Article 1904


Live Swine from Canada

Secretariat File No.

MAY 30,1995







Saul L. Sherman, Chair
Howard N. Fenton, III
Robert E. Lutz
Martin H. Freedman
W. Ian Binnie
    Joel K. Simon, Christopher M. Kane, for P. Quintaine & Son, Ltd., Pryme Pork, Ltd., and Earle Baxter Trucking LQ.; William K. Ince, Michele C. Sherman, for Canadian Pork Council and Its Members

    Edward Reisman, Stephen J. Powell for International Trade Administration, U.S. Department of Commerce

    Paul C. Rosenthal for National Pork Producers Council, Et Al.



This is a Binational Panel review of the Final Results of the Sixth Annual Review by the Department of Commerce (Commerce) under its Countervailing Duty (CVD) Order on Live Swine from Canada.1 The Petitioners challenge Commerce’s denial of separate treatment for the oldest and heaviest swine (Sows and Boars) and for a category of young, light swine (Weanlings) covered by the order.2 In all prior review periods for which separate rates have been calculated, Commerce found that these categories of swine received zero or de minimis sub-sidies under the Canadian programs being countervailed, considerably lower than the subsidies received by the class or kind as a whole.

The Petitioners are P. Quintaine & Son, Ltd., and Earl Baxter Trucking LQ (collectively, Quintaine), importers of Sows and Boars, and Pryme Pork, Ltd. (Pryme), importers of Weanlings. Support-ing Quintaine, the Canadian Pork Council (CPC) appeals the revocation of a separate subclass for Sows and Boars. Appearing in support of Commerce’s decision to rescind the separate rate for Sows and Boars is the original Petitioner on behalf of the domestic industry, the National Pork Producers Council (NPPC).

An understanding of the present issues requires an examination of relevant prior proceedings.3 The original CVD order defined the class or kind of merchandise covered as all live swine from Canada except breeding sows and boars. 50 Fed.Reg. 25,097 (June 7, 1985).4 That definition has remained unchanged, and thus the scope of the order includes Sows and Boars, and Weanlings as well.

In the First Annual Review, Quintaine appeared for the first time and sought exclusion of Sows and Boars from the scope of the order, or alternatively creation of a separate subclass (a product-specific rate) or a company-specific rate. 54 Fed.Reg. 651 (Jan. 9, 1989). NPPC did not oppose the request for a subclass, and Commerce granted the request, saying:

"The Department has considerable discretion in determining whether to differentiate among products within a class or kind of merchandise. We only differentiate among products in exceptional circumstances. Among the criteria we con-sider are the extent to which the product qualifies as a distinct product subclass within the applicable class or kind of merchandise and the extent to which the subsidy on the product differs from the subsidy on the other products within the same class or kind of merchandise." Prelim. Results, 53 Fed.Reg. 22,189, at 22,190 (1988).

This Sows and Boars subclass was carried forward without challenge or modification through all succeeding annual reviews until the present Sixth Annual Review, where Commerce, acting sua sponte, rescinded it.

In the Fourth Annual Review, Pryme appeared for the first time and sought similar relief for Weanlings—exclusion from the scope, creation of a subclass or, alternatively, establishment of a company-specific rate. Commerce found that the order’s scope included Weanlings and went on to deny the alternative relief on the ground that the evidence of record was insufficient to establish a separate Weanling subclass or rate. 56 Fed.Reg. 28,531, at 28,536 (June 20, 1991). On review, the Binational Panel (the Swine IV Panel) ordered Commerce to create a Weanlings subclass and to calculate a rate for it, based on the available evidence. 5

In the Fifth Annual Review, Commerce again found the evidence on the record as to Pryme and Weanlings insufficient. Final Results, 56 Fed.Reg. 50,560, at 50,564 (Oct. 7, 1991). After a remand in which Commerce reexamined the evidence and once more found it in-sufficient, the Binational Panel (the Swine V Panel) affirmed Com-merce’s denial of both product-specific treatment (i.e., a subclass) for Weanlings and company-specific treatment for Pryme. USA-91-1904-04 (Aug.26, 1992; June 11,1993).

In this Sixth Annual Review, Commerce first circulated questionnaires seeking information necessary to calculate a separate rate for Sows and Boars, and also a separate rate for Weanlings, and then in October of 1993 issued its Preliminary Results, proposing to eliminate the Sows and Boars subclass and to deny the Weanlings subclass. In support of this proposed action, later adopted, Commerce did not claim that the relevant facts had changed, but cited legal grounds and an internal policy memorandum dated July 13, 1993, which we discuss in detail below. Company-specific relief was also denied. See Swine VI, Prelim. Results, 58 Fed.Reg. 54,112, at 54,113-114 (Oct. 20, 1993); Final Results, 59 Fed.Reg. 12,243, at 12,255-257 (Mar. 16, 1994).


The general rules regarding the standard of review before a Chapter 19 Panel such as this are familiar from the experience under the predecessor United States/Canada Free Trade Agreement and are not in dispute. The Panel steps into the shoes of the Court of International Trade and the Court of Appeals for the Federal Circuit and is to apply the standards and the substantive law (including legislative history, case law, etc.) that those courts apply when they review a countervailing duty determination by Commerce. 6 This in turn means that the Panel is to hold unlawful "any determination, finding or conclusion found . . . to be unsupported by substantial evidence on the record or otherwise not in accordance with law." 19 U.S.C.A. §1516a(b)(1)(B). The Panel is not to substitute its own judgment, and the only question before it is whether the agency’s action had appropriate support in fact and/or law. To be in accordance with law, the agency’s interpretation of the statute need not be "the only reasonable interpretation or the one which the court would adopt had the question initially arisen in a judicial proceeding." American Lamb Co. v. United States, 785 F.2d 994, 1001 (Fed. Cir. 1986); Texas Crushed Stone Co. v. United States, 35 F.3d 1535, 1539 (Fed. Cir. 1994).

The matter is somewhat more complex when, as in this case, the agency has been following one course of action and then switches to another. It is clear that the agency’s action in changing course is entitled to deference, and that the agency is not automatically or permanently locked into its initial position:

"An initial agency interpretation is not instantly carved in stone."

Rust v. Sullivan, 500 U.S. 173, 184-87 (1991)(quoting from Chevron U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 863 (1984)); Good Samaritan Hospital v. Shalala, 508 U.S. __, 113 S.Ct. 2151, 124 L.Ed.2d 368, 382-83 (1993); Wheatland Tube Corp v. United States, 841 F.Supp. 1222, 1229 (Ct. Intl Trade, 1993); Mantex v. United States, 841 F.Supp. 1290, 1303 (Ct. Intl Trade 1993).

Indeed, to fulfill their statutory functions, administrative agencies require flexibility to enable them to adapt their policies in the light of experience and changes in circumstances. Cf. Motor Vehicle Manufacturers Assoc. v. State Farm Mutual Auto Ins. Co., 463 U.S. 29, 42 (1983)(the agency "must be given ample latitude to adapt [its] rules and policies to the demands of changing circumstances," quoting from Permian Basin Area Rate Cases, 390 U.S. 747, 784 (1968). Nevertheless, it appears from the decisions that the agency’s action is entitled to less deference when it changes its interpretation of the statute it is operating under:

"An agency interpretation which conflicts with the agency’s earlier interpretation ‘is entitled to considerably less deference’ than a consistently held agency view." I.N.S. v. Cardoza-Fonseca, 480 U.S. 421, 446 n.30 (1987)(quoting from Watt v. Alaska, 451 U.S. 259, 273 (1981)).

Often, as in I.N.S., the agency’s change of position is sustained. E.g., Good Samaritan, supra; British Steel PLC v. United States, Slip Op. 95-17, at 119-20 (Ct. Intl Trade 1995). Cf. Torrington Co. v. United States, 745 F.Supp. 718, 727 (Ct. Intl Trade 1990) (no departure from prior practice found), affd 938 F.2d 1276 (Fed. Cir. 1991). Conversely, the courts have not been unwilling to strike down agencies’ efforts to change course when the justifications ad-vanced by the agencies are found not adequate in the circumstances. Morton v. Ruiz, 415 U.S. 199 (1974); Shikoku Chemicals Corp. v. United States, 795 F.Supp. 417 (Ct. Intl Trade 1992); cf. Secretary of the Interior v. California, 464 U.S. 312, 320 n.6 (1984). In such situations:

"It is a principle of administrative law that ‘an agency must either conform to its prior norms and decisions or explain the reason for its departure from such precedent.’" Torrington, supra, 745 F.Supp. at 727, and cases there cited.

There are no mechanical rules that will automatically ex-plain all of these decisions or eliminate the need for a case-by-case assessment. The present case is unusual in that the agency does not claim that the change was occasioned by any new facts or experience, and the agency’s earlier view has been repeatedly applied to the same parties in companion proceedings. 7

We adopt the view, agreed to by all of the present parties, that the test laid down in Mantex v. United States, supra, controls here: Commerce must provide ". . . a comprehensive and reasoned analysis for re-versing its former policy . . ." 841 F.Supp., at 1303. Where no such basis of decision appears, there is present the kind of arbitrary action that this Panel, like the United States courts, is charged with curbing.


Petitioners Quintaine and CPC argue that Commerce’s reversal of policy was unreasonable and unlawful for a number of reasons. They argue that the original subclass decision was well established and time tested. They state that while jurisprudence does allow for changes in administrative policy, less deference is owed to an agency that is changing a long-standing practice, and that in any case the courts will not condone the indiscriminate exercise of agency discretion. They further state that there has been provided no adequate "reasoned analysis" as required by the court in Mantex. Finally, they argue that the parties are entitled to rely on long-continued administrative practices.

Commerce, on the other hand, argues that by revising its administrative policy in respect of the Sows and Boars subclass, it has brought its policy more closely in line with congressional intent. Commerce says that while the legislation permits the creation of sub-classes, it has concluded that the methodology used in this case (i.e., the application of the criteria approved in Diversified Products Corp. v. United States, 572 F.Supp. 883 (Ct. Intl Trade 1983)) is inapposite. Commerce reserves the right, consistent with congressional intent, to create subclasses in the future. It argues that the two memoranda contained in the present administrative record which address this issue show that the Department in fact deliberated over the issue for a substantial period of time and, accordingly, gave the abandonment of the Diversified Products criteria serious consideration.

Commerce also argues that it provided a "reasoned analysis" explaining the change. Commerce argues that it is not required to show "new factual information" in order to support such a change in policy. Finally, Commerce argues that this reversal of policy did not involve an abuse of discretion, nor did it constitute an unfair ex post facto burden on trade. The NPPC notes that the initial subclass determination was exceptional and that nothing in the relevant law or regulations requires special treatment of Sows and Boars. The NPPC further points to the provision for company-specific rates as furnish-ing a remedy for Petitioners.


We turn now to Commerce’s decision to abolish the subclass for Sows and Boars. As noted, Commerce does not claim that there are any new facts or circumstances that explain its change of policy. See Tr., 98-99. Rather, Commerce seeks to justify its reversal on purely legal grounds. This Panel finds it significant that Commerce does not claim that its experience in assessing the Sows and Boars subclass at a zero CVD rate for the preceding five years of operations produced any adverse consequences which suggest that the existence of the subclass should be terminated.

The starting point for our analysis is that in 1985 Commerce found that there were exceptional circumstances in this situation justifying the exceptional creation of a subclass. It appears that those circumstances are the gross differences between Sows and Boars and Slaughter Hogs, sufficient to meet the Diversified Products test usually applied to define a class or kind, not a mere subclass.8 Commerce could not point to any indication that these exceptional circumstances have ceased to exist (cf. Tr. 94, 107-111, 117), and thus they still provide a factual basis for the exceptional creation of a subclass. Commerce did not dispute that the Diversified Products criteria, if applicable, continue to be satisfied in this case. Commerce simply contended that the criteria were not applicable.

Commerce’s Rejection of the Diversified Products Criteria Seeking to justify its action on strictly legal grounds, Commerce said that it had made a "breakthrough" in the course of the Sixth Annual Review, which it described as follows in its Final Results:

"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. § 1677e(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 decision that the Diversified Products criteria were not appropriate for this purpose, it was not necessary to attempt to repudiate (sic) 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." 59 Fed.Reg., at 12,255-256.

Upon careful examination, ever mindful that the question is not whether we agree with Commerce, we cannot conclude that this is an explanation that satisfies the requirement of a "reasoned analysis."

The difficulties and contradictions are several. To begin with, the belated rejection of the Diversified Products test is presented as a matter of congressional intent.9 Yet Commerce points to no congressional expressions of intent to support its reading. If §1677e, with its presumption of a single rate for the entire class or kind covered by a CVD order, were said to implicitly outlaw all subclasses not expressly mentioned in the statute, that might be a tenable view; but we are repeatedly told that Commerce still reserves the right to create subclasses on other unspecified bases in the future.10  Commerce does not point to anything to support such a reading of the congressional intent, i.e., that subclasses are permitted but that the Diversified Products test is not an appropriate methodology for their creation.

Commerce argues that the Diversified Products test is inapplicable because its function is to separate larger groups— a class or kind—and not smaller, more narrowly defined groups— a subclass, such as Sows and Boars. As applied here, this view translates into the proposition that Sows and Boars cannot be treated differently from other live swine because the differences between the two groups are too great. The Panel cannot characterize as a "reasoned analysis" the conclusion that differences large enough to differentiate classes are irrelevant to the differentiation of subclasses.

A closer look at "the Diversified Products test" reveals other anomalies. "Class or kind" is a flexible concept, to be applied by the agency in shaping appropriate categories of merchandise for various regulatory purposes.11  In Diversified Products, supra, the Court affirmed Commerce’s use of four criteria in establishing a class or kind. These criteria are obvious, common-sensical considerations, to be looked at in determining whether two articles belong in the same grouping—their physical characteristics, what they are used for, how and to whom they are sold, and what purchasers look for in buying them. So basic are these criteria that counsel for Commerce conceded on oral argument that in the future Commerce might look to any or all of them in creating a subclass.12  Not only are these factors so obvious that any test that ignores all of them is likely to be defective, but all of them are matters of degree. The approach developed by Commerce and approved by the Court in Diversified Products can evaluate differences large or small. There are other criteria that might be applied for measuring smaller differences, and other ways might be devised for measuring; but the proposition that Congress did not intend the Diversified Products test to be applied in creating sub-classes is an ipse dixit that simply does not withstand careful scrutiny. 13

The Departmental Memoranda

In looking for a possible "reasoned analysis" from Commerce, this Panel carefully examined the two key internal memoranda developed during the evolution of Commerce’s position and referred to by them at one point as "building blocks" on the way to the agency’s conclusions. These two internal memoranda—one dated 9/28/92 (the 1992 Memo) and the other dated 7/19/93 (the 1993 Memo) -- both predate the October 1993 Preliminary Results, in which Commerce first announced its pro-posed elimination of the Sows and Boars subclass.

The 1992 Memo is cast largely in terms of factual circumstances and appears to have been replaced by the 1993 Memo, setting forth the essentially legal grounds noted above.14  But since the Government has at times appeared to rely on both, this Panel has fully considered both memoranda.

The 1992 Memo

The 1992 Memo, addressed to the Assistant Secretary in charge of Import Administration as a basis for deciding the issues under review here, recommends that no separate product-specific rates be calculated within the class or kind covered by any CVD order. (The Memo reports that the Policy Branch had recommended product-specific rates, based on a suggested new test, but the Memo argues that no test "can address all the potential variables [or] . . . withstand being eroded over time . . .") In support of this recommendation, three 15 "potential problems"—there is no claim that they have actually been encountered—are set forth. We discuss each in turn:

A. "If in the course of an investigation we calculate a product-specific rate and find it to be zero, we cannot speculate what the consequences would be for the ITC injury analysis."

The primary difficulty with this proposition is that Commerce acknowledged that after many years a separate zero CVD rate for Sows and Boars has caused no such problem. (Tr., 97-98.)

B. "Administering product-specific rates is likely to involve a huge administrative burden, both for IA and Customs; in the long run, the costs will be greater than the benefits."

Here again, Commerce conceded that its experience of five years with the Sows and Boars subclass provides no factual support for the concern expressed. (Tr., 97-99.) Why excluding a readily identifiable product from a duty assessment would be unduly burdensome, we are not told. Commerce did not point to any such "huge burden" that has been encountered to date. Commerce’s experience with the Sows and Boars subclass, not to mention the administrability of the far greater differentiation in rates under the anti-dumping statute, indicates the contrary.

C. "Product-specific rates may encourage foreign producers to shift production and/or exports into non-subsidized pro-ducts with the lower rate. However, one could argue that any benefit to a company is a benefit to its total production." It would seem that to the extent that exporters shift into non-subsidized products the CVD law has accomplished its purpose. But in any event there does not appear to be any serious contention that Slaughter Hogs are likely to be shifted into breeding duty in order to qualify as Sows and Boars. Once again, there is no evidence that this has occurred after five years of a separate subclass and a zero CVD rate for Sows and Boars.

Accordingly, this Panel is of the view that the 1992 Memo does not provide a "reasoned analysis" for Commerce’s change in administrative policy. The memorandum identifies two policy options, re-commends the option of abolishing all subclasses, and in support of its recommendation moots three "potential problems" for which Commerce provides no factual basis.

The 1993 Memo

The 1993 Memo takes an entirely different tack, arguing that the abolition of the Sows and Boars subclass is best explained as a matter of statutory interpretation. It is entitled "Product-specific rates in countervailing duty administrative reviews." It is signed by the Director of the Office of Countervailing Compliance, addressed to the Acting Assistant Secretary for Import Administration, and is cited in both the Preliminary and the Final Results of the Sixth Annual Review. The 1993 Memo deals initially with Pryme’s request for a separate rate for Weanlings. The body of the Memo is entitled Analysis, and Point 1 is headed: "The Diversified Products Criteria Were Not Intended To Separate Products Within a Class or Kind of Merchandise." After describing the application of the Diversified Products test in Commerce’s scoping practice, this section of the Memo concludes:

"Even in light of the Department’s broad discretion to clarify the actual scope of an order by applying the Diversified criteria, it is not reasonable to draw even more subtle dis-tinctions between products using the same criteria for the purpose of attempting to create ‘subclasses.’

* * * *

"In conclusion, if a product is found to be within the class or kind of merchandise covered by a CVD order, whether as expressed by the written descriptions of the merchandise from the original investigation or by way of a Diversified Products analysis, we determine that it is inconsistent with the purpose of the statute and the Department’s scope practice for us to rely upon the Diversified Products criteria to attempt to distinguish that product as a ‘subclass’ from the remainder of the class or kind."

The second section of the Analysis is headed: "The Statutory and Regulatory Framework With Respect To Countervailing Duty Rates." The relevant language is as follows;

"Thus, the Department has determined that the statute contains a presumption in favor of country-wide counter-vailing duty rates, and the statute and regulations are silent on whether the class or kind of merchandise subject to a CVD order may be separated into product-specific categories. However, while the Department may further analyze the issue of granting separate product-specific rates in future cases, the Department has definitely determined that the Diversified Products criteria are only appropriate for distinguishing between classes or kinds of merchandise. They are not appropriate for distinguishing between products within a class or kind of merchandise." These passages essentially repeat the assertions we have already examined above and found insufficient. The statutory presumption16  is invoked when Commerce wishes to deny a subclass, but it does not prevent the agency from creating subclasses in the future when it chooses to—so long as the standards applied are said to be other than the Diversified Products test. The 1993 Memo provides no reasons of sub-stance supporting the agency’s change of position.

Finally, it is noteworthy that in all of the material presented there is no discussion by Commerce of the objectives of the statute or of the effects various interpretations would have on the achievement of those objectives.

Since Commerce presents no valid reasons for revoking the Sows and Boars subclass, it remains,17 and this Panel remands to  Commerce to establish a separate rate for the subclass. Commerce acknowledges that it has sufficient information to do so.


Commerce’s review of the subclass issue was triggered by Pryme’s renewed application for Commerce to recognize a subclass for Weanlings. Ultimately, that application was denied not on the merits of Pryme’s request, but rather on the ground that Pryme based its claim on the same standards that had been applied to Sows and Boars; those standards having been rejected by Commerce, Pryme’s claim fell with Quintaine’s. Since we remand for reinstatement of Quintaine’s Sows and Boars subclass, this Panel is of the view that Pryme’s application must also be given appropriate consideration on remand.

On both previous occasions when Pryme sought a subclass, Commerce claimed that the information of record was not sufficient to enable it to calculate a separate rate for Weanlings. No mention was made of this point in Commerce’s Results now under appeal, and a separate questionnaire on Weanlings had been prepared, circulated, and responded to for precisely this purpose. Nevertheless, apparently in anticipation of the recurrence of such a claim by Commerce, Pryme’s brief in this review proceeding contained an appendix setting forth a detailed calculation of a rate for Weanlings. The Government’s answering brief made no mention of this subject. Upon questioning from the Panel at oral argument, Commerce stated that the Government of Canada had been unable to furnish one requested figure, the relevance of which Commerce failed to elucidate.18  If Commerce’s method of calculating the CVD rate requires information that does not exist, and the Government of Canada is not withholding information, in all of the circumstances presented here we would place a high burden on Commerce to show the need for this information. We expect that Commerce will find a way to calculate a separate rate for Weanlings, based on the available data, as they did on a previous occasion. Until the status of the subclass for Weanlings has finally been determined, Pryme’s alternative request for individual company treatment need not be addressed. The Panel therefore expresses no view at this time on the ITA’s treatment of Pryme’s request for an individual review and a company-specific rate.


We affirm in part and remand in part:

We affirm the findings that Sows and Boars and also Wean-lings are within the scope of the order.

We remand with directions to Commerce to:

I. Reinstate the Sows and Boars subclass and determine a separate CVD rate for it; and

II. Consider Pryme’s application for a subclass for Wean lings, employing the same criteria used in creating the Sows and Boars subclass, and as appropriate calculate a CVD rate for such subclass, explaining in detail any reasons that may be found to preclude the establishment of a Weanling subclass or the calcula-tion of a separate CVD rate for such subclass.






MAY 30, 1995 W. IAN BINNIE

1 Final Results, 59 Fed.Reg. 12,243 (Mar. 16, 1994). The period covered is April 1990 - March 1991. The present review is governed by Chapter 19 of the North American Free Trade Agreement (NAFTA) and 19 U.S.C.A. §1516a(g).

2 The respective categories of swine have been defined as follows:

                 Weanlings       Slaughter Hogs      Sows and Boars


Weight:   under 40 lbs.     170 to 240 lbs.        450 to 700 lbs.
Age:        6 to 8 weeks       c. 6 months               2 to 5 years.

See Prelim. Results, First Admin. Rev., Live Swine, 53 Fed.Reg. 22,189, at 22,190 (June 14, 1988). As to industry terminology generally, see ITC Industry & Trade Summary, ITC Publ. No. 2511 (AG-5) (1992).

3 The parties stipulated in the course of argument on a motion to expand the record that the Panel could take notice of the prior proceedings, as set forth in published decisions, and of any agency policy memoranda referred to.

4 The "breeding sows and boars" excluded from the scope of the order are imported to be used for breeding in the United States and must be registered with the Secretary of Agriculture for that purpose. See HTSUS, Chapter 1, Additional U.S. Note 1. They are not to be confused with the Sows and Boars involved here, which have concluded their careers as breeders in Canada and are imported into the United States to be slaughtered. To avoid this confusion, the latter are sometimes referred to as "Slaughter Sows and Boars," but unfortunately this can create confusion with Slaughter Hogs, the largest category of imports, which are also imported to be slaughtered. Slaughter Hogs reach the butcher shop or restaurant as loin, chops, ribs, ham, bacon, etc. Sows and Boars are so old and heavy by the time they are slaughtered that their meat is ground up and can be used only in sausage and like products. Loc. cit. supra, n. 2.

5 Commerce complied only after a second remand and the dismissal of a complaint addressed to an Extraordinary Challenge Committee. The challenge to the Panel’s treatment of the sub-class issue was abandoned by Commerce at the argument before the ECC. USA-91-1904-03 (May 19, 1992; July 20, 1992; Nov. 19, 1992); see ECC-93-1904-01 USA (Apr. 8, 1993), 58 Fed.Reg. 26,115 (Apr. 30, 1993).

6 The Panel is to "apply the standard of review set out in [section 516A(b)(1)(B) of the Tariff Act of 1930, as amended] and the general legal principles that a court of the [United States] otherwise would apply." NAFTA, Art. 1904(3), as amplified by Annex 1911.

7 Each CVD annual review is conducted on a separate record. Indeed, the usual purpose of the review is to take into account relevant new market information. However, Commerce carries points previously decided forward from one review to another in the absence of reason to review or reconsider. E.g., 59 Fed.Reg. at 12,250 (re carry-over of specificity findings as to subsidy programs). Cf. Shikoku, supra, 795 F.Supp., at 422, suggesting an administrative analogue to the judicial doctrine of "law of the case."

8 The only other exceptional circumstance we discern is that the petitioning domestic industry, NPPC, did not oppose the creation of the Sows and Boars subclass. (It should be noted that Weanlings are imported for the purpose of fattening them into Slaughter Hogs, so that the purchasers of imported Weanlings are U.S. hog producers.) In the present annual review before Commerce, NPPC still did not oppose continuation of the subclass, but in these review proceedings they appeared in support of Commerce’s decision to abolish the subclass.

9 ". . . we have determined that the intent of the statute 9 is that the class or kind of merchandise not be divided into sub-classes on the basis of perceived differences in products based upon the Diversified Products criteria." Prelim. Results, at 54,113.

"[W]e had been using . . . Diversified, and we determined that this is not what Congress intended. . . ." Tr., 101.

10 See, e.g., Tr. 114-115, 130-132.

11 Cf. 19 U.S.C.A. §1401a(g)(2); 19 CFR 152.105(e).

12 Tr., 113-14.

13 The Government was unable to give an example of a ground of agency decision that would not meet the "reasoned analysis" test. The only answer the Panel could elicit was—"none." Tr. 188-89. Plainly, just any explanation is not sufficient.

14 The 1993 Memo has attached to it a routing slip headed "Concurrence Record," with a note at the bottom stating:


"This [1993] memo reflects the decision [Assistant Secretary] Alan Dunn made at our meeting on September 28, 1992 [the date of the 1992 Memo]. It has undergone significant revision since it was originally circulated."

While the matter is by no means entirely clear, it would appear that the earlier version referred to is the 1992 Memo, and that the 1993 Memo was supposed to replace rather than supplement it. 

Perhaps for this reason, the 1992 Memo was not included in the record filed by Commerce with the Panel and was only produced as a result of a Freedom of Information Act inquiry and follow-up motion to expand the record by Quintaine and Pryme. The Panel denied the motion as to the other material sought, but only after it had been produced by Commerce and screened by the Panel. USA 94-1904-01, Opinion and Order on Motion to Expand the Record, Oct. 3, 1994.

15 The 1992 Memo also mentions that "Legal advises us that we have no statutory or regulatory requirement to calculate product-specific rates; nor are we expressly prohibited by the regulations from doing so."

16 19 U.S.C.A. §1671e(a)(2) establishes a presumption that a CVD order will apply to all merchandise of the class or kind from the country under investigation but provides for certain exceptions to be made by the agency. (This is evidently the source of the power claimed by Commerce to create subclasses. See n. 10, supra.)

17 See n. 7 supra.

18 Tr. 127: "We would have to go back for more information. . . . We could not because the Government of Canada couldn’t give us the number for all weanlings sold, I presume because every market hog was once a weanling."

to the top!