrowers' bank deposits on hand at these rates as of
 June 30, 1982. Therefore, all interest rates in Argentina were, in effect, regulated after
 June 1982. During the period from August 1982 through July 1983, the regulated rate
 accounted for approximately 73 percent of outstanding credit, the wholesale price index
 rate accounted for approximately 11 percent, and the "unregulated" rate accounted for 16
 percent. We consider that a weighted average of all three rates provides the best
 representation of the national average commercial rate. Using the "unregulated" rate, as
 counsel suggests, might be a measure of the rate of interest available to a company at one
 point in time but does not reflect the average rate that all borrowers were charged during
 any specific period. The "unregulated" rate does not represent a national average of
 existing credit rates. Rather, it represents an alternative commercial rate that is only one
 of the rates used by companies during this period.
 Comment 8. Counsel for the Four argues that the Department inappropriately chose the
 prime interest rate published by Morgan Guaranty as its discount rate for long-term
 loans.
 DOC Position. In our preliminary determination, we used the Morgan Guaranty prime rate
 as the discount rate for long-term peso loans. During verification we found that these
 were not fixed-rate loans, but ones tied to the wholesale price index with the interest
 charges varying monthly. Therefore, we treated these loans as short-term loans, using as
 the benchmark the "effective rate for financial entities" pulbished in Business Trends.
 There was one fixed-rate loan to Somisa during the period in which we consider it
 uncreditworthy. This loan was denominated in Swiss francs. We used a Swiss franc rate
 published in Morgan Guaranty's World Financial Markets to calculate both the benchmark
 and the discount rate.
 Comment 9. Counsel for the Four contends that certain accounting practices used in the
 financial statements of Somisa are not in accordance with U.S. "generally accepted
 accounting principles" and that such practices may be distortive of the financial position
 of the company.
 DOC Position. We used the financial statements of Somisa which were presented in
 accordance with the "generally accepted accounting principles" (GAAP) of Argentina
 and the supplemental financial statements in our determination of creditworthiness and
 equityworthiness. The financial statements are in accordance with the Argentine GAAP.
 The Argentine GAAP varies significantly from U.S. GAAP because it is inflation based.
 Given the high rate of inflation in Argentina, we consider that using the Argentine
 statements in our analyses is reasonable. When developing the specific information to be
 used for the ratios and other tests for creditworthiness and equityworthiness, we looked
 beyond the financial reporting practices to obtain the necessary information.
 Comment 10. Counsel for the Four contends that the 2 percent discount of the regulated
 interest rate provided to companies that participated in the voluntary price control
 program was not generally available because Dr. Szewach of FIEL stated in the GOA's
 amended response that only 6 to 8 percent of all loans made in the regulated market
 enjoyed this discount.
 DOC Position. Dr. Szewach's letter states that the impact of the 2 percent discount affected
 6 to 8 percent of the total regulated market. We understand that the 2 percent discount,
 which was available to firms participating in the government's price control program,
 applied only to new credit that became available in the regulated market. We verified that
 many firms representing almost every industry in Argentina were participants in the
 GOA's price control program and thus were eligible for the discount. We do note,
 however, that we did not take the 2 percent discount into account in the national average
 commercial rate used for short-term loans because the GOA provided no supporting
 documentation to verify the 6 to 8 percent impact reported by Dr. Szewach.

 Respondent's Comments

 Comment 1. Counsel for Propulsora contends that the Department should issue
 company-specific countervailing duty rates.
 DOC Position. We have used company-specific rates in this determination for the reasons
 specified supra.
 Comment 2. Counsel for Propulsora argues that Propulsora should be excluded from this
 final determination because Propulsora received no countervailable benefits.
 DOC Position. We determined that Propulsora participated in two countervailable
 programs, OPRAC 1-9 and import duty exemptions. In addition, the total ad valorem rate
 for Propulsora is not de minimis. Therefore, Propulsora cannot be excluded from this
 final determination.
 Comment 3. Counsel for Propulsora contends that the Department should use
 inflation-adjusted rather than historical data in its analyses.
 DOC Position. For our credit worthiness and equityworthiness decisions, we used the
 primary financial statements as well as the required 

*18015

 supplemental statements.
 Both sets of statements take into account, by different means and degrees, the effects of
 inflation. For the denominator used in calculating the ad valorem rate, we used
 inflation-adjusted figures. We also adjusted all benefits for inflation.
 Comment 4. Counsel for Propulsora argues that the non-itemized taxes on domestic raw
 materials in the reembolso were verified and should be considered in our determination
 of whether the reembolso confers a bounty or grant on cold-rolled carbon steel flat-rolled
 products. Counsel also contends that certain variable costs meet the Department's
 standard of "physical incorporation." Finally, counsel contends that the effective
 reembolso received is 9.7 percent, not 10 percent, because of the commissions paid to
 foreign commissionaries.
 DOC Position. In our review of taxes on domestic raw materials, we allowed those that
 were indirect and that met our standard for "physical incorporation." We also allowed the
 tax incidence claimed on certain variable cost items that we verified were physically
 incorporated into the final product.
 We did not use 9.7 percent as the reembolso rate in comparing the allowable tax
 incidence to the level of the reembolso because this is the effective reembolso rate for
 only one firm; it was not demonstrated that this is the effective reembolso rate received
 by the industry as a whole.
 Comment 5. Counsel for Propulsora contends that the statutory test for OPRAC 1-9
 post-export financing is whether it is "inconsistent with commercial considerations." As
 such, we should use a company's own short-term borrowing experience to determine the
 benchmark, and not the national average commercial rate specified in the Subsidies
 Appendix. Counsel further contends that even if a national average is used, OPRAC 1-9
 loans to Propulsora are non- preferential because the proper benchmark is the 30-day
 regulated rate, which accounted for 72 percent of the available credit from August 1982
 through September 1983.
 DOC Position. As outlined in the Subsidies Appendix, we consider that the appropriate
 benchmark for countervailable short-term loans is the national average commercial rate.
 We do not review the company's short-term loan portfolio to determine whether a
 short-term loan is inconsistent with commercial considerations because it is not likely
 that short-term loan rates would vary widely from the national average rate. Regarding
 what rate represents the national average commercial rate in Argentina during the
 relevant period, we consider that a weighted average of the three types of rates best
 represents the national average. The weighted average includes the regulated rate, the
 rate tied to the wholesale price index, and the "unregulated rate." All three of these rates
 were, in effect, regulated because either the rate itself was set by the Central Bank or the
 size of the credit pool was limited by it. We do not believe that using just the regulated
 rate set by the Central Bank, as counsel suggests, would represent the national average
 since approximately 25 percent of the existing credit was provided under the other two
 rates.
 Comment 6. Counsel for Propulsora maintains that the Department erroneously found
 Somisa to be uncreditworthy from 1976 to 1983 because Somisa meets the "ways and
 means" test and because if the ratios are calculated on inflation- adjusted figures, Somisa
 would be creditworthy.
 DOC Position. Our methodology for determining both the creditworthiness and the
 equityworthiness of a respondent is set forth in the "Analysis of Programs" section of this
 notice. We used the primary financial statements as well as the required supplemental
 financial statements in making our decision on Somisa's creditworthiness and
 equityworthiness. Using either set of statements, we would find Somisa to be
 uncreditworthy from fiscal years 1978/79 through 1981/82 and unequityworthy from
 1977/78 through 1982/83. We find that Somisa does not adequately meet the "ways and
 means" test (i.e. whether dash flow adequately covers the principal and interest
 payments) because it did not generate enough cash to service its debt.
 Comment 7. Counsel for Propulsora disagrees with the Department's calculations of
 Somisa's government loans and loan guarantees for the following reasons:
 (1) The Department did not take into account commissions and other charges on the
 guaranteed loans;
 (2) The Department used an annual Libor rate instead of the rate that reflects the actual
 timing of interest payments;
 (3) The Department did not take into account the indexed principal on Somisa's peso
 loans; and
 (4) The Department should not countervail the deferral of principal repayment on the
 peso loans because the principal is indexed.
 DOC Position. In calculating Somisa's loan guarantees, the Department did not take into
 account commissions and other charges because we have no evidence that these charges
 are also included in the benchmark interest rate, which is usually the highest spread
 available in Argentina plus the appropriate base rate. Regarding counsel's second
 argument, for purposes of the final determination, we have used the Libor rate that
 reflects the actual timing of interest payments. Regarding counsel's third and fourth
 arguments, we have taken into account as part of the actual interest charge, the indexed
 principal on Somisa's peso loans. However, because we took into account the indexation
 when comparing the loan's interest rate to the benchmark interest rate, it would be
 inappropriate to also determine that the deferral of principal repayment is not
 countervailable simply because the principal is indexed.

 Verification 

 In accordance with section 776(a) of the Act, we verified all data used in making this final
 determination.

 Suspension of Liquidation

 The suspension of liquidation ordered in our preliminary affirmative countervailing
 duty determination (49 FR 5151) shall remain in effect until further notice. The net
 bounty or grant for duty deposit purposes for each firm is as follows:
   
 ------------------------------------------------------------------------------- 
             Manufacturers/producers/exporters                Ad valorem rate    
                                                                 (percent)       
 ------------------------------------------------------------------------------- 
 Somisa--Sociedad Mixta Siderurgica Argentina (Somisa) .................... 6.42 
 Propulsora Siderurgica Saic (Propulsora) ................................. 2.34 
 All other companies ...................................................... 5.44 
 ------------------------------------------------------------------------------- 
   
 In accordance with section 706(a)(3) of the Act, we are directing the U.S. Customs
 Service to require a cash deposit in the amount indicated above for each entry of
 cold-rolled carbon steel flat-rolled products from Argentina which are entered, or
 withdrawn from warehouse, for consumption on or after the date of publication of this
 notice in the Federal Register and to assess countervailing duties in accordance with
 section 706(a)(1) and 751 of the Act.
 In accordance with section 751(a)(1) of the Act (19 U.S.C. 1675(a)(1)), we hereby give
 notice that we are commencing an administrative review of this order on April 26, 1984.
 For further information regarding this review, contact Richard Moreland at (202)
 377-2786.
 This suspension will remain in effect until further notice. This notice is published pursuant
 to sections 303 and 706 of the Act (19 U.S.C. 1303, 1671e).

 *18016

 Dated: April 18, 1984.

 William T. Archey,

 Acting Assistant Secretary for Trade Administration.

 Product Description Appendix

 For purposes of this investigation the term "cold-rolled carbon steel flat- rolled products"
 covers the following cold-rolled carbon steel products. Cold- rolled carbon steel
 flat-rolled products are flat-rolled carbon steel products, whether or not corrugated or
 crimped, whether or not painted or varnished and whether or not pickled, not cut, not
 pressed, and not stamped to non- rectangular shape, not coated or plated with metal;
 over 12 inches in width, and 0.1875 or more in thickness, as currently provided for in
 item 607.8320 of the TSUSA; or over 12 inches in width and under 0.1875 inches in
 thickness whether or not in coils, as currently provided for in items 607.8350, 607.8355,
 or 607.8360 of the TSUSA.

 Subsidies Appendix

 Certain types of capital and financial subsidies, including grants, loans, loan guarantees,
 and equity, have arisen in a number of countervailing duty proceedings. In Appendix
 2 to the "Final Affirmative Countervailing Duty Determinations on Certain Steel
 Products from Belgium" (47 FR 39304, 39316) ("Appendix 2"), we explained our
 methodology for dealing with these and other issues. Since publishing Appendix 2, we
 have recognized a number of shortcomings in the theory and application of this
 methodology. To remedy these, we proposed certain revisions in Appendix II to our
 "Preliminary Affirmative Countervailing Duty Determinations on Certain Carbon Steel
 Products from Mexico" (49 FR 5142, 5148) ("Appendix II").
 This appendix is a detailed explanation of the current countervailing duty
 methodology we use to examine grants, loans, loan guarantees, and equity. It supersedes
 the sections dealing with those issues in Appendix 2 and Appendix II. The revisions
 proposed in Appendix II have been substantially incorporated into this explanation,
 although this appendix also includes a number of further changes, principally in the areas
 of the weighted cost of capital, shape of the benefit stream, and the risk premium.
 In this process, we have proceeded largely without any clear legislative guidance.
 Although the Trade Agreements Act of 1979 ("TAA") established certain rules concerning
 the calculation of the net subsidy (see 19 U.S.C. 1677(6), the TAA did little to clarify
 issues concerning the calculation of the gross subsidy, the issue with which this appendix
 chiefly is concerned. For example, the Report of the Senate Committee on Finance on the
 TAA merely states that: "The gross subsidy is the value of the subsidy provided, or made
 available, and used." (S. Rep. No. 96-249, 96th Cong., 1st Sess. 85 (1979)). The Senate
 Report does not suggest what the "value of the subsidy" is or should be. Even on the
 question of the allocation of subsidies, the legislative history reveals nothing more
 concerete than a directive that the Department use "reasonable methods." (Id.; see also,
 H.R. Doc. No. 96-153, Part II, 96th Cong., 1st Sess. 433 (1979); H.R. Doc. No. 96-317,
 96th Cong. 1st Sess. 75 (1979)). Furthermore, the Signatories to the "Interpretation and
 Application of Articles VI, XVI, AND XXIII" ("the Subsidies Code"), which is
 implementated by the TAA, have not yet reached agreement upon rules governing the
 calculation of subsidies, as envisioned under Article 4, paragraph 2, footnote 2 ("An
 understanding among signatories should be developed setting out the criteria for the
 calculation of the amount of the subsidy.")
 Given this background, we maintain that we have "wide latitude" in which to determine the
 value of a subsidy. (See United States v. Zenith Radio Corp., 437 U.S. 443 (1978)). In our
 opinion, all that is required of us is that the methods we adopt be reasonable.
 We believe that the methodology in this appendix meet the requirement of
 reasonableness.
 We recognize that there may be alternative methods of calculating the gross value of a
 particular type of subsidy. A number of such methods were proposed by parties to these
 proceedings. As the principal institution responsible for administering the
 countervailing duty law, we selected those methods that we believe best implement
 the policies and purposes of that law.

 I. Allocating Benefits Over Time

 Funds provided under government direction or directly by the government provide a
 subsidy to the extent that the recipient pays less for the funds than it would on the
 market. In the case of a loan, this is the difference between the cash flows--the company's
 receipts and payments--on the loan under examination and the cash flows for a
 comparable commercial loan taken out by the same company. For equity, it is the
 difference between what the government paid for a share of the company and what the
 market would have paid for the share. For grants, the saving to the recipient is the face
 value of the grant; that is, the difference between what the company paid for the funds
 (nothing), and what it would have to pay on the market to receive the funds (the face
 value of the grant). The difference in cash flows can arise in a single moment, as with
 grants (complete receipt of the funds at once), or over several years, as with long-term
 loans (through periodic repayment).
 The point(s) at which the difference in cash flows occurs does(do) not always coincide
 temporarily with the economic effect of the benefit, and therefore does(do) not
 necessarily provide an appropriate schedule for assessing countervailing duties. The
 economic effect of the benefit is diffused around the time that the cash flow differential
 occurs. For example, it would be inappropriate to allocate a $1 billion grant received on
 March 17, 1981, entirely to March 17, 1981. The grant continues to benefit the company
 after that date, and thus we would not counteract the economic effect of the grant by
 assessing countervailing duties to products exported on only that single day.
 Therefore, to counteract the economic effects of such actions, we must determine an
 appropriate period over which to allocate benefits, and decide how much of the benefit to
 allocate to each subperiod (usually a year). In addition, we must choose a discount rate to
 reflect the time value of money; that is, the fact that a given nominal amount of money
 has a changing real value over time.
 The calculation of some types of subsidies, therefore, is a three step process. First, we
 must calculate the difference in cash flows between the countervailable program and the
 appropriate market alternative. Second, we must choose a discount rate for allocating
 money over time. And third, we must determine a reasonable shape and lenght for the
 stream of benefits.
 The first step is described under the separate sections below on grants, loans, and equity.
 In this section, we first consider the discount rate and then discuss the construction of the
 benefit stream.

 A. The Discount Rate

 Prior to the cases on certain steel products initiated on February 1, 1982, we allocated the
 face value of benefits with effects extending beyond the period of receipt in equal
 increments, over the appropriate time period. In each year we countervailed only that
 year's allocated portion of the total subsidy. For example, a grant of $100 million spread
 over 10 years would have been countervailed at a rate of $10 

*18017

 million per year for
 10 years, beginning in the year of receipt.
 This allocation technique was criticized for not capturing the entire subsidy in that it
 ignores the fact that money has a changing value as it moves through time. It has been
 argued that $100 million today is much more valuable to a recipient than $10 million per
 year for the next 10 years, since the present value (the value in the initial year of receipt)
 of the series of payments is considerably less than the amount if initially given as a lump
 sum. We agree with this position, and, in 1982 we changed our methodology of subsidy
 calculation accordingly (see Appendix 2). As long as the present value (in the year of
 subsidy receipt) of the amounts allocated over time does not exceed the face value of the
 subsidy, we are consistent with both our domestic law and international obligations in
 that the amount countervailed will not exceed the total net subsidy.
 The present value of any series of payments is calculated using a discount rate. The
 discount rate is a measure of the company's time preference for money. If a company is
 indifferent between receiving. $1.00 today and $1.1 next year, its discount rate for the
 intervening year is 12 percent. We choose a discount rate such that the present value of
 the cash stream remains constant. For example, if a company receives a countervailable
 grant of $1000 in 1977, we wish to countervail no more or less than 1000 1977 dollars,
 regardless of the period over which we allocate the benefit.
 A company's time preference for money is determined by it expected rate of return on
 investment and operations at the time the subsidy is received. This expected rate of
 return is often called the "opportunity cost of capital." Since this is not easily quantifiable
 or verifiable, we must choose a surrogate that accurately reflects the company's expected
 rate of return. We consider the company's actual cost of raising money, or the "weighted
 cost of capital," as the best surrogate for measuring the expected rate of return, since a
 rational company will raise money (through debt and/or equity markets) to the point at
 which the cost of raising any additional money is greater than its expected rate of return
 from those funds.
 When considering the cost of capital, we seek to determine the marginal cost facing the
 company at the apropriate time. For our purposes, therefore, the weighted cost of capital
 is the marginal cost of each type of financing used by the company (usally debt and
 equity), each weighted by its relative proportion.
 We weight-average the marginal costs of debt and equity by each company's total existing
 proportions of debt and equity. Marginal proportions of debt and equity, although more
 desirable in theory, are not readily identifiable. Because borrowing and issuing equity do
 not occur with regular frequency, any choice of a recent period to serve as a marginal
 measure of relative debt and equity usage would be arbitrary. Morever, by using total
 existing proportions, we avoid making a highly speculative guess as to how operations
 and investment would have been financed absent the subsidy. Because we cannot know
 how a company would have raised money absent the subsidy program, we assume that
 the company will use debt and equity in the same proportions at the margin as it has done
 historically.
 For the marginal cost of debt, we prefer to use the commercial rate on bonds issued by
 the company, or an interest rate on long-term commercial loans received by the
 company in the period for which we are calculating the discount rate.
 In Appendix II, we included a tax adjustment to the marginal cost of debt variable, to
 account for the fact that taxes normally are not paid on interest, thus lowering the cost of
 debt to the firm by the amount of taxes not paid. We have decided to abandon the tax
 adjustment because it is inconsistent with our policy of disregarding the secondary
 effects, including tax effects, of subsidies.
 A company's marginal cost of equity is a function of three variables: a) the risk-free rate
 on alternative investment, b) the commercial investor's anticipations of the future rate of
 return on the equity market as a whole, and c) the riskiness of investment in that
 company relative to the market. Because the latter two factors are not easily identifiable,
 we cannot calculate a company's actual marginal cost of equity. As a surrogate, we have
 used the company's marginal cost of debt (which reflects the risk-free rate and the
 company's riskiness), plus the difference between the national average rate of return on
 equity and the national average cost of debt (which reflects the anticipated future rates of
 return on the equity market). We cannot observe the anticipated rate of return on the
 equity market, and so we have used the actual rate of return on equity for calculating the
 second variable.
 Because equity normally is riskier than long-term debt, investors demand a higher return
 on equity than on lending. That is, the cost of equity (the expected rate of return of
 equity) is greater than the cost of long-term debt (its interest rate). Our use of an
 historical measure or as a surrogate for expected returns on equity, however, can lead to
 the illogical result that the measurement for the cost of equity will be less than that for the
 cost of debt.
 When investors' prior anticipations if rates if return on equity are not realized fully in the
 present period, the actual rate of return on equity may be less than the cost of debt. The
 cost of debt may also be greater than the cost of equity when less risky firms finance
 relatively more with equity, while more risky firms finance relatively more by borrowing.
 To adjust for such an anomaly, we have set a floor on the surrogate used to measure the
 national cost of equity. This floor is the national cost of debt. For our purposes, the
 weighted cost of capital in no instance will be lower than the cost of long- term debt.
 We also must consider the appropriate point in time at which to determine the weighted
 cost of capital for discount rate purposes. Where a grant or a loan is received in a year
 after the year in which agreement is reached on the principal terms of the grant or loan,
 we will determine the weighted cost of capital using the year in which agreement was
 reached rather than the year of receipt or year in which an agreement was concluded
 formally.
 We note that we may not be able to apply in all cases a discount rate based on weighted
 cost of capital as described above. This is because we may not be able to obtain
 information such as the national average long-term debt cost or rate of return on equity.
 In such cases we then fall back to the next best surrogate available based on the particular
 facts of the case. For example, for national average rate of return on equity, we may
 average returns on equity for some number of particular companies. If we believe that
 the information necessary to construct a reasonable weighted cost of capital figure is
 unavailable, we generally use instead the company's cost of long-term debt. Possible
 subsequent choices, in order of general preference, are national average long-term debt
 costs and the prime interest rate.
 Since issuing Appendix II, our experience with applying the weighted cost of capital has
 been marked by difficulties in finding the required information within the limited time
 frame for countervailing duty investigations allowed by the Act. Indeed, we have
 discovered that, in developing countries, necessary information, such as meaningful

 *18018

 national average rates of return on equity and long-term interest rates for loans
 in the home country, may not be available at all. Although we will continue our attempts
 to use the weighted cost of capital in future cases, we may be forced to change this
 practice if difficulties in finding the information systematically prohibit us from using it as
 a discount rate.

 B. Construction of the Benefit Stream

 To allocate subsidies over time, we need to construct a stream of benefits. In Appendix II,
 we continued to use an annuity-style schedule, which allocated equal nominal payment to
 each subperiod. We noted several reasons why we used the annuity-style method instead
 of the alternative preferred by one judge of the Court of International Trade, the declining
 balance method. (see Michelin Tire and Rubber Co. v. United States (6 CIT ------, Slip Op.
 83-136), December 22, 1983) ("Michelin"). First, we argued that the method that gave
 equal nominal payments was more administratively feasible than the declining balance
 method. Second, we stated that the declining balance method could result in
 countervailing duties whose present value exceeded the nominal amount of the
 subsidies in the year of receipt. Third, we argued that the declining balance method would
 require us to assign a time value to the portion of a large grant in the year of receipt,
 which was inconsistent with our decision to allocate very small grants with no time value
 adjustment to one year only. We have since adjusted the basic declining balance formula
 so that the amount of the subsidy in the first year will not exceed it s nominal value. Our
 declining balance formula also treats small grants and the first year portion of large grants
 equally. Given that the declining balance method is only slightly more difficult to apply
 than the annuity-style method, we do not consider administrative feasibility to be a
 sufficient reason in this case for our continued reliance on the latter. Although both types
 of methods are reasonable, we have decided to use the declining balance method. We use
 this method of allocation, coupled with the discount rate, to allocate certain benefits
 conferred by grants, loans, guarantees, and equity purchases.

 II. Grants

 In addition to constructing the shape of the allocation stream, we must decide the period
 over which to allocate benefits. Since the difference in cash flows for grants occurs at only
 a single moment (when the grant is received) cash flow does not provide guidance in how
 to allocate the benefit.
 It does not help to hypothesize how the company would have raised the funds absent the
 grant. Firms raise money primarily through sales, secondarily through debt, equity, and
 non-operating income. Grant money does not resemble any one of these more than any
 other; all require an obligation in return for the money, which the grant lacks. Moreover,
 for all of the alternative ways of raising money except debt, it does not make sense to ask
 the duration of the obligation; the duration of a sale is in some ways near zero, and the
 duration of equity is in some ways infinite.
 We have previously used the average useful life of renewable physical assets in the
 industry involved as the allocation period. We considered the benefit of a grant to last in
 some way as long as the average turnover of assets in the industry. However, we
 recognize first that physical assests are often a fairly small part of the costs of doing
 business, and second that even in highly capital intensive industries the benefit of funds
 received--whether from sales, debt, equity, or grants--has no particular relationship to
 the life of the machinery.
 We have concluded that there are no economic or financial rules that mandate the choice
 of an allocation period. The administering authority therefore must set a standard and
 hold to it as consistently as possible to allow its actions to be predictable. We have
 received no objections to the period proposed in Appendix II, and continue to prefer the
 average useful life of renewable physical assets as the period over which to allocate
 grants. By using the life of the company's renewable physical assets as our period of
 allocation, we are simply stating that the effects of a grant, in whatever form, can be
 spread reasonably over the time in which the company rolls over its renewable physical
 assets.
 In Appendix 2 we allocated soley to the year of receipt all grants of less than one percent
 of a company's gross revenues. As a result, under Appendix 2 we would find no subsidies
 (and make a negative determination) for a firm that received only a grant of one percent
 of the company's gross revenues, because that grant would be allocated over time.
 However, we would find subsidies for a firm receiving soley a grant of .75 percent of the
 firm's gross revenues, since that amount would be allocated soley to the year of receipt
 and would not be de minimis. To prevent this anomaly, we will total all grants. If the sum
 is less the .5 percent of all sales concerned for domestic subsidies or of all export sales
 concerned for export subsidies, we will allocate such grants only to the year of receipt.
 We will allocate all other grants over the average useful life of a company's renewable
 physical assets (equipment), as determined by U.S. Internal Revenue Service (IRS) in the
 1977 Class Life Asset Depreciation Range System (Rev. Proc. 77-10, 1977-1 C.B. 548
 (RR-38)). The use of other alternatives, such as accounting useful life as reflected in
 company records, or tax tables of the country in which the company is based, may yield
 extremely inconsistent results between companies or between countries. We continue to
 rely on IRS tax tables as our source because they provide the most practical and fair
 means of determining average useful life.

 III. Loans and Loan Guarantees

 A. Long-Term Loans and Guarantees for Companies Considered Creditworthy

 In these investigations, various loan activities give rise to subsidies. The most common
 practices are the extension of a loan on preferential terms or on terms inconsistent with
 commercial considerations where the government is either the actual lender or directs a
 private lender to make funds available, or where the government guarantees the
 repayment of the loan made by a private lender. The subsidy is computed by comparing
 what a company would pay a normal commercial lender in principal, interest, and other
 charges in any given year with what the company actually pays on the preferential loan in
 that year. We determine what a company would pay a normal commercial lender by
 constructing a comparable commercial loan at the appropriate market interest rate ("the
 benchmark") reflecting commercial terms.
 In Appendix 2, we generally used the national average commercial interest rate as the
 benchmark. We compared the loan at issue with what the average commercial borrower
 would have to pay for a loan of similar principal and duration. Upon reconsideration, we
 have decided that the benchmark for long- term loans will be company-specific, unless
 the company lacks adequate comparable commercial experience. If the latter, then we
 will use a national average loan interest rate or the debt experience of a comparable
 company as the best information available for creditworthy firms. Use of a company-
 specific benchmark for long-

*18019

 term borrowing enables us to capture the fact that
 certain companies are more (or less) risky than average, and that commercial lenders will
 take these risk characteristics into account in setting the conditions of the loan.
 For loans denominated in a currency other than the currency of the country concerned in
 an investigation, the benchmark is selected from interest rates applicable to loans
 denominated in the same currency as the loan under consideration (where possible,
 interest rates on loans in that currency in the country where the loan was obtained;
 otherwise, loans in that currency in other countries, as best evidence). The subsidy for
 each year is calculated in the foreign currency and converted at an exchange rate
 applicable for each year.
 After finding an appropriate benchmark loan, the next step in determining if, and if so, to
 what extent, a loan was given on terms inconsistent with commercial considerations is to
 calculate the payment differential between the benchmark loan and the loan at issue in
 each year. When then calculate the present value of this stream of benefits in the year the
 loan was made, using the weighted cost of capital (as described above) as the discount
 rate. In other words, we determine the subsidy value of a preferential loan as if the
 benefits had been bestowed as a lump-sum grant in the year the loan was given. This
 amount is allocated over the life of the loan, using our methodology for allocating benefits
 over time described above, to yield the annual subsidy amounts.
 If a borrowing company receives a payment holiday that is inconsistent with commercial
 considerations, the subsidy value of the payment holiday is captured in the comparison
 of the annual payments on the loan at issue with the annual payments on a normal
 commercial loan with a normal repayment schedule. A payment holiday early in the life
 of a loan can result in such large loan payments near the end of its term that, during the
 final years, the loan recipient's annual payments on the subsidized loan may be greater
 than they would have been on an unsubsidized loan. By reallocating the benefit over the
 entire life of the loan through the present value methodology described above, we avoid
 imposing countervailing duties in excess of the net subsidy.
 Because loans, unlike grants, have a readily identifiable effect on the company over time,
 we allocate the benefits over the life of the loan, even for loans expressly given to
 purchase costly capital equipment. We do not believe a contrary result for loans "tied" to
 capital equipment is required by the Act.
 Loan guarantees are countervailable only if they are provided to a specific industry or
 group of industries and only if they are on terms inconsistent with commerical
 considerations. For a creditworthy company, a loan guarantee by the government
 constitutes a subsidy to the extent the guarantee assures more favorable loan terms than
 for an unguaranteed loan. To determine if the guarantee is inconsistent with commercial
 considerations, we first compare the cost of the government loan guarantee with the cost
 of commercial guarantees. If no difference between government and commercial loan
 guarantee costs is evident, we then look to see if the government loan guarantee has
 affected the other terms of the loan.
 A special case arises when the government acts as both guarantor and principal owner or
 majority shareholder of a company. Under these circumstances, a government guarantee
 is not countervailable if it is normal commercial practice in that country for owners or
 shareholders to provide guarantees on comparable terms to their companies (see "Final
 Affirmative Countervailing Duty Determination on Carbon Steel Wire Rode from
 Trinidad and Tobago"(49 F.R. 480 (January 4, 1984)).

 B. Long-Term Loans and Loan Guarantees for Companies Considered Uncreditworthy

 In our view, a company is uncreditworthy if, absent future government support, it would
 not have been able to obtain commercial loans comparable to those which it did obtain.
 We consider a company creditworthy if it appears that it will have sufficient revenues or
 resources to meet its costs and fixed financial obligations, again absent future
 government intervention. To determine the creditworthiness of a company we analyze its
 present and past health, as reflected in various financial indicators calculated from its
 financial statements and accounts. We give great weight to the company's recent past and
 present ability to meet its costs and fixed financial obligations with its cash flow. Where
 available, we also consider evidence of the company's future financial position, such as
 market studies, country and industry economic forecasts, and project and loan
 appraisals. Because the determination is often highly complex, we consider each case
 carefully in light of the evidence on the record.
 In Appendix 2, we calculated the benefit of a long-term loan to a company considered
 uncreditworthy by treating it as a countervailable equity infusion and applying our
 equity methodology. At the time, we stated our preference for using a loan-type
 calculation to measure the benefit, with a suitable risk premium added to the benchmark.
 However, we could not find any reasonable and practical basis for selecting a risk
 premium. Thus, in 1982, we settled for the equity approach, even though we were aware
 of the flaws in our equity methodology, as discussed in Appendix II.
 We now believe that we have a practical and reasonable way to calculate a risk premium.
 For our purposes, a risk premium is the amount above the highest commonly-available
 commercial interest rate a creditworthy borrower would have to pay in order to receive a
 loan. The magnitude of the risk premium is determined solely by the lender's assessment
 of the riskiness of the company. Thus, to construct a risk premium, we need an observable
 measurement of risk, as determined by lenders. U.S. bond ratings provide us with such a
 measurement, since the difference between interest rates associated with different bond
 ratings are determined solely by risk.
 Having settled upon U.S. bond ratings, we must now determine the appropriate spread to
 adopt as our risk premium. In Appendix II, we proposed using the interest rate spread
 between the lowest commonly observed bond rating for creditworthy companies and the
 second lowest, on the basis that the last increment of risk within the range of bond ratings
 for creditworthy companies best represented the difference in risk between a marginally
 creditworthy company and an uncreditworthy company. We now consider this spread to
 be inappropriate, since the difference in risk between the least creditworthy company and
 the next-to-least creditworthy company has very little relation to the difference in risk
 between the least creditworthy company and an uncreditworthy company. The spread
 proposed in appendix II does not reflect the fact that a company's level of risk increases
 dramatically once it becomes uncreditworthy. A more appropriate measure of the risk
 between a marginally creditworthy company and an uncreditworthy company is the
 difference in interest rates associated with the difference between the least creditworthy
 and most creditworthy bond ratings. Although it is impossible to quantify the risk of an
 uncreditworthy company precisely, we beieve that this spread comes closer to measuring
 the dramatic increase in risk in lending to 

*18020

 an uncreditworthy company than the
 spread proposed in Appendix II.
 For purposes of these final determinations, we have used the difference between Moody's
 Aaa and Baa corporate bond rates. We then calculated the precentage this difference
 represents of the prime interest rate in the United States. This precentage is applied to the
 prime interest rate in the country concerned. The resulting risk premium is then added to
 the highest long-term commercial interest rate commonly available to companies in that
 country.
 We believe this method is practical. Moreover, it seems reasonable, since the spread in
 riskiness among companies in the United States, which has a highly sophisticated bond
 market, reasonably reflects market forces determining a measurement of risk. By
 applying our risk premium, we expect to achieve a meaningful measure of the value to
 uncreditworthy companies of government support in obtaining loans. As under Appendix
 2, we will not impose greater countervailing duties for a subsidized loan (to a
 creditworthy or an uncreditworthy company) than for an outright grant in the amount of
 the loan principal, because a loan cannot be worth more to a company than an outright
 grant of the same amount.

 C. Short-Term Loans

 Short-term loans (one year or less), like long-term loans, are countervailable to the
 extent that they are preferential or given on terms inconsistent with commercial
 considerations. To determine the commercial soundness of short-term loans, we compare
 the terms on the loan at issue with a benchmark; that is, a comparable commercial loan.
 Since short-term looans are received and repaid within a year, we allocate any benefits to
 one year only. Therefore, we do not need to employ present value analysis for short-term
 loans.
 For our benchmark, we use the most appropriate national average commercial method of
 short-term financing, rather than company-specific experience. We believe the
 distinction between our treatment of short-term and long-term loans is valid. Lending
 short-term generally is not as risky as long-term, because of the shorter duration of the
 repayment obligation and the greater frequency of accompanying security (for example,
 accounts receivable). Because there is little need for the lender to vary its terms to
 account for varying risk characteristics among companies, we would not expect
 company-specific short- term loan terms to vary from national average terms.
 Additionally, because of the enormous number of short-term loans involved in many
 cases, the use of company-specific benchmarks would significantly impair our ability to
 administer the countervailing duty law within the short time limits established by the
 Act.
 We do not treat uncreditworthy companies differently from creditworthy companies
 when calculating benefits from short-term loans, because of the low level of risk
 associated with short-term debt, and the frequent existence of security.

 D. Forgiveness of Debt

 Where the government has permanently assumed or forgiven an outstanding debt
 obligation, we treat this as a grant to the company equal to the outstanding principal at
 the time of assumption or forgiveness. Where outstanding debt has been converted into
 equity (that is, the government receives shares in the company in return for eliminating
 debt obligations of the company), a subsidy may result. The existence and extent of such
 subsidies are determined by treating the conversion as an equity infusion in the amount
 of the remaining principal of the debt. We then calculate the value of the subsidy by using
 our equity methodology explained below.

 IV. Equity

 It is well settled that government equity ownership per se does not confer a subsidy.
 Government ownership confers a subsidy only when it is on terms inconsistent with
 commercial considerations.
 If the government buys previously issued shares on a market or directly from
 shareholders rather than from the company, there is no subsidy to the company. This is
 true no matter what price the government pays, since any overpayment benefits only the
 prior shareholders and not the company.
 If the government buys shares directly from the company (either a new issue or
 corporate treasury stock) and similar shares are traded in a market, a subsidy arise if the
 government pays more than the prevailing market price. We strongly prefer to measure
 the subsidy by reference to market price. This price, we believe, rightly incorporates
 private investors' perceptions of the company's future earning potential and worth.
 It is more difficult to judge the possible subsidy effects of government equity purchases
 where there is no market price for the shares (as where, for example, the government is
 already sole owner of the company). In such cases, we must determine the commercial
 soundness of government equity purchases by assessing the prospects of the company at
 the time those purchases were made.
 To be "equityworthy," a company must show ability to generate a reasonable rate of
 return within a reasonable period of time. In making our equityworthiness
 determinations, we assess the company's current and past financial health, as reflected in
 various financial indicators taken from its financial statements, and, where appropriate,
 internal accounts. We give great weight to the company's recent rate of return on equity
 as an indication of financial health and prospects. Like our creditworthiness tests, our
 equityworthiness analysis also takes into account the company's prospects, as reflected in
 market studies, country and industry forecasts, and project and loan appraisals, when
 these types of analyses are available.
 For government equity purchases which we deem inconsistent with commercial
 considerations, we measure the benefit by multiplying the difference between the
 company's rate of return on equity and the national average rate (the "rate of return
 shortfall") for the review period by the total amount of the equity purchases made in
 years in which the company was unequityworthy. Under no circumstances do we
 countervail in any year an amount greater than that which is calculated treating the
 government's equity infusion as an outright grant.
 * * * * * 

 COMMENTS BY PARTIES TO THE PROCEEDING

 I Allocating Benefits Over Time

 A. The Discount Rate

 Comment 1. Petitioner, Bethlehem, and the Four argue that the cost of debt variable in the
 Department's weighted cost of capital formula should not be adjusted for taxes. This
 adjustment, they contend, is contrary to the Department's long-standing practice of not
 taking into account the tax effects of subsidies, a policy which has been upheld by the
 Court of International Trade (see Michelin). Petitioner further states that such an
 adjustment constitutes an impermissible offset to the gross subsidy.
 DOC Position. The discount rate is simply a means of allocating benefits over time when
 calculating subsidies such as grants, loans, and equity infusions. However, because it has
 an effect on the magnitude of the benefit, there is no clear distinction between the tax
 effect on the discount rate and tax effects on subsidies. Therefore, we have removed the
 tax adjustment in our weighted cost of capital formula.

 *18021

 Comment 2. Petitioner states that the company-specific debt variable of our
 weighted cost of capital formula should include a "risk premium" for all uncreditworthy
 companies.
 DOC Position. We agree. For our purposes, the weighted cost of capital is the cost of
 commercial financing facing the company. Since the risk premium, in a sense, is part of
 the commercial cost of debt facing an uncreditworthy company, we now include it in the
 marginal cost of debt variable in our weighted cost of capital formula.
 Comment 3. Bethlehem argues that the Department should not rely exclusively on its cost
 of equity surrogate for all cases. Where available, the Department should consider
 alternative measurements of the cost of equity, such as rates of return expected by
 venture capitalists and "turn around" experts, or rates of return based upon "beta," or risk
 factor, analysis.
 DOC Position. The alternative ways of estimating the cost of equity suggested by
 Bethlehem are more speculative and much more difficult to apply than the methodology
 we outline in this appendix. Therefore, we will continue to apply our methodology for
 determining the cost of equity.

 B. Construction of the Benefit Stream

 Comment 4. Petitioner and the Four contend that the Department should use the
 declining balance method, as opposed the annuity-style method, to allocate benefits over
 time. They maintain that the declining balance method is only slightly more difficult to
 apply than the annuity-style method and that the switch from one method to the other
 would require only a slight change in the Department's computer program. They further
 argue that the declining balance method does not affect small grants allocated to the year
 of receipt. The Department's argument that the use of the declining balance method may
 result in a present value whose nominal value exceeds the subsidy in the year of receipt is
 circular, in that it assumes that the correct present value is calculated by using an
 annuity-style method. In addition, the Four assert that the annuity-style method
 "backloads" the subsidy.
 DOC Position. For the reasons expressed above in this appendix, we have adopted a
 declining balance formula.
 Comment 5. Bethlehem holds that the annuity-style method of allocating benefits is
 appropriate where commercial debt and equity financing are likely financial alternatives.
 Bethlehem bases its argument on the fact that the Department has no way to estimate the
 likely benefits in different years. Thus, Bethlehem concludes that the use of a constant
 figure over time is acceptable.
 DOC Position. The choice of a method of allocating long-term benefits is not necessarily
 related to the form of repayment that the alternative financing might take. Although we
 are aware of several possible methods of allocating benefits, none of them is clearly
 superior to the others, and none of them is mandated by the Act or its legislative history.
 Having selected a reasonable method, we hold that the same method should be used to
 allocate all long-term benefits, as long as our choice is reasonable. By consistently using
 the same method, we avoid speculating on the form of the financing foregone.

 II. Grants

 Comment 6. Although not objecting in this case to the allocation of grants over the
 average useful life of the equipment, Bethlehem disapproves of the Department's focus on
 the use to which the funds were allegedly put.
 DOC Position. Again, we are aware of several reasonable choices of periods over which to
 allocate grants, but none of them is clearly superior to the others, and none of them is
 mandated by the Act or its legislative history. We originally chose the average useful life
 of the assets because we believed the benefits of a grant somehow had a life
 approximating the life of assets (see Appendix 2). We now consider this belief wrong; the
 life of the benefits is not observable. However, because we have never received
 objections to our useful life approach, and have not found any more reasonable period
 over which to allocate grant benefits, we will continue to use average useful life of the
 company's renewable physical assets.

 III. Loans and Loan Guarantees

 A. Long-term Loans and Loan Guarantees for Companies Considered Creditworthy

 Comment 7. Bethlehem states that the use of present value methodology in the
 Department's loan calculations is unnecessary in cases where the only difference between
 the terms of the countervailable loan and the benchmark loan is the interest rate.
 DOC Position. The Department applies the present value methodology in its loan
 calculations because, as explained above in this appendix, it allows us to countervail only
 the net subsidy in cases where, because of a payment holiday in the early life of the loan,
 the annual payments on the subsidized loan are greater than those on the unsubsidized
 loan. While it is true that the present value method normally does not produce a result
 that is materially different from that of a calculation based only on the difference between
 repayment schedules, it allows us to treat all loans in a like manner. For the sake of
 consistency, therefore, we continue to use the present value method in all of our
 long-term loan calculations.
 Comment 8. The Four assert that the Department should not declare a loan not
 countervailable on the basis that unspecified amounts of additional charges, such as
 guarantees, currency exchange requirements, or security, raise the effective cost of the
 loan at issue above those on the benchmark loan. Rather, the Department must choose a
 benchmark and compare it with the potentially countervailable loan.
 DOC Position. We agree. Before a loan can be considered not countervailable, we compare
 the quantifiable terms of the alleged countervailable loan with a commercial benchmark.
 Where possible, we compare the cost of the government guarantee with that of a
 commercial guarantee.

 B. Loans and Loan Guarantees to Companies Considered Uncreditworthy

 Comment 9. The Four propose an alternative, which it terms the "creditworthiness
 proxy," to our risk premium methodology as a means of valuing long-term loans to
 uncreditworthy companies. This alternative involves calculating the minimum cash flow
 available to cover interest charges that is considered necessary by lenders before they
 will give a company a commercial loan. The difference between the company's actual cash
 flow and the minimum cash flow amount should be treated as a grant. This method, the
 Four point out, is relatively easy to administer and serves as a reasonable proxy for the
 effect of government involvement on the company's ability to borrow on the market. The
 Four provide statements of experts in support of this method.
 DOC Position. Although the Four's proposal has some merit, we are concerned by the
 absence of any tangible link between the amount being countervailed (the shortfall in
 cash flow) and the program at issue (loans). For example, under this method, a loan of one
 dollar to an uncreditworthy company could result in a net subsidy of one million dollars if
 one million dollars were the difference between the company's actual cash flow and the

 *18022

 minimum cash flow. In addition, the minimum cash flow requirement chosen is
 arbitrary and does not represent a reasonable benchmark for creditworthiness across
 companies. while the requirement suggested by the Four may indeed be a common "rule
 of thumb" popular in American banking circles, we have no evidence that it is generally
 applicable to all countries. For the time being, therefore, we prefer the risk premium to
 the creditworthy proxy method.
 Comment 10. Petitioner proposes that we use the highest commonly observed interest
 rate as the risk premium, arguing that it is a "well tested rule that the risk of doing
 anything new requires an uncertainty premium at least as high as the interest rate.
 DOC Position. We have not been presented with enough evidence to assess the
 reasonableness of petitioner's proposal. At present, we do not believe that petitioner's
 method is superior to our own.
 Comment 11. Petitioner and Bethlehem maintain that, if the Department uses an interest
 rate spread derived from U.S. bond ratings for its risk premium, at the least if should use a
 larger spread, instead of the spread between the least (Baa) and next-to-least (A)
 creditworthy ratings. Both petitioner and Bethlehem contend that the spread used by the
 Department in its preliminary determinations in these investigations understates the
 subsidy, because there is no relation between the riskiness of loans to creditworthy
 companies and those to uncreditworthy companies. Petitioner argues that at a minimum,
 the Department should use the spread between A and C bonds. Bethlehem contends that
 the appropriate spread is that between Baa bonds as publishsed by Moody's and an
 average of a list of Ba, B, Caa, Ca, and C bonds, which the Department should compile by
 contacting investment banker, insurance companies, and other significant sources of debt
 capital.
 DOC Position. For the reasons stated above in this appendix, we are using the bond rating
 spread from Baa-Aaa to calculate the risk premium. This is the largest regularly published
 bond rating spread we have been able to find.

 C. Short-Term Loans

 Comment 12. The Four maintain that the Department must use effective interest rates in
 its calculation of the benefits from countervailable loans.
 DOC Position. The magnitude of the benefit from loans is a function of the difference
 between the cost of the loan under examination and the cost of the benchmark loan.
 Ideally, we attempt to quantify the total effective cost of each type of loan in our
 comparisons. However, the charges added on to the nominal interest rates for each loan
 cannot always be quantified. In these cases, we base our calculations on the difference
 between the quantifiable equivalent terms of both loans. Thus, we use nominal interest
 rates in our calculations when we are unable to quantify additional equivalent terms.
 Comment 13. Bethlehem argues that the Department should calculate the benefit from
 short-term loans by using company-specific benchmark interest rates instead of national
 average rates in cases where there is a wide variation in short-term interest rates.
 Similarly, SIDERMEX maintains that the Department should use company-specific
 interest rates in its short-term loan calculations, because national average rates do not
 accurately capture the benefit to the company.
 DOC Position. For our purposes, we believe that the use of a national average interest rate
 as the commercial benchmark for short-term loans captures the benefit to the companies
 with sufficient accuracy. Even though we have found variations in short-term interest
 rates, our overall experience indicates that such variations are generally small. The
 relatively small increase in accuracy gained by measuring short-term loan benefits by
 using company-specific benchmarks is outweighed by the burdensome increase in
 administrative burden such a policy would involve.

 IV. Equity

 Comment 14. Petitioner, Bethlehem, and the Four argue that the Department should treat
 countervailable equity infusions as grants. The Four state that treating equity infusions as
 grants has several advantages over the Department's rate of return shortfall
 methodology, including administrative feasibility and less uncertainty in application. In
 addition, the Four insist that equity infusions into unequityworthy companies have
 grant-like qualities, because the government cannot expect realistically any return from
 its investment. Should dividends be paid to government, the Four state that the
 Department can subtract them from the amount countervailed. Bethlehem maintains that
 the Department's rate of return shortfall methodology wrongly focuses on the cost to the
 donor, rather than the benefit to the recipient.
 DOC Position. We disagree that an equity infusion in an unequityworthy company has the
 characteristics of a grant. The essential difference between an equity purchase and the
 bestowal of a grant is the potential for return on equity. The domestic interested parties
 contend that the potential for any return on investment in these companies is mimimal,
 and therefore the Department should this difference in its treatment of equity infusions.
 Their argument focuses on the poor prospects for potential dividends at average rates,
 while ignoring the potential return in terms of retained earnings or increasing worth tof
 the company. Because we cannot discount this potential at the time the infusion is made,
 we should not treat equity infusions as an outright grant. To do so would raise the
 possibility of countervailing more than the net subsidy in cases where the government
 receives a return, in retained earnings or increasing worth from its investment.
 When we say that a company is unequityworthy, we are not saying that a private investor
 could not expect any return or even to recoup his original investment. Rather, we are
 simply stating that a private investor would not have purchased equity because he could
 not expect a reasonable rate of return. The treatment of equity infusions as grants,
 however, implies that the government could expect no return, in terms of dividends,
 retained earnings, or through increased worth, from its investment. We lack the ability to
 look into the future that would be necessary to make such a judgement. Because of the
 difference betwen equity purchases and grants, the treatment of equity infusions as grants
 is inappropriate. We believe that our "rate of return shortfall methodology appropriately
 measures benefits from equity by an equity- based standard. Moreover, because it is
 based on the average rate of return a company would have to pay to attract investors, it
 accurately measures the benefit to the an unequityworthy company from government
 equity infusions.
 Comment 15. Petitioner contends that the Department should subtract all domestic
 government subsidies received in prior years from the company's financial data before
 determining if government equity infusions constitute subsidies, if the company is
 uncreditworthy, and the magnitude of any subsidy from government equity infusions. To
 include domestic government subsidies in the financial data, petitioner states, would
 result in the elimination of certain non-equity benefits from the Department's
 calculations, and the violation of the Department's private investor standard for
 determining when government equity infusions are countervailable.

 *18023

 DOC Position. We have maintained consistently that subtracting out domestic
 government subsidies, contrary to petitioner's assertion, would result in the use of a
 standard different from that of a private investor to determine equityworthiness. When
 deciding to invest, a private actor will assess the financial position of the firm at that point
 in time. He will look upon any past investments, including his own, as sunk costs,
 irrelevant to his analysis of whether additional investment will earn a reasonable rate of
 return. Similarly, we assess the commercial soundness of additional government equity
 infusions by considering all prior investment and provision of subsidies as sunk costs and
 looking at the financial position of the company at the time the infusion is made.
 Petitioner's method would require us to look at the financial position of the company in a
 manner different from that of a private investor.
 Nor do we believe that our method results in our not countervailing certain subsidies.
 Petitioner uses the example of a labor subsidy, which, by decreasing the compay's cost of
 production, will decrease the size of its loss. The result is that the company's rate of return
 shortfall on equity is lessened by the amount of the labor subsidy. Consequently,the
 equity subsidy also decreases by the amount of the labor subsidy, allegedly resulting, in
 effect, in the labor subsidy not being countervailed.
 In this example, the petitioner is equating the secondary effect of the labor subsidy with
 the subsidy itself. The labor subsidy lowers the company's cost of production. By
 assessing countervailing duties against the labor subsidy, we neutralize the cost of
 production advantage. The labor subsidy also has a secondary effect on the company's
 financial position. We have consistently maintained that we do not look at the secondary
 effects of subsidies, because such analysis is highly speculative. By subtracting out
 domestic government subsidies from our creditworthy and equityworthy analyses, we in
 effect would be taking into account their secondary effect on company's financial
 position. We note that the Court of International Trade has endorsed our view that we do
 not look at the secondary effects of subsidies (see Michelin). We also note that, in another
 context, the domestic interested parties have argued against the consideration of the
 secondary effects of subsidies (see Comment 1 above).
 Comment 16. Petitioner contends that if the Department decides to apply its rate of return
 shortfall method in valuing subsidies from equity infusions, it should compare the
 company's rate of return with the national average cost of debt in those cases where the
 national average cost of debt exceeds the national average rate of return on equity.
 DOC Position. A company in which the government participates should not be expected
 to earn a rate of return greater than the national average rate. accordingly, it would be
 inappropriate to countervail against any amount greater that the national average rate of
 return on equity.
 Comment 17. Bethlehem asserts that for companies whose existing stock is publicly
 traded, the Department should not always value the subsidy from equity infusions as the
 difference between what the government paid per share and the publicly-traded price.
 The price of its existing shares, Bethlehem reasons, may have no relation to the price the
 market would pay for new shares.
 DOC Position. In calculating the subsidy from equity infusions, we strongly prefer to
 compare the government action with actual market behavior. The publicly traded price,
 we believe, is a much more accurate indicator of the company's future earnings potential
 and worth than any hypothetical measurement which we could devise. It is a much more
 reliable and accurate gauge as to whether, and if so, to what extent, government equity
 infusions are inconsistent with commercial considerations. Thus, the Department uses the
 publicly-traded share price to measure the value of government equity infusions where
 such information is available. Wherever possible, we use the public price8 immediately
 following the issuance of new shares to allow for the effect of the dilution in ownership on
 price.
 Comment 18. Bethlehem argues that if a company is considered uncreditworthy, the
 Department should find that any equity infusions are necessarily inconsistent with
 commercial considerations.
 DOC Position. Although many of the factors we consider in our equityworthiness and
 creditworthiness determinations are the same, the commercial considerations relevant to
 the granting of a loan will not be identical to those considered in a decision to make an
 equity investment. Therefore, while we recognize that we often will find uncreditworthy
 companies also to be unequityworthy, we do not believe that as a per se rule a
 determination of uncreditworthiness necessitates a determination of unequityworthiness.

 [FR Doc. 84-11312 Filed 4-25-84; 8:45 am]

 BILLING CODE 3510-DS-M