(Cite as: 59 FR 12243)

NOTICES

DEPARTMENT OF COMMERCE

(C-122-404)

Live Swine From Canada; Final Results of Countervailing Duty Administrative Review

Wednesday, March 16, 1994

*12243 AGENCY: International Trade Administration/Import Administration Department of Commerce.

ACTION: Notice of Final Results of Countervailing Duty Administrative Review.

SUMMARY: On October 20, 1993, the Department of Commerce (the Department) published the preliminary results of its administrative review of the countervailing duty order on live swine from Canada (58 FR 54,112). We have now completed that review and determine the total subsidy to be Can$0.0295 per kilogram for all live swine.

EFFECTIVE DATE: March 16, 1994.

FOR FURTHER INFORMATION CONTACT: Dana Mermelstein or Stephanie Moore, Office of Countervailing Compliance, International Trade Administration, U.S. Department of Commerce, Washington, DC 20230; telephone: (202) 482-2786.

SUPPLEMENTARY INFORMATION:

Background

On October 20, 1993, the Department of Commerce (the Department) published in the Federal Register (58 FR 54,112) the preliminary results of its administrative review of the countervailing duty order on live swine from Canada (50 FR 32,880; August 15, 1985). The Department has now completed that administrative review in accordance with section 751 of the Tariff Act of 1930, as amended (the Act). Case briefs were submitted by the National Pork Producers' Council, Petitioner, the Government of Canada (GOC), the Gouvernement du Que1bec (GOQ), the Canadian Pork Council (CPC), Pryme Pork, Ltd. (Pryme), P. Quintaine & Son (Quintaine), and Earle Baxter Trucking LQ (Baxter). Rebuttal Briefs were submitted by Petitioner, the GOC, the GOQ, and the CPC. On December 1, 1993 the Department held a public hearing at the request of Petitioner and the GOQ. In response to the comments made by the parties, the Department has recalculated benefits under the Alberta Crow Benefit Offset Program, the Feed Freight Assistance Program, and the Saskatchewan Hog Assured Returns Program. The total subsidy determined in the preliminary results of review, Can $0.0289/kg, has been recalculated. The Department now determines the total subsidy to be Can$0.0295/kilogram.

Scope of Review

The merchandise covered by this review is all live swine, except breeding swine, from Canada. Such merchandise is classifiable under the Harmonized Tariff Schedule (HTS) item numbers 0103.91.00 and 0103.92.00. The HTS item numbers are provided for convenience and Customs purposes. The written description remains dispositive. The review covers the period April 1, 1990 through March 31, 1991 and the following programs: (1) Feed Freight Assistance Program; (2) National Tripartite Stabilization Scheme for Hogs (Tripartite); (3) Que1bec Farm Income Stabilization Insurance Program (FISI); (4) Saskatchewan Hog Assured Returns Program (SHARP); (5) Alberta Crow Benefit Offset Program (ACBOP); (6) Alberta Livestock and Beeyard Compensation Program (Livestock Predator Sub-Program); (7) Ontario Farm Tax Rebate Program; (8) Livestock Improvement Program for Northern Ontario; (9) Ontario Pork Industry Improvement Plan (OPIIP); (10) Ontario Rabies Indemnification Program; (11) Saskatchewan Livestock Investment Tax Credit; (12) Saskatchewan Livestock Facilities Tax Credit Program; (13) Canada/British Columbia Agri-Food Regional Development Subsidiary Agreement; (14) Canada/Que1bec Subsidiary Agreement of Agri-food Development; (15) Canada/Manitoba Agri-Food Development Agreement; (16) Western Diversification Program; (17) Agricultural Products Board Program; (18) Canada/Alberta Swine Improvement Programs Study; (19) Canada/Ontario Canadian Western Agribition Livestock Transportation Assistance Program; (20) British Columbia Swine Herd Improvement Program; (21) Ontario Export Sales Aid; (22) Ontario Bear Damage to Livestock Program; (23) Ontario Dog Licensing and Livestock and Poultry Compensation Program; (24) New Brunswick Agriculture Development Act--Swine Assistance Program; (25) New Brunswick Swine Industry Financial Restructuring Program; (26) British Columbia Farm Income Insurance Program; (27) New Brunswick Livestock Incentives Program; (28) New Brunswick Hog Marketing Program; (29) New Brunswick Hog Price Stabilization Program; (30) New Brunswick Swine Assistance Policy on Boars; (31) Prince Edward Island Hog Price Stabilization Program; (32) Prince Edward Island Swine Development Program; (33) Prince Edward Island Interest Payment on Assembly Yard Program; (34) Nova Scotia Swine Herd Health Policy; (35) Nova Scotia Improved Sire Policy; (36) Newfoundland Farm Products Corporation Hog Price Support Program; (37) Newfoundland Weanling Bonus Incentive Policy; (38) Canada-Saskatchewan Agri-Food Development Agreement; (39) British Columbia Feed *12244 Grain Market Development Program; (40) Ontario Soil Conservation and Environmental Assistance Program; (41) Ontario Weaner Pig Stabilization Plan; (42) Nova Scotia Natural Products Act--Pork Price Stabilization Program; and, (43) Que1bec Productivity and Consolidation of Livestock Production Program. Of the above-listed programs, we found subsidies were provided to live swine producers during the review period under 12 programs. See Final Results of Review section below.

Analysis of Comments

Comment 1: Petitioner urges the Department to reexamine its practice of not finding a program de jure specific based on its availability only to the agricultural sector. Petitioner argues that the Department's practice with respect to agricultural subsidies is inconsistent with its treatment of subsidies bestowed upon other sectors of the economy, and is reminiscent of the discarded "general availability test," presuming that a program available to all of agriculture is somehow "generally available." Petitioner cites both Federal Circuit and Court of International Trade opinions in advancing the argument that the Department has considerable discretion in determining when a subsidy program is de jure specific and what practices are countervailable. Petitioner argues that the Department should exercise its discretion, and focus on factors such as the size of the agricultural sector relative to the economy as a whole, and therefore conclude that Tripartite is de jure specific because it is limited, by law, to an enterprise or industry or group of enterprises or industries. Both the GOC and the CPC argue that it would be inappropriate for the Department to now reverse a longstanding practice with regard to agricultural programs, and to do so would be tantamount to rulemaking. The GOC states that "(b)y any standard, the agricultural sector is too broad to constitute a 'specific * * * group of enterprises or industries' as required by the statute." The CPC states that what Petitioner refers to as "misguided policy" has been upheld by the Court of International Trade as a reasonable exercise of the Department's discretion. See Roses Inc. v. United States, 774 F. Supp. 1376, 1383 (CIT 1991).

Department's Position: The Department's policy with respect to agricultural programs has been incorporated into the proposed regulations, which provide that the Department "will not regard a program as being (de jure) specific * * * solely because the program is limited to the agricultural sector." Notice of Proposed Rulemaking and Request for Public Comments (54 FR 23,366, 23,380; May 31, 1989) (Proposed Regulations), at section 355.43(b)(8). See, e.g., Fuel Ethanol from Brazil (51 FR 3361; 1986). Although these proposed regulations are not final, we have determined that, for the present, it is appropriate to maintain the current policy with respect to subsidies provided to the industries within the agricultural sector. We recognize, however, that certain policies such as this one may warrant reconsideration in the future, and we agree with Petitioner that the Department's discretion permits it the authority to reverse such policies by way of the proper procedure, depending upon the policy in question. We note, in addition, that, as with other subsidy programs, in publishing the proposed regulation relating to the current agricultural sector exception, the Department emphasized in its commentary that "an agricultural program may be deemed specific if, for example, benefits under the program are limited to, or provided disproportionately to, producers of particular agricultural products." 54 FR at 23,368 (emphasis added). The use of the disjunctive "or" demonstrates the Department's recognition that an affirmative finding based upon a single factor could reasonably support a determination of de facto specificity within the meaning of section 771(5) of the Act.

Comment 2: Petitioner argues that the Department should conduct its de jure specificity analysis of the Tripartite program by focusing on the individual Tripartite plans and their implementing subsidiary agreements, rather than on the implementing legislation, Canada's Agricultural Stabilization Act (ASA), as amended by Bill C-25 to provide for Tripartite agreements. Petitioner argues that this approach is appropriate because the Tripartite Agreement for Hogs is not integrally linked with any of the other Tripartite agreements. Contrary to the Department's determination in the fourth review of this order, Petitioner argues that all Tripartite schemes were not part of one program because they are "structured pursuant to the enabling legislation and basic principles in Bill C-25 * * *" Final Results of Countervailing Duty Administrative Review; Live Swine from Canada (56 FR 28,531; June 21, 1991) (Fourth Review Final). According to Petitioner, the basis of the Department's determination appears to have been its consideration of only one factor, the purpose of the program as stated in the enabling legislation. Petitioner argues, however, that there is no evidence of a government policy to treat industries equally under the agreements because each individual agreement specifies the manner in which benefits are calculated and paid, thereby describing the class of eligible producers. Petitioner cites Certain Fresh Atlantic Groundfish from Canada; Final Affirmative Countervailing Duty Determination (51 FR 10,041, 10,049; March 24, 1986) (Groundfish), aff'd, Comeau Seafoods v. United States, 13 CIT 923, 724 F. Supp. 1407, 1416 (1989), in which the Court of International Trade (CIT) affirmed the Department's determination to examine the specificity of the Canadian Economic and Regional Development Agreements by focusing on the terms of the individual ERDA subsidiary agreements. Petitioner also argues that even if the Department examines the Tripartite schemes collectively, they are de jure limited to a specific group of industries, namely the eleven commodities covered by Tripartite agreements during this review period. Respondents counter that it is appropriate for the Department to employ an integral linkage analysis when the Department is determining whether to examine two or more programs as one. Applying the integral linkage policy here shows that the Tripartite agreements meet all of the integral linkage criteria and should therefore be considered as one program, consistent with the Department's practice in reviewing this program. According to the GOC and the CPC, the analogy which Petitioner draws between Tripartite and the regional development agreements in Groundfish provides no support for the approach endorsed by Petitioner. The GOC and the CPC also object to Petitioner's arguments on the basis that the Department has already determined Tripartite to be de jure not specific in earlier reviews of this order, and Petitioner has presented no new facts or evidence of changed circumstances which would justify reconsideration of this determination. Therefore, the Department should not revisit the question of de jure specificity.

Department's Position: For purposes of the Department's de jure specificity analysis, we have continued to treat Tripartite as a single subsidy program providing benefits to several identifiable beneficiaries through individual agreements reached between the federal government, the provincial governments and the various agricultural commodity producers. Petitioner's reliance on Groundfish and Comeau Seafoods is misplaced. In *12245 upholding the Department's determination in Comeau Seafoods, the CIT correctly identified the determinative issue as being "at what level Commerce may apply the specificity test." Comeau Seafoods, 724 F. Supp. at 1416 (emphasis in original). As the CIT found, the individual Economic and Regional Development Agreements (ERDAs) at issue there were "designed to 'establish programs, delineate administrative procedures and set up the relative funding commitments of the federal and provincial governments."' Id. at 1415 (quoting Groundfish From Canada, 51 FR at 10,049). In addition, the ERDAs were designed to provide only a procedure for "the establishment of economic development programs with stated general economic development goals." Id. at 1415 n. 13. For these reasons, the "agreements" in Groundfish were effectively separate subsidy programs, making the proper level of specificity analysis the agreements themselves. By contrast, as the Department found in the fourth administrative review, Tripartite's enabling legislation, Canada's ASA, as amended by Bill C-25, provides for established administrative procedures and funding commitments. Fourth Review Final, 56 FR at 28,532. Moreover, Tripartite's enabling legislation creates a framework for providing only one type of assistance, income stabilization to producers of agricultural commodities which establish agreements. See id. Therefore, although the record is not clear as to whether the Government of Canada retains discretion regarding when to enter into particular agreements, it is clear that Tripartite is a single program, of which the Tripartite agreements, or product-specific schemes, are "integral parts." Accordingly, the appropriate level for the Department's specificity analysis is not the individual agreements but the Tripartite program itself. In reaching this determination, we note that, contrary to the arguments of Respondents, the Department did not conduct an integral linkage analysis of Tripartite in the fourth administrative review or at any other time. See id. Finally, as we found in the preliminary results, Petitioner has not presented any new facts or evidence of changed circumstances during the present review which would warrant reconsideration of the issue of whether the Tripartite is de jure specific. Preliminary Results at 54,116. Therefore, we have declined to reconsider the Department's determination that Tripartite is not de jure specific.

Comment 3: The GOC disagrees with the Department's preliminary determination that Tripartite is not integrally linked to the other provisions of the Agricultural Stabilization Act (ASA), in accordance with the Department's proposed regulations, and that the programs examined together are not de facto specific. Furthermore, the GOC considers unreasonable the Department's reliance on non-regulatory factors such as "a documentary statement of an overall government policy to treat industries equally" and the expectation of identical treatment and benefits among the different programs at the operational level. In relying on these factors, the Department is introducing a more stringent standard than is required by the proposed regulations and is, therefore, acting contrary to law. More specifically, the GOC argues that the relationship between Tripartite and the named and designated commodity provisions of the ASA satisfies all of the Department's regulatory factors for finding integral linkage. According to the GOC, there is one statute which provides the same benefits, for the same purpose, under a centralized administration "to the producers of all agricultural commodities in Canada." The GOC further states that the Department errs by equating a policy to treat industries equally with a requirement that benefits, purposes, and administration be identical; a policy to treat industries equally is evident under the ASA, the GOC argues, because it provides every Canadian agricultural producer access to stabilization payments, when needed, in the amounts required, without regard to regional differences in a complementary fashion. The GOC further argues that, because identicality should not be expected or required, the Department's conclusion is unwarranted that the existence of Tripartite Stabilization Committees indicates that the programs are not administered in common. Equally unwarranted is the Department's distinction between Tripartite (which requires producer contributions) on the one hand, and named and designated commodities on the other (which require no producer contributions). According to the GOC, the Tripartite producer contribution requirement does not disadvantage producers because producers enter Tripartite agreements only if the benefits, such as flexibility in negotiating a payment schedule, outweigh the drawbacks. Petitioner agrees with the Department's finding that Tripartite is not integrally linked to any other support program. Citing Carbon Steel Wire Rod from Saudi Arabia; Final Results of Countervailing Duty Administrative Reviews (57 FR 8303; March 9, 1992), Petitioner argues that the GOC has failed to demonstrate that the factors considered by the Department are outside the Department's scope of authority under its proposed regulations, or otherwise not in keeping with earlier determinations. Petitioner argues that, unless the Department interprets the integral linkage standard in a strict fashion, despite the GOC's claim that the Department's interpretation is "more stringent" than that which is required by the proposed regulations, any government would be able to immunize its support programs against findings of specificity by merely articulating a very broad purpose which encompasses all programs. Petitioner also argues that the analysis of equal treatment applied by the Department is neither extraregulatory nor unreasonable. Contrary to the GOC's allegations, neither the Department's analysis in this case, nor the linkage test in general, requires identical treatment, but rather equality in receipt of benefits.

Department's Position: We disagree with Respondents and affirm our preliminary determination that the "named" and "designated" provisions of the ASA are not integrally linked to the Tripartite provision of the ASA. Contrary to the contention of Respondents, the Department's interpretation of the integral linkage policy, and the Department's integral linkage analysis in this case, are not more stringent than permitted by the Department's authority. The integral linkage policy is only an exception to the normal application of the specificity test. As the drafting of the integral linkage provision in the proposed regulations indicates, the policy was created to permit evaluating whether, in particular circumstances, the Department should deviate from its normal approach to analyzing de facto specificity in order to consider the coverage of two or more programs together instead of just one. See Proposed Regulations at s355.43(b)(6). Considering the purpose of the specificity test as a whole, we have interpreted the standard narrowly for granting an affirmative integral linkage determination. The specificity test was designed to avoid carrying the countervailing duty law to absurd results by countervailing government actions or programs such as public highways and bridges which clearly benefit the economy at large, as opposed to identifiable and specific segments of the economy. See, e.g., *12246 Carlisle Tire & Rubber Co. v. United States, 564 F. Supp. 834, 838 (CIT 1983). In implementing the appropriate standard to determine whether to permit a particular exception to the specificity test, however, such as an affirmative integral linkage finding, the Department cannot create a loophole which would allow de facto specific subsidy programs benefiting only particular segments of the economy--or particular segments of the agricultural sector--to escape the imposition of countervailing duties. Permitting respondent governments to loosely connect two or more programs which are otherwise designed to serve different purposes would create just the type of loophole the Department seeks to avoid. Besides being contrary to the Department's specificity practice, doing so would also be contrary to Congress' express requirement in the legislative history that Commerce avoid taking an "overly narrow" or "overly restrictive" view of its authority to determine specificity. S. Rep. No. 71, 100th Cong., 1st Sess. 123 (June 12, 1987). This statement, implies that Congress intended the Department to view its authority to find specificity broadly and its authority to create exceptions to its normal approach narrowly. The very fact that the programs at issue must be found to be "integrally linked" rather than merely "linked" demonstrates the limited circumstances which would warrant an affirmative finding. The evidentiary standard for establishing that two or more programs are integrally linked is two-fold. First, as we explained in the preliminary results, the government must point to an express statement in the statute or elsewhere, either at the time the first program was created or later when the additional programs were added, which reasonably documents the government's underlying intent to develop two or more programs designed as "complementary parts of an overarching governmental policy directive." Integral Linkage Analysis Memorandum, October 13, 1993 (on file in Room B-099, Department of Commerce) (quoting Carbon Steel Wire Rod From Saudi Arabia, 57 FR at 8303) (Integral Linkage Memo). The need to provide this type of objective legal evidence relates to all of the integral linkage factors set forth in the proposed regulations. The government must also provide factual evidence documenting that its original intent has been implemented, and that the programs are actually functioning in a complementary manner. This type of evidence also relates to each of the proposed factors and other relevant evidence. Contrary to the claim of the GOC, the Department does not require that the programs be "identical" in order to prevail on a claim of integral linkage. As petitioner correctly notes, however, the supporting evidence must go beyond simply identifying a broad underlying purpose encompassing several otherwise distinct programs which provide access to benefits to all or most eligible industries. For instance, in this case, the Department's linkage standard requires more than the GOC's broad statement that Tripartite and the other ASA provisions are each designed to provide income stabilization to all agricultural industries. See Integral Linkage Memo at 4. As stated above, the respondent government must demonstrate through objective record evidence that, due to an "overall policy or national development plan," it created two or more programs with the express purpose that they complement one another, not only in terms of breadth of availability and coverage, but in similarity of intent, purpose, and administration as well. Preliminary Results at 54,115 (quoting Carbon Steel Wire Rod from Saudi Arabia). Furthermore, the evidence must establish that any differences between the nature and administration of the programs are necessary because of differences in the nature of the industries being offered benefits; and despite these differences, the recipient industries are actually treated equally in terms of availability, type, and receipt of benefits. As the Department indicated in the preliminary results, the GOC was unable to point to the necessary documentation demonstrating the existence of an overall policy or development plan to create two or more complementary programs. That fact alone renders a claim of integral linkage insupportable. See id.; Integral Linkage Memo at 3-4. The Department also found that the information in the record does not establish that the named, designated, and Tripartite provisions of the ASA are administered in an equal or complementary manner. Id. In light of these basic, essential requirements, the Department's interpretation of the integral linkage policy in the preliminary results, is fully consistent with the Department's practice, proposed regulations and the legislative guidance regarding the appropriate approach to specificity analysis in general. See, e.g., Groundfish.

Comment 4: The GOC and the CPC disagree with the Department's determination that Tripartite is de facto specific. They argue that the Department's reliance on its finding that there are "too few users" of Tripartite is legally insufficient. According to Respondents, the statute and proposed regulations require consideration of all four factors enumerated in the proposed regulations at section 355.43(b)(2) before the Department can determine whether benefits under this program are provided to a specific enterprise or industry or group of enterprises or industries. The GOC argues that in reaching its preliminary results, the Department misinterpreted and misapplied Final Results of Review: Carbon Black from Mexico, 51 FR 30,385 (1986) and Cabot Corp. v. United States, 620 F. Supp. 722 (CIT 1985) (Cabot). According to the GOC, Cabot does not stand for the proposition that the Department may halt its specificity analysis upon finding "too few users" without consideration of the other regulatory factors and relevant evidence. As support, Respondents argue that a single-factor specificity test has been consistently rejected by the CIT, the Court of Appeals for the Federal Circuit, and several United States Canada Free Trade Agreement (FTA) binational panels. See Live Swine from Canada, USA-91- 1904-03, at 25 (October 30, 1992) (Second Swine IV Panel Decision); In the Matter of Softwood Lumber from Canada, USA-92-1904-01 (May 6, 1993); see also Roses, Inc. v. United States 774 F. Supp. 1376 (CIT 1991) (Roses II); and Roses, Inc. v. United States, 743 F. Supp. 870 (CIT 1990) (Roses I). Respondents point out that although the binational panel reviewing the fifth administrative review of live swine from Canada upheld the Department's specificity finding with regard to Tripartite, it did not uphold the use of a single-factor specificity test. In fact, the panel rejected the Department's finding that Que1bec's Farm Income Stabilization Insurance scheme is specific based upon only one factor. Petitioner argues that the sequential application of the specificity test is not inconsistent with U.S. law and has been held repeatedly to be a reasonable interpretation of the statute. Furthermore, according to Petitioner, Cabot supports a de facto specificity finding based solely on the existence of too few users, with no inquiry into policy or discretion. On the other hand, the binational panel decisions on which the GOC relies have no precedential value, and are only to be considered if they are "intrinsically persuasive." Accordingly, they do not supersede the binding case law which uniformly supports the Department's sequential application of the specificity test. *12247 Petitioner also notes that binational panel decisions on this issue directly contradict one another. Compare Second Swine IV Panel Decision; In the Matter of Live Swine from Canada, USA-91-1904-04 (August 26, 1992) (Swine V Panel Decision); and In the Matter of Pure and Alloy Magnesium from Canada, USA-92-1904-03 (August 16, 1993) (Magnesium). Petitioner also disagrees with the claim of the GOC and the CPC that the Department did not consider all factors in its analysis.

Department's Position: The test for determining de facto specificity requires that the Department "consider, among other things," several particular factors. Proposed Regulations at section 355.43(b)(2). Respondents misinterpret the purpose of the Department's inquiry, as set forth in the proposed regulations, when they incorrectly argue that the Department's practice "plainly calls for a finding on all four factors." As the Department has stated previously, and as the Court of Appeals for the Federal Circuit has agreed, we "must consider all of these factors in light of the evidence on the record in determining specificity in a given case." PPG Indus. v. United States, 928 F.2d 1568, 1577 (Fed. Cir. 1991) (PPG I). Moreover, while decisions of binational panels may be considered intrinsically persuasive, they are not binding on the Department. We have carefully reviewed the panel decisions cited by the GOC and do not consider them intrinsically persuasive for the reasons set forth below. See also the Department's response to Comment 12, below, regarding the specificity of Que1bec's FISI program. The GOC's reliance on the CIT's two Roses decisions is misplaced as well. In Roses, the CIT did not reject an affirmative de facto specificity determination based upon evidence relating to only one factor. Instead, the CIT rejected a finding of non-specificity which was reached without considering evidence relating to all four factors. It was in this context, after examining the Department's determination that a program was not specific based on the large number of users, that the Court properly held that the Department "does not perform a proper de facto specificity analysis if it merely looks at the number of companies that receive benefits under a program; the discretionary aspects of the program must be considered from the outset." Roses II, 774 F. Supp. at 1380. Although the CIT did not rule on the question of whether the Department could properly base an affirmative specificity determination on evidence related to only one factor, the context of the two decisions supports the Department's interpretation. See id.; see also Magnesium at 35 (cited in the Preliminary Results at 54,116). In this review, the Department determined that Tripartite provided de facto specific benefits to swine producers based upon its examination of evidence related to the first factor, the number of actual users or beneficiaries. We considered the evidence in the record regarding dominant users and disproportionate use, and the exercise of government discretion. We determined that this evidence did not detract from an affirmative de facto specificity determination on the basis of too few users. Preliminary Results at 54,116-17. Accordingly, the Department's determination is based upon substantial evidence and is otherwise in accordance with law.

Comment 5: The GOC contests the Department's failure to specifically identify, and reach a finding regarding, "a discrete, selective, targeted" class, industry or group of industries benefitting from Tripartite. The GOC cites PPG I, 928 F.2d at 1577 and PPG Indus., Inc. v. U.S., 978 F.2d 1232, 1240 (Fed. Cir. 1992) (PPG II), in support of its claim that the Department must identify a beneficiary class or industry which includes live swine producers before concluding that Tripartite is specific. Petitioner argues that neither the statute nor the regulations require governmental targeting or intent as a precondition for determining de facto specificity; the fact that the Department declined to make this finding is reasonable and in accordance with law.

Department's Position: We disagree with the GOC's contention that absent a finding that a bestowing government intended to benefit a "discrete, selective or targeted class," we may not properly find a program de facto specific regardless of how few users there are or other relevant evidence. The statute does not require and the Department's policy has not established that the Department must ascertain, or base its specificity determinations upon, the intent of the bestowing government. See 19 U.S.C. s1677(5)(B); Proposed Regulations at section 355.43(b)(2). The Department's interpretation of the statute has been expressly upheld by the CIT. Saudi Iron and Steel Co. (Hadeed) v. United States, 675 F. Supp. 1362, 1367 (1987), appeal after remand, 686 F. Supp. 914 (CIT 1988); see also Cabot, 620 F. Supp. at 732. Moreover, a binational panel in an earlier review of this order cited the legislative history underlying section 771(5)(B) of the Act to reject the GOC's same basic argument: "Under the statutory scheme, the pertinent inquiry is not whether Canada has intentionally targeted benefits to swine producers, but rather whether it has done something, intentionally or otherwise, that confers a benefit upon a 'specific enterprise or industry or group of enterprises or industries."' In the Matter of Live Swine From Canada, USA-91-1904-03, at 19- 20 (May 19, 1992) (First Swine IV Panel Decision). Similarly, the Court of Appeals for the Federal Circuit did not hold, in either PPG I or PPG II, that the Department must find intent. The court recognized that the statute provides a two-part test for specificity and that the de facto aspect is purely an inquiry into the factual question of whether, "'in its application, the program results in a subsidy only to an enterprise or industry or specific group of enterprise or industries."' PPG II, 978 F.2d at 1239 (quoting PPG I, 928 F.2d at 1576) (emphasis in original). While the court certainly did not attempt to foreclose the possibility that intent might be shown, see PPG II at 1240 n. 12, nowhere did the court indicate that the statute requires an express finding of intent in order to support an affirmative de facto specificity determination. In both decisions, the court merely used the phrase "discrete, selective, or targeted industry" to describe the industry, enterprise or group thereof that, as a factual matter, was eligible for (or should have been eligible for) or had actually received a benefit under the programs at issue. PPG II at 1240; PPG I at 1577. In this regard, we note the decision of yet another binational panel which rejected the GOC's argument by finding that the authorities cited by the GOC "generally use the term 'targeting' as a synonym for 'specific' or 'exercise of discretion."' Swine V Panel Decision at 16 n. 17. Similarly, we have interpreted the Court of Appeals' use of the same term in PPG as a synonym for "specific" or the "exercise of discretion." Therefore, no further findings are required by law to determine specificity in this review.

Comment 6: Petitioner argues that the Department's determination that there are "over 80 agricultural commodities" produced in both Canada and Que1bec understates the actual number of agricultural commodities which are eligible for benefits under the Tripartite and FISI programs, respectively. Petitioner states that the 1991 Agricultural Profile of Canada, provided to the Department by the GOC, represents the best quantification of agricultural commodities produced in Canada. It lists 131 commodities and *12248 supports the Department's determination in previous reviews that there are over 100 agricultural commodities produced in Canada. Petitioner further argues that the Department found in its memorandum on The Universe of Agriculture in Canada and Que1bec, Memorandum from Dana Mermelstein to Barbara Tillman, dated October 12, 1993 (Agricultural Universe Memo), that "the GOC has provided no indication of the criteria it applies to determine how and when a product should be listed in the (Farm Cash Receipts)," and "it is not possible to determine * * * how the GOC would reasonably and objectively determine which of the 131 commodities listed in the Profile meet these criteria." According to Petitioner, this uncertainty is a result of the failure by the GOC to provide information regarding Tripartite eligibility criteria. Therefore, Petitioner argues, the Department should draw an adverse inference and base its determination of the extent of the agricultural universe for purposes of the de facto specificity analysis on the 1991 Profile. Petitioner makes the same argument with regard to Que1bec's FISI program, alleging that the Government of Que1bec's failure to provide information about FISI eligibility requires the Department to rely on the Profile, and to make adverse inferences in determining the number of agricultural commodities produced in Que1bec. The GOC counters that Petitioner's criticisms of the Department's reasoning are invalid especially in light of the Petitioner's failure to provide substitute criteria for determining which products to include in the universe, a substitute list of products, or a definite final tally. The GOC and the CPC argue that the shortcomings in the explanation of how the Profile and the FCRs are compiled do not relate to Tripartite eligibility, nor would the law allow the Department to make the adverse assumptions Petitioner urges. The GOQ responds with three points: first, there is ample record evidence explaining and illustrating the "reasonable limitations" on FISI eligibility; second, adverse inferences are unwarranted in light of Que1bec's responsiveness to the Department's inquiries; and third, the Profile lists products at a level of aggregation which is not appropriate for defining the universe of products eligible for FISI.

Department's Position: We agree with the GOC, GOQ and CPC that Respondents' failure to provide information regarding the eligibility requirements for Tripartite and FISI is not a basis for the Department to draw an adverse inference with regard to the number of agricultural commodities produced in Canada and Que1bec, which are eligible for coverage. As the Department stated in the preliminary results, the goal of determining the number of commodities produced in Canada and Que1bec is to approximate the extent of the relevant agricultural universes and thus evaluate the coverage of the programs under consideration for the purpose of performing the de facto specificity analysis. We fully explained in the Agricultural Universe Memo how we evaluated the various sources of information in reaching the determination that there are over 80 agricultural commodities produced in both Canada and Que1bec.

Comment 7: The GOC argues that substantial record evidence does not support a finding that there are "too few users" of Tripartite; therefore, Tripartite is de facto non-specific. According to the GOC, benefits under Tripartite were provided during the review period to a "sizeable portion of the agricultural universe." With regard to the number of Tripartite users, the GOC argues that the Department's counting of the products shows that at least 9 industries or groups thereof, or 11 percent of the universe by number of products is covered by Tripartite. The GOC avers that a program need not reach all eligible users to be found not specific. The GOC points out that the binational panel ruling on the final results in the fourth review refused to sustain the Department's finding that Tripartite was specific based upon "too few users." In this review period, the Department's determination of specificity on the same grounds is all the more inappropriate because there are two more Tripartite agreements covering two additional commodities. Because Tripartite reaches more than a "trivial" number of users but less than the entire agricultural universe, the GOC claims that the Department's inquiry should extend into non-statistical factors, such as the availability of other stabilization options, and the length and complexity of the Tripartite negotiating process, to understand the reason for the limited number of Tripartite agreements. The GOC also reiterates its argument that Tripartite is an expanding program; products were added through the fifth review period, and enrollees were added in the current (sixth) review period. In addition, the CPC argues that in analyzing whether Tripartite is de facto specific, the Department must also consider the fact that commodities participating in Tripartite accounted for 33 percent of the total value of Canadian agricultural production during the review period. The Department asked for this information and, according to the CPC, cannot now simply ignore it. Petitioner rebuts that Respondents are attempting to inject into the specificity analysis several criteria that do not exist. Petitioner claims that the Department has consistently used statistical analyses in determining whether a program is de facto specific by virtue of the number of program users; in fact, the regulations require the Department to consider the number of users. Moreover, Petitioner, citing the Department's redetermination in the fifth review of this order, notes that the Department correctly does not consider that a program covering a variety of industries is necessarily de facto not specific. Petitioner further agrees with the Department's redetermination regarding the number of industries currently using Tripartite: it does not represent a variety of different types of agricultural commodities.

Department's Position: We disagree with the GOC. As we explained in the preliminary results, the Department determined that there were 11 beneficiaries of Tripartite during the review period (which the GOC now disaggregates into 13 beneficiaries), covered by eight agreements. Preliminary Results at 54,116. Tripartite's enabling legislation, Bill C-25, an amendment to the ASA, states that Tripartite benefits are available to "all natural or processed products of agriculture," thus requiring a determination that the program is not de jure specific under the Department's current policy toward agricultural subsidy programs. For purposes of its de facto specificity analysis, the Department has determined the appropriate universe of potential users in Canada against which to evaluate the number of actual users of Tripartite. That universe was comprised of over 80 agricultural commodities during the period of review. See Agricultural Universe Memorandum. Based on the Department's comparison of this evidence, we have reasonably determined that only 11 (or 13) out of over 80 is a sufficiently small number of actual beneficiaries so as to warrant a determination that Tripartite benefits a "specific enterprise or industry or group thereof" within the meaning of section 771(5)(B) of the Act. The Department disagrees with the GOC's claim that comparing the number of users to the number of potential users of a subsidy program is not probative of *12249 de facto specificity. This analysis is more than mere counting, as asserted by the GOC. The proposed regulations correctly provide that the Department will examine the number of enterprises or industries actually benefitting from a program in determining de facto specificity. See Proposed Regulations at section 355.43(b)(2). That is what the Department did here. In addition, the GOC itself acknowledges that, based upon the number of agricultural commodities, only 11 percent of the agricultural universe in Canada is covered by Tripartite. Such a finding would certainly not detract from a determination that Tripartite is de facto specific based upon the small number of users. In this same regard, we have considered the CPC's argument that the agricultural commodities participating in Tripartite accounted for 33 percent of the total value of Canadian agricultural production during the review period based on FCRs. This evidence also does not detract from a determination that Tripartite is de facto specific based upon the small number of only 11 (or 13) actual users. The statute states that a domestic subsidy is countervailable if it is limited to a specific enterprise or industry or group thereof, and the Department's proposed regulations provide that the Department will examine the number of actual beneficiaries, whether industries or enterprises, in determining de facto specificity. The Department has previously not engaged in an analysis of the percentage of production value covered by a program in making specificity determinations. However, because the CPC has raised this issue and because the proposed regulations provide that other factors may be considered, we have now considered this information in our specificity analysis. As discussed below, the Department determines that in the context of Tripartite this information has little, if any significance, in light of the relatively small number of actual beneficiaries compared to the relatively large number of eligible beneficiaries. The Department found that several of the relatively few commodities benefiting from Tripartite were produced in very small quantities during the review period. Thus, each accounted for a relatively small percentage of the total value of Canadian agricultural production. At the same time, certain Tripartite beneficiaries (e.g., swine and cattle) accounted for relatively large percentages of total agricultural production. Similarly, of the relatively large number of remaining commodities in the agricultural universe which did not receive Tripartite, some accounted for a small percentage of production value while others accounted for a large percentage. Because the relative value of agricultural production accounted for by a particular commodity is apparently, and properly, not determinative of whether it may receive Tripartite benefits, it follows that each of these non-covered commodities, whether large or small, must be equally eligible for Tripartite benefits. Accordingly, the fact that the relatively small number of commodities receiving Tripartite benefits happened to account for 33 percent of the total agricultural production value during the review period is of little, if any, significance when viewed alongside the fact that a far greater number of both large and small commodities in Canada did not receive Tripartite benefits. Finally, we note that 33 percent of production value, viewed alone, still represents only a small percentage of the eligible universe, and if that were the sole factor that we had considered, the Department would find Tripartite de facto specific. In addition, we have determined that Tripartite is not integrally linked to other income stabilization programs in Canada. Therefore, the Department is precluded from examining evidence such as that regarding the availability of other stabilization programs, which may or may not explain why there were a small number of Tripartite agreements during the review period. Similarly, we do not consider the growth of the Tripartite program during past review periods to be relevant to an analysis of whether Tripartite is de facto specific during this review period. We acknowledge that commodities were added during the fifth review period. The Department found that Tripartite was de facto specific during that review, however, based upon evidence related to the small number of users, among other things. That determination was upheld by the binational panel reviewing the Department's findings following remand. Swine V Panel Decision at 17-19. Had additional agricultural commodities been added to Tripartite's coverage during this review period, the Department would have considered that evidence and reevaluated the determination that there are too few users of Tripartite to find it not de facto specific. Furthermore, although Tripartite may have added enrollees during this review period, this evidence does not detract from the Department's finding, which properly focused upon the industries, or agricultural commodities, receiving benefits. The additional enrollees produce the same 11 (or 13) commodities that we have determined comprise a specific group of enterprises or industries. Based upon this analysis, we determine that substantial evidence supports the Department's determination that there were too few beneficiaries of Tripartite during the review period to warrant finding the program not de facto specific.

Comment 8: The GOC argues that the Department's determination that live swine producers benefit disproportionately from Tripartite improperly ignores the nature of payments under the program. The GOC claims that dollar payout levels do not show dominant or disproportionate use. First, because payouts are determined by market forces, there will always be variations in the amount of payouts to different commodities and even to the same commodity at different times. Second, the percentage of payouts received by hog producers declined substantially during the review period, suggesting that over time, the percentage of Tripartite benefits received by hog producers will return to relatively low levels. In addition, the GOC questions the value of the dominant or disproportionate use criteria in evaluating Tripartite. Because the benefits are determined by market forces, the dominant use test yields inconsistent findings regarding Tripartite's specificity. Petitioner argues that the Department's analysis of dominant or disproportionate use is supported by substantial evidence in the record, and is otherwise in accordance with law. The GOC's argument, on the other hand, is unsupported by law. Petitioner contends that the Department has previously considered arguments regarding the role of market forces in triggering payments and has concluded that these effects relate to whether a particular industry receives benefits rather than the de facto specificity of a program.

Department's Position: We disagree with the GOC. First, we note that in the preliminary results, the Department determined that Tripartite was de facto specific solely on the basis of the small number of actual beneficiaries during the review period in relation to the large universe of eligible beneficiaries. Preliminary Results at 54,116. We also found that swine producers were dominant users of Tripartite based upon the fact that they have received 70 percent of the benefits over the history of the program. In making this dominant use finding, the Department intended to demonstrate only that, assuming the *12250 Department had made no finding regarding the number of users, Tripartite could still have been found de facto specific. Id. at 54,117. Therefore, because we reasonably determined that the number of actual Tripartite users was small, no dominant use finding was required by the statute. Accordingly, inasmuch as the Department's dominant use finding was not necessary in order to support our affirmative de facto specificity finding on the basis of the small number of users, we have considered the parties' dominant use arguments only to determine whether they identify evidence in the record which would somehow detract from the Department's affirmative determination. We have determined that no such evidence has been identified. Contrary to the argument of the GOC, a dominant or disproportionate use finding could well be relevant to an income stabilization program such as Tripartite if we were unable to make a specificity finding based upon the small number of users. However, the question of whether the subject merchandise happens to constitute a large or small industry (agricultural commodity) is immaterial to the Department's specificity analysis when the Department has already determined that a program is de facto specific based on the small number of users. Assuming the number of users in a case was not small, which is not the situation here, the Department could very well determine that the subject merchandise was a dominant user regardless of its relative size. Similarly, the fact that Tripartite payments are triggered by market forces cannot be considered in determining whether the program is de facto specific. It may be that swine producers consistently receive a disproportionate share of benefits because they happen to experience consistently bad years which trigger higher payouts. Subsidies are often provided when companies or industries experience downturns in their markets, and it would be unreasonable for the Department to find that such market forces render subsidies not specific and thus not countervailable. Neither the statute nor the proposed regulations permit the Department to alter its specificity analysis on this basis.

Comment 9: The GOC also takes issue with the Department's findings that the "government of Canada may exercise discretion in the administration of" Tripartite, and that this evidence does "not detract from (our) finding of specificity" based on evidence relating to the small number of users. The GOC argues first that in relying on the legislative history of the Tripartite program to show that the Minister of Agriculture has a great amount of discretion, the Department has improperly relied on non-record evidence. According to the GOC, documents submitted by Petitioner as Tripartite legislative history were stricken from the record, and may not be considered in the Final Results. The GOC argues that, as a matter of law, the Department's proposed regulations require the Department to consider "the extent to which a government exercises discretion in conferring benefits under a program." The Department's consistent practice has been to look for the actual exercise of discretion, and the Swine V panel specifically declined to sustain the Department's approach to the contrary. Therefore, according to the GOC, the Department's finding that the government "may retain" discretion is erroneous. The GOC claims that the record on Tripartite fails to show that the GOC has ever exercised the relevant discretion, and the verification report establishes that there have been no actions limiting the availability of Tripartite agreements. Moreover, the Department persists in overlooking the extensive criteria provided in the ASA for evaluating Tripartite agreement requests. The GOC urges the Department to consider the nature of the program, which in the case of Tripartite precludes government manipulation. The government cannot control the market factors which dictate when payouts are made. Neither can the government control which producer groups will seek Tripartite agreements, and which producers will enroll once an agreement is reached. Therefore, there is no opportunity for the GOC to influence, or use its discretion in, the granting of benefits under Tripartite. The absence of evidence of government discretion must weigh against a de facto specificity determination. Petitioner claims that it is not improper for the Department to rely on the legislative history of the ASA in analyzing whether the government retains discretion. Petitioner cites the CIT decision in Central Soya Co. v. United States, 15 CIT 35, 13 ITRD 1085, 1087 (1991), which held that "the court has broad power or discretion to take judicial notice of legislative facts." Moreover, Petitioner argues that record evidence indicates that Tripartite benefits may be awarded in a discretionary manner; the negotiating process is discretionary in and of itself. The government does not automatically establish a Tripartite agreement for any producer group interested in obtaining one. Therefore, Petitioner argues that the Department's finding with regard to discretion is supported by substantial evidence in this review. Petitioner concludes that, regardless, a flawed discretion finding does not nullify the Department's specificity determination since the Department stated in the preliminary results that it "historically has not placed great emphasis on this factor."

Department's Position: We disagree with the GOC regarding the Department's approach to the evidence relating to the exercise of government discretion during this review. The Department found that the Government of Canada "may exercise discretion" in the administration of Tripartite. The Department did not base its determination of specificity on this evidence, however. As explained in the previous comments, the Department determined that Tripartite was de facto specific solely on the basis of the small number of only 11 (or 13) actual beneficiaries during the review period in relation to the universe of eligible beneficiaries. Preliminary Results at 54,116. At the same time, after reviewing all the information in the record, we were not able to identify an established, publicized and consistent review process leading to Tripartite agreements. The fact that negotiations are involved appears to indicate that the outcome may be unpredictable and inconsistent from one agreement to another. Thus, the resulting Tripartite agreements do not necessarily reflect identical terms or conditions. Preliminary Results at 54,117. We also disagree with the GOC that the Department may not rely upon the Canadian legislative history relating to the Tripartite program. First, the legislative history is arguably publicly available, published information and it may be relied on at any time during the proceeding. We determined earlier in the review, however, that it was not appropriate to permit Petitioner to add this information to the record after the deadline provided for in the Department's regulations for submitting factual information. See 19 CFR 355.31(a)(1)(ii). Regardless, the Department's regulations do not preclude the Department from adding factual information to the record at any time during a proceeding, id. at s355.31(b)(1), especially prior to the preliminary results. Therefore, the fact that the Department did not permit Petitioner to add this information to the record did not preclude the Department from adding it to the record itself and relying upon the same information in reaching *12251 its determination. Because it was plain that the Department had indeed relied upon this information, the parties had an adequate opportunity to comment upon it substantively.

Comment 10: The GOQ argues that the Department's reexamination of the FISI program, notwithstanding the decisions of two binational panels, is inconsistent with administrative practice and with the international obligations of the United States. According to the GOQ, the panels reviewing the fourth and fifth administrative reviews held that the evidence on the record did not support a determination of countervailability. By reinvestigating FISI, the Department is departing from its administrative practice not to revisit a decision absent new evidence or facts which indicate a change in the program. There is no new evidence regarding FISI; the program has remained essentially unchanged from prior reviews. The GOQ also maintains that the Department is reexamining FISI because it has never managed to compile a record sufficient to find FISI countervailable. This continuous and unjustifiable examination of FISI constitutes a restraint of international trade in violation of U.S. obligations under the General Agreements on Tariff and Trade and the FTA. Petitioner responds that the countervailability of FISI has neither been explicitly affirmed by a reviewing binational panel, nor explicitly rejected. The panel in Fresh, Chilled and Frozen Pork from Canada, USA-89-1904-06, at 19 (March 8, 1991), and Fresh, Chilled and Frozen Pork, USA-89-1904-06, at 2 (June 3, 1991) (collectively Pork), concluded that the evidence on the record was insufficient to sustain the Department's countervailability determination regarding FISI. The binational panel in the fifth review of the order on live swine ordered the Department to remove FISI benefits from its calculation for the review period because of defects in the supporting record. Thus, by examining FISI in this review, the Department has not violated its own practice of not reinvestigating a program previously found not countervailable.

Department's Position: The Department's practice is not to reexamine a specificity finding made in the investigation or in a subsequent review absent new facts or evidence of changed circumstances. In this review, however, as we explained in the preliminary results, the Department's determination to reexamine FISI is reasonable in light of new evidence compiled by the Department regarding the number of potential beneficiaries of the program and other evidence. Preliminary Results at 54,117-18. In each proceeding reviewed by a binational panel, the panel highlighted what it considered to be deficiencies either in the supporting evidence or in the Department's analysis. For instance, the Swine V panel found that the Department had failed to provide a "properly articulated rationale for determining that FISI was countervailable" based on record evidence, and ordered the Department "to remove FISI benefits from its duty calculations for that review period." The Pork panel's holding was the same. Therefore, in this review, as explained above and in the preliminary results, we have compiled new evidence.

Comment 11: If the Department does not rescind its investigation of FISI, the GOQ urges the Department not to consider FISI in isolation but together with two other Que1bec programs: Crop Insurance and Supply Management. According to the GOQ, these programs serve jointly to meet the province-wide objective of stabilizing farm income. Taken together they cover 81.2 percent of the value of Que1bec's agricultural production; they also meet the differing needs of the agricultural sector, covering each farmer's most significant risk. Furthermore, this common purpose is best demonstrated by the administrative overlap between FISI and Crop Insurance, which are both administered by the Re1gie des Assurances Agricoles du Que1bec (the Re1gie). These facts illustrate a unified provincial objective, fulfilled through complementary activities which reflect the diverse production and market risks faced by Que1bec's farmers. On this basis, the Department must conclude that FISI benefits are not de facto specific. Petitioner counters that the GOQ is really arguing that these various programs are integrally linked. Therefore, Petitioner argues, the Department should reject this argument because, having been raised only at the briefing stage of the administrative review, it is untimely. Should the Department entertain the GOQ's argument, Petitioner argues that there is insufficient record evidence to support a claim that the programs should be considered together. At the very least, the GOQ's arguments fail to address two of the factors the Department must consider when examining an integral linkage argument: funding and equality of treatment.

Department's Position: Although the GOQ did provide timely information about the programs which it now appears to contend are integrally linked to FISI, the GOQ did not present a timely allegation that these programs were integrally linked. Without a timely allegation during the investigation or administrative review that a program is integrally linked to other programs, the Department is unable to solicit and consider evidence relating to this question, and other parties are unable to comment on any determination the Department might reach. Therefore, for purposes of the Department's de facto specificity analysis, we have continued to base our determination of the specificity of FISI on the availability and use of that program standing alone. See Proposed Regulations at section 355.43(b)(6).

Comment 12: Like the GOC, the GOQ takes issue with the Department's interpretation of the statute that a de facto specificity determination may be based on only one of the factors listed in the proposed regulations. Consequently, the GOQ contests the Department's determination that FISI is de facto specific based only upon the small number of users participating in the program. It is the GOQ's view that the Department only briefly mentioned the other factors in its preliminary results, determining summarily that no other factors detracted from the specificity finding. The GOQ maintains that the Department must collect and fully evaluate all reasonably available evidence, and that it "may not rely on isolated tidbits of data which suggest a result contrary to the clear weight of the evidence." USX Corporation v. United States, 655 F. Supp. 487, 489 (CIT 1987). See also Universal Camera Corp. v. United States, 340 U.S. 474 (1950). In addition, the GOQ states that every binational panel, except one, which has examined this issue has agreed that the Department cannot find specificity after examining only a single factor. The GOQ argues that the Magnesium panel, which held that the Department may find specificity after examining only one of the de facto specificity criteria, did not face this issue squarely because it found that the Department had considered three of the four specificity criteria, and there was evidence in the record indicating specificity under the fourth. The GOQ also argues that because there are different bases for analyzing de jure and de facto specificity, the Department may not properly rely upon its practice of basing a specificity finding on the single de jure factor as a justification for relying upon a single factor to determine de facto specificity. In rebuttal, Petitioner cites Alberta Pork v. United States, 669 F. Supp. 445, *12252 451-52 (CIT 1987), the CIT decision which held that FISI is countervailable expressly because of the limited number of program users.

Department's Position: We disagree with the GOQ's interpretation of the Department's statutory and regulatory requirements as well as the GOQ's assessment of how the Department conducted its analysis of FISI. Under Universal Camera (and USX Corp.), the Department and other administrative agencies are required to base determinations upon substantial evidence "when viewed in the light that the record in its entirety furnishes, including the body of evidence opposed to the (agency's) view." Universal Camera, 340 U.S. at 488. Like the GOC, the GOQ implies that in a situation like the present one, in which the Department considers evidence regarding several evidentiary factors in reaching a determination, we are somehow required to reach affirmative findings on two or more of those factors in order to support an affirmative determination. This reading of the statute and applicable case law is mistaken. The holding of the Supreme Court in Universal Camera and other cases requires only that the Department consider all evidence. In addition, the statute does not draw a distinction between consideration of de jure and de facto evidence, as the GOQ claims. As with de facto specificity, when determining whether a program is de jure specific, the Department will consider any evidence in the record which fairly detracts from an affirmative determination. As a matter of practice and logic, however, once the Department determines that a program is de jure specific on the basis of a finding relating to certain evidence, the Department is not required to reinforce that finding with additional findings supporting an affirmative determination. Similarly, when the Department determines that a program is de facto specific based upon too few users (or evidence relating to a different factor), that finding alone warrants an affirmative specificity determination, provided the Department views the evidence "in the light that the record in its entirety furnishes, including the body of evidence opposed to the (Department's) view." Universal Camera, 340 U.S. at 488. In the present review, the Department correctly applied this standard. As the preliminary results demonstrate, we considered evidence related to all four factors outlined in the proposed regulations. As with Tripartite, we concluded that FISI was defacto specific during the review period based upon the small number of actual beneficiaries in relation to the very large number of eligible beneficiaries. Preliminary Results at 54,117-18. No evidence in the record fairly detracts from this determination. Thus, it is clear that the Department properly examined and considered all relevant evidence in the record, and its determination that FISI was de facto specific based upon the small number of users is supported by substantial evidence and is otherwise in accordance with law.

Comment 13: The GOQ challenges the Department's determination that FISI is de facto specific based upon what the Department found to be the small number of users. According to the Department's findings, FISI covered 15 products out of an eligible universe of over 80 during the review period. The GOQ states that this conclusion is flawed. First, the Department's finding that there are "over 80 agricultural commodities produced in Que1bec" is based on incorrect assumptions and is not consistent with other information in the record. While the Department defined Que1bec's agricultural universe with reference to the combined product listings applicable to both FISI and Crop Insurance, the Department never determined whether products covered by Crop Insurance are defined at the same level of aggregation as those covered by FISI. Further, the list provided by the Department in its November 4, 1993 memorandum includes 66 products and appears to have aggregated some products listed in the original documents but not others. This list includes certain products which were not produced in Que1bec during the review period, while not providing an accounting of this aggregation or the basis for combining various products. It also includes certain other products on the basis that they were produced in quantities and values similar to other livestock covered by FISI. However, there is no information about the value of production in the 1991 Agricultural Profile, and the fact that certain livestock were produced in similar quantities is not relevant to whether the products were produced at commercially comparable levels. In addition, in at least two instances, the Department double-counted: the Department should not have listed ewes and wethers separately because the Profile doesn't indicate whether both were produced in Que1bec; and the Department should not have listed bee colonies because it already counted honey (and bee colonies are not a commercial product). According to the GOQ, the 29-product listing which it provided defines the agricultural universe at the same level of aggregation as the FISI-covered products. Based on this list, FISI covered 15 out of the 29 products produced in Que1bec, which would render the program not specific based on the number of users. In addition, this simple comparison is an inadequate evidentiary basis for finding de facto specificity. The Department must examine the program coverage in terms of other factors such as the percentage of the total farm production. Agricultural commodities covered by FISI in this review accounted for 38.6 percent of the total value of agricultural production. The GOQ maintains that coverage of over a third of Que1bec's farm sector contradicts the Department's conclusion that FISI covered too few users. Petitioner responds that the assumptions the Department made with regard to Que1bec's agricultural universe are based on record evidence, and that in assessing the number of FISI-eligible products, the Department conducted extensive analysis, consulting three different alternative sources in addition to examining the undocumented list provided by the GOQ. Petitioner asserts that the GOQ's claim that the Department's classification methodology is imprecise is without merit, because the GOQ itself neglected to provide adequate guidelines to the Department. Finally, Petitioner states that the GOQ's suggested product aggregations themselves demonstrate the absurdity of their complaints.

Department's Position: We disagree with the GOQ. It is undisputed that during the period of review, FISI covered only 15 agricultural commodities under 11 schemes. As the Department explained at length in the preliminary results, in order to estimate the universe of eligible agricultural commodities in Que1bec, we examined the two different lists provided by the GOC (Farm Cash Receipts (FCRs)) and the GOQ, both of which listed 29 commodities. We determined that these estimates were not sufficiently reasonable because they disaggregated commodities much too broadly and contained unexplained inconsistencies. For instance, while listing "all vegetables for processing" as one category, the GOQ listed feeder hogs and piglets as two categories. By contrast, the actual coverage of FISI is disaggregated on a much more reasonable and consistent individual commodity basis, providing FISI *12253 schemes for such narrowly defined commodities as grain corn, sugar beets and silage wheat. See Agricultural Universe Memorandum. Therefore, as Petitioner notes, the Department relied upon several independent sources of information, including the 1990-91 Annual Report of the Re1gie des Assurances Agricoles du Que1bec (Re1gie Report) and the 1991 Agricultural Profile of Canada, and found that there are over 80 agricultural commodities in Que1bec which should reasonably be eligible for FISI schemes. We determined that compared to this relatively large number of eligible recipients, the 15 agricultural commodities actually receiving FISI benefits was a small number of recipients. In this regard, we noted that the Department considers FISI de jure not specific because, according to the FISI Act, it is supposed to be available to all "farm products" in Que1bec. The GOQ's arguments above demonstrate the difficulty of agreeing on what is the appropriate definition of "farm products" (or "agricultural commodities") for the purpose of assessing which farm products reasonably should be eligible for FISI. For instance, the GOQ appears to argue in its brief that a commodity's level of "commercial significance" bears on whether it should be eligible for FISI. However, record evidence indicates that although sugar beets remained covered by a FISI scheme during the review period, none were actually produced in the province. Similarly, the GOQ's arguments regarding wethers and ewes and bee colonies are largely unsupported in the record. Even if the GOQ is correct, the Department stressed that its estimate of the agricultural universe in Que1bec (and Canada) could not be expected to be an exact count. We also stressed, however, that the Department's estimate was conservative.

Agricultural Universe Memorandum

Finally, we have considered the GOQ's argument that commodities covered by FISI accounted for 38.6 percent of the total agricultural production value in Que1bec during the review period. We determine that this evidence does not detract from a determination that FISI is de facto specific based upon the small number of only 15 actual users. The statute provides that a domestic subsidy is countervailable if it is provided to a specific enterprise or industry or group thereof, 19 U.S.C. 1677(5)(B), and the Department's proposed regulations provide that the Department will examine the number of actual beneficiaries, whether industries or enterprises, in determining de facto specificity. Thus, although the Department has not previously engaged in an analysis of the percentage of production value covered by a program, as we explained in Comment 7 above with regard to Tripartite, we have done so here pursuant to the GOQ's argument. As discussed below, the Department has determined that, in the context of FISI, as with Tripartite, it has little, if any, significance in light of the relatively small number of actual beneficiaries compared to the relatively large number of eligible beneficiaries. Like Tripartite, FISI benefits are apparently granted and administered on an equal basis, without consideration of the commodity's relative production value. The production value of some commodities receiving FISI is small, while that of others is large. The same holds for commodities not receiving FISI. Therefore, it is reasonable to assign roughly equal significance to each beneficiary for the purpose of determining whether the actual coverage of FISI is small. Accordingly, the fact that the relatively small number of commodities receiving FISI benefits happened to account for 38.6 percent of the total agricultural production value during the review period is of little, if any, significance when viewed alongside the fact that a far greater number of both large and small commodities in Que1bec did not receive FISI benefits. Finally, we note that 38.6 percent of production value, viewed alone, still represents a small percentage of the eligible universe, and if that were the sole factor that we had considered, the Department would find FISI de facto specific. In conclusion, the Department has determined that Que1bec's arguments are unpersuasive. Accordingly, the Department determines that the relatively small number of 15 actual FISI users out of over 80 eligible agricultural commodities is small and, on that basis, FISI is de facto specific within the meaning of section 771(5)(B).

Comment 14: The GOQ argues that live swine producers are not dominant users of the FISI program, nor did they receive disproportionate benefits. The Department used "insured value" as the measure of dominant use when, in fact, this data provides no measure of the benefits which FISI participants actually receive. According to the GOQ, the fact that the insured value of live swine is greater than the insured value of other FISI-covered products does not indicate anything more than that the actual value of live swine is greater than the value of other relevant products. The actual benefit is the provincial government's share of the payouts, not the relative insured values of the products. The Re1gie Report shows that live swine received less than 20 percent of the payouts made under FISI during the review period; thus, according to the GOQ, live swine producers are clearly not dominant users of FISI. The GOQ further argues that swine producers did not receive disproportionate FISI benefits during the review period. Although the Department did not address the issue of disproportionality in its preliminary results, the GOQ asserts that it must do so now, assuming the Department finds that swine producers are not dominant users of FISI. Having received less than 20 percent of total FISI payouts during the review period, the GOQ claims that swine producers received far less than their proportional share of the payouts.

Department's Position: We agree with the GOQ that the insured value of a product is not an appropriate measure of whether a particular beneficiary is a dominant or disproportionate user of the program in question. Contrary to the assertion of the GOQ, however, it would be equally inappropriate to compare the percentage of FISI benefits received by swine producers during the review period (approximately 20 percent) to the total FISI-insured production value of live swine (approximately 51 percent) in an effort to determine whether swine producers received a disproportionate share of benefits. Most importantly, this is because FISI only benefited a small segment of the relevant universe, rendering it unnecessary to also determine whether live swine or any other beneficiary was a dominant user or received a disproportionate share of benefits. If live swine were one of two actual beneficiaries, the Department would not need to determine that one of the two was a dominant or disproportionate user in order to reasonably determine that the program provided de facto specific benefits. Similarly, even in light of all of the other evidence in the record, the fact that swine producers are one of only fifteen actual beneficiaries out of a much larger universe of over 80 eligible beneficiaries warrants a determination that FISI is de facto specific. Accordingly, inasmuch as no dominant use finding was necessary in order to support our affirmative de facto specificity finding on the basis of the small number of users, we have considered the GOQ's dominant use arguments only to determine whether they identify evidence in the record which would somehow detract from the Department's affirmative determination. *12254 We have determined that no such evidence has been identified. Only if the number of beneficiaries of a program is sufficiently large so as to call into question a determination of de facto specificity based upon the number of users would it be necessary to determine whether one or more of the beneficiaries was a dominant or disproportionate user. See Final Affirmative Countervailing Duty Determinations: Certain Steel Products from Brazil, 58 FR 37,295, 37,299 (1993). In other words, a comparison similar to that advocated by the GOQ could be meaningful in the context of "a program in which virtually every segment of the economy (or the agricultural sector) in the market naturally participates to some extent." Final Affirmative Countervailing Duty Determinations and Negative Critical Circumstances Determinations: Certain Steel Products from Korea, 58 FR 37,338, 37,343 (1993). That is not the case here, and it would not be meaningful to compare swine producers' share of FISI benefits to their proportionate share of FISI production coverage because FISI covered so few industries.

Comment 15: The GOQ argues that there is no evidence that the government exercised discretion in administering FISI: each of the propositions on which the Department relied in concluding that the FISI Act "appears to allow the GOQ considerable discretion in determining which products receive schemes" is taken out of context, inaccurate and must be reexamined. As for the Department's conclusion that discretion is evident because "schemes are established for any product * * * which the Gouvernement 'indicates,"' the GOQ argues that the producers themselves request the Re1gie to create a FISI scheme. Moreover, the GOQ claims that the Department's determination that the FISI Act contains no explicit criteria for the establishment of a scheme is clearly erroneous. For one, only farm products which are marketed under a joint plan created at the producers' discretion, and products derived from the participant's own operations are eligible for FISI. The GOQ also argues that in citing the possibility of regional FISI schemes as evidence of possible government discretion, the Department overlooked the fact that the producers themselves, not the GOQ, determine the geographic scope of a FISI scheme. Moreover, the fact that the FISI Act permits the establishment of regional FISI schemes merely ensures that all joint plans created under the Farm Products Marketing Act, even plans reflecting a collection of producers grouped by region, are eligible for FISI if the producers so desire. Finally, the record demonstrates that no regional FISI scheme has ever been created. The GOQ also contests the Department's finding of discretion on the basis that the FISI hog scheme was the only scheme during the review period which did not set a limit on the maximum level of insurance available. The GOQ contends that the ceilings were administratively burdensome and had virtually no economic impact, and that it eliminated the ceiling for the hog scheme in August 1988. Ceilings under other FISI schemes were deemed burdensome as well, and by 1992 most of them had been eliminated. Thus, the absence of a ceiling for the hog scheme during the review period, which the Department deemed to be evidence of discretion, was merely an administrative matter; its elimination cannot be cited as evidence of discretion. Finally, the GOQ argues that differences among FISI schemes in the method of computing net annual income and stabilized net annual income, and differences in eligibility and participation requirements are not evidence of discretion. The GOQ argues that these differences are necessitated because each FISI scheme experiences a unique cycle of income fluctuation, and each scheme must be self- sustaining over the life of the program. In addition, the self-sustaining level which must be achieved reflects the same income level for all of Que1bec's farmers as reflected in the average farm worker's salary in Que1bec. The GOQ argues that this is not evidence of discretion, but illustrates the non-discriminatory provision that, over the long term, all schemes will render the same degree of protection.

Department's Position: Like the GOC, the GOQ mischaracterizes the Department's findings in the preliminary results with regard to government discretion in the administration of FISI. The Department found that the FISI Act "appears to allow the GOQ considerable discretion in determining which products receive schemes." Preliminary Results at 54,118. We did not base our determination that FISI is de facto specific on this evidence, however. As explained in the previous comments, the Department determined that FISI was de facto specific solely on the basis of the small number of only 15 actual beneficiaries during the review period in relation to the universe of over 80 eligible beneficiaries. Id. At the same time, after reviewing all the information in the record, we were not able to identify an established, publicized and consistent review process leading to FISI schemes. Thus, the resulting FISI schemes do not necessarily reflect identical terms or conditions. Id.

Comment 16: The GOQ argues that the Department has incorrectly calculated FISI benefits by aggregating FISI payments paid to hog producers with those paid to piglet producers. The GOQ points out that during the review period, Que1bec exported no piglets to the United States. In addition, there is no evidence in the record which indicates that benefits paid to piglet producers are passed on to hog producers. Citing the upstream subsidies test provided for in section 771A of the Act (19 U.S.C. 1677-1(a)), the GOQ argues that the majority of piglets raised in Que1bec are sold to hog producers at arm's-length market prices. There is no evidence to support the assertion that payments received by piglet producers under the FISI piglet scheme have any effect on the price at which these piglets are sold. Therefore, the piglet payments provide no "competitive benefit" to the exported hogs, as required by 19 U.S.C. 1677-1(a), and this analysis fails on the second and third prongs of the upstream subsidies test. Should the Department determine in the final results that FISI bestows countervailable benefits on live swine, the Department should eliminate the payments under the piglet scheme and only countervail the payments under the hog scheme. Petitioner counters that payments under the piglet scheme are not upstream subsidies, but rather payments which directly benefit producers of market hogs, the merchandise which is subsequently exported. Because payments under the piglet scheme reduce the production costs in farrowing operations, the costs of producing market hogs are thus reduced. Furthermore, Petitioner rejects the GOQ's argument that "arm's-length, market price" transactions negate the benefits to hog producers from the piglet scheme: If there were no subsidies to piglets, fewer would be produced, driving up the price and therefore increasing the cost of hog production. Therefore, the Department has correctly countervailed payments to hog producers at all phases of production regardless of whether pigs are exported in all phases of development.

Department's Position: We disagree with the GOQ. Both piglets and market hogs are included within the class or kind of merchandise subject to the order on live swine from Canada. When calculating the benefits attributable to *12255 the FISI program, the Department has consistently aggregated the benefits provided under the scheme for piglets and the scheme for hogs. In this regard, the Swine IV binational panel correctly stated that "(a)n upstream subsidy inquiry is only required when benefits are provided to an input producer that does not produce the product under investigation." Swine IV Panel Decision at 73. The GOQ's argument that benefits provided by the piglet scheme should be analyzed under the statute's upstream subsidy provision is misplaced. An upstream subsidy analysis is concerned with determining the effect of benefits received by producers of a product which itself is not subject to a countervailing duty investigation or order, but which is an input into the subject merchandise. 19 U.S.C. s1677-1(a). For instance, in Final Affirmative Countervailing Duty Determination; Steel Wheels from Brazil (54 FR 15,523, 15,525-28; April 18, 1989), the Department examined whether subsidies provided to the Brazilian steel industry constituted upstream subsidies within the meaning of section 771A. The steel was an input product; it was not included in the class or kind of merchandise being investigated. As noted, piglets are subject to the countervailing duty order on live swine. Therefore, they cannot be considered recipients of an "upstream subsidy" and section 771A does not apply. Because FISI is a domestic subsidy program, because the class or kind of merchandise includes all live swine, and because live swine were exported to the United States during the review period, the fact that Que1bec did not export piglets during the review period is not relevant to the Department's analysis. Whether or not benefits to piglets benefited market hogs, domestic subsidies conferred on the class or kind of merchandise are countervailable. The benefits bestowed on the entire class or kind of merchandise, including piglets, are appropriately included in the Department's calculations.

Comment 17: The CPC, Quintaine, and Baxter argue that sows and boars are a lawful subclass, and based upon its own practice and its statutory authority, the Department should reconsider its preliminary determination to eliminate the sows and boars subclass. According to these respondents, in the first review of this order, the Department's decision to calculate a separate rate for sows and boars was compelled by what the Department referred to as "exceptional circumstances" and the "considerable" differences between sows and boars and market hogs. The Department also found that the "distinction between slaughter sows and boars and other live swine cannot be used as a means to circumvent the countervailing duty order." Furthermore, Petitioner did not object to the Department's decision. These circumstances and differences still exist, as do the Department's statutory authority and considerable discretion to establish a subclass. In the absence of a change in circumstances, Respondents argue that the Department must carefully consider whether such a change should be made sua sponte. Respondents acknowledge the Department's determination that the criteria adopted in Diversified Products v. United States, 572 F. Supp. 883 (CIT 1983), should only be used to distinguish between, not within, a class or kind of merchandise. Respondents argue, however, that the original sows and boars subclass determination was also based upon the Department's comparative analysis of the amount of subsidies applicable to sows and boars and the amount of subsidies applicable to the other products within the class or kind. While the Department explained its recent rejection of the Diversified Products criteria for distinguishing among products within a class or kind, the Department failed to explain its apparent repudiation of the second part of the test, which the statute clearly supports. According to the CPC, although the statute "creates a presumption in favor of a country-wide rate," it does provide for separate rates whenever a state-owned enterprise is involved or when there are substantial differences between companies in terms of subsidies received. Therefore, the law requires the Department to take into account extreme differences in subsidies received, and when necessary, to overcome the presumption in favor of a country-wide rate. Respondents cite section 355.47(a) of the proposed regulations to argue that the Department's statutory responsibility requires it to ensure that there is a rational connection between the countervailable benefits received by a product, and the calculation of a countervailing duty for that product. Quintaine and Baxter also cite U.S. v. Zenith Radio Corp., 562 F. 2d 1209 (1977), affirmed 437 U.S. 443 (1978), in which the Court of Customs and Patent Appeals held that "countervailing duties should equate to the true bounty actually conferred." Finally, the CPC argues that the Department's subclass methodology has been contemplated in at least two previous investigations, Certain Steel Products from the United Kingdom (47 FR 35,668; August 16, 1982) (UK Steel), and Fresh Chilled and Frozen Pork from Canada (54 FR 30,774, 30,787; July 24, 1989) (Pork). Moreover, the binational panel reviewing the Department's fourth administrative review of this order determined that the Department's initial subclass analysis was reasonable. The binational panel reviewing the fifth administrative review of this order upheld the Department's determination that information about the existence and value of benefits is necessary for the agency to make a subclass determination. Petitioner acknowledges that the Department's reconsideration of the sows and boars subclass decision is consistent with the statute and the regulations, which create the presumption in favor of country-wide countervailing duty rates.

Department's Position: We disagree with Respondents. As we explained in the preliminary results, the Department has determined that the methodology relied upon to separate the class or kind of merchandise into "subclasses" was inappropriate, and we will no longer calculate a separate rate for sows and boars or any other product on this basis. See Preliminary Results at 54,113; Memorandum on Product-Specific Rates in Countervailing Duty Administrative Reviews, from Barbara Tillman to Joseph Spetrini, July 19, 1993 (Subclass Memorandum). The decision during the first administrative review to grant sows and boars a separate countervailing duty rate based upon the subclass determination represented an exception to the Department's normal practice of calculating one rate for the entire class or kind of merchandise subject to a countervailing duty order. See 19 U.S.C. s1677e(a). The Department based its finding of a subclass exception upon a test consisting of two parts, each of which we considered necessary to warrant granting the separate rate. See Preliminary Results of Countervailing Duty Administrative Review; Live Swine From Canada (53 FR 22,189; June 14, 1989); Preliminary Results at 54,113. However, during the present review, we determined that the Diversified Products criteria, the first part of the test, "were designed to differentiate between classes or kinds of merchandise, not among products within a class or kind." Preliminary Results at 54,113. On this basis, we determined "that it was inappropriate to grant the slaughter sows and boars'subclass' exception on the basis of a Diversified Products criteria analysis." Id. Because the reversal of the subclass exception was premised upon the Department's *12256 decision that the Diversified Products criteria were not appropriate for this purpose, it was not necessary to attempt to repudiate the second part of the subclass test, i.e., the comparative analysis of the difference in benefits granted to the producers of slaughter sows and boars vis-a-vis those granted to the producers of other products within the class or kind of merchandise. See id. The CPC's reliance on UK Steel is misplaced. That investigation was terminated when the petition was withdrawn. Therefore, the Department never reached a final determination nor did it issue an order. Accordingly, the Department neither reached a final determination regarding the scope of that investigation nor fully considered the scope issues referred to by the CPC. Further, the fact that the statute provides exceptions to the presumption in favor of country-wide rates does not imply that the subclass exception should be continued simply because sows and boars receive a different amount of subsidies. As we stated in the preliminary results, the express exceptions under the statute recognize differences between individual companies (and government ownership), not between products within the class or kind of merchandise covered by the order. See 19 U.S.C. 1671e(a)(c). Therefore, the Department is only required to examine the possibility of a significant differential when the producer or exporter is government-owned. Beyond government-owned companies, the Department may examine, to the extent practicable, other producers or exporters whose benefits differ significantly from the country-wide rate. See id.; 19 CFR 355.22(d)(1). Finally, Respondents misinterpret the Department's proposed regulations with regard to the requirement that the countervailing duty rate accurately reflect the benefits bestowed on the merchandise under review. Section 355.47 of the proposed regulations only draws a distinction between subject merchandise and non-subject merchandise, and precludes the Department from countervailing benefits tied to non-subject merchandise. Sows and boars are clearly merchandise subject to the countervailing duty order on live swine from Canada.

Comment 18: Quintaine argues that the Department cannot discontinue its recognition of the sows and boars subclass and its practice of calculating a separate rate for the subclass for the following reasons. First, because the Department specifically sought information in its questionnaire with which to calculate a separate rate for the sows and boars subclass, and this information was provided by the GOC, the Department must use the information to calculate a separate rate. Second, nothing in the proceeding prior to the preliminary results indicated the Department's intention to abandon its established practice of recognizing sows and boars as a subclass and granting them a separate rate of duty on that basis. Third, the Department's methodology for calculating the de minimis threshold specifically contemplates the differences between sows and boars and the rest of the class of live swine and uses sales data specifically pertaining to sows and boars as the basis for achieving a weighted-average price for all live swine. Quintaine and Baxter also argue that in abandoning its subclass practice, the Department has acted without notice and created an ex post facto burden on trade not contemplated by the parties at the time of export. Sows and boars which entered during the review period were subject to a product-specific deposit rate substantially lower than the rate for other live swine. The producers and exporters did not contemplate that these entries would be liquidated at the much higher live swine rate determined in the preliminary results in light of the Department's recognition of the sows and boars subclass since the first administrative review of the order. Therefore, Respondents claim that the Department's abandonment of its subclass practice is unfair, inequitable, unprecedented, and an arbitrary abuse of the Department's discretion. Quintaine, Baxter, and Pryme add that the implication in the Department's Subclass Memorandum, that it may further analyze the use of product-specific rates in future cases, will likely result in a product-specific application of the countervailing duty law. Thus, although sows and boars will no longer be entitled to subclass treatment, other products may enjoy such treatment in the future.

Department's Position: We disagree with Respondents. Although the Department collected the information necessary to calculate a separate rate for sows and boars, we subsequently determined that doing so was not appropriate for the reasons articulated in the Subclass Memorandum, the preliminary results and the above comment. After the preliminary results, all parties had ample opportunity to comment on the Department's decision. Respondents provided comments, which we have fully considered. Respondents are also mistaken in claiming that the Department is precluded from changing its policy in this area. "The mere fact that an agency reverses a policy, or a statutory or regulatory interpretation, does not indicate the agency's decision is unreasonable, arbitrary, or capricious." Mantex, Inc. v. United States, Slip Op. 93-242 at 27 (CIT December 22, 1993) (citing Rust v. Sullivan, ------ U.S. ------, 111 S. Ct. 1759, 1769 (1991)). The courts have long recognized that agency policies must be permitted to evolve under judicial supervision. See, e.g., Motor Vehicle Mfrs. Assn. of United States v. State Farm Mutual Auto. Ins. Co., 463 U.S. 29, 42 (1983). An agency "is not required to 'establish rules of conduct to last forever,"' Rust v. Sullivan, 111 S.Ct. at 1769 (citations omitted), but rather "must be given ample latitude to adapt (its) rules and policies to the demands of changing circumstances." Motor Vehicle Mfrs. Assn., 463 U.S. at 42. The Supreme Court has repeatedly upheld the fundamental principle that an agency's "revised interpretation deserves deference because '(a)n initial agency interpretation is not instantly carved in stone' and the 'agency, to engage in informed rulemaking, must consider varying interpretations and the wisdom of its policy on a continuing basis."' Rust v. Sullivan, 111 S.Ct. at 1769 (citations omitted). It is clear that this necessary decision-making process may be accomplished on a case-by-case basis, permitting the Department to adapt its policy during successive reviews, with the only limitation being that it "justi(fy a) change of interpretation with a 'reasoned analysis."' Id. (citations omitted). As explained by the Department in the previous comment and elsewhere, the record in this proceeding reflects the Department's "reasoned" analysis and the justification for its change of interpretation. See, e.g., Subclass Memorandum. Furthermore, contrary to Respondents' claims, the Department's change in policy does not create an unjustified ex post facto burden for exporters and importers of slaughter sows and boars. It is not uncommon for a product covered by an order to enter with a low (or zero) cash deposit rate and to ultimately be assessed a much higher rate as a result of an administrative review covering those entries. Such entries are also routinely assessed interest as required by the regulations. See 19 CFR 355.24. This is a reasonable contingency of which importers and exporters are well aware *12257 when entering merchandise under an order and making deposits of estimated duties. Moreover, the Department's statement in the Subclass Memorandum that we "may further analyze the issue of granting product-specific rates in future cases" in no way qualified the Department's rejection of the subclass policy. With this statement, the Department indicated that it had not determined whether to consider product-specific rates on some other basis, outside the framework of the rejected subclass analysis. Therefore, we affirm our determination in the preliminary results that one country-wide rate will be assessed on all subject merchandise. Finally, we also disagree with Quintaine regarding the de minimis calculation methodology. Because countervailing duties on live swine are calculated on a per-kilogram basis, rather than ad valorem, we must determine what de minimis is on a per-kilogram basis. Our methodology for determining this merely accounts for the price differences between sows and boars and the rest of the class or kind of merchandise within the order on live swine. We must recognize this difference, just as we recognize and account for the difference in provincial prices of other live swine, in order to establish an overall weighted-average price per kilogram for the subject merchandise, from which we then determine the de minimis value, in Canadian dollars (i.e., 0.5 percent of the weighted average price per kilogram). The subject merchandise includes slaughter sows and boars. Therefore price data for sows and boars must be factored into that calculation. However, mere recognition that sows and boars sell at a different price level for purposes of this calculation does not require the Department to calculate a separate rate for sows and boars, as Quintaine would suggest.

Comment 19: For many of the same reasons given above, Pryme argues that the Department must recognize a subclass for weanlings. First, the recognition of subclasses has been an established and consistent expression of the Department's analysis since the determination in the first administrative review to calculate a separate countervailing duty rate for sows and boars. In the case of weanlings, in the fourth and fifth reviews of this order, the Department concluded that it lacked sufficient information in the record to calculate a subclass rate. See Fourth Review Final at 28536; see also Final Results of Countervailing Duty Administrative Review; Live Swine from Canada (56 FR 50,560, 50,564; October 7, 1991) (Fifth Review Final). Pryme argues that the Department's statement in the Final Results of the fifth review that "(t)he Department has considered Pryme's request, but determines that further information would be required to reach a determination, and that it would be inappropriate to delay the processing of the review to solicit such information," indicates that a timely request and the proper information could have resulted in the finding of a weanling subclass in the fifth review. By virtue of Pryme's timely request in this review, the Department solicited and Pryme and the GOC provided information in order to establish a subclass for weanlings. Therefore, provided the established subclass criteria are met, Pryme argues that the weanling subclass should be granted.

Department's Position: As we determined in our preliminary results, and as explained in Comments 17 and 18 above, we have determined that it is inappropriate to establish subclasses within the class or kind of merchandise covered by an order, as the Department previously did with regard to sows and boars. The fact that the Department denied Pryme's requests to establish a subclass for weanlings in two earlier reviews, based on the untimeliness of the requests and insufficient information with which to conduct the two-part analysis, is not relevant to the issue of whether to grant weanlings a subclass in this review. The Department may alter its practice provided it gives a reasoned analysis for doing so, as explained above. Furthermore, as discussed in Comment 18 above, the Department is not required to establish a subclass for weanlings merely because Pryme made a timely request and responded to the Department's requests for information in this review.

Comment 20: Pryme argues that the Department's failure to recognize the weanlings subclass results in an inaccurate assessment of countervailing duties in contradiction of the purposes of the countervailing duty law. Citing Zenith, Pryme argues that countervailing duties must be equivalent to the benefits conferred. Pryme argues that weanlings qualify for substantially different benefits than the other live swine covered by this order because it falls within the company-specific exception to the presumption in favor of country-wide rates provided for in the statute. See 19 U.S.C. 1671e(a)(2). Pryme argues that benefits received by weanling exporting companies as compared with those received by other exporters of live swine demonstrate the significant difference in the subsidies received by the companies.

Department's Position: We disagree with Pryme that the statute requires the Department to calculate a separate rate for weanlings. Pryme's reliance on the statute's language allowing the Department to determine "that there is a significant difference between companies receiving subsidy benefits" to support this argument is misplaced. This provision requires the Department to consider whether to distinguish among companies receiving different subsidies, not among different products included in the class or kind of merchandise covered by an order. Pryme's request for a weanling subclass is not premised upon its status as a company, but upon its status as a weanling exporter. See Department's Position at Comment 18, above.

Comment 21: Pryme argues that it has met all of the Department's requirements for a company-specific rate. Pryme made a timely request for an individual review, and provided the Department with information with which to calculate a company-specific rate. Record evidence indicates that Pryme received no benefits on its exports of live swine during the review period, and any benefits which Pryme did receive during the review period were de minimis. According to Pryme, in its preliminary results, the Department improperly declined to calculate a company-specific rate for Pryme based on what the Department referred to as an "incomplete" or "incorrect" certification. Pryme argues that this finding ignores the fact that there is no prescribed form of certification in the statute or the regulations. See 19 CFR 355.22(a). The Department's verification report states that the certifications were accurate as presented with regard to weanlings, but notes that the Department found that, during the review period, Pryme had received Tripartite benefits on market hogs sold in the quarter prior to the review period. Pryme argues that these benefits were de minimis; therefore, the certifications were neither incorrect nor incomplete, since Pryme received no cognizable benefits. In addition, Pryme argues that the Department should not be concerned with the Tripartite payment received by Pryme during the review period because it was made on merchandise sold prior to the review period. As support, Pryme cites the Department's regulations, which provide that an "administrative review * * * normally will cover entries or exports of the merchandise during the most recently completed *12258 reporting year of the government of the affected country." 19 CFR 355.22(b).

Department's Position: We disagree with Pryme. In addition to the subclass request addressed above, Pryme made two other requests. First, Pryme requested what it referred to as a "company-specific rate," i.e., "individual rate" in accordance with section 706(a)(2)(A) of the Act and s355.22(d) of the Department's regulations. As we explained to Pryme after receiving its request, because of the very large number of exporters of live swine, the Department conducts reviews of this order on an aggregate basis and does not collect individual sales and export data. Therefore, we have determined that it is not practicable to examine whether a significant differential exists between the country-wide rate and the net subsidies received by individual producers. See 19 CFR 355.22(d); 53 FR at 52,325-26 (December 27, 1988) (commentary to the proposed regulations). In addition, Pryme requested an individual review, in accordance with s 355.22(a)(2) of the Department's regulations, which requires that several conditions be met before the Department may review an individual producer or exporter. First, a person requesting an individual review must provide the Department with a certification that the person did not apply for or receive benefits on the subject merchandise from any programs which the Department had previously found countervailable, and that the person will not do so in the future. The person must also provide certifications from the government of the affected country stating that no benefits were provided to the person requesting the review or to any of the person's suppliers. Finally, the person must provide the certifications of its suppliers of the subject merchandise, and of the government regarding those suppliers, stating that they did not apply for or receive benefits under the countervailable programs, and that they will not do so in the future. 19 CFR 355.22(a)(2). The Department must then verify that all certifications "are complete and accurate." Id. at s355.22(f)(2). If the Department determines that the certifications are complete and accurate, that is, there was no net subsidy received on the merchandise covered by the request, as provided for in s 355.22(f)(1), that person is assessed a zero rate and a corresponding zero cash deposit rate. Pending the verification required pursuant to s355.22(f)(1) of the Department's regulations, we accepted Pryme's timely filed certifications which stated that Pryme "did not apply for or receive any net subsidy on the merchandise, i.e., Weanlings, swine weighing less than 40 kg., under the National Tripartite Scheme" during the review period. Although weanlings are part of, but not the entire class or kind of merchandise, the Department accepted the certifications, pending verification, based on the assumption that during the review period, Pryme produced and sold only weanlings and had not received any subsidies on any of the subject merchandise during the review period. However, at verification, we found that during the review period, Pryme had sold market hogs and had received benefits under the Tripartite program, based on market hog sales prior to the review period. See Verification Report at 4. Pryme argues that because the Tripartite payments it received during the review period were based on sales prior to the review period, it is inappropriate for the Department to examine these Tripartite payments. We disagree. The Department's standard practice is to countervail benefits when they affect the cash flow of the company. See Proposed Regulations at s 355.48(a). In all reviews of this order since the inception of the Tripartite program, the Department has requested, and the GOC has provided information regarding Tripartite payments made during the review period. The record shows that quarterly payments are made based on hog sales and hog prices in the prior quarter. Therefore, the payments made in the first quarter of the review period regularly reflect sales and prices in the quarter prior to the review period. Under the Department's methodology the benefits associated with these payments are countervailed during this review period. Similarly, Tripartite payments for hog sales in the fourth quarter of a particular review period are made in the following quarter, outside the review period. They are not examined by the Department until the next review period. Accordingly, we properly accounted for Tripartite payments Pryme received during this review period and determined that Pryme's certifications were not complete and accurate with regard to the subject merchandise. In a memorandum on Pryme Pork's request for an individual review, dated April 7, 1993 (on file in Room B-099, Department of Commerce), we stated that "although Pryme's certifications were accurate with regard to weanlings (i.e., Pryme received no benefits on its sales of weanlings), the discovery that Pryme did receive benefits on sales of market hogs, other subject merchandise, rendered Pryme's certifications * * * incomplete." In the preliminary results, we stated that Pryme's certification was incorrect, effectively terminating the individual review of Pryme. Preliminary Results at 54,113. In addition, although we stated in the preliminary results that "we found that, during the review period, Pryme sold only weanlings" (Preliminary Results at 54,113), we have reexamined the record evidence, and it shows that weanlings were the subject merchandise exported by Pryme during the review period, but that Pryme also sold market hogs in April, June and November, 1990, and January and March, 1991 (Verification Report at 7). The Department therefore concluded that Pryme's certifications did not cover Pryme's sale of market hogs during the review period, or Pryme's receipt of benefits on the sale of market hogs during the review period. Although Pryme argues that there is "no prescribed form of certification," the regulations clearly provide that the certifications must state that the "person did not apply for or receive any net subsidy on the merchandise." 19 CFR 355.22(a)(2) (emphasis added). Pryme's certifications, inasmuch as they only applied to weanlings, when in fact, Pryme also sold market hogs, were incomplete. Furthermore, in reexamining the record pursuant to Pryme's arguments after the preliminary results, we have determined that ManitobaPork, est., which administers Tripartite in Manitoba, declined to certify that Pryme had not received Tripartite payments during the review period. Therefore, Pryme's request for an individual review was not properly accompanied by the government certifications required under s355.22(a)(ii) of the Department's regulations. In the case of incomplete or inaccurate certifications, the regulations make no provision for further examination of existing benefits, thus precluding the Department from reaching the issue of whether the benefits received by Pryme are de minimis. In objecting to the Department's preliminary determination, which effectively terminated the individual review of Pryme, Pryme contends that although its certifications were not complete and accurate, they were close. Therefore, in Pryme's view, the Department should have accepted them. We disagree. The Department addressed this issue when promulgating the regulations, and stressed that "we must be reasonably satisfied that the producer or exporter is entitled to a zero rate. Thus, we require *12259 the requester's and the government's certifications that the requester is so entitled." 53 FR at 52,328. As described above, the certifications provided by Pryme were not complete and accurate, as required by the regulations. On that basis, the Department should not have initiated an individual review. Once it did, and once the Department determined that Pryme's certifications were not complete and accurate, we properly terminated, in effect, the individual review.

Comment 22: The CPC argues that the Department should reconsider its determination that the Ontario Rabies Indemnification Program is specific to livestock producers and therefore countervailable. The benefits provided under this program reimburse livestock producers for the value of animals which a federal inspector requires to be destroyed because they are determined to be rabid. The CPC argues that such rabid animals are destroyed in the interest of public health and safety; the loss which livestock producers incur is in the interest of a larger, more general good. The CPC cites the General Agreement on Tariffs and Trade to support its proposition that this type of government action is an exception to the countervailing duty laws of member countries: "Nothing in this agreement shall be construed to prevent the adoption or enforcement by any contracting party of measures: * * * necessary to protect human, animal or plant life or health * * *." General Agreements on Tariff and Trade, 1947, Art. XX, T.I.A.S. 1700.

Department's Position: We disagree. The reimbursements provided under the Ontario Rabies Indemnification Program are limited by law to livestock producers, and therefore, contrary to the CPC's argument, this is a de jure specific program. Once an animal is determined to have rabies, the producer has a clear incentive to destroy the animal in order to protect the remaining livestock. It is also in the interest of the public to have the animal destroyed. However, it is unclear to the Department how the fact that the government then compensates the producer could be viewed as also in the interest of the public health and safety. Because the government payment does not create an incentive to destroy the animal that is not already present (i.e., since the payment is not necessary to ensure destruction of the animal), we determine that the payment serves no preventive health or safety purpose whatsoever. Payments for the value of the animal cannot be construed to be "necessary to protect human, animal or plant life and health." The payment is, instead, a countervailable benefit under U.S. law and GATT.

Comment 23: Petitioner argues that the Department should revise its calculation methodology for the Alberta Crow Benefit Offset Program (ACBOP) to account more accurately for grain consumed by swine in Alberta. Specifically, Petitioner argues that the Department's current methodology does not accurately account for grain eaten by breeding sows and boars. The sows and boars adjustment which the Department currently uses to determine the grain eaten by hogs only accounts for an additional weight gain by a sow or boar of 2.1 kilograms; according to Petitioner, this adjustment is insufficient to reflect the grain eaten daily by sows and boars as an integral part of swine production. Petitioner argues that the Department has the discretion to revise its ACBOP calculations, and should do so using another Alberta Agriculture study provided by the GOC in the questionnaire response. Petitioner maintains that this study is a reliable source for feed and grain consumption information because it is recent, comprehensive, and published by Alberta Agriculture. Petitioner has provided an alternative methodology using information in this study, which Petitioner argues more accurately accounts for grain consumed in the production of swine in Alberta. Petitioner also argues that its methodology simplifies the Department's attempt to account for the difference in weight between market hogs and slaughter sows and boars by recognizing that the grain fed to sows and boars to bring them up to market weight (which they surpass during their breeding careers), as well as the grain they consume during their breeding careers are inputs into the production of live swine. The CPC counters that Petitioner's proposal is an unsupported and illogical attempt to increase the ACBOP benefit by double-counting the grain consumed by sows and boars. The CPC maintains that the production figures used by the Department already account for grain consumed by sows and boars. Petitioner's methodology also ignores the fact that the Department has carefully examined the issue of average weights for market hogs versus sows and boars in the first review of this order. Those averages accurately reflect the much higher weights and much lower production of sows and boars vis-a-vis market hogs. The CPC also takes issue with Petitioner's proposal that ACBOP benefits should be allocated on the basis of hog production rather than hog marketings. The CPC argues that such a change in the Department's calculation methodology requires the Department to examine the census of the entire Canadian hog population during the review period rather than relying on a simple accounting of all hogs marketed, as it did in the calculations for the preliminary results and all other reviews. The CPC further argues that Petitioner's proposed methodology misuses data from two entirely different sources and is flawed by an inaccurate conversion from pounds to kilograms. Finally, the CPC notes that the ACBOP methodology has evolved over time; its present incarnation has been upheld by the binational panel reviewing the fifth review of this order and Petitioner has not advanced any evidence which warrants the Department's reconsideration of the ACBOP methodology.

Department's Position: We agree with the CPC regarding the alternative methodology Petitioner proposes. The Department fully analyzed the record document relied upon by Petitioner before rejecting it in favor of the source document which the Department has relied upon in the past. We determine that the study relied upon by Petitioner is not comprehensive, as Petitioner asserts, and therefore the Department chose not to use it in the ACBOP calculation. Petitioner acknowledged that its proposed alternative study does not include information about the composition of "starter" diets, which is necessary to the ACBOP calculation. The study on which the Department did rely, "Diets for Swine," includes complete information about hog diets at all stages of growth. Moreover, we agree with the CPC that it is inappropriate to "mix and match" information from these two distinct sources, because they are based on different underlying assumptions regarding the composition of hog diets. We also disagree with Petitioner regarding the manner in which the ACBOP methodology accounts for all grain consumed in the production of live swine in Alberta. The sow and boar weight adjustment, while seemingly small, provides an average weight which accurately reflects the much higher weight of sows and boars but the much lower production level. This adjustment enables the Department to accurately account for the additional grain consumed by sows and boars during their breeding careers, and the Department's ACBOP methodology overall reasonably and accurately accounts for grain consumed in the production of swine in Alberta. In addition, we agree with the CPC that *12260 Petitioner's reliance on production rather than marketings represents too great a departure from the Department's methodology in this case for us to consider it at this late stage in the review. Moreover, Petitioner's failure to illustrate that the Department's methodology is flawed or unreasonable further supports the Department's decision not to change its methodology.

Comment 24: The CPC alleges that the Department's preliminary ACBOP calculations contain significant clerical errors which must be corrected: the Department must use the correct figures for the number of live swine produced and for the amount of barley, wheat, and oats grown in Alberta. The correct figures were reported in the questionnaire response, and must be used.

Department's Position: After examining the CPC's allegation, we found minor clerical errors, and have corrected our calculations accordingly. We now determine that the ACBOP benefit is Can$0.0027 per kilogram for all live swine.

Comment 25: Petitioner argues that the Department should adjust its calculations for the Saskatchewan Hog Assured Returns Program (SHARP) to account for the deficit in the stabilization fund accrued over the life of the program. Petitioner maintains that because SHARP was terminated during this review period, with a large cumulative deficit, the Department must address additional benefits which should have been accounted for in earlier reviews. The size of the deficit indicates that in every year in which the program was operational, payouts to hog producers exceeded contributions by the hog producers and the Province of Saskatchewan. This deficit was financed by loans from the provincial government to the stabilization fund; no repayments appear to have been made. Petitioner argues that, in prior reviews of this order, the Department should have countervailed total payouts to producers, net of any producer contributions into the fund. Thus, the remainder of the fund deficit (the total fund deficit minus the amount of the deficit countervailed in this review) constitutes a subsidy that has never been countervailed. Because the program has been terminated, there is now no hope that the deficit will be repaid with future contributions. Petitioner argues that the record shows that the Government of Saskatchewan has decided to write off this deficit, and forgive the loans which financed it. Petitioner now urges the Department to treat the deficit amount, less any amounts previously countervailed, as a grant to Saskatchewan swine producers during the review period. Petitioner further argues that this grant does not constitute the full benefit realized by swine producers. The Department must also calculate the benefit attributable to the apparently interest-free nature of this loan since October 31, 1989, the date of an Order- in-Council which provided that no interest will accrue on the loans. The CPC, in rebuttal, submits that there is no basis for the Department to countervail the entire SHARP deficit. While the SHARP account remains in deficit, without a final decision about the resolution of the fund, there is clearly no benefit to any party, including Saskatchewan live swine producers. The CPC further argues that in its preliminary results, the Department has incorrectly calculated SHARP benefits, by adopting a methodology, without explanation, which is a departure from that established in earlier reviews. The CPC argues that the facts support the Department's use of the earlier established methodology: the Department countervailed one-half of the total stabilization payments made to live swine producers, which accurately reflected the equal contributions made by the provincial government and the live swine producers into the SHARP fund.

Department's Position: Prior to its termination, SHARP provided stabilization payments to hog producers in Saskatchewan at times when market prices fell below a designated "floor price." Hog producers provided one-half of the funds for the SHARP program and the provincial government provided the remaining one- half. Therefore, the Department's practice, in past reviews, has been to countervail one-half of all SHARP payouts to hog producers. In accordance with the establishment of the Tripartite Scheme for Hogs, SHARP was terminated on March 31, 1991, during the review period. Whenever the balance in the SHARP account was insufficient to cover stabilization payments to participants, the provincial government loaned the needed funds to the program at terms consistent with commercial considerations. As of its termination date, the SHARP fund had a sizeable deficit, representing the cumulation over the operating years by which SHARP payouts were greater than the producers' and government's contribution to the SHARP fund. Therefore, the SHARP deficit represents payments already made to hog producers, half of which the Department has already countervailed in prior reviews. The Department has reconsidered the calculation methodology used in the preliminary results, and has determined that we will countervail one-half of the SHARP payouts for the current review period, as in previous reviews. While the SHARP account remains in deficit, however, without a final decision on the resolution of the deficit, there is no benefit to Saskatchewan live swine producers beyond the interest not accruing on the deficit. Thus, there is no reason for the Department to conduct a benefit analysis of the deficit as Petitioner suggests. If the Department learns in a later review that the deficit has been forgiven by the Government of Saskatchewan, it will at that time determine whether the loan forgiveness constitutes a countervailable benefit and apply the appropriate methodology to measure it. However, we have information on the record indicating that effective October 31, 1989, interest stopped accruing on this deficit. We determine that interest not accrued constitutes a benefit to live swine producers. To measure that benefit, we are treating the deficit as a short-term loan. See Memorandum on SHARP Calculation Methodology, from Swine Team to Barbara Tillman, on file in Room B-099, Department of Commerce. To determine the benefit, we first calculated the average amount of the deficit during the review period by taking a simple average of the balance of the deficit at the beginning and the end of the review period. We then multiplied the benchmark interest rate by half of the average deficit. We used as our benchmark interest rate the simple average of the monthly rates (for the review period) reported as "Typical Short-Term Interest Rates" in the Financial Statistics Monthly, Section 2, Domestic Markets--Interest Rates, published by the Organization for Economic Cooperation and Development, February, 1991, and January 1992. We then added this interest-related benefit to the payout- related benefit (one-half of the SHARP payments to live swine producers during the review period, consistent with our methodology in previous reviews). We divided this amount by the total weight of live swine produced in Saskatchewan. We then weight-averaged the benefit by Saskatchewan's share of total Canadian exports of live swine to the United States. On this basis, we preliminarily determine the benefit from SHARP to be Can$0.0022 per kilogram for all live swine during the review period.

*12261

Comment 26: The GOQ and the CPC allege that the Department incorrectly allocated the benefits attributable to the Feed Freight Assistance (FFA) program. According to Respondents, the Department recognized, in the first part of its calculations, that not all swine production in the provinces covered by FFA is eligible to receive benefits under this program. However, when the Department weight-averaged the per-kilogram benefit by the respective provinces' share of total Canadian exports of live swine, the Department erroneously assumed that all exports of swine from the FFA-eligible provinces were eligible for assistance. To correct this error, Respondents urge the Department to apply the same ratio it uses to determine FFA-eligible production for the purpose of determining FFA-eligible exports. The Department should then weight-average the per kilogram benefit by the share of total Canadian exports accounted for by this adjusted export figure. Petitioner argues that the Department's calculation methodology correctly translated the FFA benefits provided on a per-kilo basis of hog production to the applicable proportion of exports of live swine to the United States. Petitioner argues that following the Respondents' methodology, which requires adjusting provincial exports downward, results in the "double-subtraction" of the exports used to weight-average the benefit.

Department's Position: We agree with Respondents that the methodology used to calculate FFA benefits was flawed. However, we are correcting the flaw using a different approach. Although we recognize that FFA availability is limited to certain areas within the participating provinces, we determine it is not appropriate to adjust provincial production downward, as we did in the past. This adjustment is not required because the appropriate denominator for this federal program available in only some provinces is the total production in the provinces in which FFA operates. We determine that adjusting the denominator as we did in the past results in overstating the FFA benefit. To determine the FFA-benefit per kilo of live swine we first divided the amount of feed transportation assistance to all live swine producers by the weight of all live swine produced in all FFA-eligible provinces. We then used the ratio of the total amount of exports from the provinces in which the FFA is available to total Canadian exports of live swine in order to calculate the weighted benefit. The result is accurate because in doing the calculations we weight-averaged all the benefits for each province by the total amount of exports from that province. We then summed the resulting weighted benefits to determine the country-wide rate. Having discontinued the adjustment in production, there is no need to adjust the exports in the manner Respondents suggest. Using this methodology, we have calculated the FFA benefit to be Can $0.00018 per kilogram for all live swine.

Comment 27: The CPC argues that two provincial programs, the New Brunswick Hog Price Stabilization Program, and the Prince Edward Island Hog Price Stabilization Program should be added to the Department's list of terminated programs. Proper documentation of these programs' terminations was provided in the questionnaire response.

Department's Position: We agree with the CPC regarding the Prince Edward Island Hog Price Stabilization Program. The GOC provided documentation that this program was terminated, and that documentation indicates that no residual benefits will accrue to hog producers. Therefore, we will include this program in our list of terminated programs and will no longer examine it. However, we disagree with the CPC regarding the New Brunswick Price Stabilization Program. While the New Brunswick provincial government stated that the program was terminated, the GOC has provided neither adequate documentation of the program's termination, nor information regarding residual benefits. Therefore, we will continue to list this program as "not used" until such evidence is provided in a future review.

Comment 28: The CPC argues that the Department should issue a final determination which, as in past reviews, directs Customs to use the exchange rate in effect on the date of entry of the subject merchandise for both deposit rates and final assessments. In the preliminary results, the Department proposed using two different exchange rate methodologies: for the cash deposit rate, Customs will convert the assessment amount in Canadian dollars using the exchange rate in effect on the date of entry; for the final assessment of entries made during the period of review, Customs will convert using a simple annual average exchange rate. To institute two different methodologies for these calculations which have always shared the same methodology would constitute a retroactive change in prior agency practice. Petitioner argues that the use of a simple average exchange rate by the Department is not contrary to its regulations. Petitioner claims that pursuant to 19 CFR 353.60 there was no "sustained change" in the prevailing exchange rate during the review period that would materially distort the value of the Customs assessment. Consequently, the Department's method is acceptable under the regulations and should be retained in its final determination.

Department's Position: After consideration of the CPC's argument, we will instruct Customs to assess duties on live swine during the review using the appropriate exchange rate in accordance with Customs' regulations. Petitioner has misapplied section 353.60(b) of the Department's regulations, which guides the Department's use of exchange rates in antidumping proceedings.

Final Results of Review

As a result of our review, we determine the net subsidy to be Can$0.0295 per kilogram for the period April 1, 1990 through March 31, 1991. The net subsidy determined for each program is as follows:

------------------------------------------------------------------------------
Program Rate per
kilo
------------------------------------------------------------------------------
(1) Feed Freight Assistance Program .................................. $0.00018
(2) National Tripartite Stabilization Scheme for Hogs ................. 0.01910
(3) Que1bec Farm Income Stabilization Insurance Program ............... 0.00420
(4) Saskatchewan Hog Assured Returns Program .......................... 0.00221
(5) Alberta Crow Benefit Offset Program ............................... 0.00268
(6) Alberta Livestock and Beefyard Compensation Program
(Livestock Predator Sub-Program) .................................... 0.00000
(7) Ontario Farm Tax Rebate Program ................................... 0.00000
(8) Livestock Improvement Program for Northern Ontario ................ 0.00000
(9) Ontario Pork Industry Improvement Plan ............................ 0.00043
(10) Ontario Rabies Indemnification Program ........................... 0.00000
(11) Saskatchewan Livestock Investment Tax Credit ..................... 0.00045
(12) Saskatchewan Livestock Facilities Tax Credit ..................... 0.00028
Total .................................................................. 0.0295
------------------------------------------------------------------------------

Therefore, the Department will instruct the Customs Service to assess countervailing duties of $Can0.0295 per kilogram on all shipments from Canada of the subject merchandise exported on or after April 1, 1990 and on or before March 31, 1991. Further, as provided for by section 751(a)(1) of the Act, the Department will collect cash deposits of estimated countervailing duties of $Can0.0295 per *12262 kilogram on all shipments of the subject merchandise from Canada, 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 Act (19 U.S.C. 1675(a)(1)) and 19 CFR 355.22.

Dated: March 9, 1994.

Joseph A. Spetrini,

Acting Assistant Secretary for Import Administration.

(FR Doc. 94-6001 Filed 3-15-94; 8:45 am)

BILLING CODE 3510-DS-P