Understanding National Treatment: The Particpatory Vision of the WTO UNDERSTANDING NATIONAL TREATMENT: THE PARTICIPATORY VISION OF THE WTO Peter M. Gerhart* and Michael S. Baron* I. INTRODUCTION: THE INTERPRETIVE PROBLEMS Rules against discrimination are easy to state at a general level but are devilishly difficult to apply in particular cases; the gulf between articulating principles of non-discrimination and applying them is wide.' So it is with the national treatment provisions of Article III of GAIT. 2 At a general level, the national treatment principle is sensible, self-evident, and seemingly straightforward. Whether stated in the principle's general and formal version-that a member country must not treat foreign products less favorably than domestic products (without justification under Article XX)-or in one of the common variants-that a WTO member may not discriminate on the basis of the national origin of the product (without justification under Article XX)-the principle appears to be self-applying. Yet the general principle, a bedrock of the WTO system, gives little guidance to help us see whether a domestic measure treats imports less favorably than domestic goods or discriminates on the basis of national origin. Naturally, we look to the purpose of the anti-discrimination provision to help us apply it, but moving from general purpose to a specific test is also problematic. By all accounts, the national treatment principle is designed to interdict "hidden protectionism" and to prohibit measures that are equivalent * Professor of Law, Case Western Reserve University School of Law. ** J.D. 2003, Case Western Reserve University School of Law. The authors would like to thank Mel Durchslag and Johnathan H. Aoller for their helpful comments and critiques. 1. For example, the Equal Protection Clause of the U.S. Constitution simply states that "[n]o State shall... deny to any person within its jurisdiction the equal protection of the laws." U.S. CONST. amend. XIV, § 1. Yet, the Supreme Court chooses from at least three different levels of scrutiny to determine the validity of a state or federal statute that discriminates against a group of people. See ERWIN CHEMERINSKY, CONSTITUTIONAL LAW: PRINCIPLES AND PouciEs 526, 526-33 (1997) (giving an overview of Equal Protection analysis). See generally Tristin K. Green, Discrimination in Workplace Dynamics: Toward a Structural Account of Disparate Treatment Theory, 38 HARv. C.R.-C.L. L. REv. 91 (2003) (arguing that the current test for identifying unlawful discrimination must be changed in order to provide equity in the workplace); Regina E. Gray, Comment, The Rise and Fall of the "Sex-Plus" Discrimination Theory: an Analysis of Fisher v. Vassar College, 42 HOw. L.J. 71 (1998) (discussing gender discrimination in the workplace under Title VII); Rebecca Hanner White, Modem Discrimina- tion Theory and the National Labor Relations Act, 39 WM. & MARY L. REv. 99 (1997) (comparing discrimination theories under the National Labor Relations Act and Title VII). 2. The General Agreement on Tariffs and Trade 1994, art. III, Apr. 15, 1994, Marrakesh Agreement Establishing the World Trade Organization, Annex 1A, THE LEGAL TEXTS: THE RESULTS OF THE URUGUAY ROUND OF MULTILATERAL TRADE NEGOTIATIONS 17 (1999), 33 I.L.M. 1154 (1994) [hereinafter GATT 94]. IND. INT'L & COMP. L. REV. to tariff barriers,3 with the goal of protecting the commitments that WTO members have made to reduce tariff and other trade barriers and to insure equality of competitive conditions. But identifying hidden protectionism or measures that circumvent the rules against trade barriers is tricky business.4 3. According to the Appellate Body, [Tihe broad and fundamental purpose of Article III is to avoid protectionism in the application of internal tax and regulatory measures. More specifically, the purpose of Article III 'is to ensure that internal measures not be applied to imported or domestic products so as to afford protection to domestic production.' Toward this end, Article 1 obliges Members of the WTO to provide equality of competitive conditions for imported products in relation to domestic products... Article III protects expectations not of any particular trade volume but rather of the equal competitive relationship between imported and domestic products. WTO Appellate Body Report on Japan-Taxes on Alcoholic Beverages, WT/DS8/AB/R, WT/DS1O/AB/R, WT/DSl l/AB/R, at 15 (Oct. 4, 1996) [hereinafter Japan-Alcohol (AB)] (citations omitted). This anti-protectionist thrust is supported by Article 111: 1, which provides a statement of general interpretive purpose: "The contracting parties recognize that internal taxes and other internal charges, and laws, regulations and requirements affecting the internal [distribution of products] should not be applied to imported or domestic products so as to afford protection to domestic production." GATT 94, supra note 2, art. 3, para. 1. The Appellate Body has recognized that this "general principle" from Article III: 1 informs Article I1:4, although Article 111:4 does not explicitly refer to the general principle. WTO Appellate Body Report on European Communities-Measures Affecting Asbestos and Asbestos-Containing Products, WT/DS135/AB/R, para. 98 (Mar. 12, 2001) [hereinafter EC-Asbestos (AB)]. "[T] here must be consonance between the objective pursued by Article III, as enunciated in the 'general principle' articulated in Article HI: 1, and the interpretation of the specific expression of this principle in the text of Article 111:4." Id. There has been some confusion about the relationship between Article 111:I and Article 111:4. An earlier Appellate Body decision seemed to indicate that this general principle informs the various provisions of Article II in different ways. WTO Appellate Body Report on European Communities-Regime for the Importation, Sale and Distribution of Bananas, WT/DS27/AB/R, para. 216 (Sept. 9, 1997) [hereinafter EC-Bananas]. Accordingly, some have seen EC-Asbestos to be a change in the way that Article H1I: 1 informs Article II:4. For example, Professor Regan interprets the Appellate Body in EC-Bananas to be saying that Article 111: 1 is not to be looked at in interpreting HI:4. Donald H. Regan, Regulatory Purpose and "Like Products" in Article 1114 of the GA7T (With Additional Remarks on Article III:2), 36 J. WORLD TRADE 443, 446-47 (2002) [hereinafter Regan, Regulatory Purpose]. However, in EC-Bananas, the Appellate Body merely pointed out that "a determination of whether there has been a violation of Article 11:4 does not require a separate consideration of whether a measure 'afford[s] protection to domestic production."' EC-Bananas, supra para. 216. This was a reaction to the panel's decision to apply the "design, architecture and structure" test in its 111:4 analysis. See id. para. 215-16. The statement in EC-Bananas that Article HI: I does not present a separate test is consistent with the statement in EC-Asbestos that I: 1 informs the interpretation of the tests that are set forth in Article 111:4. See also WTO Panel Report on Japan-Measures Affecting Consumer Photographic Film and Paper, WT/DS44/R, para. 10.369 (Adopted Apr. 22, 1998) [hereinafter Japan-Film] (using Article HI: 1 in interpreting 1I:4 but not separately considering "so as to afford protection"). 4. Article III is not the only WTO treaty provision that tries to interdict hidden pro- tectionism or unreasonable barriers to trade. The General Agreement on Trade in Services (GATS) also contains a national treatment provision. See General Agreement on Trade in Services, Apr. 15, 1994, Marrakesh Agreement Establishing the World Trade Organization, Annex 1B, THE LEGAL TEXTS: THE RESULTS OF THE URUGUAY ROUND OF MULTILATERAL TRADE NEGOTIATIONS 284, 33 I.L.M. 1167 (1994), art. 17. Moreover, two other treaties, the [Vol. 14:3 WTO NATIONAL TREATMENT The task is not made easier by the way the national treatment principle is articulated in the WTO treaties. The general Article I1:4 invocation to give "like" imported goods no less favorable treatment than domestic products is simple enough, but it requires us to squeeze the relevant analysis into a few words, and the words are neither defined in GATT nor self-defining. Moreover, the Article I prohibitions present a series of puzzles in themselves. Why are the rules against discrimination in tax measures (under I1:2) different from those applicable to other regulatory measures (under 11:4)? Why have two separate tests for tax measures, one for taxes on like products (the first sentence of 11I:2) and another for taxes on directly competitive or substitutable products (the second sentence and Ad Article of 1:2)? What is the significance of the Delphic instruction in Article mH: 1 that measures "should not" (rather than must not) be applied "so as to afford protection to domestic production"? Finally, if, as some believe, one cannot assess discrimination without looking at the purpose of the regulation, what is the relationship between Article II, which does not mention regulatory purpose, and Article XX, where regulatory purpose is central to the analysis? Generally, WTO scholarship and the popular view of the WTO assume that the national treatment standard has substantive content-that is, that it requires the analyst to evaluate, in some way, the appropriateness of a country's regulatory scheme to see whether the regulatory scheme is consistent with the values that make up the WTO's free trade regime.5 This substantive orientation inevitably leads analysts to advocate some version of an aims and effects test-some inquiry into the purposes of the measure (to see whether, on the one hand, it is protectionist, or, alternately, whether it advances some Agreement on Sanitary and PhytoSanitary Standards and the Agreement on Technical Barriers to Trade have roughly the same purpose. See Agreement on the Application of Sanitary and PhytoSanitary Measures, Apr. 15, 1994, Marrakesh Agreement Establishing the World Trade Organization, Annex 1A, THE LEGAL TEXTS: THE RESULTS OF THE URUGUAY ROUND OF MULTILATERAL TRADE NEGOTIATIONS 59,33 I.L.M. (1994); Agreement on Technical Barriers to Trade, Apr. 15, 1994, Marrakesh Agreement Establishing the World Trade Organization, Annex 1A, THE LEGAL TEXTS: THE RESULTS OF THE URUGUAY ROUND OF MULTILATERAL TRADE NEGOTIATIONS 121, 33 I.L.M. (1994). 5. The common focus of national treatment analysis is on a framework that strikes the appropriate balance between the regulatory autonomy of member states and the suppression of hidden protectionism. See generally JOHN H. JACKSON, THE WORLD TRADING SYSTEM: LAW AND POLICY OF INTERNATIONAL ECONOMIC RELATIONS 212 (1997) (referring to the "clash of policies" inherent in the national treatment provision); RAJ BHALA & KEVIN KENNEDY, WORLD TRADE LAW: THE GATI-WTO SYSTEM, REGIONAL ARRANGEMENTS, AND U.S. LAW 90-105 (1998) (discussing the national treatment obligation); Frieder Roessler, Diverging Domestic Policies and Multilateral Trade Integration, in 2 FAIR TRADE AND HARMONIZATION 1 (Jagdish Bhagwati & Robert E. Hudec, eds., 1996) ("[Ihe rules of [GATT] primarily aim at the reduction of barriers between markets, not at the harmonization of competitive conditions in markets. They therefore impose in principle only constraints on trade policies, but leave the contracting parties free to conduct their domestic policies."). GAETAN VERHOOSEL, NATIONAL TREATMENT AND WTO DISPUTE SETTLEMENT: ADJUDICATING THE BOUNDARIES OF REGULATORY AUTONOMY 2 (2002) (portraying the national treatment analysis as turning on the desire to liberalize trade without requiring deeper market integration or harmonization). 2004] IND. INT'L & COMP. L. REV. legitimate and non-protectionist purpose), some inquiry into the measure's effects on international trade or foreign producers, and some notion of how to balance legitimate purpose and adverse effects.6 Although analysts use a wide variety of verbiage to articulate these tests,7 these substantive approaches are grounded in the common notion that the WTO is overseeing a country's domestic measures to consider how they stack up in light of the impact of the measure on the values of the WTO regime.8 6. The most developed of these approaches is the "aims and effects" test, which under the traditional understanding regulatory purpose is analyzed under the "like product" inquiry under Article 11I, but only when the regulation at issue is origin-neutral. See Won Mog Choi, Overcoming the "Aim and Effect" Theory: Interpretation of the "Like Product" in GATT Article III, 8 U.C. DAVIS J. INT'LL. & POL'Y 107, 115 (2002). Simply put, "aims and effects" asks "whether they [internal regulatory measures] have a bona fide regulatory purpose and whether their effect on conditions of competition is protective." Robert E. Hudec, GATT/WTO Constraints on National Regulation: Requiem for an "Aim and Effects" Test, 32 INT' LLAWYER 619, 626 (1998) [hereinafter Hudec, Requiem]. According to Hudec, such an analysis brings Article lII jurisprudence more in tune with the policy goals of GATT, as stated in Article HI: 1. Id. Hudec also believes that regulatory purpose and trade effects of a measure are the two most important aspects of distinguishing valid regulation from protectionism. Id. at 628. Also, by bringing regulatory justification into the "like product" inquiry, valid regulation will not be made invalid by the harsh rigors of Article XX analysis. Id. The "aims and effects" test, applied to "like products," received support in two GATIT panel decisions. See Robert E. Hudec, "Like Product": The Differences in Meaning in GATT Articles I and III, in REGULATORY BARRIERS AND THE PRINCIPLE OF NON-DISCRIMINATION IN WORLD TRADE LAW 101 (Thomas Cottier & Petros C. Mavroidis eds., 2000) [hereinafter Hudec, "Like Product"], (citing United States-Measures Affecting Alcoholic and Malt Beverages, B.I.S.D., (39th Supp.) at 206 (1993), and United States-Taxes on Automobiles, GATT, GATTr Doc. DS.31/R (Oct. 11, 1994) (unadopted)). However, the "aims and effects" test employed by these two decisions are rejected under current WTO case law. See discussion infra note 9. The "alms and effects" approach has also found a home among commentators under the "so as to afford protection" requirement of Article I11:2 second sentence, and even the "no less favorable treatment" requirement of Article 111:4. See Robert Howse & Donald Regan, The Product/Process Distinction-An Illusory Basis for Disciplining 'Unilateralism' in Trade Policy, 11 E.J. INT'L L. 249, at 267 (2000) ("[I]n its discussion of 'affording protection,' the Appellate Body in Japanese Alcohol may or may not have rejected 'the aims and effects test,' but it clearly did not reject consideration of aims and effects."); Roessler, supra note 5 at 29. See also Lothar Ehring, De Facto Discrimination in World Trade Law: National and Most- Favored-Nation Treatment-or Equal Treatment?, 36 J. WORLD TRADE 921, 945 (2002) (arguing against reading "aims and effects" into the requirement of "no less favorable treatment" in Article 111:4). 7. See Edward S. Tsai, "Like" is a Four-Letter Word-GAIT Article III's "Like Product" Conundrum, 17 BERKELEY J. INT'L L. 26 (1999); Kazumochi Kometani, Trade and Environment: How Should WTO Panels Review Environmental Regulations Under GAIT Articles III and XX?, 16 Nw. J. INT'L L. & BUS. 441 (1996); Choi, supra note 6, at 111 (designating a "proportional tax differentiation based on transparent criteria" test). 8. More recently Geatan Verhoosel has recommended a necessity test for determining the scope of the national treatment provision. See VERHOOSEL, supra note 5, at 2. Under this test, a panel or the Appellate Body would determine whether the restriction on trade that was inherent in the measure was necessary to achieve the purpose of the regulatory system. Id. If it were not necessary, the regulation would be found to have violated the national treatment [Vol. 14:3 WTO NATIONAL TREATMENT These substantive orientations toward the national treatment principle have led to some difficulties in interpreting the Article MI decisions of the WTO panels and the Appellate Body. Although seeming to eschew any aspect of the aims and effects test,9 the Appellate Body has called for an examination of the "design, the architecture, and the revealing structure of a measure,"10 when assessing tax measures, a standard that looks to many to be a test that focuses on the purpose of the measure." And the Appellate Body has provision. Id. Upon analysis this approach also requires a substantive review of the clash between trade values and domestic regulatory values. Although the approach focuses on the connection between the purposes and the effect of the regulation, by assuming that the decision maker can recognize both lawful purposes and adverse effects, it subsumes a form of the aims and effects test. This book is reviewed in Recent Publications: Globalization of Law and Capital, 28 YALE J. INT'L L. 275, 295 (2003) (reviewed by John David Lee) and in Simon Lester, Book Review, 2003 J. INT'L ECON. L. 291 (2003). 9. According to the Appellate Body, [T]he third inquiry under Article 111:2, second sentence, must determine whether 'directly competitive or substitutable products' are 'not similarly' taxed in a way that affords protection. This is not an issue of intent. It is not necessary for a panel to sort through the many reasons legislators and regulators often have for what they do and weigh the relative significance of those reasons to establish legislative or regulatory intent. If the measure is applied to imported or domestic products so as to afford protection to domestic production, then it does not matter that there may not have been any desire to engage in protectionism in the minds of the legislators or the regulators who imposed the measure. Japan-Alcohol (AB), supra note 3, at 27-28. Japan-Alcohol (AB) also rejected the "aims and effects" approach to "like products" under 111:2 first sentence. Hudec, Requiem, supra note 6, at 630. The Appellate body has rejected resort to legislative intent and purpose in other contexts as well. See, e.g., WTO Appellate Body Report on United States-Continued Dumping and Subsidy Offset Act of 2000, AB-2002-7, 16 Jan. 2003 (no need to inquire into legislative intent when interpreting measure that allowed complaining domestic industry to recover dumping duties). 10. Japan-Alcohol (AB), supra note 3, at 29. The Appellate Body has applied the "design, structure and architecture" test in all HI:2 second sentence cases since Japan-Alcohol. See WTO Appellate Body Report on Canada-Certain Measures Concerning Periodicals, WT/DS31/AB/R (June 30, 1997) [hereinafter Canada-Periodicals]; WTO Appellate Body Report on Korea-Taxes on Alcoholic Beverages, WT/DS75/AB/R, WT/DS84/AB/R (Jan. 18 1999) [hereinafter Korea-Alcohol]; WTO Appellate Body Report on Chile-Taxes on Alcoholic Beverages, WT/DS87/AB/R, WT/DS 1 10/AB/R (Dec. 13, 1999) [hereinafter Chile- Alcohol]. 11. See Hudec, Requiem, supra note 6, at 631-632 (stating in the context of III:2, second sentence, "neither the Appellate Body's insistence on different words nor its insistence on objective analysis serve to mark a clear distinction between its 'protective application' concept and the 'aims and effects' analysis.... The decision in the Japan-Alcoholic Beverages case itself did not make clear just how far the Appellate Body's rejection of the 'aim and effect' approach would be carried."). EC-Bananas, supra note 3, by preventing application of design, architecture and structure test to 111:4, effectively limited that test to only 111:2, second sentence. See Hudec, "Like Product", supra note 6, at 117-18 (claiming that under EC-Bananas "the aims and effects test was rather summarily rejected as an incorrect application of the 'like product' test under Article 111:4."). 2004] IND. INT'L & COMP. L. REv. explicitly seemed to endorse a purposive interpretation in one recent case,12 raising new questions about the role of purpose and effects in interpreting Article In.13 Commentators have also suggested that the Appellate Body use an "effects test," suggesting that the Appellate Body examine the proportionate burden of the measure on domestic and foreign products; if the burden on foreign products is disproportionate to the burden on domestic products, the measure can be said to have a protectionist effect. 4 For example, they view "design, architecture, and structure" as an effects test. 5 And some commenta- tors see both purpose and effects in analysis in the cases. 6 Any version of the aims and effects test is problematic, in part because the text of Article III does not support it.'7 Moreover, this substantive orientation toward identifying and interdicting "hidden protectionism" is a 12. See Chile-Alcohol, where the Appellate Body stated that it examines [T]he design, architecture and structure of a tax measure precisely to permit identification of a measure's objectives or purposes as revealed or objectified in the measure itself. Thus, we consider that a measure's purposes, objectively manifested in the design, architecture and structure of the measure, are intensely pertinent to the task of evaluating whether or not that measure is applied so as to afford protection to domestic production. Chile-Alcohol, supra note 10, para. 71. 13. See, e.g., Regan, Regulatory Purpose, supra note 3, at 443 (in Chile-Alcohol "the Appellate Body has told us that... in deciding whether a measure is applied 'so as to afford protection,' we must consider 'the purposes or objectives of a Member's legislature and government as a whole'-in other words, the regulatory purpose of the measure."). However, Regan misinterprets why the Appellate Body looks to "design, architecture and structure," See discussion infra accompanying note 147. 14. For example, Lothar Ehring assesses two possible tests for determining the effect of a measure-the "diagonal test" and the "asymmetric impact test." Lothar Ehring, De Facto Discrimination in World Trade Law: National andMost-Favored-Nation Treatment--or Equal Treatment?, 36 J. WORLD TRADE 921, 924 (2002). Under the "diagonal test," the inquiry is "whether there are any imports receiving less favourable treatment than any like domestic products." Id. Under the "asymmetric impact test," the inquiry is whether imports as a whole are treated less favorably than domestic products as a whole. Id. at 924-25. While suggesting the asymmetric approach to effects is the better approach, Ehring states that a finding of asymmetric impact would not be necessary for finding less favorable treatment. Id. at 925,928 (stating that other facts could lead to a violation, such as the application of the measure or its objective design). 15. See id. at 938 (discussing Chile-Alcohol). See also Simon Lester & Kara Leitner, Dispute Settlement Commentary, European Communities-Asbestos (Appellate Body Report) 14 (2001), available at www.worldtradelaw.net/dscsamples/index.htm (last visited Mar. 9, 2004) ("A discriminatory effect approach appears to have been applied by the Appellate Body in the context of Article 111:2, second sentence in Chile-Alcohol."). 16. See Hudec, Requiem, supra note 6, at 631 (discussing the panel decision in Japan- Alcohol as calling for an effect test rather that an "aims and effects" test in the context of Article 111:2 second sentence). Hudec then states the Appellate Body in Japan-Alcohol called for protective effect plus "protective application .... which for all the world looked like an objective analysis of regulatory purpose." Id. 17. See id. at 628-29 (discussing that lack of textual basis for "aims and effects" approach in "like product" analysis is clearest in 111:2 first sentence). See also Choi, supra note 6, at 117 ("[T]he aim-and-effect theory cannot overcome its critical weaknesses-namely, the lack of textual basis and the ample risk of circumvention."). [Vol. 14:3 WTO NATIONAL TREATMENT major source of friction between notions of national sovereignty and the WTO and creates a public relations problem for the WTO. WTO critics with a substantive orientation see the WTO as interfering with the ability of a country to embrace non-economic goals-as a threat, for example, to environmental or safety values-while street protestors see it as symbolic of undue interference from Geneva, perhaps driven by the overwhelming influence of multinational corporations. Even supporters of the WTO, although staunchly defending the need for rules against "hidden protectionism," must-under the substantive view--concede some room for either purpose or effects to be taken into account in order to mesh WTO and national values,' 8 albeit without any consistent way of understanding how to define either purposes or effects, or how to balance them. In this article we suggest that this substantive-based understanding of the national treatment provision should be, and is being, replaced by a proce- durally oriented understanding, one that largely avoids a judgment about the substantive values underlying national regulation or the clash between the free trade values of the WTO and national regulatory values. When properly understood, the interpretive standards that the Appellate Body has set up are not an endorsement of an aims and effects review. Instead, the Appellate Body is moving, seemingly deliberately, toward a vision of the national treatment principle that emphasizes process values, specifically the importance of protecting domestic lawmaking processes that allow domestic interests to provide "surrogate representation" 9 for adversely affected foreign interests. This interpretation of the national treatment principle puts the Appellate Body in the position of looking at domestic legislation to see whether domestic forces that have interests identical to the interests of foreigners (and would therefore give surrogate representation to foreign interests within the domestic lawmaking process) have in fact been silenced or had their role impaired. This is the surrogate representation rationale of the national treatment principle. This article, by expanding on the surrogate representation rationale, reorients our understanding of the national treatment provisions of Article III from a substantive to a procedural perspective. This reorientation is faithful to the jurisprudence of the Appellate Body interpreting Article III, and shows how the Appellate Body has consistently steered away from a substantive review of national legislation under Article I1 and away from examining either the substantive aims or their relationship to the external effects of domestic regulatory measures, even as it has given real teeth to the national treatment provision. This reorientation is also faithful to the central interpre- 18. See, e.g., Hudec, Requeim, supra note 6, at 620 ("The policing activity of domestic regulatory measures is a delicate task, one that requires reaching an acceptable balance between the trade objectives of the regime and the legitimate regulatory claims of members states."). 19. The term is taken from Laurence H. Tribe's discussion of the same rationale under U.S. dormant Commerce Clause jurisprudence. LAURENCE H. TRIBE, 1 AMERICAN CONSTITU- TIONAL LAW, § 6-5, 1055 (2001). The concept has also been called "virtual representation." JOHN HART ELY, DEMOCRACY AND DISTRUST 82 (1980). 2004] IND. INT'L & COMP. L. REv. tive principle of Article 111, the principle of equality of competitive condi- tions,2° and it provides answers to the puzzles that we have already noted about the relevant WTO provisions and thus leads to a more coherent WTO jurisprudence. Moreover, this reorientation is consistent with, and supports, the central function of the WTO in the international system, which is to enable countries to participate effectively in the policymaking of other countries.2 Finally, this reorientation will ease the perceived tension between the values of the trade regime and domestic regulatory values, because it gets the WTO out of the position of overseeing the clash between those values. This article reflects and transposes in the context of the WTO national treatment jurisprudence an ongoing debate in U.S. constitutional jurisprudence over the appropriate basis for courts to invalidate state regulation that affects interstate commerce. Like the national treatment provision, the idea behind this so-called dormant Commerce Clause jurisprudence has led the Supreme Court to strike down state regulations that discriminate or burden interstate commerce. 22 One view, similar to the dominant interpretation of the WTO's national treatment provision, gives the dormant Commerce Clause substantive content by emphasizing the role of dormant Commerce Clause jurisprudence in protecting against state legislation that would interfere with a common market in the United States, emphasizing the economic goals of the juris- prudence. 23 Another view, and the one highlighted in this article, is grounded in political theory-namely that the purpose of the dormant Commerce Clause jurisprudence is to protect out-of-state citizens from harmful decisions made 20. Under the reasoning of this article, the "equality of competitive conditions" test is the same as the inquiry into surrogate representation. Because the "equality of competitive conditions" test is better supported by the text of Article Ell over any purpose-driven test, the surrogate representation inquiry is also better supported by the text of Article 11I. 21. See infra text accompanying notes 33-43. 22. Although the U.S. Constitution does not explicitly limit state power in this respect, the affirmative grant of power to the U.S. Congress is thought to impliedly limit the power of states, even when the exercise of Congressional power is unexercised and thus lies dormant. A substantial body of thought questions whether this implied limitation on state power is an appropriate role for courts to exercise, especially given the fact that Congress can always limit state power through preemptive legislation. See generally CHEMERINSKY, supra note 1, at 403-06 (summarizing the arguments, but noting that the dormant Commerce Clause is "firmly established"). Martin H. Redish & Shane V. Nugent, The Dormant Commerce Clause and the Constitutional Balance of Federalism, 1987 DuKE L. J. 569, 573 (1987) (the dormant Commerce Clause "lacks any basis in constitutional democratic theory"). 23. Jack L. Goldsmith & Alan 0. Sykes, The Internet and the Dormant Commerce Clause, 110 YALE L.J. 785, 795 (2001) ("The primary justification is that the dormant Commerce Clause ensures free trade among the states and thereby secures the associated econo- mic benefits."); Richard A. Posner, The Constitution as an Economic Document, 56 GEO. WASH. L. REV. 4, 17 (1987) ("When so interpreted, the commerce clause becomes a charter of free trade."). But see Donald H. Regan, The Supreme Court and State Protectionism: Making Sense of the Dormant Commerce Clause, 84 MICH. L. REv. 1091, 1267 (1986) ("[T]alk of the Nation as an economic unit, talk of free trade, and talk of free access to markets may reflect nothing more than vehemence in the condemnation of protectionism."). [Vol. 14:3 WTO NATIONAL TREATMENT without their representation or representation by a surrogate.24 The two views, of course, are not mutually inconsistent,25 and various commentators have attempted to synthesize them in their own analysis.26 However, the distinction between a substantively-based and a process-based rationale is crucial not only for the freedom that it gives states to regulate their local economies, but also for the legitimacy of the enterprise of interfering with local decisions. 27 It is not surprising then that adherents to one rationale or the other continue to dispute their relative merits.28 It may be appropriate to foreshadow some of the doctrinal conclusions of this analysis. First, the aims and effects test is indeed dead. When the Appellate Body refers in its analysis to the purpose of a measure, it is doing so not to distinguish protectionist from non-protectionist purposes on sub- stantive grounds. Instead, it is looking at the measure in a far narrower way- namely, to determine whether the purpose of the particular classification chosen by the regulatory authority was to buy domestic support for the measure by imposing disproportionate costs on foreign producers. 29 Similarly, although national treatment analysis necessarily looks at the degree to which imports are adversely affected by a measure, this is not a substantive effects 24. TRIBE, supra note 19, at 1051 ("[S]tate and local lawmakers are especially susceptible to pressures that may lead them to make decisions harmful to the commercial and other interests of those who are not constituents of their political subdivisions."). See Julian N. Eule, Laying the Dormant Commerce Clause to Rest, 91 YALE L.J. 425 (describing and analyzing the process based approach but recommending that analysis be moved from Commerce Clause jurisprudence to the Privileges and Immunities Clause); Daniel A. Farber & Robert E. Hudec, Free Trade and the Regulatory State: A Gatt's Eye View of the Dormant Commerce Clause, 47 VAND. L. REv. 1401, 1405 (1994) ("Local legislatures may be well suited to weigh the importance of gains in terms of the costs they are willing to pay, but there is no reason to think that they have any capacity to make an honest weighing of the balance between their own gains and the costs to outsiders within the larger community."); Mark V. Tushnet, Rethinking the Dormant Commerce Clause, 1979 Wis. L. REv. 125, 125 (1979) ("[Jludicial displacement of legislative judgment is appropriate when it seems that the legislative process has operated in a distorted way-for example by excluding some affected interest from the legislative process."). The process-based theory is endorsed as the rationale for overseeing state taxation in Ernest J. Brown, The Open Economy: Justice Frankfurter and the Position of the Judiciary, 67 YALE L.J. 219, 229, 232 (1957). 25. See, e.g., CHEMERINSKY supra note 1, at 404 ("These justifications, of course, are not mutually exclusive, but quite consistent."). 26. See, e.g., Tushnet, supra note 24 (combining the notion that the dormant Commerce Clause contains a kind of substantive due process of free trade with the political process theory); Goldsmith & Sykes, supra note 23 (purporting to unify the efficiency and the process justifications for the dormant Commerce Clause). 27. We expand on this point infra section IV. 28. See, e.g., TRIBE, supra note 19, at 1058 ("[A]lthough the Court's Commerce Clause opinions have freely employed the language of economics, the decisions have not interpreted the Constitution as establishing the inviolability of the free market."). But see, e.g., Redish & Nugent, supra note 22, at 613 ("[T]he democratic process model... proves too much. Once we agree that the key factor is the lack of representation in the legislative process, any state regulation affecting the residents of other states ('foreign residents' )--whether discriminatory or not-is rendered suspect."). 29. See infra the discussion of Chile-Alcohol text accompanying note 147. 2004] IND. INT'L & COMP. L. REv. test; this test does not seek to identify the trade-distorting effects of the measure in order to balance the trade-distorting effect against the non-trade purpose of the measure. Instead, an examination of the impact of the measure on foreign producers is an attempt to measure the extent to which foreign producers and their domestic surrogates have been effectively eliminated from the domestic debate about the substantive wisdom of the measure under consideration. Section II of this article explains the surrogate representation rationale that underlies rules against discrimination like those embodied in Article Im. We argue that the WTO's primary function is to allow countries to represent the interests of their producers and exporters when the laws of foreign countries impede those interests, and that this function is important because otherwise those interests might be underrepresented when foreign countries formulate their policies. This is what Gerhart has elsewhere called the participatory vision of the WTO.3° We then point out that the surrogate representation rationale, which is identical to the rationale underlying the dormant Commerce Clause in U.S. constitutional jurisprudence,3 recognizes that interests in the regulating country, including consumers and those domestic producers who will be subject to the regulation, can serve as a proxy for those foreign interests, providing surrogate representation to the foreign interests. When that occurs, the participatory deficit32 that is inherent in a system of territorially bound government can be overcome. One function of the WTO, and specifically of the national treatment provisions, is to insure that the possibility of surrogate representation is not nullified or disarmed in the regulating country. Section III then reviews the Appellate Body's jurisprudence under Article III to show that the surrogate representation rationale is indeed guiding the Appellate Body as it shapes the national treatment provisions. In this discussion, we show how other understandings of national treatment, and particularly those that would look to include expansive tests of purpose or effect of a regulatory measure, are misinterpreting what the Appellate Body is doing. Section IV, the concluding section, summarizes some of the implications of this analysis for our understanding of the role of the WTO and its evolving jurisprudence. 30. Peter M. Gerhart, The Two Constitutional Visions of the World Trade Organization, 24 U. PA. J. INT'L ECON. L. 1, 3 (2003). 31. See TRIBE, supra note 19, at 1057. 32. This deficit is different than the "democratic deficit" that exists between citizens of the world and direct involvement with international organizations. See Gerhart, supra note 30, at 9-11. The participatory deficit is expounded infra text accompanying notes 34-40. [Vol. 14:3 WTO NATIONAL TREATMENT It. THE PARTICIPATORY VISION OF THE WTO AND SURROGATE REPRESENTATION The national treatment interpretation that is advanced here reflects the role of the WTO as an institution of international federalism.33 In that role, the WTO gives participatory rights to adversely affected foreign interests that would otherwise be unrepresented when a country makes its policy. The national treatment provision plays a vital part of that role because it allows the W'TO to oversee the lawmaking processes in member countries to make sure that those processes do not devalue or ignore forces within the country that could give the interests of foreign producers surrogate representation when policy is made. Gerhart has written elsewhere in greater length about the participation- enhancing function of the WTO. 4 Briefly, this function responds to a signifi- cant problem of democratic representation in a globalized, interconnected world. The problem is that even though national lawmaking often has effects outside the country, lawmakers generally have insufficient incentives to take those effects into account because adversely affected people are outside the lawmaking polity.35 When lawmaking has external, transnational effects that 33. See, e.g., Farber & Hudec, supra note 24, at 1404-05. The conventional explanation of the extraordinary legal protection given to free trade policy is that, unlike most other policy measures, trade restrictions cause direct and immediate harm to 'outsiders' who actually are members of the same wider community. External controls are required, the argument goes, because local units will not properly take into account these harms to other community members. In a community consisting of several smaller units of government (a United States consisting of individual states, or a GATT consisting of individual nations), the ultimate question is whether the gain of the regulation for insiders outweighs the harm it causes to outsiders. Local legislators may be well suited to weigh the importance of gains in terms of the costs they are willing to pay, but there is no reason to think that they have any capacity to make an honest weighing of the balance between their own gains and the costs to outsiders within the larger community. Indeed, human experience tells us that, in a democracy, they have every reason not to do an honest job. Id. 34. Gerhart, supra note 30. 35. As has been said in connection with the dormant Commerce Clause: "The checks on which we rely to curb the abuse of legislative power-election and recall-are simply unavailable to those who have no effective voice or vote in the jurisdiction which harms them." TRIBE, supra note 19, at 1052. "The representation-enforcing approach commands judicial intervention where the mechanisms of participatory government have failed to operate, but it also requires deference where no such defect appears." Eule, supra note 24, at 442 (discussing the process-based surrogate representation approach to the dormant Commerce Clause). Analysts of the dormant Commerce Clause identified strands of surrogate representation spread throughout Supreme Court decisions. See generally Gerhart, supra note 30, at 38-48. "[S]tate regulations are rarely struck down for the explicit reason that they are the products of unrepresentative political processes. Rather, this political defect should be seen as underlying the forms of economic discrimination which the Supreme Court has treated as invalidating certain state actions with respect to interstate commerce." TRIBE, supra note 19, at 1057. Discriminatory trade measures appear in two cases. In case one, either there are no surrogates 2004] IND. INT'L & COMP. L. REV. are not adequately given weight in the lawmaking process, crucial aspects of democratic representation are threatened, for democratic principles of participation and accountability posit that all those who are adversely affected by the policy will participate in making the policy.36 The WT'O restores to national law-making a balance of participation and accountability, and thus of democratic acceptability, by restraining national lawmaking that would adversely affect foreign interests without having to take those interests into account. Under this vision of the WTO, the members of the WTO are not imposing substantive values on one another, nor are they giving trade values transcendent weight in public policy. Participation and accountability are not about outcomes or substantive standards, but about processes.37 Naturally, a regulatory decision-maker must take into account, and balance, the interests of competing groups of producers, as well as the interests of consumers and the broader society. When all those with relevant interests are represented in the forum that sets up the regulatory regime, we accept the legitimacy of the regulatory regime as a reflection of the public interest even if we argue against the wisdom of the regulation. Debate about the regulation either accepts its for outside producers inside the regulating polity or there are surrogates inside the regulating polity, but their interests are altered by the enacted measure such that they are no longer viable surrogates. "[Wihen the regulation is of such a character that its burden falls principally upon those without the state, legislative action is not likely to be subjected to those political restraints which are normally exerted on legislation where it affects adversely some interests within the state." S. C. State Highway Dept. v. Barnwell Bros., 303 U.S. 177, 185 (1938). In case two, there are surrogates inside the regulating polity, and they are affected the same as those outside the polity; therefore, the court must let the measure stand. See Minnesota v. Clover Leaf Creamery Co., 449 U.S. 456, 473 (1981) ("'The existence of major in-state interests adversely affected by the Act is a powerful safeguard against legislative abuse."). 36. See DAVID HELD, DEMOCRACY AND THE GLOBAL ORDER 16(1995). Throughout the nineteenth and twentieth centuries theorists of democracy have tended to assume a 'symmetrical' and 'congruent' relationship between political decision-makers and the recipients of political decisions. In fact, symmetry and congruence have often been taken for grated at two crucial points: first, between citizen voters and the decision-makers whom they are in principle able to hold to account; and secondly, between the 'output' (decisions, policies, and so on) of decision-makers and their constituents-ultimately, the 'people' in a delimited territory. Id. See also Markus Krajewski, Democratic Legitimacy and Constitutional Perspectives of WTO Law, 35 J. WORLD TRADE 167, 171-72 (2001). ("[A] decision can be called democratic if those affected by the decision were the participants in the decision-making process... Accordingly, those who have to comply with the decision-or in other words: who are governed by it-have to be the decision-makers.") (citation omitted). 37. See, e.g., ROBERTA. DAHL, ON DEMOCRACY 37 (1998); Jack L. Walker, A Critique of the Elitist Theory of Democracy, 60 AM. POL. Sci. REV. 285, 288 (1966) ("Although the classical theorists accepted the basic framework of Lockean democracy, with its emphasis on limited government, they were not primarily concerned with the policies which might be produced in a democracy; above all else they were concerned with human development, the opportunities which existed in political activity to realize the untapped potential of man."). [Vol. 14:3 WTO NATIONAL TREATMENT legitimacy and focuses on the regulation's substantive wisdom or criticizes the procedural legitimacy of the measure's enactment.38 When adversely affected persons-such as foreigners-are not included in the lawmaking forums, however, the procedural concerns are especially acute. The probability that foreign interests will be ignored or displaced is especially great when member nations are crafting their regulatory regimes. 39 This is so because countries have a natural tendency to buy off domestic opposition to regulatory proposals by imposing cost on foreigners; the domestic industry is likely to have less opposition to the costs of a regulatory regime when the regime imposes disproportionately higher costs on foreign rivals. Indeed, Ralph Nader, for one, has argued that imposing costs on foreign rivals is an important aspect of the regulatory state.' The national treatment provision, like its counterpart in the dormant Commerce Clause doctrine of the U.S. Constitution, is designed to oversee the political process in member countries to insure that the interests of foreigners are not denigrated or ignored. This is the basis for the participatory, process- based "representation reinforcement ' 41 rationale for the external supervision of state political processes that underlies the dormant Commerce Clause, and, we believe, the WTO's national treatment provision. The rationale has, however, been misinterpreted, for it does not invalidate state legislation merely because foreign interests are not represented in state lawmaking forums, as some have mistakenly thought.42 It is not the "inherently limited constituency" 43 of national lawmaking by itself that creates the problem. Such a basis for invalidating regulation would, as the critics maintain, be too broad a principle, invalidating regulatory measures merely because foreign interests were adversely affected. The rationale behind the surrogate representation understanding of the national treatment provision is to oversee state law- making processes to determine when the process has in fact co-opted those political forces that would otherwise provide surrogate representation for foreign interests. 38. See, e.g., Gerhart, supra note 30, at 27-33. Under public choice theory, commentators sometimes seek to question the substantive wisdom of aregulation byquestioning its procedural legitimacy. Because those efforts frequently rest on precarious assumptions about how voters define the public interest, they are rarely successful in our view. 39. TRIBE, supra note 19, at 1051-52. "[T]he Court's rigorous tests.., underscore the recognition implicit in the Commerce Clause that state and local lawmakers are especially susceptible to pressures that may lead them to make decisions harmful to the commercial and other interests of who are not constituents of their political subdivisions." Id. 40. Ralph Nader, Statement at the Uruguay Round Trade Negotiations, Hearings before the Senate Committee of Finance 240, 252 (Mar. 16, 1994) (claiming that domestic laws such as laws on the export of raw logs are necessary to buy the loyalty of domestic industry in exchange for accepting conservation limits on logging). 41. TRIBE, supra note 19, at 1054. 42. See, e.g., Goldsmith & Sykes, supra note 23, at 795-96; Redish & Nugent, supra note 22, at 614-15. 43. TRIBE, supra note 19, at 1052. 2004] IND. INT'L & COMP. L. REv. To see that point, we must recognize that foreign interests are not necessarily under-represented in national lawmaking processes. Surrogates- that is, people within the national polity who share the interest of foreigners and who will represent those interests when the regulatory framework is set up-represent foreign interests. In general, foreign producers have two sets of domestic proxies when domestic regulators consider the scope and form of the regulation. First, domestic consumers represent the interests of foreign producers; when foreign producers offer reasonable substitutes to domestic products, the interests of domestic consumers and foreign producers are symmetrical and identical.' Consumers seek to generate consumer surplus by finding better goods at cheaper prices. When they do, the sales generate producer surplus for those producers who are able to supply the goods that generate the most consumer surplus. When no barriers to exchange exist, consumers tell us when certain foreign products compete with domestic products. In their search for better products at lower prices, consumers naturally represent the legitimate interests of producers anywhere in the world.45 Trade barriers, on the other hand, make it difficult for consumers to recognize, and therefore to represent, the interests of foreign producers. Admittedly, consumers will not be perfect proxies for the interests of foreign producers. Consumers face well-known collective action problems that make it difficult to represent their own interest, let alone the interest of foreign producers. When consumer interests are small and dispersed, con- sumers will have trouble organizing. 46 We should not, however, over- emphasize the collective action problems of consumers. Often "consumers" are not the ultimate consumers of goods. Instead, "consumers" tend to be large manufacturers or retailers who depend on foreign sources of supply. Additionally, even for less powerful groups of consumers, advances in communications and the rise of consumer advocacy have helped overcome the collective action problems.47 44. See id. at 1955 for a discussion on potential consumer surrogacy in the context of the dormant Commerce Clause. See also Tushnet, supra note 24, at 133, 138-39. 45. John 0. McGinnis & Mark L. Movsesian, The World Trade Constitution, 114 HARV. L. REV. 511, 572-89 (2000). In suggesting their own version of a process-oriented test for determining the existence of hidden protectionism, Professors John McGinnis and Mark Movsesian develop a test that capitalizes on a flipped notion of surrogate representation, emphasizing the importance of foreign producers representing the interests of domestic consumers in the domestic regulatory process. Id. McGinnis and Movsesian suggest that a transparency requirement would allow affected industries to comment on regulations. Id. at 573. These industries, then, would represent the diffuse consumer groups who would benefit from a lack of regulation, but are not well represented in the regulatory process. Id. at 574-75. Also, if a regulation places burdens on the domestic industry as well, "it gives foreign producers some virtual representation in the domestic political processes that lead to the regulation and provides some assurance that the regulation is not discriminatory." Id. at 574. 46. See generally MANCUR OLSEN, THE LOGIC OF COLLECTIVE ACTION (1965). 47. See Robert V. Percival, Environmental Legislation and the Problem of Collective Action, 9 DuKE ENVTL. L. & POL'Y F. 9, 19 (1998) (stating the Environmental Defense Fund uses the internet and "latest communications technology to rally public support for their causes."). See also Peter H. Schuck, Against (and for) Madison: an Essay in Praise of Factions, [Vol. 14:3 WTO NATIONAL TREATMENT But the major point is not that consumers are always good surrogates for foreign producers. The point is that they can be, and when they are, this surrogacy is worth protecting. A second group that provides surrogate representation for foreign producers consists of domestic producers who seek to resist regulation that they feel is too costly or burdensome. Although domestic producers and foreign producers often have antagonistic competitive interests, when they are similarly situated from a regulatory standpoint, they share a common interest in reducing the adverse effects of regulation. Moreover, even when national regulation affects producers differently, those domestic producers who are in the same position as foreign producers will represent the interests of the foreign producers, even if consumers are neutral concerning the outcome of the regulatory struggle. Consider a hypothetical case used by Professor Regan.48 Imagine that a country is deciding whether to impose a tax on producers of plastic containers (but not cardboard containers) in the belief that plastic containers (but not cardboard containers) damage the environment. This regulation would benefit the makers of cardboard containers, because it would put them at a competitive advantage and would disadvantage the makers of plastic containers. Even though foreign makers of plastic containers are outside the lawmaking jurisdiction, the domestic makers of plastic containers, if they are numerically strong enough and able to organize, can adequately represent the foreign interests. Because their interests are identical, the domestic group can represent the foreign interest if the circumstances are right. Such surrogate representation-by either consumers or domestic pro- ducers with similar interests-is an important mechanism by which the democratic principles of participation and accountability are advanced in a world where policymaking is territorially confined and decentralized. As a result, the WTO has a vital role to play in making sure that members do not interfere with the mechanism of surrogate representation. When foreign interests are effectively represented through consumer or producer surrogates within the country undertaking the regulation, their representation effectively 15 YALE L. & POL. REv. 553,566-67 (1997) (noting success of public advocacy groups despite public choice theory). 48. Regan, supra note 3, at 447. See Minnesota v. Clover Leaf Creamery Co., 449 U.S. 456 (1981). The facts in Professor Regan's example appear to be drawn from Clover Leaf Creamery, where the regulatory measure was upheld. Id. In that case, Minnesota banned "the retail sale of milk in plastic non-returnable, non-refillable containers," while allowing the sale of milk in other such containers, like paperboard cartons. Id. at 458. See also TRIBE, supra note 19, at 1054. The pulp-wood industry within Minnesota received a benefit from this measure because its sales increased. TRIBE, supra note 19, at 1054. Also, all producers of plastic resins (who were disadvantaged by the regulation) resided outside the state. Clover Leaf Creamery Co., 449 U.S. at 473. In the course of its decision, the Supreme Court claimed there were adequate surrogates within the state to represent the non-resident interests. Id. Although the claim of adequate surrogacy may have been incorrect, see TRIBE, supra note 19, at 1055, the theory nonetheless buttressed the Court's decision that Minnesota had not violated the dormant Commerce Clause. Id. at 1054. 2004] IND. INT'L & COMP. L. REV. reduces the deficit in participatory lawmaking that would otherwise occur because foreigners are not present in the lawmaking jurisdiction. Although foreigners' voices are not heard, their interests are, and often effectively. Preserving those mechanisms of surrogate representation from domestic legislative interference becomes an important role for the national treatment provision to play, one which helps to knit otherwise parochial lawmaking units together in a federal system. Often, of course, when domestic proxies for foreigners do not exist, or when their representation is inadequate, no effective surrogate representation can make up for the exclusion of foreigners from the domestic lawmaking process.49 Consider first the situation in which foreign interests are un(under)represented domestically. Taking the plastic/cardboard container example, if all makers of plastic containers were foreigners, and if no indus- tries inside the country relied on use of plastic containers in their business, then the regulation would not adversely affect any domestic producer and domestic producers could not represent foreign producer interests. A regulation taxing or banning the sale of plastic containers might be in the public interest, but the public interest would be determined without having the views or information of the makers of plastic containers represented in the policy debate. Only consumers would represent the interests of the makers of plastic containers, and their interests would be torn between their interests as consumers in cheaper products and their interests as citizens in a cleaner environment. Under these circumstances, the regulation of plastic containers may threaten the participatory principle that those adversely affected by the regulation should be able to participate in the debate about whether the regulation should be imposed. Next, consider the case where the domestic proxies represent foreign interests but the representation is inadequate. The concept of "inadequate representation" must be carefully delineated, of course. We cannot assess the quality of surrogate representation in some abstract way by trying to evaluate the quality of the arguments or the effectiveness of the surrogate's communica- tions. Nor can we evaluate the adequacy of surrogate representation by evaluating the results of the regulatory lawmaking, for that would effectively be a review to see who "should" prevail, and that would be akin to reviewing the substantive merits of the regulation. However, the notion of "inadequate representation" can be sensibly understood in a non-substantive way by focusing on the objective ways in which consumers and similarly situated domestic producers may be inadequate proxies for foreign producers. 49. That is why the disproportionate impact of a regulation is relevant to its validity. As Justice Stone said in South Carolina State Highway Dep't v. Barnwell Bros., 303 U.S. 177, 185 (1938): "[W]hen the regulation is of such a character that its burden falls principally upon those without the state, legislative action is not likely to be subjected to those political restraints which are normally exerted on legislation where it affects adversely some interests within the state." Conversely, "the fact that [the regulations] affect alike shippers in interstate and intrastate commerce in large numbers within as well as without the state is a safeguard against their abuse." Id. at 187. See also Southern Pacific Co. v. Arizona, 325 U.S. 761, 767 (1945). [Vol. 14:3 WTO NATIONAL TREATMENT As we have already alluded to, consumers may be ineffective surrogates because of the problem of organizing or because their interests are not purely commercial. Domestic producers may be inadequate proxies for foreign pro- ducers because they are too few in number to have a meaningful voice.0 More to the point, even if they are numerically sufficient, domestic producers may be inadequate proxies for similarly situated foreign interests because, as we have seen, domestic proxies can so easily be "bought off' within the context of the regulatory decision-making by providing the domestic surrogate some competitive advantage over otherwise similarly situated foreign producers." An example of this occurred in U.S.-Gasoline.2 The U.S. Clean Air Act of 1990 required that pollutants in gasoline meet certain requirements in relation to 1990 gasoline "baselines."53 Domestic refiners had three possible methods of determining their 1990 baseline, but foreign refiners had only one method to determine their baseline, 4 and if a foreign refiner could not use that method, it had to use a statutory method. Under this system, even when imported gasoline was chemically identical to domestic gasoline, foreign but not domestic producers would be forced to further clean their gasoline in order to remain in compliance with EPA standards under the Act.56 Foreign refiners would then have to make "cost and price allowances because of their need to import other gasoline with which the batch could be averaged so as to meet the 50. See Clover Leaf Creamery Co., 449 U.S. at 458. This appears to have been the situation in Clover Leaf Creamery. Although there were no producers of plastic resins in Minnesota, other groups adversely affected by the ban might have represented their interests. Id. Looking at the plaintiffs in the case suggest who the surrogates were, and they included "a Minnesota dairy that owns equipment for producing plastic non-returnable milk jugs, a Minnesota dairy that leases such equipment .... a Minnesota company that produces plastic non-returnable milk jugs .... [and] a Minnesota milk retailer ....... Id. Although the court found these to be a safeguard against abuse, the strength of these surrogates may have been overstated. TRIBE, supra note 19, at 1055. 51. Similarly, Mark Tushnet has pointed out the danger that logrolling within a state may mean that legislators systematically protect in-state interests from out-of-state competition. Tushnet, supra note 24, at 137. 52. WTO Appellate Body Report on United States-Standards for Reformulated and Conventional Gasoline, WT/DS2/AB/R (Apr. 29, 1996) [hereinafter U.S.-Gasoline (AB)]. Other cases in which the Appellate Body has struck down regulatory measures because they imposed disproportionate costs on foreigners are: WTO Appellate Body Report on Korea-Measures Affecting Imports of Fresh, Chilled and Frozen Beef, WT/DS161/AB/R, WT/DS 169/AB/R (Dec. 11, 2000) [hereinafter Korean-Beef (AB)]; and WTO Appellate Body Report on Turkey-Restrictions on Imports of Textile and Clothing Products, WT/DS34/AB/R, 39 I.L.M. 159 (Oct. 22, 1999) available at http://www.wto.org (last visited Feb. 16, 2004) [hereinafter Turkey-Textiles]. The cost-shifting aspects of these cases are discussed in Gerhart, supra note 30, at 56-61. 53. U.S.-Gasoline (AB), supra note 52, at 5. Pollutants in reformulated gasoline had to be reduced, while pollutants in conventional gasoline could remain but not go higher than 1990 levels. Id. at 4-5. 54. WTO Panel Report on United States-Standards for Reformulated and Conventional Gasoline, WT/DS2/R, para. 6.2-6.3 (Jan. 29, 1996) [hereinafter U.S.-Gasoline (Panel)]. 55. Id. para. 6.4. 56. Id. para. 6.10. 2004] IND. INT'L & CoMP. L. REV. statutory baseline., 57 As the Appellate Body said in striking down this discrimination, "to explore adequately [alternative means of achieving its goal of clean air] means ... to count the costs for foreign refiners that would result from the imposition of statutory baselines [as the United States had for domestic refiners]."" One can see the surrogate representation model at work in this case. Normally domestic refiners serve as surrogates for the foreign refiners because they have the same interests in the marketplace. However, because domestic refiners gained an advantage in the marketplace over foreign refiners, domestic refiners were less likely to represent the foreigners in the regulatory bodies. They were bought off, and therefore, altered the normal surrogacy foreign refiners would have enjoyed. Because the foreign interests affected by the measure were not represented in the domestic forum, a process failure occurred; and the regulatory scheme could not survive scrutiny under Article XX. This is the broader lesson of the U.S.-Gasoline case. Even if foreign and domestic interests are perfectly aligned initially, the regulatory process can change that alignment by driving a regulatory wedge between domestic and foreign producers. 9 If the regulation imposes disproportionate costs on foreign producers-even similarly situated ones-the domestic producers will no longer act as proxies for the foreign producers. Or if the regulatory scheme gives benefits to domestic producers that are not given to foreign producers, the proxy relationship that should have protected the interests of foreign producers would break down. Domestic producers would no longer be able to adequately represent the foreign interest because they would get a benefit of the regulatory regime not given to the foreigners. When we examine the decisions of the Appellate Body in the next section we will see further examples of ways in which the legislative process can drive a wedge between the interests of foreign and domestic producers. This shows the essence of the surrogate representation rational. When the legislative process has been shown to interfere with the process by which foreign interests can be represented in national lawmaking forums by national surrogates, the legislation is procedurally objectionable and ought not to stand. 6° 57. Id. 58. U.S.-Gasoline (AB), supra note 52, at 27. 59. Professor Tushnet has the most extended discussion of this phenomenon, noting both the possibilities of buying the loyalty of domestic interests, see Tushnet, supra note 24, at 132, and the limits of this kind of analysis. Id. at 140. 60. It may also be helpful to recast the basic surrogate representation argument in somewhat different terms in order to illustrate its breadth. Under the analysis given here, the problem of tariffs is not just that tariffs are economically inefficient. As Gerhart has argued in his earlier work, in terms of participatory democracy, tariffs impose a cost on foreigners under circumstances where foreign producers cannot participate effectively in the decision-making process. See Gerhart, supra note 30, at 21-25. A related point is relevant to an analysis of national regulation under Article Ill. Tariffs allow domestic regulatory policy to be made under circumstances in which we can no longer depend on consumer interests to act as a proxy for [Vol. 14:3 WTO NATIONAL TREATMENT Given the importance of the right of participation to promoting harmony among nations and the importance of surrogate representation in affirming participation, the WTO role and the role of the national treatment provision are clear. The WTO review of domestic regulation under Article ImI must be oriented to uncover those situations in which domestic proxies for foreign interests are either non-existent or have been compromised in some way. When this occurs, national regulation has disrupted the mechanism by which domestic proxies will represent the interests of foreign producers, and the WTO has a legitimate function in either invalidating the regulation on that ground or at least making sure that the regulatory regime is supported by a valid justification under Article XX (which, incidentally, also depends on protecting the interests of foreign producers not to be excluded from a market without some effective participation in the decision).61 Several aspects of this approach to the national treatment provision are attractive. First, this approach says that the national treatment provision is not concerned about differential treatment of imported products in the abstract, or in comparing that impact with the regulatory goals of the measure. Instead, it is concerned with differential treatment that is proven to result when the surrogate representation by domestic producers that should protect foreign interests has been compromised. This interpretation avoids the clash between the domestic values that the regulation seeks to achieve and the trade effects of the regulation, and gives foreign interests no greater power to overturn regulatory measures than domestic interests have.62 If the foreign interests are foreign producer interests. Tariffs prohibit consumers from gaining the surplus available from foreign production, thus driving an economic and political wedge between consumers and foreign producers. Theoretically, consumers should still have an interest in foreign production, but because of high search costs, it may be difficult for consumers to recognize this. Because tariffs eliminate a portion of foreign production from consumer's choice set, governments that have imposed tariffs have removed any incentive consumers would otherwise have to argue against regulation that adversely affects foreign producers. This problem is not necessarily ameliorated when the tariffs come down because the lingering effects of the tariffs would continue to make it difficult for consumers to recognize and understand their options. For some time, information costs would still be high and marketing and delivery channels from foreign producers would still have to be constructed. Lawmaking in this atmosphere might still take place in a situation where consumers could not act as effective surrogates for the interests of foreign producers because they would not be able to understand their own options. 61. See Gerhart, supra note 30, at 66-70. "[In the landmark Shrimp-Turtles decision, the Appellate Body made the procedural rights of foreigners the touchstone for the application of the general exceptions of Article XX of GATT." Id. at 66. In that case, the Appellate Body required the United States to negotiate in good faith and non-discriminatorily and required transparent and predictable processes in the administration of regulations. Id. at 69. 62. Analysts who believe that the dormant Commerce Clause contains the substantive value of free trade seem to be confusing the power given to the U.S. Congress with the power denied to the states. Without a doubt, the Congress was given power over interstate commerce in order to protect the common market of the United States from state or private interference. But that does not make economic efficiency a Constitutional value; it only operates to confer on Congress the power to take efficiency values into account when Congress exercises its powers. Moreover, this does not imply a limitation on the regulatory authority of the states; as 2004] IND. INT'L & COMP. L. REv. adequately represented in the policymaking forum but are overridden by other policy considerations, the WTO has no authority to question the decision. It is only when there is evidence that the foreign interests are not represented that WTO intervention is warranted. Moreover, the surrogate representation understanding of the national treatment principle is also consistent with the role of governments in regulating markets. The attraction of well functioning markets is not merely that they improve economic efficiency but that they allow the consumers of a country to represent foreign interests. In well functioning markets, if foreign interests are not successful it is because consumers have decided that the foreign products do not meet consumers' criteria of selection. But in markets free of restrictions on trade, consumer purchases of foreign products indicate the existence of foreign interests in having access to the market and therefore in how the market is regulated. Markets do not allow discrimination against foreigners unless different treatment is justified by consumer choice.63 was just made clear the Commerce Clause is not a value-laden provision but only an empowering provision. The true relationship between the free market in the United States and Constitutional restrictions on state power is just the opposite of what those who espouse efficiency content for the Commerce Clause believe it to be. Because Congress has allowed interstate commerce to flourish, the instances in which state actors are called on to be surrogates for out-of-state actors has grown, thus making it more important than ever to invoke the dormant Commerce Clause to strike down state legislation. The important role the dormant Commerce Clause follows from increasing economic interdependence, but it does not cause that interdependence. Even the case that has come to symbolize the efficiency-based view of the dormant Commerce Clause, H.P. Hood and Sons v. DuMond, 336 U.S. 525 (1949), makes this analysis clear. The most quoted part of that opinion is: Our system fostered by the Commerce Clause, is that every farmer and every craftsman shall be encouraged to produce by the certainty that he will have free access to every market in the Nation that no embargoes will withhold his exports, and no foreign state will by customs duties or regulations exclude them. Likewise, every consumer may look to the free competition from every producing area in the Nation to protect him from exploitation by any. Such was the vision of the Founders; such has been the doctrine of this Court which has given it reality. Id. at 539. Even aside from the fact that this quote refers to the system "fostered" by the Commerce Clause rather than the system "commanded" by the Commerce Clause, this quote follows language that more nearly captures the process based rationale of the dormant Commerce Clause. In particular Justice Jackson noted "the established interdependence of the states only emphasizes the necessity of protecting interstate movement of goods against local burdens and repressions." Id. at 538. In other words, it is economic integration that leads to the need to police local burdens and repressions, not the policing of local burdens and repressions that leads to economic integration. Moreover, it is repressions-and presumably repression of political interests-that is the focus of the prohibition. 63. Consumers may, of course, be prejudiced against foreign goods in a way that leads to less favorable treatment of otherwise "like" goods. As long as we endorse consumer sovereignty and the market mechanism, however, we must be willing to say that consumer decisions are final (in the absence of a market failure) and that the ignorance or prejudice of consumers can be overcome only by education and more knowledge, not by government action at the national or international level. In situations where a potential competitive relationship exists but consumers fail to take advantage of that relationship we can ask governments to take [Vol. 14:3 WTO NATIONAL TREATMENT Consumers in well-functioning markets are the best authority to tell us whether foreign producers have an interest in the market that needs to be protected when the government that regulates the market is determining the scope and nature of its regulatory program. Sometimes, of course, the government needs to intervene in markets to carry out important government functions-to overcome market failure and to raise revenue, for example. Under the interpretation offered here, the goal of Article Im is to make sure that, during these interventions, the interests of foreigners are represented in the same way that the interests of domestic producers are represented. Where the interests of the domestic industry and the foreign industry are identical, the foreign industry is represented by the domestic industry. So if the burden of any regulation is distributed evenly over the producer population, the domestic industry and the foreign industry interests are aligned and domestic producers can represent foreign producers. When surrogate representation is preserved, government intervention in markets is substantively sound and preserves the role of consumers as the moving force behind economic decisions. Before moving on to see how the Appellate Body has built its inter- pretation around the surrogate representation rationale, we can profitably address two possible objections to the rationale. Superficially, one might object that because some members of the WTO are not functioning democracies in the Western model, it would be wrong to presume that some participatory ideal or vision underlies the WTO' s work. But a moment's thought will convince us that such an objection is misplaced. In the first place, the WTO is the successor organization to GATT and GATT started as an organization driven primarily by the Western democracies. 6 It is quite plausible to believe that the "founding" countries were influenced by the need to provide a forum by which one country could object to the policies adopted by other countries that adversely affected their export producers,65 no action that facilitates or augments that prejudice, but cannot expect governments to compensate for that prejudice. 64. "Although the GATIT was not formed at the 1944 Bretton Woods Conference, nevertheless the Bretton Woods Conference contemplated the necessity of an International Trade Organization." John H. Jackson, THE WORLD TRADING SYSTEM 27-28 (1989) (considering GATT as part of the Bretton Woods System). See also BHALA & KENNEDY, supra note 5, at 1-3 (1998). The GATT is actually a by-product of a failed effort to create the International Trade Organization (ITO), through the Havana Charter. Id. at 2. The Prepatory Committee that worked on the Havana Charter had representatives from: Australia, Belgium, Luxembourg, Brazil, Canada, Chile, China, Cuba, Czechoslovakia, France, India, Lebanon, the Netherlands, New Zealand, Norway, South Africa, the USSR, the United Kingdom, and the United States. Id. at 1. The USSR was the only member that did not become a contracting party to the GATT 1947. Id. 65. See Peter M. Gerhart, WTO History Reexamined: The Participatory Vision (forthcoming). See also THOMAS ZEILER, FREE TRADE, FREE WORLD: THE ADVENT OF GATT (1999) (confirming that GATT was motivated by assumption that cooperation on trade would lead to cooperation on political issues), CATHERINE BARBIERI, THE LIBERAL ILLUSION: DOES FREE TRADE PROMOTE PEACE? (2002) (testing political hypothesis animating GATT, that interconnected economies foster peace). 2004] IND. INT'L & COMP. L. REV. and, therefore, that the animating motivation for the national treatment provision was shaped by the participatory vision of the WTO. Moreover, we should take note of Ann-Marie Slaughter's reminder that the Bretton Woods institutions, including GATT, were designed to allow transnational regula- tion.' A system set up to enhance international regulatory law (in order to overcome international market failures) is not likely to impose stringent substantive limitations on national regulation designed to overcome market failures. A second objection to the surrogate representation rationale, one carefully articulated by Professor Regan, is that under any circumstances consumers in the country adopting the regulatory measure will provide positive surrogate representation for foreign producers. Under this view, because the surrogate representation rationale is superfluous, it cannot provide a theoretical basis for understanding federalist legal restraints on regulatory activity. Professor Regan's view is that as long as the regulation is not protectionist, we can be sure that when a regulatory body protects all local interests it will simultaneously protect all foreign interests. Accordingly: If the legislature adopts legislation that optimizes with respect to all affected in-state interests, then the overall result will be efficient with respect to all interests, local and foreign. I shall refer to this property of our examples as "local/global equiva- lence." To say that a sort of regulation exhibits "local/global equivalence" is to say that if a regulation of that sort opti- mizes "locally" (over all in-state interests) it will necessarily optimize "globally" (it will lead to an outcome that is efficient with respect to all interests, local and foreign.67 In a nutshell, local/global equivalence-where it exists-completely undercuts the virtual representation argument.6" In this view, the function of federalist review of regulatory measures is to determine whether the local political process has served local interests. If it has, then it has also served foreign (outside) interests; if it has not served local interests, then it should be struck down for that reason (although doing so incidentally protects foreign interests). This view essentially equates the service of local interests with non-protectionism, and protectionism with the non-service of local interests. And because it equates the protection of local 66. Ann-Marie Burley Slaughter, Regulating the World: Multilateralism, International Law, and the Projection of the New Deal Regulatory State, in MULTLATERALISM MATrERS: THE THEORY AND PRAXIS OF AN INSTITUTIONAL FORM 125 (John Ruggle ed., 1993). 67. Donald H. Regan, Judicial Review of Member-State Regulation of Trade Within a Federal or Quasi-Federal System: Protectionism and Balancing Da Capo, 93 MICH. L. REV. 1853, 1859-60 (2001) (footnote omitted) [hereinafter Regan, Judicial Review]. 68. Id. This view then provides a crucial argument in his analysis of the application of the national treatment standard by the WTO. See Regan, Regulatory Purpose, supra note 3, at 452. [Vol. 14:3 WTO NATIONAL TREATMENT interests with an appropriate local process-that is, with one that is not captured by special interests-it is focused exclusively on whether special interests have captured the local legislative process. As Regan argues: Protectionist legislation normally does not optimize over all local interests. It normally does result from a failure of the political process with respect to local interests. Protectionist legislation standardly results from local producer interests wielding excessive power in the political process, which allows them to distort disorganized consumer interests. So, in any case where there is a significant suggestion of protec- tionism, it is appropriate for the court to consider whether the political process has gone awry in its treatment of local interests. But if the answer is no (if the law is not protec- tionist), there is no justification for balancing to protect foreign interests.69 There is much in Regan' s analysis that turns out to be congruent with the surrogate representation analysis that we present here. Because Regan recognizes that local interests can represent outside interests, Regan is in effect endorsing the premise that surrogate representation is an important feature of local regulatory measures. We agree that where the local/global equivalence holds, there is no reason to intervene to overturn regulatory measures. Where we part company with Regan however, is in how we define whether the local/global equivalence holds. Regan equates protectionism with the absence of the local/global (or surrogate representation) identity and then defines the presence or absence of the local/global identity in terms of local capture by special interests. His motivation for doing this is to counter the notion that review of state (or national) regulatory measures should involve a balancing of in-state and out-of-state interests, and thus a weighing of competing interests. In his view, the only issue should be whether there is a legitimate purpose behind the statute, and that can be determined by assessing whether the process has been captured by special interests. This attempt to equate special interests with parochial interests and determine the presence of special interests by looking at regulatory purpose is ingenious, but ultimately inappropriate for the WTO. By equating protectionism with "capture by special interests," Regan is unduly narrowing the scope of federalist review of domestic measures. Here, Regan is falling into a trap that is endemic to much of the dormant Commerce Clause literature-the assumption that the anti-protectionist thrust of the dormant Commerce Clause can by equated with review to avoid "capture by special interests." In fact, the protectionism that is invalidated under the dor- 69. Regan, Judicial Review, supra note 68, at 1861. 2004] IND. INT'L & COMP. L. REV. mant Commerce Clause is far broader than simple "special interest capture legislation." Domestic regulatory measures may be protectionist not just because special interests capture the regulatory apparatus but, in a wider sense, because they are parochial. That is, domestic regulation may systematically ignore the impact of regulation on foreign producers and therefore result in regulation that is procedurally invalid. To see this, assume that consumers want to regulate plastic containers for environmental reasons and pulpwood producers want to regulate plastic containers to suppress competitive alternatives. Legislation that results from the confluence of these interests can hardly be called special interest legislation because consumers are seeking to represent their own interests, not those of pulpwood producers. Yet consumers in that situation can hardly be thought to represent the interests of out-of-state producers of plastic containers. This is precisely the situation where some oversight of the legislative process to protect the interests of out-of-state producers would be called for; a situation where both consumers and producers are acting parochially because they do not represent the interests of out-of-state producers. The aim of the dormant Commerce Clause analysis, and, correspondingly, national treatment analysis, is not special interests but parochial interests. Regan's statement that any regulatory body that takes into account all local interests will also take into account out-of-state interests is flawed because it is based on the view that consumers care only about efficient laws and, as a result, consumers will lobby against regulation that is inefficient. This view is apparently based on the assumption that when it comes to policymaking, consumers will acts as consumers and vote for policy that is in their economic self-interest. Under this view, if there is no efficiency- motivated reason for regulation, then consumers adequately represent the interests of foreign producers and can act as good surrogate representatives for the foreign producers. On the other hand, if the regulation in question is itself efficiency enhancing-because it addresses an important market failure-then the regulation has a non-protectionist purpose and is not protectionist. In the latter case-where regulation is needed to overcome a market failure-the consumer may not be a good surrogate for foreign manufacturers (because the regulation will adversely affect foreign manufacturers), but the consumer is a good surrogate for a non-protectionist interest (because the regulation is needed not for protectionism but to increase market efficiency). In this way, the surrogate representation rationale is superfluous. If there is a good purpose for the regulation (that is, an efficiency-enhancing purpose) it is not pro- tectionist, and if it is protectionist, we can tell from that conclusion that foreign producers (like domestic consumers) have been undercut by special interests. One problem with this analysis is that it assumes we can identify purpose and use that analysis as the fulcrum on which to base our finding or illegality. Although Professor Regan' s discussion of this difficulty is quite sophisticated, [Vol. 14:3 WTO NATIONAL TREATMENT many have not been persuaded that distinguishing protectionist from non- protectionist purpose is easily done.7° A more fundamental objection to this analysis, however, is that it assumes consumers as voters are interested only in efficiency; will represent the interests of foreign producers when that is the most efficient interest; and the interest underlying the regulatory measure when that is the most efficient interest. The equation of the consumer-as-voter interest with the efficient interest is, of course, erroneous. When determining their positions on public policy, it is just as likely as not that voters will ignore their personal interest in efficient outcomes and advocate instead for non-efficient outcomes.71 Voters often advocate policy not on the basis of their narrow economic in- terest, but on the basis of non-economic values that might underlie the regula- tory measure. Consumers, to be sure, are self-interested in their commercial dealings, but can act as citizens when it comes to public policy matters. 70. See Regan, Regulatory Purpose, supra note 3, at 458-64 (discussing objections to a tribunal's ability to identify regulatory purpose). Regan stipulates that tribunals are not to look into the collective mind of a legislature, but rather look for "what political forces are responsible for the measure under review." Id. at 459. Regan suggests the Appellate Body could create a rebuttable presumption that a regulation is non-protectionist if there is a plausible non- protectionist purpose for the regulation. Id. at 459-60. While objective evidence is important to rebutting the presumption, Regan also points to ministerial statements (of the kind discussed in Canada-Periodicals, supra note 10) as an example of the type of evidence that could refute the presumption of non-protectionist purposes. Id. at 459. Objective evidence, offered by the complaining country, will often be enough to shift to the defendant country the burden of going forward with the evidence, usually by asserting a non-protectionist regulatory justification. On the other hand, if there is relevant "subjective" evidence in the form of ministerial statements, or legislative committee reports, or whatever, the tribunal should consider that too .... remembering always that even such "subjective" evidence is still just evidence. Id. at 460. See also Regan, Judicial Review, supra note 68, at 1890-94 (discussing inquiring into legislative purpose in the dormant Commerce Clause context). Choi lists several problems with determining legislative purpose. First, there are often many reasons for a certain piece of legislation, and determining which one(s) should be used for Article 111 is a difficult task. Choi, supra note 6, at 119 (citing WTO Panel Report on Japan-Taxes on Alcoholic Beverages, WT/DS8/R, WT/DS 1O/R, WT/DS 1 l/R, para. 6.16 (July 11, 1996) [hereinafter Japan-Alcohol (Panel)]). Second, the complete legislative history of a regulation may be impossible to access, and "could be manipulated by both proponents and opponents of the legislation." Id. at 119 (citing Japan-Alcohol (Panel), para. 6.16). Third, Choi suggests difficulties relating to deter- mining how to value "prepatory work" and circumstances surrounding the regulation. Id. at 119 (citing the Vienna Convention on the Law of Treaties as an example of how "supplementary means" might be handled by a tribunal). Although not willing to concede that determinations of regulatory purpose cannot be successfully accomplished by panels and the Appellate Body, Tsai states, "The need for research and study into this area of establishing the proper standards for evaluating regulatory aim is indeed extensive." Tsai, supra note 7, at 58. 71. See Gerhart, supra note 30, at 27-33. 20041 IND. INT'L & COMP. L. REv. III. THE APPELLATE BODY DECISIONS The Appellate Body has crafted an analytical understanding of the Article III-XX combination that fully reflects the participation-enhancing role of the WTO.72 This jurisprudence provides a coherent set of tests under Article DI that can be explained only by the surrogate representation rationale.73 For analytical purposes, Article Ill contemplates two parallel, though slightly distinct, inquiries for two subjects it regulates: tax regulation and non- tax regulations. The first inquiry seeks to identify the universe of relevant products-the "like" product inquiry, in Article m:4 (applicable to non-tax regulations) and the "like" or "directly competitive or substitutable" product test in Article 111:2 (applicable to taxes). The second general inquiry in both Article m1:2 and Article 111:4 is a "less favorable treatment" inquiry. For Article 111:4 the measure must treat imports no less favorably than domestic goods. In Article 1I:2 the taxes on imports must not exceed taxes on domestic products (if the products are like) or "not similarly taxed" and "applied so as to afford protection" (if the pro- ducts are directly competitive or substitutable). By examining the like product and less favorable treatment standards sequentially, we can see how they together demonstrate the surrogate representation rationale underlying the Appellate Body's interpretation of the national treatment standards. M. A. THE LIKE PRODUCT ANALYSIS The test for determining whether imported products are either "like" or "directly competitive or substitutable" fully reflects the surrogate representa- tion rationale. The basic inquiry concerns the competitive relationship between foreign and domestic products, which is tantamount to an inquiry to determine whether the imported goods are sufficiently competitive with domestic products that consumers can serve as surrogates for the interests of foreign producers. The competitive relationship test stems from the Border Tax Adjust- ments74 case as incorporated into WTO jurisprudence and interpreted in 72. Under other points of view that seek an inquiry into regulatory purpose, the case law appears inconsistent. See VERHOOSEL, supra note 5, at 52 ("[A] number of egregious inconsistencies can be observed in the current case law defining the interface between WTO law and domestic regulation."). 73. This analysis, therefore, responds to the criticisms of those who argue that the Appellate Body case law appears to be inconsistent. See VERHOOSEL, supra note 5, at 52. In our view, that criticism is flawed because it seeks to understand the national treatment provision in terms of substantive law. 74. Report of the Working Party, Dec. 2, 1970, GATT B.I.S.D. (18th Supp.) at § 18S/97- 109 (1972). [Vol. 14:3 WTO NATIONAL TREATMENT Japan-Alcohol.5 Several criteria determine whether imported products are "like" domestic products: the product's end-uses in a given market, con- sumers' tastes and habits, the physical properties of the products, and common tariff classifications. 76 As the panel in Japan-Alcohol declared: "[T]he wording [of Article I and of the Interpretative Note ad Article III] makes it clear that the appropriate test to define whether two products are 'like' . .. is the marketplace. 77 It is understood that the word "like" need not be applied in the same way in Article 111:2 as it is in Article 111:4, 7 ' although, as the following analysis shows, the underlying inquiries are similar. By concentrating on competitive relationships, the national treatment provision focuses on the relationship between the interests of consumers in the domestic market and foreign producers to determine how closely aligned they are. If consumers treat the imported and domestic products as close substitutes, the products are "like" for the purposes of Article Im, which also tells us that consumers have the potential to provide surrogate representation for the interests of foreign producers.79 Under these circumstances, when countries interfere with the process by which consumers might represent the interests of foreign producers, they decrease the surrogate representation that 75. WTO Report on Japan-Taxes on Alcoholic Beverages, WT/DS8/AB/R,WT/DS 10/ AB/R, WT/DS 1 1/AB/R (Oct. 4, 1996) [hereinafter Japan-Alcohol (AB)]. See Hudec, supra note 6, at 112-13 (commenting on the originally unofficial nature of the Working Party criteria). See also Regan, Regulatory Purpose, supra note 3, at 465 (claiming that the criteria's "canonical status should be reconsidered"). 76. Japan-Alcohol (AB), supra note 3, at 20-21. Tariff classifications were added in 1987. Robert Howse & Elisabeth Tuerk, The WTO Impact on Internal Regulations-A Case Study of the Canada-EC Asbestos Dispute, in THE EU AND THE WTO: LEGAL AND CONSTI- TUTIONAL ISSUES 293 (Grainne De Burca & Joanne Scott eds., 2001) (citing Report on Japan--Customs Duties, Taxes and Labelling Practices on Imported Wines and Alcoholic Beverages, (Nov. 10, 1987) U6216 B.I.S.D. 34S/83. 77. Japan-Alcohol (Panel), supra note 71, para. 6.22. This conclusion was affirmed by the Appellate Body. See Japan-Alcohol (AB), supra note 3, at 20. 78. According to the Appellate Body, The concept of "likeness" is a relative one that evokes the image of an accordion. The accordion of "likeness" stretches and squeezes in different places as different provisions of the WTO Agreement are applied. The width of the accor- dion in any one of those places must be determined by the particular provision in which the term "like" is encountered as well as by the context and the circumstances that prevail in any given case to which that provision may apply. Japan-Alcohol (AB), supra note 3, at 21. "It follows that, while the meaning attributed to the term 'like products' in other provisions of the GATT 1994, or in other covered agreements, may be relevant context in interpreting Article m:4 of the GATr 1994, the interpretation of 'like products' in Article II1:4 need not be identical, in all respects, to those other meanings." EC-Asbestos (AB), supra note 3, para. 89. It is widely understood that the term "like" product in Article :4 can be determined by drawing a wider circle than is true for the term "like" in 111:2, but that the circle is not as wide as the combination of like and directly competitive and substitutable in Article 111:2. Id. para. 99. See, Sydney M. Cone, I, The Asbestos Case and Dispute Settlement in the World Trade Organization: The Uneasy Relationship Between Panels and the Appellate Body, 23 MICH J. INT'LL. 103, 124 (2000) (pointing out that para. 99 is dicta). 79. See discussion supra, notes 44-46. 2004] IND. INT'L & COMP. L. REv. is an important part of participatory lawmaking and is therefore suspect. By contrast, if foreign goods do not sufficiently compete with domestic goods, then domestic consumers cannot act as surrogates for foreign producers. The interests of domestic consumers are not aligned with those of foreign producers, and there is no need to inquire about less favorable treatment to see whether the measure was passed in a way that disrupted the process of surrogate representation. The competitive relationship test is also used in applying the second sentence of mH:2 to determine whether imported products are "directly competitive or substitutable" with domestic goods, which again reflects the importance of recognizing consumers as surrogates for the interests of foreign producers. The test is similar to the "like" products test, but casts a wider net by expanding the range of products where the consumer can represent foreign producers. 80 The factors that are relevant to this inquiry are similar to those used in the "like product" inquiry: physical characteristics, common end-uses, tariff classifications, and the marketplace. 8' But here the Appellate Body summed them up by seeking an inquiry into common end-uses or "elasticity of substitution. 8 2 In other words, similar to the interpretation of the word "like" in the first sentence, the first inquiry concerns consumer behavior and identifies instances in which consumers might serve as effective political proxies for foreign interests. The Appellate Body's elaboration on this competitive relationship test further demonstrates the surrogate representation view of the national treatment provision. Pre-existing barriers to foreign producers may be relevant to the analysis of competitive relationships because consumer perceptions about the marketplace may be influenced by prior restrictions on foreign producers that made it difficult for consumers to recognize their joint interest with foreign producers. In Korea-Alcohol8 3 the Appellate Body stated that a potential competitive relationship could buttress a finding of a direct competitive relationship 84 and agreed that the inquiry must include not only 80. "How much broader that category of 'directly competitive or substitutable products' may be in any given case is a matter for the panel to determine based on all the relevant facts in that case." Japan-Alcohol (AB), supra note 3, at 25. 81. Id. "Marketplace" referring to competition in the relevant market. 82. Id. [The decisive criterion] seems to be whether two products have common end- uses as shown by the demand cross-price elasticity of the two products. That is, if for every sale of the import there is one lost sale of the domestic product, then the two products are perfect substitutes and in direct competition. In a case of perfect substitutability, the imported and domestic products are like products and are covered under Article 111:2, first sentence. Instances of less-than-perfect substitutability are addressed under Article 111:2, second sentence." BHALA & KENNEDY, supra note 5, at 97. 83. WTO Report on Korea-Taxes on Alcoholic Beverages, WT/DS75/AB/R, WT/DS84/AB/R (Jan. 18, 1999) [hereinafter Korea-Alcohol (AB)]. 84. Id. para. 113, 120. [Vol. 14:3 WTO NATIONAL TREATMENT existing substitutability, but also the capacity for substitutability-a concept it termed "latent demand."85 By clarifying the term "directly" in "directly competitive or substitutable,""6 this analysis takes into account that prior regulation might have hindered or prevented consumers from recognizing the interest they have in foreign goods. In order to identify when consumers have the potential to provide surrogate representation to foreign producers, a panel must determine what the competitive relationship would have been without prior restraints on that relationship.87 Further, because this counterfactual is so difficult to determine, Korea- Alcohol allowed the use of evidence from a third market to establish that consumers have an economic and, by implication, surrogate interest in the foreign goods. "[E]vidence from other markets may be pertinent to the examination of the market at issue, particularly when demand on that market has been influenced by regulatory barriers to trade or to competition. 88 Where consumers have been prevented from speaking for foreign interests, the inquiry turns to whether consumers in other countries identify their interests with foreign as well as domestic producers. The conclusion that the competitive relationship test reflects the role of consumers as potential surrogates for foreign interests is also supported by what the Appellate Body has said about the role of purpose in applying the competitive relationship test and in the Appellate Body's treatment of regulatory measures that facially discriminate against foreign goods. A. 1. The Role of Purpose-The Asbestos Case In EC-Asbestos,89 the Appellate Body made it look as if the purpose of the regulatory measure was relevant to applying the "like product" tests, thus giving support to those who would read purpose into the analysis of Article HI. However, a proper understanding of that opinion shows that the purpose of a measure has no role other than to help apply the competitive relationship test. In 1997, France prohibited the manufacture, processing, sale, and importation of asbestos fibers and products containing asbestos fibers, although it allowed 85. Id. para. 114. 86. Id. para. 109. 87. [S]tudies of cross-price elasticity... involve an assessment of latent demand. Such studies attempt to predict the change in demand that would result from a change in the price of a product following, inter alia, from a change in the relative tax burdens on domestic and imported products. Id. para. 121. 88. Korea-Alcohol (AB), supra note 83, para. 137. 89. EC-Asbestos (AB), supra note 3. See generally Jochem Wiers & James Mathis, The Report of the Appellate Body in the Asbestos Dispute: WTO Appellate Body Report 12 March 2001, WT/DS135AB/R, European Communities-Measures Affecting Asbestos and Asbestos- containing Products, 28 LEGAL ISSUES OF ECON. INTEGRATION 211 (2001) (discussing the EC-Asbestos report). 2004] IND. INT'L & COMP. L. REV. the production and sale of asbestos substitutes.9" Therefore, the ban clearly benefited domestic producers of asbestos substitutes over their foreign asbestos-producing competitors. In the context of the discussion of "like" product, the Appellate Body said that health risks are to be considered in the Article I1:4 "like product" inquiry.9 1 Purpose-theorists seized upon this indication and suggested that the Appellate Body was acknowledging that if the legislature can advance a non-protectionist purpose for the legislation then the products would be found to be not "like." 92 However, a close reading of the Appellate Body's opinion shows that, in fact, regulatory purpose is not an independent reason for finding that products are not "like." Instead, it is simply a fact that helps us understand the competitive relationship between imported and domestic goods. The Appellate Body integrated a consideration of health factors into two of the Border Tax Adjustments criteria: physical properties and consumers' tastes and habits.93 Thus, when determining which physical properties are relevant to the "like product" inquiry, "panels must examine those physical properties of products that are likely to influence the competitive relationship between products in the marketplace. 94 The health risks are relevant to the physical property inquiry not because they might show a non-protectionist desire to protect human health, but rather because health-risks are likely to influence consumers' decisions and thus are relevant in determining whether 90. EC-Asbestos (AB), supra note 3, para. 1-2 (there were a few exceptions). 91. Id. para. 113. 92. See Regan, supra note 3, at 467 ("[T]he Appellate Body's attempts to rely solely on competitive relationship, without bringing in regulatory purpose, either have an otherworldly air, or else require reference to regulatory purpose to complete them. Perhaps the Appellate Body thought WTO insiders were not yet ready for explicit appeal to regulatory purpose."). EC-Asbestos is the only Appellate Body decision with a concurring opinion. In it, the concurring member argues that scientific evidence of the health risks is so abundant that the Appellate Body should have declared definitively that asbestos fibers are not like the substitute fibers. EC-Asbestos (AB), supra note 3, para. 152. That is, the concurring member could not "imagine what evidence relating to economic competitive relationships as reflected in end-uses and consumers' tastes and habits could outweigh and set at naught the undisputed deadly nature of... [the asbestos fibers]." Id. Although the concurring member would limit his suggestion to this case alone, "the other Members of the Division feel unable to take [this step] because of their conception of the 'fundamental,' perhaps decisive, role of economic competitive relationships in the determination of the 'likeness' of products under Article 111:4. Id. para. 153. Second, the concurring member questions how fundamental an economic interpretation of likeness under 111:4 ought to be and warns that "fundamentally" might become one and the same with "exclusively." Id. para. 154 (concluding such a decision should be left for a different time). Although Regan suggests these statements by the concurring member leave room for possible consideration of regulatory purpose in a different case, the concurring member did not refer to regulatory purpose in his opinion. 93. EC-Asbestos (AB), supra note 3, para. 113. 94. Id. para. 114. [Vol. 14:3 2 WTO NATIONAL TREATMENT consumers can be effective surrogates for foreign producers.95 Stressing the point further, "evidence relating to health risks may be relevant in assessing the competitive relationship in the marketplace between allegedly 'like' products." 96 Similarly, health risks play an important role in consumers' tastes and habits because these tastes and habits "are very likely to be shaped by the health risks associated with a product which is known to be highly carcinogenic." 9' Here too, the analysis focuses on competitive relationships98 and whether or not consumer surrogates tell us that foreign producers have a viable interest in the market. In short, attention to the health related properties of a product-and how those health related properties affect competitive relationships-is relevant to analyzing the role of consumers as surrogates for the interests of foreign producers. If consumers do not consider the products to be competitive sub- stitutes because the products have different health related properties, then a consumer cannot act as a surrogate for the interests of foreign producers. When this is true, finding the goods to be "not like" simply reflects the fact that treating those goods differently cannot take away any representation of foreign interests that consumers would otherwise provide. As a result, from the standpoint of the participatory-enhancing function of the WTO, the regulation is less suspect as a process for driving a wedge between consumer interests and adversely affected foreign interests. If the interests were not that close in the first place (because of the health related properties of the products) the function of the WTO in policing surrogate representation by consumers is not impaired. 95. See Howse & Tuerk, supra note 76, at 288-89 (acknowledging "the approach of the Appellate Body was to introduce the fundamental human interests at stake not through an examination of regulatory purpose, but rather by making those interests relevant to an analysis of the competitive relationship between products in the market place."). One argument Regan presents for considering regulatory purpose under likeness can be summarized as follows: (1) If a plastic container harms the environment and cardboard containers do not, they "are not 'like' in any ordinary sense"; (2) existence of "harm" is determined by the regulating government; (3) therefore harm depends on regulatory purpose; (4) therefore likeness depends on regulatory purpose. Regan, supra note 3, at 448-49. In other words, Regan distinguishes the physical effects of a product from the harm that product may cause. In the context of the Asbestos case, a physical effect of asbestos is that it causes cancer. However, asbestos does not harm unless the regulating state determines that cancer is not worth the benefits of asbestos products. However, Regan's argument relies on the presumption that "harm" determines likeness, which is clearly contrary to the Appellate Body's focus on health risks, without any discussion of the benefits of asbestos. That is, the Appellate Body report focuses on the effects of asbestos on health in determining likeness. 96. EC-Asbestos (AB), supra note 3, para. 115. 97. Id. para. 122. 98. Id. para. 117. 2004] IND. INT'L & COMP. L. REv. Ill. A. 2. SPECIAL TREATMENT OF FACIALLY DISCRIMINATORY MEASURES Domestic regulation sometimes discriminates against foreign products on its face by putting foreign and domestic products into different regulatory categories by explicitly identifying one regulation for foreign products and a different regulation for domestic products. When that occurs, the analysis can dispense with any examination of the competitive relationship between imported and domestic products; the foreign and domestic products are automatically considered to be "like" products. For example, in Argentina- Bovine Hides,99 Argentina established a tax system based on factors wholly unrelated to the nature of the products or their competitiveness, but dependent only on the national origin of the producer and whether the product was being sold inside Argentina. Applying the hypothetical product test,"° the panel stated that it was therefore "inevitable... that like products will be subject to [the taxes at issue],"' 1 and that it was not necessary to prove separately either the "like product" requirement of Article 111:2, first sentence 0 2 or even that "trade involving like imported products actually exist[ed]."' 3 99. WTO Panel Report, Argentina-Measures Affecting the Export of Bovine Hides and the Import of Finished Leather, WT/DS155/R (adopted Feb. 16, 2001) [hereinafter Argentina-Bovine Hides]. 100. The hypothetical product test draws its support from the Section 337 case. There the United States applied different procedures when foreign goods were alleged to have violated a U.S. patent than it did when domestic goods were alleged to have violated a patent. The imported infringing goods were not necessarily the same as the domestic infringing goods and in many cases would have no competitive relationship. Although the panel in the section 337 case was not interpreting the word "like," it had no problem holding that the United States could not escape from its obligations under national treatment for that reason. The panel noted that if competitive products were infringing domestically and as imports they would have been treated differently and that a hypothetical circumstance was enough to bring the measure within the purview of section 337. Panel Report on United States-Section 337 of the Tariff Act of 1930, GATT B.I.S.D. 36S/345 (Nov. 7 1989) [hereinafter Section 337]. 101. Argentina-Bovine Hides, supra note 99, para. 11.169. 102. Id. See also WTO Panel Report, Indonesia-Certain Measures Affecting the Automobile Industry, WT/DS54/R, WT/DS55/R, WT/DS64/R para. 14.113 (adopted July 23, 1998) ("[An origin-based distinction in respect of internal taxes suffices in itself to violate Article 111:2, without the need to demonstrate the existence of actually traded like products."). WTO Panel Report, Korea-Measures Affecting Imports of Fresh, Chilled and Frozen Beef, (July 31, 2000) WT/DS 161/R, WT/DS 169/R, para. 627 [hereinafter Korean-Beef (panel)]. Any regulatory distinction that is based exclusively on criteria relating to the nationality or the origin of the products is incompatible with Article 1II and this conclusion can be reached even in the absence of any imports (as hypothetical imports can be used to reach this conclusion) confirming that there is no need to demonstrate the actual and specific trade effects of a measure for it to be found in violation of Article 111. Id. Although the panel was overruled by the Appellate Body on the issue of whether facial dis- crimination necessarily results in a violation of Article III, the finding that there did not need to be actual like products was not disputed on appeal. Korean-Beef (AB), supra note 52, para. 133. 103. Argentina-Bovine Hides, supra note 99, at 11.169. [Vol. 14:3 WTO NATIONAL TREATMENT This approach also conforms to the surrogate representation rationale. When a measure is facially discriminatory, we can automatically say that the regulatory authority did not account for the role of domestic consumers as surrogates for the interest of foreign producers. If the regulatory authority had represented the role of consumers as surrogates for foreign producers, they would not have singled out foreign products for special treatment on account of their foreignness. The products can be presumed to be "like" because foreign products identical in every relevant respect with domestic goods would have been treated differently, a sure sign that consumers have not served as effective surrogates for foreign interests. In sum, the Appellate Body's approach to the determination of like products fully implements the surrogate representation rationale by seeking to identify those classes of cases in which domestic consumers will function as good surrogates for the interests of foreign producers. Of course, preserving this surrogacy is a necessary but not a sufficient condition for the application of Article m11; the analysis must go on to inquire into whether the foreign producers receive less favorable treatment. III. B THE LESS FAVORABLE TREATMENT STANDARD The approach developed by the Appellate Body to determine whether imports are treated less favorably than domestic goods (under Article 111.4) or (equivalently) taxed differently so as to afford protection (under Article 111.2), also reflect the surrogate representation rationale. III. B. 1. Facial Discrimination When the measure distinguishes on its face between domestic and foreign products, the Appellate Body has had to determine whether the mere fact that foreign goods are treated differently from domestic goods is enough to infer less favorable treatment and, if not, what additional facts must be proven. The answer to each question is revealing from the standpoint of the surrogate representation rationale. The Appellate Body addressed both issues authoritatively in Korean- Beef , 4 in which the Appellate Body reviewed a dual retail system for beef products. Imported beef had to be sold in different outlets from domestic beef or (for larger stores) from different locations within the store. Although the measure facially distinguished between like products on the basis of national origin, the Appellate Body determined that such differential treatment was not 104. Small retailers that were a "Specialized Imported Beef Store" could sell any meat except domestic beef. Korean-Beef (AB), supra note 52, para. 143. Any other small retailer could sell any meat other than imported beef. Id. A large retailer could sell both, so long as the imported and domestic beef were sold in different sales areas. Id. 2004] IND. INT'L & COMP. L. REv. unlawful in itself; the complaining country still had to prove that the differential treatment was also less favorable treatment. °5 This conclusion is, of course, not only logical and suggested by the structure of Article 111.4 (which, after all, makes "less favorable treatment" a required part of the analysis), but it is also consistent with the surrogate representation model. First, the refusal to automatically invalidate facially discriminatory measures shows that the Appellate Body is not engaging in substantive review of national regulatory measures to determine whether their purpose or effect is to discriminate, effectively rejecting the notion that facial discrimination shows an impermissible purpose. 6 This is true because the fact of differential treatment of foreign and domestic goods does not necessarily mean that foreign producers are not adequately represented in the domestic decision-making process. Domestic producers could provide ade- quate surrogate representation even if different regulations apply to foreign goods if the foreign producers are more favorably advantaged or if the different treatment reflects more than the different circumstances of the foreign producers that are relevant to the regulatory scheme.0 7 That might be the case, for example, where the regulatory measure specified the safety features for products but allowed foreign products to be admitted if they met the different 105. Id. para. 135. The Appellate Body explicitly stated that the different treatment under the measure need not be a formal difference (i.e., facial discrimination). Id. para. 137. 106. Some commentators, for example, would make facially discriminatory measures an automatic violation of the national treatment provision on the ground that the fact of discrimination shows an unlawful purpose. See, e.g., Regan, Regulatory Purpose, supra note 3, at 455. The Appellate Bodies rejection of that position is further evidence that they are rejecting purpose as a substantive test for national treatment. Although "[c]ases of explicit discrimination stand out because the explicitly different treatment is viewed as evidence that discrimination against foreign goods is a deliberate policy... GATr/WTO legal texts have not created separate rules for explicitly discriminatory regulatory measures." Hudec, Requiem, supra note 6, at 621-22. 107. The national treatment principle may also forbid formally identical treatment in certain circumstances. Section 337, supra note 100, para. 511. For example, a procedural requirement that applies to both domestic and foreign producers may be found to be unreasonably burdensome on the foreign producers, and thus a violation of Article II. Since domestic producers would not suffer as harsh a burden, they would be poor surrogates for foreign interests. See BHALA & KENNEDY, supra note 5, at 100. Exposure of imported products to the risk of discrimination is itself a form of discrimination prohibited under Article II. In the panel report, Import, Distribution and Sale of Certain Alcoholic Drinks by Provincial Marketing Agencies, the panel concluded that Canadian minimum price regulations for beer undermined one of the fundamental purposes of Article 111:4, which is to ensure that internal regulations do not dilute or eliminate the benefit of Article 11 tariff concessions. Moreover, the panel report establishes that equality of treatment of imported products vis-t-vis the domestic like product still may be a national treatment violation. Even though the two products are treated identically (e.g., as in the case of minimum price regulations), a national treatment violation nevertheless exists if the imported product could undersell the domestic like product but for the minimum price control. [Vol. 14:3 ]WTO NATIONAL TREATMENT safety regulations of the home government. Such a regulation based on the principle of mutual recognition would be differential treatment but not an instance in which the incentive of the domestic firms to resist the regulation on behalf of the foreign producers had been compromised. How then are we to know when facially discriminatory measures treat foreign goods less favorably? The cornerstone of the analysis of less favorable treatment is the concept of equality of competitive conditions-the single consistent value that runs throughout the Appellate Body jurisprudence.1' As the Appellate Body said in Korean-Beef, "whether or not imported products are treated 'less favorably' than like domestic products" depends on "whether a measure modifies the conditions of competition in the relevant market to the detriment of imported products." "0 The test centering on equality of competitive conditions invokes the image of a level playing field and summarizes the basic commitment of the WTO to remove barriers to competition and allow markets to work across borders. WTO jurisprudence makes it clear that this test does not focus on the impact of the measure on trade flows," 0 and that it is not an effects test in that sense. Instead, it is a test that looks at the costs imposed by a measure to deter- mine whether the regulatory costs are evenly distributed between domestic and foreign producers."' The equality of a competitive conditions test expresses the central and foundational wisdom of the surrogate representation rationale: domestic producers can never be effective representatives of the interests of foreign producers if they stand to gain a competitive advantage to offset the cost of regulation. 108. The purpose of this first sentence of 11:2 is to protect "expectations on the competitive relationship between imported and domestic products." Panel Report. United States-Taxes on Petroleum and Certain Imported Substances, GATT B.I.S.D. 34S/136, para. 5.1.9 (June 17, 1987) [hereinafter U.S.-Petroleum]. "Article 1:4, which is the parallel provision of Article III dealing with the 'non charge' elements of internal legislation, has to be construed as serving the same purpose." Section 337, supra note 107, para. 5.13 (Nov. 7, 1989). "The words 'treatment no less favourable' in ... [I1:4] call for effective equality of opportunities for imported products in respect of the application of laws, regulations and requirements affecting the internal sale, offering for sale, purchase, transportation, distribution or use of products." Id. para. 5.11. "[T]his standard of effective equality of competitive conditions on the internal market is the standard of national treatment that is required, not only with regard to Article II generally, but also more particularly with regard to the 'no less favourable treatment' standard in Article I:4." Japan-Film, supra note 3, para. 10.379. 109. Korean-Beef (AB), supra note 52, para. 142. 110. Korean-Beef (AB), supra note 52, para. 137. "Article RI protects expectations not of any particular trade volume but rather of the equal competitive relationship between imported and domestic products." Japan-Alcohol, supra note 3, at 16. See also, U.S.-Petroleum, supra note 108, para. 5.1.9. See also supra text accompanying note 108 (discussing development of the standard of equality of competitive conditions). 111. Although the Appellate Body referred to the "fundamental thrust and effect of the measure," Korean-Beef, supra note 52, para. 142, it evidently did not mean the effect of the measure on trade flows but the effect of the measure on costs, a central condition of competition. Id. para. 145. 2004] IND. INT'L & COMP. L. REV. When it applied this cost-based test to the Korean regulatory structure, the Appellate Body found that the vast majority of small Korean retailers chose to sell domestic beef rather than foreign." 2 Therefore, imported beef required new channels to reach consumers if imported beef was to compete with domestic beef." 3 "The central consequence of the dual retail system can only be reasonably construed ... as the imposition of a drastic reduction of commercial opportunity to reach, and hence generate sales to, the same consumers served by the traditional retail channels for domestic beef."'"1 4 Further, "what is addressed by Article 11:4 is merely the governmental intervention that affects the conditions under which like goods, domestic and imported, compete in the market within a Member's territory."'" 5 Because the measure clearly imposed costs on foreign producers that were not imposed on the domestic beef industry, the domestic industry could not be relied upon to represent foreign interests in the domestic regulatory process. The flaw with the dual retail system was that by imposing greater costs on imported than on domestic goods, the scheme itself showed that the regulatory process failed to preserve the role of domestic producers as surrogates for foreign interests. By turning the test focused on "equality of competitive conditions" into a test focusing on the differential impact of the costs of the measure, the Appellate Body has avoided reintroducing the effects part of the "aims and effects" test. In its place it has invoked a test that determines whether the regulatory process has imposed differential costs on the foreign and domestic producers, because that would itself be a sure sign that domestic producers have not been successful surrogates for foreign producers. Admittedly, although meat producers might not have been good surrogates for foreign beef producers, meat retailers in Korea might have been good surrogates and foreign interests might have been fully represented in the regulatory process. As a general matter, retailers represent consumer interests because retailers enhance their own welfare by generating benefits for their customers; their goal as distributors is to enhance consumer surplus-the net benefits that consumers get from the low prices and high quality that competitive markets provide. Assuming that Korean consumers consider imported and domestic beef to be close substitutes, one might have expected the retailers to argue against measures that would make comparison shopping more difficult. This would have provided foreign beef producers with the surrogate representation they needed to provide the pro-consumer, pro-foreign producer point of view when the measures were being adopted. But in the context of the Korean-Beef case, the possibility that retailers in Korea would provide effective representation for the interests of foreign producers was weak. Retailers, being numerous and diverse, would have 112. Id. 113. Id. 114. Id. 115. Id. para. 149. (Vol. 14:3 WTO NATIONAL TREATMENT obvious difficulties in organizing to protect their (and consumer) interests. The fact that the measure would raise retailer costs uniformly, and thus put none at a competitive disadvantage versus other retailers, decreased the incentive to organize in opposition to the measure. Moreover, the incentive for Korean retailers to organize was blunted by the preexisting structure of the beef market. Prior import quotas on foreign beef, and its high cost, meant that the market share of foreign beef was low. As a result, retailers in Korea did not have an established and defined interest in promoting foreign beef, and therefore did not have an accurate assessment of the consumer surplus that could be generated from selling domestic and foreign beef side by side. Had the Korean measure been changing a well-established pattern of equal access, rather than disrupting an emerging distribution pattern, the retailers would have been injured to a far greater extent. We can speculate that they would have provided greater regulatory resistance, and therefore a heightened level of surrogate representation. Moreover, retailers who sold only domestic beef may have supported the segregation of foreign beef, feeling that they could gain over their rivals by not having to respond to competitive pressures to carry both domestic and foreign beef. Retailers that would have to absorb additional expense if the market were allowed to work would not be averse to seeing the market mechanism disrupted, and would therefore "defect" from any retailer coalition to oppose the measures." 6 Retailers were thus inadequate surrogates for foreign beef producers. III. B. 2. Facially Neutral Measures Often the domestic regulatory measure in question will not single out imported products for special treatment; it will be neutral concerning the origin of the goods."' Under these circumstances it is especially difficult to design a test for the less favorable treatment standard that is not either over-inclusive or under-inclusive. A test that relies only upon disproportionate effects would be over-inclusive by unduly impinging on the freedom of a country to regulate in the public interest when the regulation has international effects. A test that ignored effects-that is, one that exempted origin neutral measures from 116. It is also instructive to examine the restaurant market as an outlet for foreign beef. Forty-five percent of the foreign beef sold in Korea was sold through restaurants. Korean-Beef (panel), supra note 102, para. 618. As far as we can tell from the record in the case, the Korean government imposed neither labeling requirements nor a requirement that the menus separately list the foreign and domestic beef. Id. In that segment of the beef market, where preexisting arrangements did not segregate the domestic and foreign beef, the power of the restaurants was apparently great enough to represent effectively the foreign beef producers and overcome any attempt to segregate the market for foreign and domestic beef. Id. 117. Interestingly, the problem of facially neutral measures has arisen in only one Appellate Body case under Article HI:4, EC-Asbestos, supra note 3. Most of the analysis of facially neutral measures has been with respect to tax measures, where the Appellate Body has developed a sophisticated and nuanced approach that fully reflects the surrogate rationale. 2004] IND. INT'L & COMP. L. REv. m1:4-would be under inclusive and in fact, would make it easy for a country to engage in protectionist measures by making its regulation look facially neutral. "8 The Appellate Body has made clear that taxing imports and domestic products the same within a certain fiscal category does not absolve the regulating country of its obligations under Article Ill, 9 but what test determines when a facially neutral measure should be struck down? Any extended look at the external effects of a measure would require the analyst to consider the trade effects in light of the purpose of the measure, and would therefore require a substantive balancing of trade and non-trade values under some form of an aims and effects test. The Appellate Body has largely avoided such a substantive review by constructing tests for less favored treatment (and the equivalent standard in Article mH:2) that focus on whether the measure appears to have been designed in a way that effectively co-opted domestic producers from acting as surrogates for the interests of foreign producers. Again, the equality of competitive conditions concept looks at the quality of the surrogacy. 20 Under the first sentence of 111:2, the inquiry into the treatment of imports is relatively straightforward, for the provision says that any differential in the tax on like goods is impermissible; no de minimis test or complex inquiry is necessary to determine whether the differential treatment upsets the competi- tive balance or otherwise impairs competitive conditions. Any difference is conclusively presumed to impair competitive conditions. The analysis of this standard under the surrogate representative rationale is straightforward. If imports are taxed in excess of domestic "like products" then we can assume that the domestic producers with interests most similar to the foreign producers were a poor group of representatives of the foreign interests. We can assume this because the tax favors domestic producers and that would occur only if the regulating government has bought off the domes- tic producers by offsetting higher taxes by relieving them from foreign competition. Alteration of competitive conditions in favor of domestic pro- 118. See Howse & Tuerk, supra note 76, at 285 (discussing how hidden discrimination in facially neutral measures requires an interpretation that allows Article Ell to reach instances of de facto, as well as de jure, discrimination). 119. Chile--Alcohol (AB), supra note 10, para. 52. 120. Although the central thrust of Article El is sometimes portrayed as having multiple purposes, equality of competitive conditions remains the foundational concept. For example, in Korea-Alcohol, the Appellate Body identified three objectives of Article III: "avoiding protectionism, requiring equality of competitive conditions and protecting expectations of equal competitive relationships .. " Korea - Alcohol (AB), supra note 10, para. 120. It appears, however, that those measures that are protectionist are a strict subset of those that violate the requirement of equality of economic conditions. That is, if WTO jurisprudence prohibits the latter, the former will always be eliminated. Such a view may be supported by the Appellate Body's later statement in the same opinion reducing the three objectives to one: "the object and purpose of Article miI is the maintenance of equality of competitive conditions for imported and domestic products." Id. para. 127. This single requirement of providing "equality of economic conditions" applies to Article m:4 as well. Japan - Film, supra note 3, para. 10.369. [Vol. 14:3 WTO NATIONAL TREATMENT ducers through tax measures tells us that domestic producers reduced their resistance to the tax because the disadvantage of the tax was offset by the advantage of the freedom from foreign competition. By invalidating the tax, the first sentence of Article 111:2 corrects the participatory deficit that reduces the surrogate voice of foreign interests in domestic policymaking. One might ask whether it is appropriate to assume that mere differential impact of a tax measure is enough to justify a finding that the measure was in fact favoring domestic producers so that they would reduce their opposition to the tax. We have already seen that differential treatment is not necessarily less favorable treatment.12' Tax categories, however, are generally constructed on the basis of revenue needs rather than regulatory needs. The process of determining the categories into which the taxed products falls is generally determined by the revenue requirements of the government rather than by any aspect of the product itself. That makes it relatively easy to conclude that the revenue to be derived from like products ought to not be a basis for dis- tinguishing between the products. When a revenue distinction is nonetheless made, it is appropriate to draw the conclusion that creating differential revenue streams probably does not reflect differences in the product and therefore must reflect the fact that in the decision-making process domestic producers were able to relieve some of the burden of taxation on themselves by inducing the decision-maker to impose a relatively greater burden on foreign producers (the very entities for whom the domestic producers should have been acting as surrogate). Of course, not all tax legislation is designed only with revenue in mind. Sometimes tax authorities create differential tax categories in order to discourage consumption of one type of product. They may distinguish between high and low nicotine cigarettes, for example, or between the high and low alcohol content of liquors in order to promote products that are perceived to be safer. That possibility should not change the analysis of tax measures. First, the precise rule invalidating any differential taxation of like products is fully justified because tax classification based on product content is rare. Moreover, when tax classifications have a non-tax purpose, the differential (and the differential effect) created to achieve that purpose can be justified, and thus allowed under Article XX. The Appellate Body need not complicate Article II analysis by taking safety objectives of revenue measures into account at that point of the analysis. For the second sentence of Article 111:2, the less favorable treatment inquiry revolves around two standards: whether the imports are "not similarly taxed" and whether they are taxed "so as to afford protection."'' 2 The "not similarly taxed" inquiry requires a showing that the differential is more than de minimis. "Dissimilar taxation of even some imported products as compared 121. See discussion supra Part ILI.B.1, notes 105-107. 122. Japan-Alcohol (AB), supra note 3, at 27-29. 20041 IND. INT'L & COMP. L. REv. to directly competitive or substitutable domestic products is inconsistent with" this standard.123 The second inquiry, whether the tax is "applied so as to afford protection," requires the analyst to examine "the design, the architecture, and the revealing structure of the measure."'124 Just as is true under Article I, this too relates to the effect of the measure on the equality of competitive conditions. In some instances, "[t]he very magnitude of the dissimilar taxation in a particular case may be evidence of such a protective application."' 25 But in other instances, additional unspecified factors might be relevant.'26 Significantly, the Appellate Body has cited, with apparent approval, a 1987 Panel statement that the "so as to afford protection" test was a matter of looking at factors that could show sufficient evidence of fiscal distortions of the competitive relationship between ". . .imported and domestic products 'affording protection to the domestic production .... "',127 In other words, "so as to afford protection" means nothing more than affecting the equality of competitive conditions in favor of domestic products. Commentators have interpreted the "design, architecture and structure test" as the equivalent of the test for determining the purpose of the tax classification and therefore as introducing a substantive inquiry into the evaluation of the tax classification. 28 Analysis shows, however, that the Appellate Body had a more sophisticated and less confrontational view in mind. They look at the design, architecture, and structure of the measure to determine whether the tax categories have been constructed to disrupt the natural alliance between domestic and foreign producers in opposition to the tax measure. 129 In Korea-Alcohol, the very large difference in taxation was enough to justify a finding that the tax classification was "so as to afford protection."1' Even beyond that simple conclusion, however, the Appellate Body elaborated on the design, structure, and architecture test. The tax operated: 123. Canada-Periodicals, supra note 10, at 31. 124. Japan-Alcohol (AB), supra note 3, at 29-3 1. 125. Id. at 30. 126. Id. at 30. 127. Id. at 28. 128. See discussion, supra note 6. 129. Admittedly, in Canada-Periodicals, supra note 10, at 30-32, the Appellate Body looked at a government report and two statements by government officials to support its conclusions about the design, architecture and structure of the classification, which has been construed by purpose theorists to be a basis for determining purpose. See, e.g., Regan, Regulatory Purpose, supra note 3, at 459. Their resort to this legislative evidence was in the context of different treatment that was said to be "beyond excessive, indeed it is prohibitive." Canada-Periodicals, supra note 10, at 30. Excessive disparity has been held to invalidate measure on its own, making this legislative purpose something of dicta. See Japan-Alcohol (AB), supra, note 3, at 30-3 1. 130. Korea-Alcohol (AB), supra note 10, para. 150. [Vol. 14:3 WTO NATIONAL TREATMENT ... in such a way that the lower brackets cover almost exclusively domestic production, whereas the higher tax brackets embrace almost exclusively imported products. In such circumstances, the reasons . ..as to why the tax is structured in a particular way do not call into question the conclusion that the measures are applied "so as to afford protection to domestic production."'31 Because the favored and disfavored categories were virtually conterminous with the distinction between domestic and foreign produced products, and because the differential between the classifications was large, the classifications were easily interdicted. The tax authorities drew the lines between favored and disfavored categories in such a way that segregated the interests of the domestic and foreign producers of liquor and made it impossible for domestic producers to act as surrogates for the foreign producers. The design, structure, and architecture inquiry was in fact an inquiry into the quality of the surrogacy. The relevant analysis was a great deal harder in Chile-Alcohol-the most recent Appellate Body decision applying the design, architecture and structure test-because there the favored and disfavored categories contained both domestic and imported products.'32 Again, the way the Appellate Body interprets the Article mI:2 standards shows that it is analyzing the factual background of the measure to determine whether domestic producers provided effective surrogate representation for the interests of domestic producers when the measure was adopted. The arguments of the parties turned on the fact that the favored tax brackets contained some imported goods (i.e., not all imports were disadvantaged) while the disfavored tax brackets included domestic goods (i.e., the adversely affected group was not only imported products). Although almost all of the relevant imports were taxed in the highest bracket, 133 and even though the vast majority of the comparable domestic products were taxed in the lowest bracket 134 a large proportion of the disfavored group of products included domestic goods. Moreover, in the higher, disfavored brackets, imports were relatively small, and domestic goods made up a major portion of the sales. Accordingly, Chile could easily argue that the tax brackets were not 131. Id. The only domestic product that fell into the disfavored tax classification was distilled soju, and that accounted for less than one percent of Korean production of the relevant product. Id. para. 147. Moreover, in the favored tax category "[t]here is virtually no imported soju so the beneficiaries of this structure are almost exclusively domestic producers." Id. para. 150 (quoting the panel decision). 132. Chile-Alcohol (AB), supra note 10, para. 1. 133. Id. para. 67 (stating almost ninety-five percent of directly competitive or substitutable imports were in the highest bracket). 134. Id. para. 67 (stating seventy-five percent of all domestic production was taxed at the lowest rate). 2004] IND. INT'L & COMP. L. REV. designed "so as to afford protection;" otherwise the disfavored category would not have included such a large proportion of domestic producers, so that within that category imported and domestic goods were similarly taxed, rather than "not similarly" taxed. 135 In the context of the surrogate representation analysis, Chile's argument was tantamount to the claim that foreign interests were adequately protected because the adversely affected domestic interests could represent them. The Appellate Body's two-prong test must be understood as responding to these claims by weighing the adequacy of the surrogate representation. The Appellate Body applied the "not similarly taxed" and "so as to afford protection" tests to explore whether the regulatory process had kept adversely affected domestic producers from effectively representing the interests of foreign producers. 136 The first test, "not similarly taxed," looks at the distribution of the burdens and benefits of the regulatory scheme between domestic and imported goods. Noting that 95% of the imported goods were taxed at the higher rate and 75% of the domestic products at the lower rate, 137 the Appellate Body concluded that: "the tax burden on imported products, most of which will be subject to a tax rate of 47 percent, will be heavier than the tax burden on domestic products, most of which will be subject to a tax rate of 27 percent." 138 In other words, at least on an aggregate level, the distribution of burdens and benefits is such that the adversely affected domestic producers seem to have a disproportionately smaller interest than foreign producers in avoiding the higher tax. This differential impact, however, was not enough to show that the measure was designed to afford protection. It demonstrated that the class of domestic producers who could represent the interests of the foreign producers (those in the higher tax category) was small in relation to the entire class of domestic producers. However, that fact by itself would not necessarily indicate that domestic producers could not effectively represent the interests of foreign producers. They may have been effective, but unsuccessful, representatives. Accordingly, the Appellate Body looked to the "so as to afford protection" prong of the analysis to assess the effectiveness of the domestic representation of foreign producers. 135. See id. para. 12. 136. The European Communities did present evidence that the Chilean government bought off domestic producers. WTO Panel Report, Chile-Taxes on Alcoholic Beverages, WT/DS87/R, WT/DSl O/R, para. 7.121 (June 15, 1999). The EC alleged that the Chilean government's preservation of preferential treatment of lower alcohol content products (the majority of domestic production being in the lower alcohol content bracket) allowed a higher tax on domestic products with higher alcohol content. Id. However, the panel did not engage in an inquiry of this alleged deal between the government and the domestic industry. Id. para. 7.122. 137. Chile-Alcohol (AB), supra note 10, para. 50. 138. Id. para. 53. [Vol. 14:3 WTO NATIONAL TREATMENT First, the Appellate Body addressed the fact that Chilean products constituted the "major part of the volume of sales in [the disfavored] bracket,"' 39 which would seem to indicate that as spokespersons against the higher taxes the domestic producers had more at stake than foreign producers. The Appellate Body acknowledged the direction but not the weight of that point, noting that "This fact, does not by itself, outweigh the other relevant factors, which tend to reveal the protective application of the New Chilean system." 140 The fact that the larger proportion of producers in the disfavored class were domestic-not foreign-tells us a great deal about the small number of imports but very little about the effectiveness of the domestic producers in representing the interest of producers of liquor with that alcohol content. The "other factors" alluded to by the Appellate Body show how the Appellate Body has embraced and applied the surrogate representation rationale. Two factors indicated that the classification adopted by Chile was designed to undermine surrogate representation. First, the tax rate rose steeply for liquor with an alcohol content above 35 proof and liquor with an alcohol content of 39 proof,14 1 and, second, "approximately half of all domestic production has an alcohol production of 35 [proof] and is, therefore located on the line of the progression of the tax at the point immediately before the steep increase in tax rates. .. .""' The conclusion from this tax structure is clear. Chile drew its tax classification to minimize the number of domestic producers who would be in the disfavored categories and therefore minimized the group of producers who would have an identity of interest with the foreign producers. Had Chile set the tipping point for the large jump in tax rates at products with an alcohol content of 34 proof, it would have had a large number of domestic producers aligned with the foreign interests. Instead, Chile effectively neutralized the opposition of that large group of domestic producers by including them in the lower rate and their foreign competitor in the higher rate. Chile also effectively neutralized the mechanism by which foreign producers might have had their interests represented by domestic producers, which is the very problem that the WTO is working to solve. If Chile wants to segregate natural allies in the political process, it must do so for some overwhelming regulatory purpose encompassed within Article XX. Consistent with the surrogate representation rationale, the Appellate Body rejected the broad claim that past de jure discrimination would be used as the basis for supporting a finding of bad faith. To equate past de jure discrimination with an appraisal of the present measure "would come close to a presumption of bad faith." 143 In terms of surrogate representation, one cannot support a finding of present protectionism with a finding of past 139. Id. para. 67. 140. Id. 141. Id. para. 63. 142. Id. para. 64. 143. Chile-Alcohol (AB), supra note 10, para. 74. 2004] IND. INT'L & COMP. L. REv. protectionism unless one can link the past discrimination to a present impairment of the mechanism by which domestic producers and consumers represent foreign interests. However, in analyzing the quality of the proxy representation of foreign interests by domestic interests, the Appellate Body noted that the "comparatively small volume of imports consumed on the Chilean market may, in part, be due to past protection."'" Here too, the Appellate Body is applying the surrogate representation rationale. Past discrimination of imports-which Chile accomplished by taxing different types of liquor at different rates-means that the country had already disrupted the market mechanism by which consumers protect the interests of foreign producers. Because prior discriminatory taxes had denied consumers the opportunity to express their preferences for foreign products in the marketplace, the taxes also diminished consumers' opportunity to evaluate and express their preferences for foreign products in the policymaking arena. The regulatory process needs to be especially protected when prior discrimination has impaired its ability to function.'45 In this connection, we can see the relationship between the test that looks at design, architecture, and structure and the legislative purpose behind the measure. The Appellate Body affirmed its prior ruling that justification for the unlawful discrimination was not to be relevant to determining whether Article III was violated. They would not examine "the many reasons legislators and regulators often have for what they do."' 6 But they would look at whether there were explanations for the design, architecture, and structure of the measure that were unrelated to protectionism. If the country could explain how the design, architecture, and structure came about for reasons that were unrelated to the need to buy the loyalty of domestic producers, it could refute an inference of unlawful discrimination, even though at this stage it would not be appropriate to ask whether the discrimination met the goals and tests of Article XX. In other words, the inquiry is not into whether the purpose of the regulation is permissible or substantively valid in some way, but only to determine whether there was a reason to negate the inference that the surrogate representation had been impaired. This is a limited use of purpose, geared only to determine whether the design, architecture, and structure show that there was no attempt to disrupt surrogate representation. 144. Id. para. 68. 145. By contrast the Appellate Body rejected the panel's conclusion that a finding of unlawful discrimination could be based on the "interaction of the New Chilean System with the Chilean regulation which requires most of the imports to remain at the highest tax level without losing their generic name and changing their physical characteristics." Id. para. 73. Those regulations were not of the type that could impair the ability of domestic manufacturers and consumers to represent the interests of foreign producers. Id. 146. Id. para. 71 (citing Japan-Alcohol (AB), supra note 3, at 27). [Vol. 14:3 WTO NATIONAL TREATMENT III. IMPLICATIONS AND CONCLUSION As the foregoing has demonstrated, the Appellate Body has developed an interpretive framework for the national treatment provision of Article III that is consistent with the process-oriented role of the WTO, and re- emphasizes it as an institution whose central mission is to insure that when a member country takes regulatory action affecting foreigners, the interests of the foreigners are not ignored in the decision-making process. The implications of this interpretation for our understanding of the WTO and its role as an international organization are significant. The process-based interpretation presented in this article sees the national treatment provision as a mechanism by which the WTO's dispute resolution process can determine whether the interests of foreigners that would normally be represented by surrogates within a lawmaking jurisdiction have in fact been undercut and stymied. When foreign goods are in close enough competition with domestic goods to satisfy the "like" or "directly competitive" test, we know that under ordinary conditions the interests of foreign producers will be represented by domestic producers or consumers (or both), and this identifies a situation in which it is important to preserve that surrogate representation. When, however, analysis of the regulatory measure-its design, architecture and structure or its comparative treatment of foreign producers-reveals that those surrogates have been undercut in the regulatory process (for example, because the regulation imposes disproportionate costs on foreign producers), then the regulatory measures impermissibly impairs the participatory function that the WTO is designed to uphold. The process-based account of national treatment gives Article Ill a coherent content, and furnishes answers to the kinds of issues that were raised at the beginning of this article. The key phrases of Article I1 take on a consistent meaning, focusing either on the role of consumers as surrogates for foreign producers (the "like" or "directly competitive" tests) or on the imposition of disproportionate costs on foreigners in order to ameliorate domestic opposition (the "less favorable" treatment test). The injunction that no member should apply measures so as to afford protection is a general statement of the surrogate representation rationale. Taxes are treated differently from other regulatory measures because most often they are used for revenue and not regulatory purposes and therefore can tolerate a broader notion that different classifications can be attributed to "buying off' the surrogate representation of domestic producers. Finally, this reading maintains a healthy relationship between Article III and Article XX. Under this reading, Article III deals only with whether the surrogate representation mechanism has been impaired; Article XX tells us whether the purpose for doing so outweighs the loss of political participation by surrogates for the foreign producers. This process-based account of the national treatment provision suggests that most WTO analysts have been looking in the wrong direction when seeking a meaning for Article Ill. Previous analysis of the national treatment 2004] IND. INT'L & CoMP. L. REv. provision has assumed that it had some substantive content, and that it therefore required the analyst to balance the values of a free trade system against the values inherent in a regulatory system. That approach set up a natural clash between the WTO and the trade regime, on the one hand, and national regulatory sovereignty on the other. The surrogate representation rationale, by contrast, does not assume that the national treatment provision elevates any substantive value (such as free trade) above other substantive values. It assumes only that the WTO enforces a process value-the process value of allowing those who represent the interests of foreign producers to do so without being co-opted in the course of the legislative process. The implications of the shift from a substantive account of the national treatment provision to a procedural account are significant. The two accounts have vastly different implications for our understanding of international federalism, for the role of the WTO, and for the division of lawmaking authority between the members of the WTO and the panels and Appellate Body. The substantive view of the national treatment provision inevitably posits a conflict between free trade values and national regulatory agendas. It assumes that the WTO and its members are engaged in a prolonged debate about how to interject free trade values into national regulatory agendas, and therefore results in a search for tests that will achieve the correct "balance" between regulatory autonomy and the international trading system. Accordingly, the various tests that have been devised to chart the border between trade values and regulatory values reflect the political proclivities of the analyst and the personal trade-offs made by the analyst when considering the appropriate goals of regulation. This has led to a wide and indeterminate range of opinions about how the balance should be struck. The substantive view of the WTO therefore naturally raises questions about the scope of global federalism, the process by which trade values were made ascendant over other values within that federal system, and the appropriateness of moving decision-making authority away from democratic governments. Inevitably, therefore, the substantively based view leads to distrust of the WTO by those who support sovereignty and national regulatory autonomy, and puts the friends of the WTO in a defensive position. It leads to attacks on the WTO for displacing national regulatory choices with trade values enforced by an unelected and distant group of decision-makers. By contrast, the process-based view appeals to values that are widely shared and that do not threaten the goals of regulatory regimes. The process- based view suggests that the only value at stake in national treatment cases is one that is widely shared, rather than contested-and that is the value of having the interests of those affected by a regulation be represented within the lawmaking forum that enacts the regulation. This value not only appeals to widely shared values of participatory lawmaking, but it is one that regulatory bodies can meet easily without sacrificing their regulatory goals. They need simply respond to affected interests directly rather that by reducing the [Vol. 14:3 WTO NATIONAL TREATMENT objection to the regulation by domestic producers. (And even if they do not, they can still justify the interdiction of surrogate representation if they meet the standards of Article XX). The substantive view of the national treatment provision also raises troubling issues about the division of power between the member states and the Appellate Body that are less significant under the process-based view. Naturally, some interpretive function is inevitable. There is simply no way for the WTO members to adopt a code against protectionist measures. A significant issue under the substantive view is the legitimacy of delegating lawmaking power to the unrepresentative and unaccountable members of the Appellate Body. By what right do they seek to overturn national legislation and how do they develop the expertise to evaluate and balance purpose and effects? The process-based view avoids this difficulty by positing that the role of the Appellate Body is not to balance trade values against local regulatory values, but simply to police the process by which national regulatory decisions are made, a role which is more highly suited to unelected and unrepresentative decision-makers. The Appellate Body has wisely limited its review under Article III to issues of process, for these are the kinds of decisions that bodies like the Appellate Body have a comparative advantage in addressing. The question of who should make which decisions in a federal system is a significant one. In the context of the national treatment provision-just as in the context of the dormant Commerce Clause-an underlying issue is who should have the burden of seeking federal legislative review of the judicial interpretation. Under the Commerce Clause, Congress can always overturn the decision of a court because Congress is the ultimate arbitror of interstate commerce. Therefore, as many accounts of the dormant Commerce Clause emphasize, judicial review is really determining which party should have the burden of going to Congress to have the legislation overturned. The same is true under the national treatment provision, of course, but the stakes are even higher, for, as many have noted, the possibility of overturning the decision is weaker. Decisions of the panels and Appellate Body cannot be overturned unless all the members agree to a new standard. The process-based view, more than the substantive view of national treatment, respects this aspect of WTO lawmaking by limiting the scope of review to process based matters and therefore preserves the authority of the members to set the substantive standards under which they will be governed. Finally, the substantive view of the national treatment provision restricts national autonomy in ways that the process-based view does not. Presumably, if national values conflict with the trade values of the WTO because the effect of regulation on trade outweighs the national values, then no change in the legislation can preserve the national values unless the measure can be justified under Article XX. By contrast, under the process-based view, national regulation is not permanently forestalled or subjected to the tests of Article XX. A national regulatory body that runs afoul of the national treatment 2004] IND. INT'L & COMP. L. REv. provision can continue to address the regulatory need, reformulating its regulatory process to restore the potency of the surrogate representatives. Korea can still tax liquor and the United States can still regulate to clean the air. In other words, along several important dimensions, the process-based view of the national treatment provision is superior to the substantive view. It is a more conservative function for an international institution to perform; it fits more closely to the institutional competency of judges of the panels and Appellate Body; and it appeals to values of participatory democracy that are more widely accepted and value neutral than the substantive values that underlie free trade. At the same time that the process-based review fits more comfortably within the lawmaking structure of the WTO, it is not an impotent or pro-forma exercise. The review remains searching; it is just not substantively intrusive. By serving to preserve surrogate representation in the lawmaking process, this review performs the same important role that process performs in any law- making setting. It gives those who are adversely affected a stake in the debate and in the outcome. It reduces tensions and bad feelings generated when opportunities to participate are limited. It helps knit together the policy- making machinery that in our system of nation-states is otherwise territorially confined. Most of all, it insures that economic interdependence is managed in a way that encourages participatory interdependence so that the tensions from economic interdependence do become political tensions as well. By avoiding the parochial, it protects the ideal of participatory democracy in a global economy. [Vol. 14:3