Certain Cotton Yarn Products From Brazil; Final Results of Countervailing Duty

Administrative Review

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

ACTION: Notice of final results of countervailing duty administrative review.

SUMMARY: On March 23, 1988, the Department of Commerce published the preliminary results of its administrative review of the countervailing duty order on certain cotton yarn products from Brazil. We have now completed that review and determine the net subsidy to be de minimis for five firms, 12.15 percent ad valorem for Cotonoficio Guilherme Giorgi, and 2.56 percent ad valorem for all other firms during the period May 18, 1984 through December 31, 1984. We also determine the net subsidy to be 22.30 percent ad valorem for Kanebo, 7.75 percent ad valorem for Cotonificio Guilherme Giorgi, and 12.82 percent ad valorem for all other firms during the period January 1, 1985 through December 31, 1985.

EFFECTIVE DATE: February 1, 1990.

FOR FURTHER INFORMATION CONTACT:Philip Pia or Paul McGarr, Office of Countervailing Compliance, International Trade Administration, U.S. Department of Commerce, Washington, DC 20230; telephone: (202) 377-2786.

SUPPLEMENTARY INFORMATION:

Background

On March 23, 1988, the Department of Commerce ("the Department") published in the Federal Register (53 FR 9465) the preliminary results of its administrative review of the countervailing duty order on certain cotton yarn products from Brazil (42 FR 14089; March 15, 1977). We have now completed this administrative review in accordance with section 751 of the Tariff Act of 1930 ("the Tariff Act").

Scope of Review

The United States, under the auspices of the Customs Cooperation Council, has developed a system of tariff classification based on the international harmonized system of Customs nomenclature. On January 1, 1989, the United States fully converted to the Harmonized Tariff Schedule (HTS), as provided for in section 1201 et seq. of the Omnibus Trade and Competitiveness Act of 1988. All merchandise entered, or withdrawn from warehouse, for consumption on or after that date is now classified solely according to the appropriate HTS item number(s).
Imports covered by this review are shipments of Brazilian yarn, carded but not combed, wholly of cotton. During the review period, such merchandise was classifiable under items 301.01 through 301.98, inclusive, and under item 302.--with statistical suffixes 20, 22, and 24 of the Tariff Schedules of the United States. This merchandise is currently classifiable under HTS items 5205.11.10, 5205.11.20, 5205.12.10, 5205.12.20, 5205.13.10, 5205.13.20, 5205.14.10, 5205.14.20, 5205.15.10, 5205.15.20, 5205.31.00, 5205.32.00, 5205.33.00, 5205.34.00, and 5205.35.00. The HTS items are provided for convenience and Customs purposes. The written description remains dispositive.
The review covers the period May 18, 1984 through December 31, 1985 and 20 programs: (1) CACEX export financing; (2) an income tax exemption for export earnings; (3) the IPI export credit premium; (4) CIC-CREGE 14-11 financing; (5) BEFIEX; (6) Price Equalization Program; (7) FST financing; (8) incentives for trading companies ("Resolution 883"); (9) accelerated depreciation for Brazilian-made capital goods; (10) tax reductions on export production equipment ("CIEX"); (11) export financing under Resolution 68 ("FINEX"); (12) duty-free treatment and tax exemption on equipment used in export production ("CDI"); (13) export financing under the Fundo Nacional de Participacoes ("FUNPAR"); (14) exemption from state-administered value-added taxes ("CM") on domestic sales; (15) export production financing ("PROEX"); (16) benefits from import substitution ("PROSIM"); (17) financing for the storage of merchandise destined for export ("Resolution 330"); (18) Green- Yellow drawback; (19) federal cotton auctions; and (20) federal stock (EGF) loans. Our analysis and findings for each of these programs is the same as described in our preliminary results notice except where stated within the comments.

Analysis of Comments Received

We gave interested parties an opportunity to comment on the preliminary results. We received comments from the Government of

*3443

Brazil and the American Yarn Spinners Association ("AYSA").

Comment 1: The Brazilian government argues that the Department should use for its short-term loan benchmark the annual interest rate in effect on the date that each loan was obtained instead of the average annual rate for the review period. In a high-inflation economy, such as exists in Brazil, an average rate for the review period distorts the actual interest differentials on each loan. In Canned Tuna from the Philippines; Final Results of Countervailing Duty Administrative Review (52 FR 43758; December 4, 1986), the Department used quarterly, rather than annual, average rates to derive a more accurate benchmark for a high-inflation economy. Similarly, in Final Affirmative Countervailing Duty Determination and Countervailing Duty Order; Certain Steel Wire Nails from Thailand (52 FR 36987; October 2, 1987), the Department stated that: "Using six-month benchmark rates * * * is not warranted in this instance, because no major fluctuation in the Thai economy occurred during the time the loans were obtained." Since there was a "major fluctuation" in Brazilian interest rates during the review period, the Brazilian government argues that the Department should depart from its normal practice of using annual average benchmark rates.

Department's Position: In Thai Nails, we found that the relative stability in Thai credit markets justified the use of an annual average interest rate as a benchmark rather than using the commercial interest rates available at the time the company actually borrows. Although it is true that in Brazil, unlike in Thailand, interest rates fluctuated considerably during the review period, we continue to believe that an average annual interest rate is the appropriate benchmark in this case.
The predominant short-term lending instrument commercially available in Brazil during the review period was a 90-day discount of accounts receivable. Our benchmark is a yearly average of these discount rates, which are published weekly in Analise/Business Trends. In order to borrow a given amount for one year during the review period, a commercial borrower in Brazil would have been required to roll over a 90-day discounted loan three times. Each rollover would consist of discounting the original principal by the commercial discount rate prevailing on the day of the rollover. Given that discount rates varied considerably from one week to the next in Brazil during the review period, it is unlikely that any of the rollover rates would be equal. To assume that the discount rate in effect on the day the loan was originally contracted would apply to the three subsequent rollovers is unfounded. If the loan were contracted at a time that the discount rate was below the yearly average, applying the less-than-average rate to the entire year would understate the benchmark. Likewise, if the loan were contracted at a time that the discount rate was above the yearly average, applying the greater-than-average rate to the entire year would overstate the benchmark. Therefore, an annual average benchmark is the best reflection of the actual commercial cost of borrowing at variable rates over a twelve-month period.
In Philippine Tuna, we used the quarterly commercial benchmark rates because there were major fluctuations in Philippine interest rates during the review period and, unlike in Brazil, one-year fixed-rate loans were a commercially- available alternative.

Comment 2: The Brazilian government contends that the Department should use as its short-term loan benchmark the average commercial bank lending rates published by Morgan Guaranty Trust Company in its World Financial Markets instead of the average of weekly accounts receivable discount figures published in Analise/Business Trends. Commercial bank lending practices are most similar to Resolution 674/882 financing, the source of Morgan Guaranty's figures.

Department's Position: We have considered and rejected this argument in other Brazilian countervailing duty cases. See, e.g., Certain Carbon Steel Products From Brazil; Final Results of Countervailing Duty Administrative Review (52 829; January 9, 1987). The Brazilian government has provided neither new evidence nor new arguments which convince us to reconsider this issue. We continue to believe that an average of the weekly accounts receivable discount rates is the appropriate basis for deriving our benchmark.

Comment 3: The Brazilian government argues that the Department overstated the short-term loan interest rate benchmark by compounding monthly interest rates. If the Department continues to use the annual average discount rate as its benchmark, it should adjust its compounding methodology. Since discounting requires advance payment of interest, compounding a nominal monthly interest rate to obtain an annual effective interest rate results in an artificial benchmark that overstates the benefit. The Department should calculate an effective rate for the ninety-day discount of accounts receivable and multiply this rate by four to annualize the benchmark. This calculation would take into account the quarterly rollover of principal.

Department's Position: After further review, we have adjusted our methodology to better reflect commercial bank lending practices in Brazil during the review period. The most common form of short-term commercial financing in Brazil during the review period was a 90-day discount of accounts receivable. Typically, a loan was quoted at a nominal monthly discount rate, meaning that interest was paid up front. However, if a firm in Brazil wanted to secure a loan for one year, a bank would roll over the loan each quarter by discounting the original principal based on the monthly nominal rate in effect at the beginning of each quarter that the loan is outstanding. Therefore, we have calculated an annual effective interest rate benchmark by following the methodology prevalent among Brazilian commercial banks during the review period. We multiplied the average monthly discount rates in 1983 and 1984 by three to obtain a 90-day discount rate. We then converted the 90-day discount rate to an interest rate. Finally, we compounded this rate four times to obtain an annual effective interest rate. On this basis, we have revised our benchmarks from 159.73 percent to 183.55 percent for 1983, and from 324.12 percent to 428.02 percent for 1984. Using the revised benchmark, we determine the benefit from CACEX export financing for 1984 to be 9.75 percent ad valorem for Cotonificio Guilherme Giorgi, and 1.30 percent ad valorem for all other companies in 1984, except those companies with de minimis aggregate benefits. For 1985, we determine the benefit to be 5.33 percent ad valorem for Kanebo, 1.07 percent ad valorem for Cotonificio Guilherme Giorgi and 3.29 percent ad valorem for all other companies. See also our responses to Comments 9 and 10.

Comment 4: The Brazilian government claims that, in calculating the short-term interest rate benchmark, the Department should not include the tax on financial transactions (the "IOF"). The IOF functions as an indirect tax, and neither the exemption nor the rebate of an indirect tax is considered a subsidy under the General Agreements on Tariffs and Trade and U.S. law. Inclusion of the IOF in the benchmark improperly countervails an exemption of an indirect tax applicable to exports.

Department's Position: We have considered and rejected this argument in

*3444

other Brazilian countervailing duty cases. See, e.g., Certain Castor Oil Products From Brazil; Final Results of Countervailing Duty Administrative Review (48 FR 40534; September 8, 1983).The Brazilian government has provided neither new evidence nor new arguments which convince us to reconsider this issue.

Comment 5: The Brazilian government argues that the Department failed to take into account a change in the program of preferential working-capital financing for exports, administered by CACEX and the Banco do Brasil, in calculating the duty deposit rate. During the review period cotton yarn exporters made interest payments on loans from this program authorized under Resolutions 674, 882, 950 and 1009. Each successive Resolution superseded and amended the previous one.
Under Resolution 674, there was an interest rate ceiling on these loans of 60 percent. On January 1, 1984, Resolution 882 changed the payment date for both interest and principal to the expiration date of the loan without changing the interest rate ceiling. On August 21, 1984, Resolution 950 made these loans available at prevailing market interest rates less an equalization fee of 10 percentage points paid by the Banco do Brasil to the lending bank.
On May 2, 1985, Resolution 1009 changed the equalization fee to 15 percentage points. Under the Department's methodology, this results in an interest differential of 16.5 percent (the equalization fee plus the 1.5 percent IOF). The Department should calculate the deposit rate based on borrowing patterns of each firm in 1985 at the applicable interest differential of 16.5 percent.

Department's Position: We agree and have adjusted our calculations accordingly. On this basis, we determine that for purposes of cash deposits of estimated countervailing duties the benefit from this program is 1.44 percent ad valorem for Cotonificio Guilherme Giorgi, 3.45 percent ad valorem for Kanebo, and 1.28 percent ad valorem for all other firms.

Comment 6: The Brazilian government claims that the Department incorrectly allocated the benefits from the income tax exemption for export earnings program over export sales instead of total sales. Since the program rebates direct taxes, it is a domestic subsidy, which requires the Department to allocate the benefit over total sales. Further, effective January 1, 1988, the Government of Brazil decreed that export earnings are no longer fully exempt from income taxes. The Department should take into account this change in calculating the cash deposit rate for this program.

Department's Position: We have considered and rejected the first argument in other Brazilian countervailing duty cases. See, e.g., Certain Carbon Steel Products From Brazil (op. cit.). The Brazilian government has provided neither new evidence nor new arguments which convince us to reconsider this issue. Regarding any changes in this program, we do not have sufficient information to recalculate the cash deposit rate. The Government of Brazil provided neither the official decree establishing the change nor an explanation of the effect of the change on this case.

Comment 7: The Brazilian government contends that the Department incorrectly found that the export tax offsets only the countervailable benefit from the IPI export credit premium. The Brazilian government instituted a staged reduction of the IPI export credit premium from 11 percent in 1984 to zero effective May 1, 1985. During this period, however, the companies paid an export tax of 11 percent, which was greater than the average IPI export credit premium received. Since section 771(6) of the Tariff Act requires the Department to consider the full amount of export taxes paid during the review period, the Department should count the overpayment as an offset to the gross subsidy from all programs. In recent, notices, the Department calculated the total amount of net subsidy by subtracting the total amount of export taxes paid. See, e.g., Certain Carbon Steel Products From Brazil (op. cit.), Certain Stainless Steel Products From Brazil; Final Results of Countervailing Duty Administrative Review and Renegotiation of Suspension Agreement (51 FR 45371; December 18, 1986), and Certain Tool Steel Products From Brazil; Final Results of Countervailing Duty Administrative Review and Renegotiation of Suspension Agreement (51 FR 45376; December 18, 1986). The Department should follow these precedents in this case.
Department's Position: We disagree. Section 771(6)(C) of the Tariff Act provides that:
for the purpose of determining the net subsidy, the administering authority may subtract from the gross subsidy the amount of * * * export taxes, duties or other charges levied on the export of the merchandise to the United States specifically intended to offset the subsidies received (emphasis added).
The export tax that the Brazilian government is referring to was specifically intended to offset the subsidy received by each company from the IPI export credit premium. Therefore, we have offset completely only the subsidy received by each company from the IPI export credit premium program.
In the suspended investigations of Tool Steel and Stainless Steel, the Brazilian government imposed an export tax with the intent of offsetting the net subsidy on the merchandise exported to the United States. This export tax was established in accordance with section 704(b)(1) of the Tariff Act, which provides that:
The administering authority may suspend an investigation if the government of the country in which the subsidy practice is alleged to occur agrees, * * * to eliminate the subsidy completely or to offset completely the amount of net subsidy, with respect to that mechandise exported directly or indirectly to the United States, within 6 months after the date on which the investigation is suspended, * * *
Therefore, we allowed the export taxes in these cases to offset the total net subsidy.
In Carbon Steel, the Brazilian government, after unilaterally imposing a 27.42 percent export tax on shipments of certain carbon steel products between publication of the preliminary and final determinations, requested that the Department enter into a suspension agreement. While we exercised our discretion not to enter into a suspension agreement based on an export tax, we did allow the export tax because it was specifically intended to offset the total net subsidy on the merchandise exported to the United States.

Comment 8: The Brazilian government contends that export taxes are specifically contemplated by U.S. law and that the Department should not reconsider whether these taxes are permissible offsets. Furthermore, as section 771(6)(C) of the Tariff Act clearly indicates, Congress has concluded that export taxes, levied with the specific intent to offset subsidies, do serve the "larger purpose of the countervailing duty law," a phrase the Department used in its preliminary results. The Department may not defy clear congressional intent.

Department's Position: We determine for purposes of this review that the IPI export tax meets the criteria set forth in section 771(6)(C) and is, therefore, a permissible offset. Nonetheless, we believe that we have the discretion to determine whether offset taxes serve the larger purpose of the countervailing duty law. Section 771(6)(C) states: "(f) or the purpose of determining the net subsidy, the administering authority may subtract * * * export taxes, duties, or other changes * * *" (emphasis added).

*3445

Therefore, we believe that Congress did not intend the export tax offset to be automatic.

Comment 9: The Brazilian government argues that loans issued pursuant to the Banco do Brasil's CIC-CREGE 14-11 circular do not constitute a government program and, therefore, cannot confer a subsidy on exports of cotton yarn. The Banco do Brasil receives no financial support from the Government of Brazil and operates the program in a manner consistent with commercial considerations. Even assuming, arguendo, that the program is countervailable, the Department has overstated the benefit by using an incorrect benchmark (see Comment 3).

Department's Position: We have considered and rejected the first argument in other Brazilian countervailing duty cases. See, e.g., Final Affirmative Countervailing Duty Determination; Brass Sheet and Strip From Brazil (51 FR 40837; November 10, 1986). The Brazilian government has provided neither new evidence nor new arguments necessary for us to reconsider this issue. Regarding the second argument, we revised our benchmark, as explained in our response to Comment 3, and adjusted our calculations accordingly. In addition, we discovered an error in our calculations that understated the benefit. After adjusting for both the error and the revised benchmark, we determine the benefit from this program for 1984 to be 2.29 percent ad valorem for Cotonificio Guilherme Giorgi, and 0.01 percent ad valorem for all other companies in 1984, except those companies with de minimis aggregate benefits; and 0.03 percent ad valorem in 1985 for all companies except Kanebo and Cotonificio Guilherme Giorgi, who had no benefits attributable to this program.

Comment 10: The Brazilian government argues that the Department has overstated the benefit from the FST financing program by using an incorrect benchmark (see Comment 3).

Department's Position: After making adjustments using our revised benchmark (see our response to Comment 3), we determine the benefit from this program for 1984 to be zero for Cotonificio Guilherme Giorgi, and 0.03 percent ad valorem for all other companies in 1984, except those companies with de minimis aggregate benefits; and 0.01 percent ad valorem for all other companies in 1985, except Kanebo and Cotonificio Guilherme Giorgi, who had no benefits attributable to this program.

Comment 11: The Brazilian government argues that the Department incorrectly determined that the Price Equalization Program (PEP) is countervailable. The PEP, which was implemented as an import substitution program, did not confer any benefit on cotton yarn producers. The PEP was established to eliminate a temporary domestic oversupply of raw cotton by encouraging the use of domestic cotton in exports. It was operated in a manner consistent with item (d) of the Illustrative List of Export Subsidies annexed to the Agreement on Interpretation and Applications of Articles VI, XVI, and XXIII of the General Agreement on Tariffs and Trade ("GATT"), which states:
The delivery by government or their agencies of imported or domestic products or services for use in the production of exported goods, on terms or conditions more favorable than for delivery of like or directly competitive products or services for use in the production of goods for domestic consumption, if (in the case of products) such terms or conditions are more favorable than those commercially available on world markets to their exporters (emphasis added).
The Department has determined in past countervailing duty cases that the intent of the exception to item (d) is to permit an import substitution program without giving rise to a subsidy. In the Final Negative Countervailing Duty Determination; Certain Stel Wire Nails from the Republic of Korea (47 FR 39549; September 8, 1982), the Department held that "price preferences for inputs to be used in the production of export goods constitute a subsidy only if the preference lowers the price of that input below that which the input purchaser would pay on world markets." Similarly, in the Final Affirmative Countervailing Duty Determination; Certain Steel Products From the Federal Republic of Germany (47 FR 39345; September 7, 1982), the Department states: "* * * the price charged for subsidized FRG coal certainly does not undercut the freely available market price * * * Therefore, non-FRG purchasers of subsidized FRG coal do not benefit from FRG coal subsidies."
In fact, in its preliminary results notice the Department agreed in principle that the PEP itself does not constitue a subsidy according to item (d):
If Brazilian exporters of cotton yarn had actually imported raw cotton in commercial quantities during the review period, we could have considered such imports as the commercially available alternative to domestic raw cotton (in which case the PEP might have been consistent with item (d)). However, because of import restrictions imposed by the Government of Brazil, cotton yarn exporters did not import raw cotton during the period of review.
Respondents argue that it is improper for the Department to assign such a restrictive interpretation to item (d). Even if a government instituted a program that banned imports of a good and offered exporters the same good at the world market price, such a program would not be a subsidy.
However, even with this restrictive interpretation, the Department's rationale for finding that the PEP confers a countervailable subsidy is based on the erroneous assumption that raw cotton was not imported. No restrictions on the importation of raw cotton existed during the review period. As shown by numerous import documents submitted with the comments on the preliminary results, cotton yarn producers imported raw cotton in amounts exceeding those which were distributed under the PEP. Raw cotton was impoted duty-free by cotton yarn producers under Brazil's suspension drawback system, which provides exemption from all duties if the raw cotton is physically incorporated into the final export product. The suspension drawback system allowed producers to import raw cotton at the world market price. Therefore, the commercial alternative to the PEP program was to import raw cotton at the world market price.
Conversely, AYSA argues that the Department correctly found the PEP countervailable. The PEP provided a specific quantity of raw cotton to yarn producers at a price that was significantly lower than both the internal market price and the world market price, and only exporters were able to purchase the raw cotton through the PEP. Furthermore, the Department was correct in using the internal price of raw cotton as its benchmark for purposes of calculating the benefit.

Department's Position: We disagree with respondent. The Brazilian government's decision to insulate its cotton producers from foreign competition placed domestic users of cotton at a disadvantage vis-a-vis competitors abroad by raising the price of domestic cotton. During the review period, Brazilian yarn exporters were able to overcome this competitive disadvantage in two ways: duty drawback and the PEP.
Imported cotton in Brazil is subject to normal cutoms duties. By using duty drawback, a practice acceptable under U.S. countervailing duty law and the GATT, Brazilian yarn exporters were exempted from normal customs duties on raw cotton provided that the cotton was processed into yarn and re-exported. Alternatively, under the PEP, the Brazilian government sold cotton

*3446

exclusively to yarn exporters at a price well below the price commercially available in the domestic market. The PEP was an instrument used by the brazilian government to ameliorate the deleterious effects of high- priced domestic cotton on a specific group of downstream users.
The circumstances in both FRG Steel Products (op.cit.) and Korean nails (op.cit.) differ from those in this case. In FRG Steel Products, there was no evidence of preferential pricing of imputs. Rather, the coal subsidies and coal import restrictions were part of a comprehensive program designed to assist the Federal Republic of Germany's coal industry, from which steel producers received no benefits. In fact, steel producers and other users of coal in the Federal Republic of Germany were required to pay a slight premium over world prices.
In Korean Nails, the Korean producers of nails for export had access to wire rod from foreign as well as domestic sources at comparable prices. Although afforded the opportunity through tariff protection to charge high prices for wire rod used in the manufacture of products sold domestically, POSCO (an integrated steel producer which is largely government owned) and other Korean producers of wire rod chose to lower their prices to exporters of nails and compete with foreign-sourced wire rod purchased under duty drawback. We concluded that "the different prices for purchasers do not arise from a scheme to subsidize exports, but rather are a commercial response to a segmented market, one segment being protected and the other fully open to foreign competition." We further stated that "this dual pricing system reflects strictly economic motivations [of the wire rod producers] rather than a desire of the Government of Korea (the owners of POSCO) to subsidize nail exports."
We noted in addition that our conclusion regarding the dual pricing system was consistent with the principle contained in item (d). However, our decision not to countervail the Korean pricing scheme was not made solely on the basis of item (d). Rather, our decision was based in large part on a determination that POSCO was acting in a commercially reasonable fashion by instituting a dual- pricing system. As support for this, we stated that two privately-owned Korean wire rod producers also had dual-pricing systems in place. These facts led us to conclude that the Korean government was not acting to subsidize exports.
In this case, the fact pattern is different. Private sellers of raw cotton in Brazil did not institute a dual-pricing scheme. Instead, the Brazilian government intervened to ensure that Brazilian yarn exporters could continue to use domestically-sourced cotton while cotton producers continued to enjoy the full benefits of tariff protection. Thus, the Brazilian government's decision to establish the PEP and sell raw cotton destined for export production at low prices made possible exports that otherwise would not have occurred. Without this direct government action, cotton yarn exporters would have had to pay the high domestic price for Brazilian raw cotton.
In determining whether item (d) is applicable to the identification and measurement of an export subsidy from this type of program, we have examined the law and its legislative history. Section 771(5) of the Tariff Act states, in relevant part: "the term 'subsidy' has the same meaning as the term 'bounty or grant' as that term is used in section 303, and includes, but is not limited to, the following: (i) any export subsidy described in Annex A to the Agreement (relating to the Illustrative List of Export Subsidies) * * * " (emphasis added). While Congress incorporated the Illustrative List in the statute, it did not limit the definition of export subsidy to the practices outlined in the List. The legislative history of the TAA explains, "The reference to specific subsidies in the definition is not all inclusive, but rather is illustrative of practices which are subsidies within the meaning of the word as used in the bill. The administering authority may expand upon the list of specified subsidies consistent with the basic definition." S. Rep. 96- 249, 96th Cong., 1st Sess. 85 (1979). See, also, Trade Agreements Act of 1979: Statements of Administrative Action, H.R. Doc. No. 96-153, Pt. II, 96th Cong., 1st sess. 432 (1979). The Illustrative List is not, therefore, controlling of the identification and measurement of export subsidies, but must be considered along with other provisions of the statute and its legislative history, administrative practice and judicial precedent.

We consider direct government provision of an input at a price lower to exporters than to producers of domestic products to confer a subsidy within the meaning of section 771(5) of the Tariff Act. It is irrelevant whether the PEP is consistent with item (d) or whether cotton yarn exporters could have imported raw cotton at world market prices. We are concerned with the alternative price commercially available in the domestic market.
An analogy to the PEP program is the case of export loans. In this case, as in many others, we have determined that export loans at preferential interest rates constitute a subsidy. In measuring the subsidy, we do not concern ourselves with whether firms could have borrowed money at commercial rates in international credit markets. The fact that, as a result of a government program, they borrowed from domestic sources at rates below those commercially available in the domestic market leads us to determine that a subsidy is bestowed.
To calculate the benefit from the PEP, we multiplied the amount of cotton purchased by each firm at the PEP price by the average internal market price between April and December 1985. The benefit is the difference between this amount and the amount paid for raw cotton under the PEP. We allocated the benefit over each firm's total exports for 1985. On this basis, we determine the benefit from this program to be 6.68 percent ad valorem for Cotonificio Guilherme Giorgi, 15.24 percent ad valorem for Kanebo and 7.79 percent for all other firms.
Since the PEP was eliminated in February 1986, we preliminarily determine that for purposes of cash deposits of estimated countervailing duties there is no current benefit from this program.

Comment 12: The Brazilian government argues that the Department incorrectly overstated the net subsidy by failing to include those companies with de minimis levels of benefits in calculating the weighted-average country-wide subsidy rate. Since countervailing duties are applied on all products of a country subject to a countervailing duty order, any country-wide average must, by definition, include all producers. While de minimis companies may still obtain a zero rate, the calculation of the weighted average must consider all producers. Furthermore, the Court of International Trade has directed the Department to include all companies under review in its calculations of average countervailing duty rates. See, Fabricas El Carmen, S.A., de C.V., et al. v. United States, Slip Op. 87-113 (CIT October 7, 1987).

Department's Position: The court vacated its initial decision in Fabricas El Carmen. Nevertheless, even the moot decision is irrelevant to this case. The court remanded the case because it was dissatisfied with the Department's having calculated a country-wide rate based on the responses of only 16 out of more than two thousand companies

*3447

subject to the countervailing duty order. See, Final Affirmative Countervailing Duty Determination: Certain Textile Mill Products from Mexico (50 10824; March 18, 1985). The court stated:
* * * ITA's approach of dividing those benefits by the exports of those same 16 firms is unreasonable in that it will not yield a national average rate for all 2,000 firms, but rather, it will only yield an average rate for those 16 firms.
Section 607 of the Tariff and Trade Act of 1984 establishes a statutory presumption in favor of country-wide countervailing duty rates, with the possibility of company-specific rates if the Department determines that a significant differential exists. The Department's policy is to use a single weighted-average country-wide rate unless there is a significant differential between an individual company rate and the weighted-average country-wide rate. According to the Department's regulations, 19 CFR 355.20(d)(3), we have interpreted a significant differential to be a difference of five percentage points or 25 percent from the weighted-average country-wide rate, whichever is greater. We also consider a de minimis rate for an individual company to be, by definition, "significantly different" from the weighted-average country-wide rate.
Contrary to the Brazilian government's assertion, we did include those companies with de minimis benefits in calculating the weighted-average country- wide rate. This country-wide rate served as the basis for comparison with individaul company rates to determine whether significant differentials existed. Having found significant differentials between the level of subsidies received by individual exporters of cotton yarn, we used a combination of company-specific rates and an "all other" rate for assessment and duty deposit purposes. An "all other" rate is different from a country-wide rate because, by definition, it cannot be based on all companies. We do not include the weights of companies with significantly different benefits in calculating the "all other" rate because to do so would misstate the benefit for the "all other" companies.
In any countervailing duty case involving companies that receive significantly different benefits, the designation "significantly different" and "country- wide" are mutually exclusive. Once any company receives a separate rate, the country-wide rate is no longer applicable when assessing countervailing duties. The duty assessed on the exports of the remaining "all other" companies must be different from the duty that would have been assessed had there been only a country-wide rate. This is so because the amount of the net subsidy on the merchandise exported to the United States remains constant regardless of whether we set a country-wide rate or a combination of company- specific rates and an "all other" rate. With a country-wide rate in this case, we would assess duties on the exports of a greater number of companies. With company-specific de minimis rates and an "all other" rate, we assess a greater rate of duty on the exports of fewer companies. In either case, the amount of countervailing duties collected by the Customs Service is the same.

Firms Receiving De Minimis Benefits

We determine that the following firms received de minimis benefits during the period May 18, 1984 to December 31, 1984:
(1) Brasital S.A. Para a Industria E.O. Comercio;
(2) Cia. Industrial e Agricola Boyes;
(3) Fiacao Amparo S.A;
(4) Lanficio Amparo Ltda; and
(5) Unitika do Brasil Industria Textil Ltda.

Final Results of Review

After reviewing all of the comments received, we determine the net subsidy to be de minimis for five firms, 12.15 percent ad valorem for Cotonificio Guilherme Giorgi, and 2.56 percent ad valorem for all other firms for the period May 18, 1984 through December 31, 1984. The Department determines the net subsidy to be 22.30 percent ad valorem for Kanebo, 7.75 percent ad valorem for Cotonificio Guilherme Giorgi, and 12.82 percent ad valorem for all other firms for the period January 1, 1985 through December 31, 1985.
The Department will instruct the Customs Service to liquidate, without regard to countervailing duties, shipments of Brazilian carded cotton yarn from the five firms with de minimis benefits in 1984, and to assess countervailing duties of 12.15 percent of the f.o.b. invoice price on shipments of this merchandise from Cotonificio Guilherme Giorgi, and 2.56 percent of the f.o.b. invoice price on shipments from all other firms entered, or withdrawn from warehouse, for consumption on or after May 18, 1984 and exported on or before December 31, 1984. The Department will also instruct the Customs Service to assess countervailing duties of 22.30 percent and 7.75 percent of the f.o.b. invoice price on shipments from Kanebo and cotonificio Guilherme Giorgi, respectively, and 12.82 percent of the f.o.b. invoice price on shipments from all other firms exported on or after January 1, 1985 and on or before December 31, 1985.
The Department will instruct the Customs Service to waive cash deposits of estimated countervailing duties on shipments of this merchandise from three companies: Fiacao Amparo, Novo Odessa and Cotonificio Guilherme Giorgi; and to collect a cash deposit of estimated countervailing duties of 2.38 percent of the f.o.b. invoice price on shipments of this merchandise from all other firms entered, or withdrawn from warehouse, for consumption on or after the date of publication of this notice. This deposit requirement shall remain in effect until publication of the final results of the next administrative review.
This administrative review and notice are in accordance with section 751(a)(1) of the Tariff Act (19 U.S.C. 1675(a)(1)) and 19 CFR 355.22
Dated: January 26, 1990.

Eric I. Garfinkel,

Assistant Secretary for Import Administration.