Mission Possible: Reciprocal Deference between Domestic Regulatory Structures and the WTO

Volume 40 Number 1Winter 2007

Article 5

Mission Possible: Reciprocal Deference between Domestic Regulatory Structures and the WTO

Elizabeth Trujillo [email protected]

Part of theLaw Commons

Mission Possible: Reciprocal Deference Between Domestic Regulatory Structures and the WTO

Elizabeth Trujillot

Introduction .....................................................

I. Protectionism and Domestic Regulation-Is There a

Palpable Difference under Article III of GATT? ........... 211 A. Reconciling Domestic Regulation with Free Trade ...... 211 B. Article III at Cross Purposes ........................... 214 C. EC-Asbestos: A Step in a Different Direction ........... 221 D. Legitimate Domestic Regulation and National Treatment

II. Finding Answers in WTO Member States .................

Violations-The Slippery Slope ........................

225 227 A. The U.S. Dormant Commerce Clause Analogy ......... 227 B. The Regulatory Model versus the Antidiscrimination

M odel ................................................

C. Deference to the Transparent Regulatory Structures of

M em ber States ........................................

111. Lessons for National Treatment from the NAFTA Chapter

11 Tribunals ............................................. A. Deference and "Like Circumstances" under Chapter 11

of N AFTA .............................................

B. A Closer Look at Article 1102 ......................... C. Comparing System s ................................... IV. Creating Bridges to Regionalism ..........................

A. WTO Adjudication of Domestic Regulation After

M ethanex .............................................

B. Reciprocal Deference Encourages Bridge-Building ...... C. Applying Reciprocal Deference .........................

C onclusion ......................................................

239 240 247 249

249 256 259 262

Introduction

Many think that a true global society is impossible to achieve either because nations ignore the decisions of international tribunals or choose simply to dismiss their obligations under international agreements altogether. Sadly, this rings true at times, at least in the context of international trade. U.S. compliance with free trade agreements, including the General Agreement on Tariffs and Trade (GATT), can be challenging when faced with regulatory policies that, on the one hand, appear protectionist and, on the other, are authorized within the context of legitimate domestic regulatory processes.

In adjudicating domestic regulatory policies, the World Trade Organi-

zation (WTO) decides the validity of those policies according to the GATT' and its Covered Agreements;2 in doing so, however, the international agency directs member states in matters of domestic regulatory policy.3 In this way, the WTO functions as an internal player in the formulation of regulatory policy as much as an external adjudicator of those policies. In a similar way, the WTO cannot ignore that its decisions and policies are ultimately enforced within those same domestic regulatory structures.4 WTO panels and domestic regulatory institutions can become more aligned and work toward a more cohesive multilateral trade regime in a recognizable procedure of reciprocal deference.

This paper asserts that a superior approach to adjudicating internal regulatory policy encourages deference to domestic government structures that implement legitimate domestic policy, but only to the extent that such deference illuminates for the WTO panels the non-protectionist purpose of a particular measure. While internal policies that directly discriminate among products based on nationality or origin clearly violate the national treatment requirements of Article III of GATT,5 it is the facially neutral

regulatory measures with protectionist and discriminatory effects that are more difficult to assess, even within transparent regulatory processes.6 In characterizing the national treatment analysis along the lines of the products themselves, however, the WTO panels ignore the possibility that a regulatory measure may have legitimate domestic purposes, even if it places "incidental burdens" on trade.7 A procedural mechanism that places the burden on those domestic regulatory processes closest to the measure in question, namely the legislature, would encourage transparent implementation of each measure and ensure state compliance with WTO agreements. Furthermore, WTO panels must consider the impact of their decisions regarding national treatment on regional tribunals. In certain instances, some measures may be better assessed at the regional level. The Chapter 11 tribunal under the North American Free Trade Agreement (NAFTA) recently stated in Methanex Corporation v. Government of the United States that provisions adjudicated under GATT were not per se "to be transported to investment provisions" under NAFTA.8 This is an important assertion by the NAFTA tribunal, especially considering that several prior NAFTA decisions previously deferred to WTO interpretations of national treatment under GATT in adjudicating alleged national treatment violations under NAFTA.9 The extent to which regional tribunals, and subsequently domestic courts, should and will defer to GATT jurisprudence is unsettled. Clearly, Methanex represents both an attempt to distinguish the regional tribunal adjudicatory process from that of the WTO in the NAFTA context and a reminder of the lack of coordination that exists between the two regimes.10 In proposing a reciprocal deference approach for the WTO's handling of regulatory measures, I will draw from parallel debates in U.S. constitutional law, as well as decisions from the NAFTA tribunals. 11 There

are also important distinctions to be drawn among these legal regimes.

A WTO Panel decision lays out the problem at the regional level. The High Fructose Corn Products case began as a dispute under the NAFTA agreement, first as an antidumping case and more recently as an investorgovernment dispute.12 The U.S. investor-claimant, Corn Products International, brought a claim against the Mexican government in response to a federal tax passed by the Mexican legislature on soda bottlers using high fructose corn syrup as a sweetener instead of sugar.13 No such tax was passed for using sugar.14 The investor, a U.S. corporation with the largest market share within the Mexican high fructose corn syrup industry, claimed that the Mexican government passed the tax in a way that violated NAFTA Chapter 11 national treatment requirements.'5 The case is currently pending before a NAFTA Chapter 11 arbitration panel. Interestingly, on June 10, 2004, the U.S. government submitted a claim to the WTO Panel to decide whether this Mexican tax violates Article III because it treats "like and directly competitive or substitutable products" differently. 16 Unlike respondents in similar WTO cases, Mexico did not address whether the products were "like" or not. Rather, Mexico alleged that the WTO did not have jurisdiction over the NAFTA tribunal in deciding this case, and it claimed that, as a developing country, it was entitled to special and differential treatment.17 Mexico also argued that the tax was justified because of an ongoing sugar dispute involving Mexican access to the U.S. sugar market.18 In 2005, the WTO Panel decided that the tax measure was indeed a violation of Article III of GATT.19 The extent to which this WTO

decision will impact the outcome of the Chapter 11 case is unclear, but it is likely that, after Methanex, the WTO decision will be less persuasive than in prior NAFTA decisions, in which the tribunals looked to GATT decisions for guidance.20 Notably, the damages under a NAFTA Chapter 11 dispute differ from those available in a WTO dispute.21 Although the WTO may be correct in finding the tax protectionist, this case raises interesting questions regarding the adjudication of national treatment disputes, WTO removal from issues better settled at the regional level, and WTO deference to regional tribunals and the domestic political processes when dealing with questions of legitimate internal regulatory policies.

The GATT agreement, as amended to include other Uruguay Round agreements, is the document that lays out the obligations of member states regarding tariff and subsidy reductions.22 Article III of GATT recognizes that regulatory measures outside the scope of tariffs and subsidies can amount to non-tariff barriers to trade (NTBs) and, therefore, undermine multilateral efforts to reduce trade barriers more generally.23 One of the primary goals of Article III is to invalidate domestic regulatory measures, including taxes that amount to NTBs.24 Such barriers undermine the economic goals of free trade.25 The GATT Dispute Settlement Body, which today falls under the aegis of the World Trade Organization, and the WTO Appellate Body are the adjudicating bodies for determining whether the actions of member states are in compliance with the obligations under the WTO agreements.26 Generally, in dealing with domestic regulatory measures under Article III of GATT, the WTO panels focus on the "likeness" of the products in question.27 More specifically, the panels care about the extent of the competitive substitutability between the products in question.28 WTO panels make their determinations according to the covered agreements and do not look to other sources of international law except in

rare instances.29 Many specific regulatory measures, including those dealing with health, technical and licensing matters, intellectual property, and investment measures, are also adjudicated under the covered agreements.30 They also do not generally defer to the decision-making power of regional tribunals in making their determinations.31 Regional agreements such as NAFTA32 incorporate mechanisms through which Parties may convene regional tribunals to remedy alleged violations.33 NAFTA, however, does specifically allow for Parties to settle some issues before the WTO dispute settlement body.34 Furthermore, NAFTA Parties tend to defer by analogy to WTO determinations in interpreting provisions under NAFTA, and NAFTA tribunals have recognized this deference.35 For this reason, WTO determinations can impact adjudication of similar matters at the regional level and, therefore, WTO panels must factor this into their own adjudicatory formulae.

Recent WTO panels generally deal with domestic regulatory measures in Article III of GATT, where the panels tend to focus on the "likeness" of the products affected and the discriminatory nature of the regulatory measure in question.36 Furthermore, unlike in the context of U.S. commerce clause jurisprudence, WTO panels do not distinguish between measures with discriminatory effects and those applied even-handedly but with "inci-

dental burdens" on trade.37 There are other types of regulation that the "like products" test cannot properly settle. For example, the test does not distinguish between regulatory measures in place to create economic development and those that are adopted merely to oust competition.38 Furthermore, the test cannot accommodate measures deriving from traditionally regulated domestic markets that begin to deregulate in certain sectors.39 Free trade agreements thrive, at least in the short-term, in part by encouraging competition in those market sectors that have traditionally been regulated. In the case of Mexico, for example, the sugar industry has been highly regulated in the past and remains regulated in some aspects.40 Likewise, the Mexican government has required that its primary telecommunications service provider, Telefonos de Mexico, S.A. de CV. (Telmex) open its networks for use by other authorized Mexican companies, two of which are U.S. affiliated.41 With respect to U.S. electricity markets, there has been a push for pro-competition policies that convert a traditionally regulated monopoly into a "partially" regulated market.42 Within this type of regulatory framework, even U.S. antitrust law seems unable to distinguish precisely between anti-competitive policies and pro-competitive ones.43 The WTO panel's attempt to define "anti-competitive" practices in Mexico-Measures Affecting Telecommunications Services may have over-reaching and unintended effects.44 At the same time, the Article III "like products" test is even less equipped to settle the issue. WTO panels run the risk of invalidating otherwise legitimate regulatory policy, particularly via facially

neutral measures that affect or disadvantage outside competition but do not necessarily directly or intentionally protect inside competition. In doing so, the WTO may unwittingly take on a more sovereign role, setting trade norms and creating to domestic regulatory measures.

law with respect

In a regulatory model approach to the WTO, this increased sovereignty for setting standards may be a desirable outcome.46 Uniform standards regarding health, labor, safety, and environmental issues could help harmonize the domestic regulatory measures of member states.47 Even if uniform regulatory standards are a long-term aspiration for free trade and globalization, uniformity may produce resentment among members. Ultimately, the WTO depends on its member states to implement and abide by these standards.48 Under a regulatory model approach for eliminating NTBs, WTO panels will inevitably alienate member states from the GATT multilateral regime and encourage a push towards regionalism.49 Furthermore, it will widen the present disconnect between the GATT regime and trade regimes under regional trade agreements.50 The challenge for WTO panels is to respect legitimate domestic regulatory policies while still implementing international standards that allow for predictable adjudication of domestic regulatory policy. Although it is not the role of WTO panels to decide whether domestic measures are legitimate as a matter of substantive law, they must have some procedural basis upon which to assess their legitimacy. A regulatory model would be weak in formulating these assessments. Procedural mechanisms that place the burden on member states to maintain transparent political processes and prove the need for regulatory measures would promote coordination among the WTO, member states, and regional tribunals. Under such reciprocal deference, the WTO may have an impact at the domestic level rather than solely as an outside enforcer of free trade.

Part I of the paper will discuss protectionism, as seen through the eyes of WTO dispute resolution bodies. It will consider important issues arising under domestic regulatory law that Article III of GATT is not equipped to resolve. It will also take a closer look at the principles of national treatment, and demonstrate that WTO applications of this doctrine are at cross purposes. On the one hand, WTO panels have exhibited a willingness to delve into the purpose of domestic measures to determine whether they are protectionist.51 On the other hand, recent WTO panels also seem more inclined to focus on the competitive substitutability of the products and on whether the measure treats them differently.52 Under the latter approach, the legitimacy of the regulatory measure itself is no longer factored in the in treatment equation. This part will demonstrate national EC-Asbestos, the WTO panels seem more willing to consider the regulatory measure in question as part of the larger equation for "like products" analysis without burdening themselves with the task of finding a legitimate purpose behind the regulatory measure. Finally, Part I will examine the difficulty in differentiating national treatment violations from legitimate regulatory policy.

that

Part II attempts to find solutions examining similar debates in U.S. law. It will take a closer look at the questions that the "like products" test of Article III of GATT cannot address. It will draw upon comparisons to the U.S. dormant Commerce Clause jurisprudence in dealing with these issues and propose ways that the WTO panels may find these comparisons useful. I suggest, however, that because the WTO and the U.S. courts deal with different sets of concerns, important distinctions between the two systems in dealing with trade must be also be considered. Finally, this part will recommend that WTO panels distinguish between facially neutral measures with discriminatory effects and those with only "incidental burdens" on trade. It will examine two main approaches for the WTO: the regulatory model and the anti-discrimination model.53 This part will borrow from both models in proposing that WTO panels defer to domestic regulatory processes in dealing with those facially neutral, internal measures that have other legitimate purposes. In doing so, the WTO panels should not decide the legitimacy of these measures as a matter of substantive domestic law. When in doubt, however, they should look to the domestic regulatory processes in place and ask the legislatures to clarify the legitimate purpose and demonstrate that the measure in question is in fact the "least restrictive means" of implementing domestic policy and otherwise complies with the state's WTO obligations.54 In this way, the WTO encourages domestic internal regulatory processes to be transparent but

also places the burden on those same domestic political institutions to prove the need for the policy.

Part III will consider parallel discussions regarding domestic regulation under NAFTA. It will also illuminate relevant questions unanswered by the WTO panels' interpretation of Article III of GATT. It will emphasize that the WTO panels can better align the interests of the multilateral trade regime with those of regional agreements by considering the effects of its decisions on the outcomes of similar disputes at the regional level. Furthermore, Part III will demonstrate the willingness of the NAFTA tribunals to look to domestic regulatory processes when deciding matters of regulatory policy. It will also show that Chapter 11 NAFTA tribunals implement at times a balancing test which factors the need for the regulation at the domestic level, against its potential interference with the objectives of NAFTA.55 There are lessons here for the WTO panels in adjudicating regulatory policy under Article III of GATT.

Finally, this article will conclude with an in-depth look at the impact of Methanex on the future decision-making power of regional tribunals in the context of national treatment violations. More specifically, Part IV will consider the impact that Mexico-Tax Measures may have on the NAFTA tribunal's decision on whether Mexico would owe damages to the U.S. for a national treatment violation.56 It will bring together lessons for the WTO panels in the U.S. dormant Commerce Clause jurisprudence and the NAFTA jurisprudence in dealing with internal regulatory measures. It will also demonstrate that it is in the questions asked where the WTO panels may learn the most. Furthermore, it will propose an alternative application of Article III, which begins by focusing on the impact of a legitimate regulatory measure on the competitiveness of the products and incorporates the measure into the "like products" test. Finally, in dealing with facially neutral measures, Part IV will propose a reciprocal deference approach, in which WTO panels defer to domestic regulatory structures in finding answers to questions of legitimacy. Part IV does not propose to give the domestic governments the last word on issues of legitimacy; it does assert, however, that proof of legitimate measures can be better ascertained at the domestic level, close to the institutions that implement the policy.

Therefore, by implementing procedural mechanisms in assessing inferences of legitimacy, WTO panels can encourage more transparent domestic internal regulatory structures and make legislatures accountable by requiring proof of both the need of the measure and the absence of alternative means more aligned with its commitments under WTO. Furthermore, WTO panels can decide that certain matters would be better addressed at the regional level, since regional agreements may better address the issues at hand, and the regional tribunals are closer to the domestic regulatory

structures in question. In conclusion, through a different means of adjudicating regulatory measures under Article III, the WTO panels can ensure transparency within domestic regulatory structures and promote coordination and cohesiveness between the multilateral trade regime of GATT and the regional and domestic regimes.

1. Protectionism and Domestic Regulation-Is There a Palpable

Difference under Article III of GATT?

A. Reconciling Domestic Regulation with Free Trade

In principle, Article III of GATT was designed to eliminate domestic measures that in their application would discriminate between imported and "like" domestic products and thereby effectively protect domestic producers at the expense of foreign producers.57 However, in their desire to eliminate NTBs, the WTO Panels are setting norms for defining national treatment violations based solely on product distinctions, their competitive substitutability, and equal treatment.58 This kind of analysis, while focused on preserving equal competitive opportunities and encouraging market access, fails to consider that virtually any domestic regulatory measure is protectionist to some degree.59

Traditional notions of globalization and the spread of free markets have recently encountered obstacles which, ironically, regulation can help amend. For example, when Japanese car companies decide whether to build a plant in the United States or in Ontario, Canada, they consider the costs of providing private health care and training U.S. employees versus Canadian employees.60 Interestingly, the Canadian government's ability to offset healthcare costs has been a decisive factor in companies' choices to invest in Canada rather than in the United States.61

Presumably, legitimate regulatory policy supports the public welfare of a state and accounts for market failures.62 Furthermore, some domestic

57. See JACKSON, supra note 25, at 279-80. 58. See Panel Report, Japan-Customs Duties, Taxes and Labeling Practices on Imported

policies that impact consumer preferences may actually contribute to creating new markets. For example, a gas guzzler tax placed on domestic and imported automobiles exceeding a specific miles per gallon ratio is a fiscal measure intended to encourage consumers to buy more fuel-efficient cars and, in turn, push manufacturers to produce more fuel-efficient cars.63 United States-Taxes on Automobiles presented this problem.64 In this case, the panel determined the "likeness" of the automobiles in question based on the "aim and effect" of the fiscal measures.65 The panel focused on whether the measure was designed to discriminate between domestic and foreign automobiles.66 It used an "inherence" principle, determining that the measure was not aimed at protecting a domestic industry because the ability to produce fuel-efficient cars was not "inherent" in any one of the parties involved.67 Such a measure was not aimed at protecting the U.S. automobile industry, even if it impacted some of Europe's less fuel-efficient In the spirit of Malt-Beverages, the panel looked at cars at the time. whether there was a bona fide regulatory purpose, and whether the measure would protect the domestic industry so that it would retain its competitive advantage.68 The panel decided that the fiscal measures were applied even-handedly, and in a non-discriminatory fashion, and that they furthered legitimate regulatory purposes other than to protect the U.S. car industry.

Furthermore, domestic policy intended to promote competition in certain markets traditionally regulated or "protected" by the state may fall prey to WTO scrutiny. Conflicts may arise between free trade agreements and domestic policy when domestic industrial policy protects a domestic industry temporarily from foreign competition.70 The "Emergency Action" provision of GATT Article XIX actually contemplates and upholds certain

types of trade restrictions that help a distressed domestic industry restructure itself.71 These protections apply, however, primarily to measurable injury due to imports.72 Of particular difficulty is when these conflicts arise outside the jurisdictional scope of Article XIX of GATT.73 Where domestic government intervention is legitimate under domestic law in the form of regulation or market participation, domestic industry may be inherently eligible for more protection than foreigners.74 Antidumping laws are not structured to deal with these issues.75

Article III of GATT is not equipped to deal with these issues either. Free trade agreements thrive on implementation of domestic pro-competitive policies in traditionally regulated markets. One example lies in the U.S. regulatory structure for electricity markets. U.S. electricity firms traditionally regulated both at the federal and state levels have been subjected to deregulatory policy in certain sectors and states.76 There has been inconsistency among U.S. courts in determining which kind of anticompetitive behavior is still regulated and which is not, producing counterintuitive for pro-competition policies. For example, expansive applications of the U.S. state sovereign immunity doctrine that exempts state-regulated industries from antitrust liability may actually further the effects of capture in dominant suppliers already established in a traditionally regulated market.77 The "like products" test of Article III, as imple-

results

mented by WTO panels, focuses on the products themselves rather than the regulatory measure or the regulatory framework.78 With this narrow focus, the test cannot cope with domestic deregulatory policies or the impact that such policies may have on foreign competition within those markets.

In Mexico-Tax Measures, it is not irrelevant that the need for access to U.S. markets for surplus sugar production arose from a regulatory structure in Mexico that traditionally protected Mexican sugar growers.79 The Mexican sugar industry was attempting to cope with the transition of opening its market to private investment, particularly foreign investment.80 Mexico alleged that U.S. non-compliance with prior agreements to provide market access to the Mexican sugar industry justified its adoption of a sugar tax to handle the sugar surplus problem.8' In the NAFTA context, the dispute in the Methanex case arose out of a California regulation banning the use of methanol in reformulated gasoline because the state government found the additive to be a health and environmental risk.82 Neither the WTO panel nor the regional tribunals seem equipped to deal with these kinds of questions, and the "like products" test does not make room for these inquiries either. Perhaps the founders of the WTO and other regional tribunals did not design the panels to solve these issues at all. Yet some regional agreements, such as NAFTA, contemplate trade in areas traditionally subject to government regulation, sometimes even more so than in the United States.83 The WTO Panel, in its adjudication of Article III, should recognize these complexities within the framework of its covered agreements and encourage member states to better deal with the issues at the domestic level.

B. Article III at Cross Purposes

The GATT dispute settlement and appellate bodies have struggled over the years with the application of the "like products" test under Article III of GATT.84 Paragraphs one and two of Article III are the internal tax provisions.85 The former provides a general principle of non-protectionism

with respect to internal taxes and regulations.86 The latter is much more specific as to the obligations of member states when dealing with internal taxation and "like domestic products."'7 Article III, paragraph four addresses non-fiscal measures that treat products of foreign origin "less favorably" than "like products" of national origin.88 In writing Article II1, the drafters sought to prevent internal measures that would discriminate between imported and "like" domestic products, creating a "constructive tariff' on imports and effectively benefiting domestic producers.89 The differences in the first two sentences of Article III's second paragraph result in a two-tiered objective with problematic results.90 On the one hand, Article III attempts in nature.91 In doing this, however, it also seems to disallow measures that have protectionist effects, even if the measures do not necessarily discriminate between imported products and like domestic products.92 Because this seems contrary to Article IIl's purpose, the panels have focused on the first sentence of the second paragraph, which is clearly intended to prevent discriminatory measures on imported products also produced domesti-

internal measures that are protectionist

to eradicate

cally.93 In principle, the scope of the fourth paragraph is broader than the second paragraph.94 However, they work in tandem since fiscal measures, as well as non-fiscal ones, may frustrate trade in similar ways. In sum, the purpose of Article III is to prevent member states from passing fiscal and non-fiscal regulatory measures that are intended to protect domestic products at the expense of their foreign counterparts. Arguably, GATT does not go so far as to eliminate all market protection; rather, it performs a monitoring function and guarantees that such protections work within the strict parameters of GATT as enforced by the WTO.95 Therefore, part of the purpose of Article III is also to ensure a level playing field when it comes to competition by eliminating NTBs that protect domestic production at the expense of a comparable foreign import.96

While interpretations of GATT clarify that an internal tax levied on both imported and "like" domestic products is an internal tax even though the tax is levied on imported goods at the time of importation, GATT provides no guidance where "like" products are not involved.97 At times, GATT panels focused solely on the products themselves, specifically on whether products were treated differently and were competitively substitutable.98 In this way, the panels narrowed the scope of the test while broadening the meaning of "like." At other times, the panels broadened the scope of the test by considering the purpose and the effects of the regulatory measure at hand in assessing the "likeness" of the products themselves.99 This perspective shifts the focus from the competitiveness of the

products to the legitimacy of the regulatory measure-a question that WTO panels may not be in a position to assess. Under either approach, however, the claimants, in asserting a national treatment violation, must first establish whether a "like product" is involved.100

For the most part, WTO decisions regarding national treatment under Article III distinguish between regulatory measures that discriminate on their face against foreigner (de jure discrimination) and facially neutral measures that do not discriminate based on origin but nevertheless have a "disproportionate impact" on foreign goods (de facto discriminatory).10 However, the same rules apply to both: if the goods or services in question are "like," they are in violation of Article III unless they can be justified as "necessary" violations under Article XX'02 In determining "likeness" under Article III, however, panels have collapsed facially non-discriminatory measures that only have protectionist effects on imports with those that have protectionist effects but are justified as furthering other legitimate regulatory objectives.103 Arguably, panels deal with this distinction under Article XX, which creates specific exemptions for a measure that would otherwise be a national treatment violation. Under Article XX, the respondent has the burden of proving that the measure is "necessary" per Article XX.104

Until European Communities-Measures Affecting Asbestos and Asbestos Containing Products,'05 GATT panels were divided in dealing with national treatment problems between the approach in Japan -Tax on Alcoholic Bever-

ages (the 1987 decision)106 and that in United States-Measures Affecting Alcoholic and Malt-Beverages.107 In Japan-Alcoholic Beverages, the U.S., European Union, and Canada claimed that a Japanese tax scheme, in which certain imported alcoholic beverages were subject to a higher internal tax than the domestically produced schochu, discriminated against foreign exporters to Japan.108 Whereas Japan offered a highly restrictive interpretation of "like product," claimants proved that imported spirits were "like products" to the Japanese spirits, schochu, by virtue of being "directly competitive and substitutable."109

the possibility of legitimate domestic measures

Despite this tendency of WTO panels to focus solely on the competitive substitutability of products, and the unequal treatment of a regulatory measure as to those products, one panel decision demonstrated the WTO's willingness to consider legitimate domestic measures as part of the equation. United States-Measures Affecting Alcoholic and Malt Beverages considthat, though ered discriminatory as between domestic and imported products, may not necessarily violate Article III because of its protectionist intent.10 This case involved U.S. tax schemes on alcoholic beverages at both the federal and state levels, as well as other local restrictions on transportation and distribution of alcoholic beverages. I11 Most notable was a Mississippi excise tax that set a lower rate for wines made from a specific grape, the vitus rotundifola grape.112 This measure benefited the Mississippi producers that relied heavily on the grape. 113 Therefore, the effect of the tax measure was to provide lower tax rates on domestic beer and wine than on "like" imported products.'14

Unlike 1987 Japan-Alcoholic Beverages, the panel took a different approach in determining whether this tax violated Article III, an approach seemingly more consistent with the purpose of Article III. It did not solely rely on the determination of whether the products in question were "like."

Rather, in making this determination, the panel considered the purpose of the domestic measure in question, and whether it was enacted with the "aim and effect" of protectionism.1 15 Although the panel invalidated the tax as violating Article III, it arrived at this conclusion in a much different manner than it did later in 1996 Japan-Alcoholic Beverages. The panel's analysis of whether the measure was discriminatory against "like domestic products" did not only turn on whether the products shared certain physical characteristics or were "directly competitive or substitutable." Instead, it stated that this determination must consider whether the purpose of the measure was itself protectionist.116 The panel concluded that the purpose of Article III was not to prevent member states from "using their fiscal and regulatory powers [in ways] other than to afford protection to domestic production."117 Furthermore, it stressed that the focus of the "like products" analysis must be whether such a distinction furthers the purpose of a nondiscriminatory regulation that is not intended to be a non-tariff trade restriction, even if such regulation has the effect of restricting trade.118 Though the panel in U.S. -Malt Beverages provided no guidance as to the types of measures that could be legitimate because they are not intended to "afford protection," the panel did recognize that distinctions can be drawn between regulatory measures intended to protect and those having protectionist effects but serving other legitimate regulatory objectives.119

Of course the most obvious problem with the analysis in U.S.-Malt Beverages is that it requires a Panel to discern the legislative intent of the measure.120 The Panel's decision regarding the Mississippi tax was aided by a U.S. concession that the differing tax schemes served no regulatory purpose "other than to subsidize small local producers.'121 However, the U.S. -Malt Beverages decision reinforces the notion that Article III does not

necessarily require absolutely equal treatment of "like" products, particularly when domestic measures are adopted for reasons other than to protect domestic industry.'22 Rather, it targets those measures that treat imported goods less favorably than "like" domestic goods for the purpose of granting an advantage to the domestic goods at the expense of the imported one. 123 The panel recognized the importance of making "the like product determination in the context of Article III... in such a way that it not unnecessarily infringe upon the regulatory authority and domestic policy options of contracting parties."124

The subsequent 1996 Appellate Body decision of Japan-Taxes on Alcoholic Beverages, however, decided that the "aims and effects" approach was not the proper interpretation of Article III's second paragraph.125 Instead, it reinforced the panel's decision that the Border Tax Adjustments were the guiding criteria, along with competitive substitutability.'26 Therefore, the focus in determining "substitutability" or "likeness" shifted to the physical characteristics of the products, their end-uses, consumer tastes, and the tariff classification of the products.127 If a regulation treats the "like products" differently, notwithstanding the purpose of the regulation, then it violates Article III. This analysis does not contemplate the purpose of the regulation.128 Rather, it collapses facially discriminatory measures with facially neutral ones that have only discriminatory effects; focuses on the equal treatment of the "like products"; and ignores the possibility of legitimate, non-protectionist regulatory objectives.129 While the Appellate Body

stated that the first sentence of Article 111:2 should be read in light of the purpose laid out in paragraph one-"to avoid protectionism"-its emphasis on the Border Tax Adjustments and the "likeness" of the products seems to undermine this purpose.130 The WTO panels' analysis does not distinguish between, on the one hand, regulatory measures meant to protect domestic products, and on the other, measures that are facially non-discriminatory and intended to further other legitimate "non-protectionist" goals, but, nevertheless, have incidental protectionist effects. The common result is that where there is unequal treatment, the panels almost always find a national treatment violation. This emphasis on equal treatment of "like products" seems to miss the point. Article III is concerned with measures that afford protection to domestic producers at the expense of foreigners.131 It does not ensure equal treatment, especially where equal treatment to foreigners and domestic producers may be impossible with certain regulatory measures.'32 It was not until EC-Asbestos that the WTO panels found an alternative to the model espoused in 1996 Japan-Alcoholic Beverages.1

C. EC-Asbestos: A Step in a Different Direction

In European Communities-Measures Affecting Asbestos and AsbestosContaining Products,134 the Appellate Body opened the door for expanding the meaning of "like products" without reinstating the "aims and effects" test of U.S. -Malt Beverages. Canada brought this case against France, challenging a French decree that prohibited importation of asbestos and products with asbestos fibers and imposed penalties for violations of the

decree.135 The claimants alleged that this decree violated the Agreement on Technical Barriers to Trade ("TBT Agreement") as well as various provisions of GATT, including Article III. The European community asked the Appellate Body to reconsider the panel decision.136

The European community claimed that the panel erred in finding chrysotile asbestos fibers to be "like" polyvinyl alcohol cellulose and glass fibers ("PCG fibers"), and that it erred in finding chrysotile-cement products, containing asbestos fibers, to be "like" fibro-cement products.137 In its reading of Article III, the European Community made an interesting argument that the need to look into the legitimacy of regulatory policy is itself within the purpose of Article 111.138 That is, the European Community claimed that the panel mistakenly ignored the "very reason why the [d]ecree single[d] out asbestos fibers; namely, the fact that asbestos fibers [were] carcinogenic." 139 It claimed further that the comparison should have been based on a larger category than the products themselves; specifically, the French decree denied competitive opportunities equally to "all carcinogenic asbestos [fibers]."'140 In this way, the determination of "likeness" would not establish the discrimination; rather, a conclusive finding of "likeness" would only lead a panel to the next question regarding the impact on trade due to the discriminatory effect of the regulatory measure. This reading of Article III allows a panel to balance a measure's impact on trade against its regulatory significance.

The Appellate Body agreed with the European Community's understanding of "like products.'141 It stated that the question of "likeness" between the products involved is actually a question "about the nature and extent of a competitive relationship between and among products."'142 It also stressed, however, that the kind of competitive relationship matters, for just because products are in "some competitive relationship" does not in itself amount to the products being "like."'143

The importance of the Appellate Body's decision in this case is in narrowing the scope of the "like products" test under Article III. It allows for comparison of categories of "like" imported products against categories of

"like" domestic products.144 The factor that distinguishes these categories may include the regulatory measure itself, as long as it is legitimate and contains a non-protectionist objective that furthers legitimate domestic goals. In the EC-Asbestos case, the Appellate Body characterized the Border Tax Adjustments as a "framework" within which it may initially find "likeness" among the products but, nevertheless, described this initial finding as non-conclusive.145 The analysis went one step further: it considered the regulatory measure in question and decided whether it was connected to the core of the product, which would distinguish it from another product with similar end-uses.146 That is, in the case of chrysotile asbestos fibers, the fact that they were carcinogenic, and that the regulatory measure was intended to address this harmful feature, indicated that these fibers were in fact different from the PCG fibers.147 Therefore, they were not "like products." '48

In EC-Asbestos, the WTO Appellate Body did not focus on the purpose of the regulatory measure, which would come closer to the "aims and effects" test, but rather incorporated the legitimate regulatory measure in its determination of whether chrysotile asbestos fibers and PCG fibers were "like."'149 It examined the regulatory measure's effects without scrutinizing the legislature's motive or intent.150 Similar its ruling in 1987 Japan-Alcoholic Beverages, the Appellate Body focused on the products themselves, particularly their competitive relationship and market substitutability.151 In EC-Asbestos, however, it found that the effect of the regulatory measure on the targeted products' market could impact their substitutability.'52 That is, consumer choices among the fibers would change if they knew that one contained asbestos and the other did not. Consumer choices, therefore, would affect fibers' end-uses as well. Finally, the measure had a reasonable nexus to the distinctive aspect of the product that was in need of regulation, namely the carcinogenic nature of chrysotile asbestos. 153

to

EC-Asbestos is a step in the right direction for WTO recognition of legitimate regulatory measures. The WTO demonstrated its willingness to be a forum for domestic concerns while promoting market access and a reduction of trade barriers. More recently, WTO panels seem more inclined to allow different treatment of "like products", provided the different treatment is something less than less favorable treatment to the foreign product. For example, in Dominican Republic-Measures Affecting the Importation and Internal Sale of Cigarettes, the Appellate Body approved the panel's conclusion that a bond requirement, levied on both domestic and foreign cigarette producers for tax payment purposes, did not violate Article 111:4, even though it imposed an "extra burden" on foreign importers due to their smaller tax bracket.154 The Appellate Body focused on the bond requirement's "detrimental effect" on foreign producers and concluded that it was not enough of a burden to amount to less favorable treatment.155 However, the Appellate Body did not define "detrimental effect" or distinguish it from "less favorable treatment."'56 It reasoned that the smaller market for foreign cigarettes effectively diminished the tax liability of foreign producers, even though the bond requirement formally treated both foreign and domestic producers identically.'57 To avoid confusion, the Appellate Body should have stated that, regardless of the disparate effects of the bond requirement, the foreign and domestic cigarettes were essentially "unlike" since the size of their market, which determines their tax liability, was different. Instead, the Appellate Body seemingly narrowed the application of Article 111:4 with respect to de facto or facially non-discriminatory measures by finding that differential treatment did not amount to treating foreign cigarette producers less favorably than domestic producers. 158 The Appellate Body, in implementing this "detrimental effect" standard, suggested that a legitimacy determination of the regulatory measure is itself necessary. Unfortunately, this approach may further confuse the application of Article III.

With respect to alleged national treatment violations, the WTO can consider the legitimacy of domestic discriminatory measures that are facially neutral and balance them against the burdens on free trade when deciding the "likeness" of products.159 The WTO panels are not, however,

in the best position to assess these issues as a matter of substantive law, especially with respect to regulatory measures that use science as measure for legitimacy.160 Instead, the WTO panels can incorporate procedural mechanisms that encourage deference to domestic regulatory structures, which may be illuminating on questions of legitimacy.

D. Legitimate Domestic Regulation and National Treatment Violations-

The Slippery Slope

While enforcing its non-discrimination policies, the WTO panels still recognize the need for regulatory measures to foster domestic economic development.16' They do not, however, establish criteria to distinguish measures intended for domestic economic development from those that are merely protectionist and, therefore, a national treatment violation. In a 1958 GATT Report entitled Italian Discrimination Against Imported Agricultural Machinery, the panel considered the Italian government's establishment of a revolving credit fund, which granted special credit benefits for the purchase of Italian agricultural machinery.162 The fund was to exist for a limited period of time in order to encourage the purchase of Italian agricultural machinery;163 however, this favorable credit scheme did not exist for the purchase of foreign agricultural machinery.164 The panel considered Italy's arguments that such a law existed to help develop the Italian economy.165 It concluded that the original intention of the GATT was not "to limit the right of a contracting party to adopt measures which appeared to it necessary to foster its economic development or to protect a domestic industry, provided that such measures were permitted by the terms of the General Agreement."'166 Though the panel found that the domestic objectives could be achieved in ways more in line with Italy's commitments under GATT, it distinguished between domestic measures passed solely to protect domestic production and those that aimed to promote domestic economic development.167 It recognized that GATT does not prevent domestic governments from passing laws that foster domestic economic development.'16 The panel did not provide guidelines for distinguishing

between laws that foster domestic economic development from those that are per se protectionist. In failing to draw these distinctions, the WTO cannot clearly articulate the difference between a "permissible" regulatory measure under GATT and one that amounts to protectionism.

that violate national

The understanding that the WTO will not tolerate anticompetitive practices by domestic governments or private entities is evident in its condemnation of measures treatment obligations. Anticompetitive practices seemingly authorized by domestic governments can nevertheless be difficult to assess, even at the domestic level. In Mexico-Measures Affecting Telecommunications Services, the WTO panel attempted to deal with the definition of "anticompetitive practices" for the first time.169 This interesting case concerned the telecommunications industry in Mexico, which had been traditionally regulated prior to legislative efforts to open it to competition, much like in the U.S. telecommunications industry.170 The dispute between the U.S. and Mexico dealt with Mexican laws that governed the supply of services in the context of telecommunications, a market that continues to be "partially" regulated in Mexico.17' The U.S. claimed that interconnection rates charged by Mexico's largest telecommunications service supplier, Telmex, failed to comply with the General Agreement of Trade in Services (GATS) and its Reference Paper, which required "cost-oriented" and reasonable rate bases.172 Also, the U.S. claimed that the Mexican government failed to prevent Telmex from engaging in anticompetitive behavior.173 While finding Mexico in violation of its commitments, the panel defined "anticompetitive practices" on a broad basis.174 It did not, however, deal with either the Mexican policy of regulating the telecommunications sector and allowing for market dominance or the degree of regulation that would be construed as "anticompetitive."'75 Mexico attempted to show that the policies construed as "anticompetitive" were actually procompetitive.176 Although this case did not deal specifically with the issue of "like products," it did raise issues surrounding "legitimate" regulatory policies designed to deregulate

sectors where a natural monopoly existed, and the ability of WTO panels properly to adjudicate such policies at the supranational level.

II. Finding Answers in WTO Member States

A. The U.S. Dormant Commerce Clause Analogy

Under the current WTO interpretation of Article III, the focus is not on legitimate regulatory policy but on the products themselves, and whether or not they are "like" according to the criteria set out by the WTO panels. Once a panel determines that a "likeness" between products is demonstrated, and the regulatory measure treats one differently from the other, it is very difficult for a respondent to prove that there is no intent "to afford protection to domestic production." In other words, after 1996Japan Alcoholic Beverages, a WTO panel will treat measures that are facially discriminatory (de jure) and neutral measures with discriminatory effects (de facto) in the same manner.177 If discriminatory, either kind of measures will violate Article III of GATT.178

In this way, U.S. law may aid WTO panels in their application of Article III. Under U.S. law, legal principles governing the application of the dormant Commerce Clause account for state measures that regulate in facially non-discriminatory ways to achieve a legitimate local purpose.179 Also, the U.S. Supreme Court will invalidate state laws that are facially discriminatory towards other states' products.180 The presumption that a facially discriminatory state measure is unconstitutional can be rebutted by showing that it furthers some other legitimate non-economic state interest or is necessary and no alternative means are available.'8' Where a state measure is facially neutral, the Court examines its purpose and

effects, very much like the "aim and effects" test under US-Malt Beverages.182 Unlike the WTO, however, U.S. case law distinguishes between measures that have a discriminatory effect on outsiders and those that are implemented even-handedly but create "incidental burdens" on interstate commerce.183 In the former scenario, a state may rebut the presumption of a commerce clause violation if it demonstrates a legitimate state interest, and that the measure is necessary to further that interest.184 With the latter cases, the Court incorporates a balancing test weighing the local benefits of implementing such a measure against the costs to interstate commerce.185 In this way, the Court addresses those measures that have an "incidental" or "indirect" burden on interstate commerce but do not entail an obvious protectionist purpose. 186 In applying the balancing test, the Court must first find a legitimate state purpose advanced by the measure. 187

Whereas U.S. law considers several factors dealing with discriminatory intent to give a competitive advantage to local producers, GATT law focuses on the likeness of the products, including their substitutability, irrespective of whether there is a local production producing the same products.188 In the context of the dormant Commerce Clause, although a legitimate state interest is not presumed and the tendency is to favor "free trade" among the states, the U.S. Supreme Court does not ignore the possible existence of a legitimate state interest.189 Rather, it attempts to balance

therefore should be eliminated.195 The U.S. Constitution authorizes Congress to regulate interstate commerce through the Commerce Clause.196 The dormant Commerce Clause is a judicially created inversion of the commerce power, recognizing state power to regulate intrastate commerce as long as it does not interfere with Congress' authority to regulate interstate commerce.197 However, the idea behind tempering state power in this way is the preservation of Congress' ultimate authority to regulate interstate commerce, which, in turn, is of interest to all the states as part of a federalist system.198

The GATT and the more modern WTO do not claim to have such regulatory power over global commerce. Instead, the WTO adjudicates the trade arrangements of its member states and guarantees a multilateral framework that will facilitate free trade and encourage economic interdependence.199 Many will argue that this multilateral regime is in fact a federalist system or, at minimum, the GATT principles are a form of constitutional law.200 Whether the WTO is viewed as a lawmaking institution or just the guarantor of globally accepted trade principles, the WTO panels undeniably adjudicate to preserve a multilateral trade regime created and accepted by its members, and to facilitate free trade among member states.201 It is arguable if this is also enforcement of any one law. Despite this fundamental difference in the two systems, they can still learn from one another while implementing the principles in ways more aligned with the institutions they are trying to preserve.

What can the WTO panels learn from the debate surrounding the dormant Commerce Clause? Perhaps the most important lessons are found where regimes deal with similar sets of concerns.202 In both regimes, the

challenge is not the facially discriminatory regulatory measure, but the "indirect" regulatory, or the facially neutral, measure that affects or disadvantages outside competition.203 Where the WTO may learn the most from the dormant Commerce Clause jurisprudence is in the questions asked in deciding which domestic regulatory measures are protectionist enough to be in violation of Article Ill. In addition, it is important for the WTO to recognize that sovereign nations are in the best position to regulate domestically and to address social objectives, even if they employ the least efficient means of achieving those goals.204 The WTO must also ensure, however, that domestic measures-even those with legitimate objectivesdo not hamper the goals of free trade. A first step in achieving some balance between the two sets of interests is in recognizing that some regulatory measures have the competitiveness of outsiders in only "incidental ways." Furthermore, when ruling on those measures with clear discriminatory effects, the WTO should consider whether the measure offers any alternative means of achieving the same objective. Perhaps this is where the WTO should begin its distinctions and weave them into the "like products" test rather than within the stricter Article XX.

legitimate domestic goals

affect

that

Article XX creates an exemption for domestic regulatory measures that would otherwise be a GATT violation but are justified because they deal with the protection of health and safety, exhaustible resources, public morals, and other areas.205 GATT panels have struggled to reconcile Article XX with Article Ill, especially with measures that are facially neutral but have discriminatory effects.206 Some of this conflict has been based on the language of GATT. Article XX states that some measures are justifiable as long as they are not applied arbitrarily or in unjustifiably discriminatory ways, and if they do not amount to "disguised restrictions on international trade."207 Presumably, all violations of Article III, however, are "disguised restrictions on international trade"; therefore, it seems difficult to find any justification under Article XX. Finding compliance under Article III does not render Article XX inoperative.208 Notably, however, Article XX is very limited in scope and places the burden on the respondent to prove that the given regulatory measure is the least restrictive, and that no alternative

exists.209 The "necessity" requirement in Article XX is confined to regulatory measures that are inherently compelling.210 For these reasons, respondents should have the opportunity to prove the legitimacy of the given regulatory measure in the context of Article III rather than under Article XX, which presumes a national treatment violation.

In making determinations regarding the legitimacy of a regulatory measure, WTO panels should look to transparent domestic regulatory structures for answers. More specifically, they should require those institutions closest to the measure, namely the legislatures, to answer to the WTO regarding the need for such a measure while remaining committed to their obligations under GATT. To some extent, the Chapter 11 tribunals under NAFTA are already dealing with regulatory measures in this way. However, in deciding more normative issues such as the meaning of "like" for purposes of national treatment, NAFTA tribunals tend to defer to WTO decisions.

B. The Regulatory Model versus the Antidiscrimination Model

The creators of the GATT system did not intend to create an international institution that would replace domestic governance; however, they did want to discourage member states from engaging in protectionist activities.212 The founders intended to create a multilateral trade regime that would promote peace through economic interdependence and sustainable development.213 What is unclear is if they intended to follow a regulatory model approach by establishing an international agency that would take an autocratic stance on issues of free trade law, or if they envisioned a fluid system that would set standards but ultimately let the member states execute and incorporate these standards into their national systems, an approach much more in line with the antidiscrimination model.214 Assessing the intent of the original founders is perhaps less important than establishing a realistic framework through which WTO panels may best deal with domestic regulatory measures. One approach is the "regulatory

model", which portrays the WTO as a regulatory commission that proposes universal standards for regulatory measures dealing with issues of labor, environment, health, and safety.215 One of the problems with this model is its vulnerability to special interest capture.216 Under this approach, the WTO panels adjudicate according to their own universal standards.217 At some level, the WTO does this by addressing issues that only come under the Covered Agreements.218 Even in embracing those universal standards, however, the WTO ultimately relies on domestic laws and local democratic frameworks in its application and enforcement of those standards.219 The present regulatory model fails to take into account that in dealing with domestic measures, the WTO must work with domestic regulatory structures by permeating them with a web-like network of influence to ensure compliance with GATT agreements.

John 0. McGinnis and Mark L. Movsesian, who advocate an antidiscrimination model for the WTO rather than a regulatory model, discuss a procedural mechanism within the larger context of developing a "jurisprudence of covert protectionism.1220 They set up a legal and procedural framework using an antidiscrimination model of the WTO, whereby WTO panels may more clearly assess "covert protectionism" at the domestic level.221 However, the jurisprudence they develop applies primarily to regulatory measures in the areas of labor, the environment, health and safety.222 The authors also propose "determinate rules," based on the requirements of transparency,223 performance orientation, and consistency, to distinguish legitimate regulations by member states from illegitimate ones.224 They also rely, however, on an "objective evidence requirement" for member states seeking to prove the legitimacy of such measures.225 Such objective evidence would be loosely based on scientific

V Vol. 40

studies.226 This approach, while helpful for measures whose risk assessments can be achieved through scientific studies, may fail to assess other legitimate reasons for regulations which science cannot measure, such as public choice or economic development. They recognize that even measures meeting the three requirements may nevertheless create an inference of covert protectionism.227 For these instances, the authors outline a "procedure-oriented approach" that uses "objective evidence" based on scientific risk assessments to determine a measure's legitimacy.228 It is here where they embrace a deferential approach to national risk assessment mechanisms as well as a "least restrictive means test."'229

C. Deference to the Transparent Regulatory Structures of Member

States

The GATT was only one part of a larger, faulty engine, the International Trade Organization (ITO),230 which was designed to administer the GATT.231 Against the backdrop of a devastated post-war Europe, the Bretton Woods system, the framework within which the ITO, the GATT, and other bodies would function, was originally intended to propel global economic integration and development.232 The ITO failed to materialize, however, leaving the GATT as the ill-equipped, main international trade agreement.233 The WTO was therefore formed as an institutional remedy for the failures of the ITO.234

Unlike most modern federalist systems, the WTO does not obtain its authority through general principles of sovereignty.235 Rather, it gathers its powers from the willingness of member states to participate in the global process; thus, the WTO has the role of adjudicator and facilitator rather than agent, of a sovereign entity.236 In this role, WTO panels must balance the domestic interests of its member states with the interests of the

global community.237 In doing so, WTO panels must enforce the economic goals of the global community while also preserving non-economic objectives of the WTO, including development, peace, transparency, and just distribution of economic assets through economic interdependence and multilateralism.238 Ultimately, given that the WTO can only work within the parameters of its Covered Agreements, the WTO has trouble addressing these multilateral goals alone.239 For issues not addressed by these agreements, the WTO system must also rely on its member states to promote these multilateral goals at the domestic level.

Rather than distort the balance of the multilateral trade regime with strict formalism and broad interpretations of disguised trade restrictions, the WTO panels can become a forum for discussion of legitimate regulatory measures by adopting a contextualist approach in its adjudication of those measures.240 In its current strict formalism, the WTO arguably empowers itself with "constitutional" authority, setting legal norms for a network of governmental participants that have not necessarily delegated such authority to this international institution.24' This is not to say that the beginnings of such a constitutional framework have not been set; however, this is a separate, albeit related, question. In entering into GATT and later acquiescing to the panel decisions of the WTO, member states did not delegate to the WTO absolute authority to rule and decide normative questions regarding free trade.242 Most of the sovereign member states already have administrative processes to manage trade issues, and many have entered into their own regional and bilateral trade agreements.243 Member states do, however, look to the WTO both for political support to criticize restrictive trade measures of their counterparts and for affirmative rulings

on those policies.244 In this way, the WTO is as much an internal player in domestic trade and regulatory policy-making as it is an external adjudicator.245 Inherent in facilitating free trade, however, is enabling member states to make and regulate domestic decisions for their constituents.

The institutions closest to the execution of the domestic regulatory measure are in the best position to judge the legitimacy of regulatory measures as a matter of domestic law. At the same time, those very institutions may also be most interested in furthering those measures for reasons other than acceptable public purposes.246 Therefore, objective criteria should drive domestic institutions' implementation of regulatory measures. Although scientific risk assessment for measures dealing with health, safety, or the environment creates an objective framework within which the "need" for such measures can be assessed,247 science is often inconclusive. Therefore, other public assessments of risk involving the democratic, deliberative process are also important in establishing such measures.248 Questions of legitimacy are complicated even further when international tribunals must judge regulatory measures as a matter of international law or, in the case of the WTO, GATT law.249 For example, the Appellate Body in EC-Measures Concerning Meat and Meat Products (Hormones), found that European bans on U.S. beef imports on grounds that national regulations prevented hormone use in beef was in violation of the SPS Agreement because the European Community failed to base its measure on a scientific risk assessment.250 The Appellate Body did not, however, go so far as to find the hormone ban to be a disguised trade restriction.251 Because of

domestic political processes and the interest groups at play, an international trade standard that is too lenient in upholding regulatory measures may in fact further protectionist measures that become disguised trade restrictions.252 Furthermore, certain special interest groups at a domestic level may also aid in eliminating protectionist measures by lobbying against those interest groups that espouse protectionist measures.253 This "Darwinian strategy" assumes that the export interest groups are the strongest groups and will eventually defeat those groups pushing for protectionist measures. It is unclear whether such a battle maximizes consumer welfare. The question of who should determine the legitimacy of a regulatory measure is an important one and is beyond the scope of this paper. Member states are understandably reluctant to delegate too much authority to an international institution such as the WTO in dealing with domestic issues.254 Suffice it to say, however, the WTO panels do not have the resources to investigate the domestic intent of regulatory measures or their legitimacy.255 They can only assess these questions substantively according to the Covered Agreements.256

Therefore, it is within domestic regulatory processes that protectionist measures may best be curtailed. It is also within domestic governmental structures that local needs and interests may be determined. As explained in the Sutherland Report, the WTO "sets some strict disciplines under which governments may choose to respond to special interests.''257 The WTO provides a framework within which trade liberalization and economic interdependence may flourish, and it helps to shape internal policy. In addition, compliance under regional agreements adds another level to this multilateral framework. Regional tribunals tend to look to WTO jurisprudence in applying provisions under these agreements.258 Coordination under these circumstances is a further challenge, and it is important that the WTO panels recognize these factors when adjudicating regulatory measures. To do so, the panels can defer as a matter of procedure to the domestic institutions closest to the measures at hand in assessing issues of legitimacy. Some may ask if a strategy that encourages deference to domestic institutions just incites institutions to pass protectionist measures

under the guise of legitimate and transparent democratic processes. In other words, are we not putting the fox in the henhouse?259 While there is concern for protectionist capture at the local level, the regulatory model does not necessarily eliminate it.260 Furthermore, special interest groups at the domestic level will battle against each other, and the democratic process will ultimately decide how best to comply with free trade commitments.261 A regulatory model may run the risk of capture as well, particularly if it does not encourage fluidity and change. One must remain realistic about the extent of WTO enforcement of its decisions. The WTO's ability to influence the behavior of its member states rests primarily in its ability to authorize members to retaliate against other members not in compliance with GATT agreements.262 Furthermore, the WTO may shape pubto the lic perception of a nation's willingness to remain committed international regime as a whole.263 Ultimately, however, the WTO relies on its member states to enforce its decisions.264 Too much retaliatory behavior by member states leads to isolationism, which is exactly what the GATT was intended to prevent.265 The reciprocal deference approach I propose is just one piece of the larger puzzle of cohesiveness and coordination between WTO adjudicatory power and domestic implementation of WTO policies. It is not the final solution, but it provides a procedural safeguard for WTO adjudication of internal regulatory measures. For example, if the panel finds that as a matter of GATT law the measure in question is facially neutral yet has some protectionist effects on trade, the panel should ask the legislature to demonstrate the necessity of the measure. It will then be up to the WTO panel to make a final determination as to compliance with GATT and the Covered Agreements.

III. Lessons for National Treatment from the NAFTA Chapter 11

Tribunals

Through Article XXIV of GATT, the WTO system permits the formation of regional integration agreements that may incorporate their own dispute resolution mechanisms.266 Whereas WTO panels do not defer to regional tribunals or domestic court decisions in adjudicating the GATT provisions, NAFTA tribunals do tend to defer to WTO decisions when

applicable.267 More specifically, the signatories of NAFTA ("Parties") reaffirmed their rights and obligations under GATT.268 Article 102 states that the Parties "shall interpret and apply" the NAFTA treaty provisions "in law."'269 Moreover, accordance with applicable rules of international national treatment obligations for market access under NAFTA are specifically in accordance with Article III of GATT.270 As under the GATT, however, the notion of "like circumstances" is applied in various contexts throughout NAFTA, and NAFTA tribunals struggle to understand its meaning.

A. Deference and "Like Circumstances" under Chapter 11 of NAFTA

In dealing with national treatment issues for foreign investments, Chapter 11 NAFTA tribunals apply the operative phrase "like circumstances" to determine national treatment violations under Article 1102.272 Article 1102 states that NAFTA Parties must treat the investments of other Parties in a manner

no less favorable than it accords, in like circumstances, to its own investors with respect to the establishment, expansion, management, conduct, operation, and sale or other disposition of investments.273

Defining what exactly constitutes "in like circumstances" is the challenge, for the NAFTA treaty never explicitly defines this phrase. Prior NAFTA cases have dealt with the issue and for the most part have chosen to look to "like products" under Article III of GATT for answers.274 It is, nevertheless, important to remember that, while Chapter 11 arbitration panels do establish binding determinations with enforceable awards, just as the WTO, they are not bound by stare decisis.275 Therefore, each panel decides every case on its own merits and on an individual basis according to all of the surrounding circumstances.

By allowing investors to sue party-governments, thereby safeguarding foreign investment, NAFTA marked a major development in international

investment law.276 Pushed by their concern about the threat of government expropriation or nationalization of their investments, the parties were quite thorough in drafting Chapter 11 with specific provisions for damages and arbitration of disputes under international arbitration rules of ICSID or UNCITRAL.277 The drafters likely never imagined, however, that litigants would question regulatory measures with such vigor as under Chapter 11 investor disputes. Despite their deference to Article III of GATT, the tribunals have also established some interesting ways of dealing with domestic regulatory measures.278 A more detailed examination of the Chapter 11 decisions will reveal similarities and differences with comparable decisions by the WTO panels under Article III.

B. A Closer Look at Article 1102

In S.D. Myers, Inc. v. Canada, the U.S. investor S.D. Myers, an Ohio corporation that treated Canadian PCB waste in its U.S. facility through Myers Canada, brought a Chapter 11 action against Canada, challenging Canada's prohibition on polychlorinated biphenyls (PCB) exports.279 Among other allegations, S.D. Myers claimed that the Canadian export restrictions violated national treatment provisions of Art. 1102.280 The claimant asserted that the export ban effectively barred it from competing for PCB waste disposal business in Canada, while not similarly restraining Canadian companies, which had access to processing facilities located in Canada.28' The tribunal agreed with the investor.282 In making this determination, however, it looked to the intent of the Canadian export restrictions to determine whether it was protectionist.283 After reviewing records from the Canadian Ministry of the Environment at both the federal and provincial levels, the tribunal found that the intent of the export ban was to bar the export of PCBs to the U.S. "to protect the Canadian PCB disposal industry from U.S. competition.'284 The tribunal recognized that there could have been an "indirect environmental objective," but it held that there were alternative means of reaching these objectives more in line with

With respect to the issue of national treatment under Article 1102 of NAFTA, the tribunal deferred to the WTO's analysis of Article III under the decision Japan Alcoholic Beverages decision.286 The tribunal stated that a finding of "like" under GATT did not necessarily invalidate the measure; instead, it would just "set the stage for an inquiry into whether the different treatment of situations found to be 'like' [was] justified by legitimate public policy measures that [were] pursued in a reasonable manner.'287 The tribunal also looked to the OECD practices in dealing with foreign investments in "like situations.'288 The tribunal considered the larger objectives of NAFTA, which included the Environmental Side Agreement that expressed the parties' concerns for the environment as well as the elimination of trade barriers among the three countries.289 The tribunal stated:

The assessment of "like circumstances" must also take into account circumstances that would justify governmental regulations that treat them differently in order to protect the public interest." The concept of "like circumstances" invites an examination of whether a non-national investor complaining of less favorable treatment is in the same "sector" as the national investor.290

In comparing this interpretation of "like circumstances" to the WTO panels' interpretation of "like products," the NAFTA tribunal appears to be much more willing than the WTO to understand "likeness" as falling within certain categories, which may or may not include certain regulatory measures.291 This type of interpretation is reminiscent of the "aims and effects" test and closer to the EC-Asbestos application of Article III than the narrower test in Japan Alcoholic Beverages.

The tribunal in S.D. Myers stated that the determination of whether there was a national treatment violation under Chapter 11 of NAFTA turned on two factors: (1) whether the measure has the effect of discriminating on the basis of nationality; and (2) whether the measure is facially discriminatory along the same lines.292 This broad sweep on domestic reg-

ulatory measures was not without some deference to the legitimate goal of the measure-to comply with the Basel Convention and protect the environment.293 While recognizing the possibility that the measure was legitimate, the tribunal did require proof that the measure was the "least restrictive" means of achieving the regulatory goal.294 This burden was not met and the tribunal decided that there were alternative means of achieving the same results without having such a discriminatory effect and violating 1102.295

After S.D. Myers, it appeared clear that NAFTA Chapter 11 tribunals, while recognizing legitimate regulatory policies, would tend to coincide with WTO panels in collapsing measures with discriminatory effects with those intended to discriminate between products of nationals and nonnationals. At the same time, S.D. Myers demonstrates the willingness of NAFTA Chapter 11 tribunals to incorporate legitimate regulatory measures into the analysis of "like circumstances," focusing the comparison on larger categories that may incorporate those measures rather than the competitive substitutability of the products involved.296 In making these determinations, however, the NAFTA tribunals also seem more inclined to defer to the domestic regulatory framework from which the measures arose.297 This framework can help tribunals begin to understand the legitimacy of the measures in place and better determine whether alternative measures more aligned with the objectives of NAFTA could have been implemented. Another Chapter 11 dispute between Canada and the U.S. took deference one step further.

Pope & Talbot, Inc. v. Canada,298 while agreeing with S.D. Myers in one respect, added a new and useful dimension in dealing with domestic regulatory measures in the context of national treatment.299 In Pope & Talbot, the investor Pope & Talbot, Inc., a publicly traded corporation incorporated under the laws of Delaware and engaged in the manufacture and sale of softwood lumber through its Canadian wholly owned subsidiary Pope & Talbot International Ltd., brought a NAFTA Chapter 11 claim against the Canadian government for 1102 violations as well as other Chapter 11 violations.300 The majority of the sales of softwood lumber by its Canadian

subsidiary was being exported to the United States. The backdrop to this case was the 1996 Softwood Lumber Agreement signed by Canada and the U.S., under which Canada was required to place softwood lumber on the Export Control List under Export and Imports Permits Act.30' Under this Agreement and relevant laws, Canada required federal export permits on each exportation to the U.S. of softwood lumber first manufactured in certain Canadian provinces covered under the Agreement.302 The Canadian government instituted fee structures for the issuance of permits for the U.S. export of softwood lumber that was first manufactured in the covered provinces for quantities above the established base in a given year.303 Provinces placed on the export control list included British Columbia, Alberta, Manitoba and Quebec.304 All exports to the U.S. from these provinces would require an export permit and the payment of fees.305 The claimants alleged that the implementation of these fee structures and permit requirements was intended to protect Canadian softwood lumber and hurt the U.S. investment.306 The NAFTA tribunal issued a final merits award dismissing claims under Article 1102 and concluded that the regime was a reasonable, albeit flawed, implementation of the Canada-U.S. Softwood Lumber Agreement.307

In defining national treatment in Pope & Talbot, Canada tried to incorporate a "Disproportionate Disadvantage Test", which focused on the determination of whether there were any Canadian-owned investments accorded the same treatment as the investor.308 The tribunal looked at GATT cases applying Article 111:4 and determined that panels in those cases did not look into whether "imported products were subjected to disproportionate disadvantage.'309 Canada clarified that if the measure was dejure discriminatory, there would clearly be an 1102 violation.310 If the measure was facially neutral (de facto), however, the question should have been whether, behind the neutrality, "the measure disadvantage[d] the foreign-owned investment" in a disproportionate way.311

In making its final decision, the tribunal looked to its decision in S.D. Myers, where Canada similarly alleged that the issue in 1102 was "whether the practical effect of the measure [was] to create a disproportionate benefit for nationals over non-nationals .... 312 The Pope & Talbot tribunal disagreed with Canada and clarified that even in S.D. Myers, the focus should be on whether the claimant and its Canadian competitors were "in like circumstances."313 Once the answer was in the affirmative, "the finding of a denial of national treatment was a foregone conclusion."314 Weighing disproportionate advantages was not, therefore, required.

This emphasis on the "likeness" of the position of the investments involved is again reminiscent of WTO panels' adjudication of "like products" in the context of Article III. In rejecting the "Disproportionate Advantage" test, however, the NAFTA tribunal took an interesting approach to discerning "like circumstances." It stated that in deciding which investments were "in like circumstances," it would consider the entire background on the softwood lumber trade between the U.S. and Canada leading up to this dispute.315 In this way, the tribunal deferred to the domestic regulatory processes at play. It concluded that the comparison of "like circumstances" should be between the foreign owned investment and the domestic investment "in the same business or economic sector."'316 In making these comparisons, the tribunal emphasized that whether they were "like" in the latter respect was just the first step.317 The more important question was whether there were differences in their treatment; if there were, then the presumption should be that in fact there was a violation of national treatment.318

So far, the analysis is much like the WTO application of Article III in the context of "like products." Ultimately, the NAFTA tribunal in Pope & Talbot was concerned, like the WTO panels, with the way in which application of a regulatory measure affected the competitive relationship of investments "in like circumstances."319 However, the Pope & Talbot tribunal took the analysis one step further: it converted the presumption of national

treatment into a rebuttable presumption. The tribunal concluded that if the differences in treatment "have a reasonable nexus to rational government policies that 1) do not distinguish, on their face or de facto, between foreign-owned and domestic companies, and 2) do not otherwise unduly undermine the investment liberalizing objectives of NAFTA," the measure is in compliance with 1102.320 Essentially, this test balances the legitimacy of the governmental policy against the burdens it places on the free trade objectives of NAFTA. What constitutes a legitimate policy is left up to the discretion of the tribunal; however, it may be framed along the lines of (1) whether the measure discriminates (de jure or defacto) based on origin, and (2) whether the measure is reasonably aligned with the governmental policy.32' With such a test, the tribunal admittedly runs the risk of having to decide on issues of legitimacy. The Pope & Talbot tribunal was clear, however, in placing the burden on the respondent for justifying regulatory policies "not motivated by preference of domestic over foreign owned investments."322 More importantly, the tribunal clarified that the issue of "justified discrimination" should be included in determining whether the investments are "in like circumstances."323 This approach allows for distinctions to be made along the lines of larger categories, which is similar to the suggested approach in S.D. Myers. In contrast, the WTO approach would require that justification for discriminatory measures occur within the stricter parameters of Article XX.324

In another Chapter 11 dispute, the NAFTA tribunal again considered domestic regulatory processes to determine whether the regulatory measure in question was legitimate. In Gami Investments, Inc. v. The Government of the Mexico, the Chapter 11 NAFTA tribunal considered whether governmental expropriations of a U.S. investor's sugar mills amounted to expropriation under NAFTA and a violation of national treatment under 1102.325 Though the tribunal did not apply the same balancing test as in Pope & Talbot, it looked closely at the administrative processes in Mexico that decided to expropriate the sugar mills.326 Under the heading of "maladministration" in its Final Award, the tribunal considered the regulatory framework surrounding the sugar industry in Mexico, an industry that the Mexican government has traditionally protected.327 After owning Mexico's sugar mills for nearly twenty years, the Mexican government implemented

a privatization program through the Sugarcane Decree of 1991.328 The purpose of this program was to promote private investment in the Mexican sugar industry, while allowing the government to regulate all the different aspects of the sugar mill industry, from planting to refining.329 The program established guidelines as to exclusive territorial agreements for the sale of sugar to certain mills, sugar production ceilings, reference prices and price adjustments, and the establishment of export quotas.330 Among their allegations, the claimants asserted that the Mexican government failed to properly implement these regulations, thereby necessitating a shutdown of the sugar mills.331 The tribunal deferred to the regulatory framework itself in deciding whether the investor could have a cause of action against the government, but it clarified that it would only consider whether Mexico's actions (or omissions) were a violation of international law.332 The tribunal refused to determine whether the regulations in place were legitimate and recognized that the regulatory framework called for governmental supervision.333 The tribunal instead focused on whether the government intentionally caused the failures in the sugar program, and whether those failures directly caused injury to the U.S. investor.334

On the question of national treatment, the GAMI tribunal did not focus on whether the nationalizations of the sugar mills were intrinsically wrong because of their protectionist effects. Instead, it looked to whether the nationalizations were implemented differently vis-A-vis the U.S. claimant compared to other investors "in like circumstances.'335 It noted that some of the sugar mills that were expropriated were domestic corporations with no foreign investors.336 After a brief analysis and rejection of GAMI's argument that discrimination had occurred, the tribunal decided that the motivation behind Mexico's decision to expropriate certain mills while sparing others was based on the government's regulatory policy that considered it in the public interest to salvage financially troubled sugar mills to prevent their insolvency.337 It stated that "Mexico perceived that mills operating in conditions of effective insolvency needed public participation

in the interest of the national economy in a broad sense.'338 The NAFTA tribunal looked to the regulatory structures in place for clarification, deferring to the judgment of the government and holding that because the ends pursued were legitimate, the means used were not discriminatory.339

The GAMI decision was informed by the tribunal's initial decision that the policy was legitimate, thereby making it possible to defer to the government's regulatory policy throughout the opinion. If nationalization was of only foreign-owned investments, the question may have been more difficult for the tribunal. In the latter scenario, the tribunal might have made a determination along the lines of Pope & Talbot. That is, it would have had to first determine whether the foreign-owned and domestic investments were "in like circumstances.'340 Then, the tribunal would decide whether there was a reasonable nexus to a rational nondiscriminatory governmental policy and balance this against the burdens to the free trade objectives of NAFTA.341

C. Comparing Systems

Despite the challenging endeavor of unveiling consistencies in international trade tribunals, the NAFTA tribunals seem to be aligned with legal applications of the U.S. dormant Commerce Clause in a few ways. First, they recognize the distinction between facially discriminatory measures (de jure) and neutral measures with discriminatory effects (de facto).342 NAFTA Tribunals and WTO jurisprudence under Article III are similar in this respect. Furthermore, all three would agree that facially discriminatory measures would clearly violate national treatment unless otherwise justifiable. For example, a Chapter 20 NAFTA tribunal determined that a "blanket refusal" by the U.S. to comply with its NAFTA obligations and allow access to Mexican trucking firms was a violation of national treatment.343 The tribunal conceded that the U.S. regulatory framework already in place to review these applications for access could deny Mexican firms access if they did not comply with U.S. licensing and environmental standards, as long as those standards were not arbitrary or discriminatory in unjustifiable ways.344 The tribunal treated the U.S. actions as facially

discriminatory measures intended to protect domestic trucking companies at the expense of Mexican trucking firms.345

Second, NAFTA tribunals consider that even those neutral measures with discriminatory effects may serve non-protectionist and legitimate regulatory goals.346 The WTO panels would likely concede this point as well 347 However, the NAFTA tribunals, albeit a in the context of Article 111. more ad hoc system than the WTO Appellate Body, seem more aligned with U.S. courts in finding that these measures may be valid for purposes of national treatment only if there are no less restrictive alternatives available to achieve the desired goal.348 The WTO panels would settle these questions under the stricter framework of Article XX.349

this consideration against the burdens

Third, NAFTA tribunals will look for a nexus between a facially neutral regulatory measure and the legitimate government objective and then balance to free trade under NAFTA.350 It is unclear if NAFTA tribunals place much emphasis on distinguishing between facially neutral measures with protectionist effects and those that only "incidentally burden" the NAFTA objectives. For the purposes of this article, it is sufficient to note that NAFTA tribunals presume a national treatment violation where a measure is intentionally protectionist or has protectionist effects.351 More importantly, NAFTA tribunals recognize the possibility of legitimate regulatory policies and defer to domestic regulatory processes in deciding their legitimacy under international law.352 In Pope & Talbot, for example, the tribunal gave much deference to the legitimacy of the 1996 Softwood Lumber Agreement critical in the dispute.353

This deference, however, has its limits. In Metalclad Corporation v. United Mexican States, Metalclad Corporation, a U.S. investor, alleged that a municipal order requiring the investor to stop all building activities because it lacked appropriate licenses was a violation of Mexico's obligation under Chapter 11 of NAFTA.354 According to the U.S. investor, the Mexican federal government had assured Metalclad that no such licenses were required.355 Aside from the interesting federalism problem that the tribunal attempted to resolve, the Chapter 11 tribunal incorrectly decided that the lack of transparency in the domestic regulatory processes of Mexico, both at the federal and state levels, were a violation.356 Despite errors in the tribunal's interpretation of Articles 1105 and 1110, the interesting outcome of the early Chapter 11 case demonstrated the uninhibited willingness of the tribunal to delve into the Mexican regulatory framework at play. Mexican counsel argued that the tribunal was neither equipped to decide on matters of Mexican domestic law nor was it the role of the tribunal to do so.357 The tribunal did, however, begin by looking at the domestic regulatory framework at hand in deciding whether the measures arising from that framework violated Mexican obligations under NAFTA.358 While transparency is an important objective for NAFTA, the Chapter 11 tribunals will not enforce it as an obligation under NAFTA.359

IV. Creating Bridges to Regionalism

A. WTO Adjudication of Domestic Regulation After Methanex

Let us return to the initial illustrative problem concerning the Mexican sugar industry and high fructose corn syrup (HFCS). In Corn Products, a U.S. investor, Corn Products International, Inc., brought a Chapter 11 claim against the Mexican government for passing a tax on the soft drink industry for beverages containing fructose, specifically HFCS-55, that allegedly was intended to discriminate against the claimant and protect the Mexican sugar industry.360 The Mexican government adopted the tax,

which became effective on January 1, 2002, as an amendment to an older excise tax scheme, Impuesto Especial Sobre Producci6n y Servicios ("Special Tax on Products and Services"), which placed taxes on particular products and services such as tobacco and alcoholic beverages, in order to supplement income tax revenues.36' The tax affected those industries using HFCS-55 in their beverages-primarily the soda bottler industry.362 It did not affect products using sugar as the sweetener.363

The history of sugar cane producers in Mexico reflects the paradox of Mexican policy toward U.S. investment in sweeteners. On one hand, the Mexican government encourages US investment in and production of high fructose corn syrup; on the other, the Mexican government vigorously protects its traditional sugar cane production. U.S. investors, through Mexican subsidiaries, are the chief producers of HFCS, a product derived from yellow corn that must be imported from the U.S because Mexico primarily grows white corn.364 Though HFCS must undergo several processes, an important component of the product, HFCS-55, is also imported from the U.S.365 The Mexican tax measure in question was a 20% tax placed not on HFCS itself, but on the soda industry using it as a sweetener in its soft drinks and other beverages.366 The tax measure affected the transfer of these beverages, services related to their transfer (a "distribution tax"), and other imposed "booking requirements.'367 As a result, the excise tax mea-

sure applied "indirectly" to HFCS, in excess of how it was applied to a "like" domestic product, namely sugar cane, the use of which was not taxed at all.

Corn Products International alleged that the Mexican government violated its commitment to equal national treatment under Chapter 11 of NAFTA.369 The success of this claim turns on whether it can prove that investment in the high fructose corn products industry is "in like circumstances" to investment in the sugar industry. It is unclear what force the 2005 WTO decision regarding the tax will have on the present Chapter 11 dispute. Judging from prior Chapter 11 decisions, the WTO decision should significantly impact the tribunal's decision regarding whether the two investments, sugar and high fructose corn syrup, are "in like circumstances." The tribunal may consider the historic sugar dispute between Mexico and the U.S.; however, without a formal agreement upon which to rely, the historic dispute will most likely not help Mexico. Whether the sugar dispute should be resolved under a Chapter 20 dispute resolution panel is beyond the scope of a Chapter 11 investor-dispute tribunal, and the WTO did not address this issue in its final decision.370 If the tribunal finds a national treatment violation under Article 1102, the U.S. investor will likely present a claim against the Mexican government for monetary damages. This raises the question of the degree to which WTO panels should adjudicate disputes arising from government policies applicable at the regional level. In other words, not only should the WTO defer to transparent domestic regulatory processes in determining issues of procedural legitimacy of regulatory measures, but perhaps it should remove itself entirely from certain issues and defer to the regional tribunals for final adjudication. The Methanex case, decided in 2005, seems to allude to such a solution.371

368. Id. 'H 8.26-36. The panel focused on the U.S.'s second submission, in which the U.S. primarily argued that non-cane sweeteners such as beet sugar were "like" cane sugar products but were treated differently:

The United States has submitted claims regarding the treatment that Mexico accords both to imports of soft drinks and syrups and to imports of non-cane sweeteners, such as beet sugar and HFCS. The United States emphasizes that although the measures at issue are imposed by Mexico on soft drinks and syrups, this is a dispute which fundamentally concerns the treatment accorded to sweeteners.

looked

Methanex was a NAFTA Chapter 11 case where the Canadian investor alleged that California bans on the use of methanol for reformulated gasoline violated several commitments under the investment protection chapter of NAFTA, including national treatment (1102),372 fair and equitable treatment (1105),373 and expropriation (1110).374 In its national treatment claim, the claimant attempted to defer to WTO decisions regarding "like products" in arguing that ethanol investments and methanol investments like circumstances.'375 Specifically, Methanex are "in to EC-Asbestos as well as S.D. Myers to argue that competition was the litmus test for likeness.376 Methanex claimed that methanol producers are "in like circumstances" to ethanol producers because they both compete in the oxygenate market, and customers perceive the two as interchangeable.377 Then, applying the test in Pope & Talbot, the claimant stated that once less favorable treatment had been established, the burden shifted to the respondent to justify such treatment on the basis that it furthered a legitimate regulatory objective, in this case, environmental protection.378 The tribunal decided that the "comparator" in this case should be other methanol producers, and, as to them, the ban was uniform and applied to all MTBE producers.379 Therefore, there was no 1102 violation because there was no less favorable treatment.380 In agreeing with the U.S., the tribunal considered the differences between ethanol and methanol and concluded that the two products were not "like products" because competition between them was not sufficient.381 Furthermore, the ban on methanol was authorized by U.S. federal law, and the two products did not share the same tariff classification under the Harmonized System of Tariffs.382 Interestingly, the U.S. argued that even in applying the EC-Asbestos test for "like products," there was no less favorable treatment because the measure's purpose-to prevent a health and environmental hazard-was legitimate, and this fact should be considered in the "likeness" test.383 Though this argument seems to coincide with current WTO jurisprudence regarding the adjudication of regulatory measures, the tribunal made a surprising turn.

It stated that the "like products" test used by the WTO panels did not necessarily apply to "like circumstances" under the Chapter 11 of NAFTA.384 to WTO jurisprudence in this The tribunal thus chose not to defer instance. Instead, the tribunal looked at other uses of "like" in NAFTA and concluded that the drafters did not intend that "trade provisions ... be transported to investment provisions."385 This decision seemed to assert the authority and the independence of the regional tribunal in light of cases similar to those already resolved under WTO jurisprudence.

In Mexico-Tax Measures, the WTO panel noted that Mexico did not respond to the U.S. claims of sugar and HFCS being "like" products.386 Mexico actually conceded that HFCS and cane sugar were substitutable products "in certain applications.'387 Mexico took the position, however, that the tax measure was "necessary" to obtain U.S. compliance with NAFTA and its prior agreements with Mexico regarding sugar market access.388 Furthermore, Mexico claimed that the measure was not a disguised trade barrier, because any protectionist intent in the measure was justifiable under Article III, due to its dispute with the U.S. regarding U.S. market access for its sugar exports, as well as social and economic difficulties for the Mexican sugar industry resulting from this dispute.389 The WTO panel decided it did have jurisdiction to decide the matter, and it did not defer resolution of the problem to a Chapter 20 NAFTA dispute resolution body.390 As a matter of GATT law, the WTO panel had jurisdiction; however, after Methanex, its decision may not necessarily be dispositive for the final resolution of Corn Products under a Chapter 11 NAFTA tribunal. In making its final decision, the WTO Panel in Mexico-Tax Measures primarily followed Japan-Alcoholic Beverages in its analysis regarding Article 1II, concluding that sweeteners made of cane sugar and those made of non-cane, such as HFCS, were "like" products according to the Border Tax Adjustments factors.391 The panel proceeded to decide affirmatively that Mexico was applying the internal tax "in excess of those applied, directly or

indirectly to like domestic products."392 Finally, the panel examined whether there was a violation of Article 111:2, sentence 2.393 It decided, as in its analysis of Article 111:2 sentence 1, that (1) the products were "directly competitive or substitutable products" according to the Border Tax Adjustments; and (2) the "like" products were "not similarly taxed."'394 Then the panel proceeded to conclude that this tax was indeed implemented "so as to afford protection to domestic production.'395 In doing so, the panel first characterized the products as "like" by again applying the Border Tax Adjustments criteria.396 Finally, the panel concluded that the tax also violated Article 111:4 by treating non-cane sweeteners less favorably than cane sweeteners.397 Consistent with EC-Asbestos, the panel found that "the scope of 'like' in Article 111:4 is broader than the scope of 'like' in Article 111:2, first sentence."398 Therefore, there was no need to determine their "likeness" under Article 111:4 once their likeness had already been established under Article II1:2.399 The panel did not consider whether the fiscal measure was legitimate under the circumstances.

The effect of Mexico-Tax Measures with respect to Article III is threefold. First, it solidifies the WTO's views on Article III, at least in the context of fiscal measures under Article 111:2, and its determination of "like products," as established by the Appellate Body in Japan-Taxes on Alcoholic Beverages. Second, it applies this national treatment standard not only to the "like products" themselves but also to the components of finished products that are competitive and interchangeable.400 Even more so than Japan-Alcoholic Beverages, the panel did not distinguish between the intent of the tax measure and its effects, probably because the measure was clearly protectionist from its inception, and even Mexico conceded this point to some extent.401 Rather, in its determination of whether the tax was passed "so as to afford protection to domestic production," the panel based its conclusion on whether the products were "like" and treated

unequally.402 The panel never considered whether its aim was based on legitimate regulatory objectives or the extent to which the effects of the tax furthered such an aim. Furthermore, the panel did not take into account the regulatory framework of the Mexican government regarding sugar, as was done in GAMI Investments in the Chapter 11 NAFTA context.403 If the WTO panel had focused on the regulatory framework for sugar in Mexico, the outcome would probably have been similar to the GAMI case. Actually, a closer look at the Mexican legislature's argument for passing the tax would have just reinforced the idea that the tax was indeed protectionist.404 Perhaps in this case, the measure was clearly de jure protectionist and not facially neutral; therefore, no deference was even necessary. Under the "deference" model, however, the adjudicatory process would have required any questions of legitimacy (if there were some) to be "remanded" back to the Mexican democratic process. If the democratic institution closest to the measure, namely the Mexican legislature, were unable to prove either a legitimate purpose for the measure or the existence of alternative means more aligned with its GATT commitments, the WTO panel should strike down the measure. In this way, WTO panels can encourage transparency within the regulatory structures at the domestic level and have a greater impact on the implementation of domestic policy without taking an omniscient position. Furthermore, the WTO panels can suggest that certain issues, such as the U.S.-Mexico sugar dispute, be better settled at the regional level through bilateral negotiations and regional tribunals that are closer to the problems at hand.

This brings us back to the question of whether the WTO panels should defer to its member states in dealing with domestic regulatory measures under Article III. More specifically, should the WTO panels defer to the relevant regional tribunals in the final resolution of a specific dispute? The WTO panel in Mexico -Tax Measures concluded that it had jurisdiction to decide the issue in this case.405 More importantly, however, it declined to defer final resolution of the dispute to the NAFTA dispute resolution body.406 The complexity of this omission lies in the fact that its determination of the validity of the measure under national treatment standards may impact the decision at the regional level. Methanex seemed to assert the authority of a regional tribunal to resolve national treatment problems concerning foreign investment independently of WTO adjudication of regulatory policies affecting those investments.407 In doing so, the regional tribunal seemed to recognize the close relationship between private investment and government behavior through regulation.408 Herein lies the disconnectedness of the multilateral trade regime as it pertains to Article III.

B. Reciprocal Deference Encourages Bridge-Building

Reconciling the domestic need for regulation with the GATT objective of eliminating trade barriers is not an easy task for the WTO panels. However, there are various instances where they are forced to entertain this apparent conflict. Article III is an important provision in dealing with this issue, particularly because of its relevance to comparable provisions in WTO Covered Agreements as well as in regional agreements such as NAFTA.409 Regional tribunals, while tailoring GATT interpretations of "like products" to the specific regional concerns and issues, will nevertheless defer to WTO decisions in this regard.410 Therefore, in making its decisions, the WTO panel should consider the effects of these judicial determinations. It is difficult to say, for example, how the recent HFCS case decided by the WTO will impact Mexico's obligations to the U.S. investor under Chapter 11 of NAFTA. While not necessarily deciding issues relevant to domestic regulatory policy or to regional concerns, WTO panels can help create cohesiveness within the multilateral trading regime by becoming a forum for discussion of legitimate regulatory policies. EC-Asbestos was one step in this direction.411

A balancing test, such as that used by U.S. courts in the context of the dormant Commerce Clause, may not be so helpful in adjudicating regulatory measures within the context of Article III. Such a balancing test would purport to balance the need for the regulatory measure as a means of furthering legitimate non-protectionist domestic objectives against the burdens such a measure places on trade.412 A "cost-benefit" balancing test of this kind is probably unrealistic for the WTO and trade agreements in general.413 Such a test cannot, for example, distinguish between less organized special interest groups and more influential organized groups that may benefit the most from passing certain regulations.414 Furthermore, it would require WTO panels to make substantive determinations as to the legitimacy of a regulatory measure per domestic law.415 While the WTO dispute settlement system is probably not in the best position to make such

determinations or to weigh the benefits of a certain regulation to a member state against the net costs of such a regulation to other member states, the WTO can provide a forum for discussion of regulatory measures through its adjudication of Article Ill. WTO panels can provide a procedural mechanism that allows for deference to its member states for facially non-discriminatory regulatory measures with discriminatory effects.416 In this instance, the mechanism would task member state respondents with proving the legitimacy of their regulatory measures according to their own transparent regulatory schemes.417 Furthermore, it would allow for recognition that certain regional disputes are best resolved at the regional level. In the spirit of Professors McGinnis and Movsesian, an "antidiscrimination approach" to WTO adjudication of domestic regulatory measures is helpful in the application of Article III of GATT, though to a lesser degree if the contested measure does not deal with labor, the environment, health, and safety.418 This approach is illuminating in that it distinguishes between facially discriminatory measures and non-discriminatory measures with incidental effects on trade, attempting to deal with only those that amount to "covert protectionism." In deciding whether products in question are "like," WTO panels may consider the regulatory measure and its effects on the competitive relationship of the products. At the outset, panels should treat de jure and de facto discriminatory measures differently. With de jure discriminatory measures, the current application of the "like products" test may be sufficient, as in cases like Mexico-Tax Measures.419 Any exemption to Article III may then be pursued by respondents under the stricter Article XX analysis.420 In cases with facially neutral measures (de facto), however, WTO panels should distinguish between those that have discriminatory effects and those that are non-discriminatory but place "incidental burdens" on trade. WTO panels should create a procedural framework to better assess these covert measures. The framework must function within the context of Article III and the definition of "like," where the question of legitimacy first arises. If the regulatory measure is (1) transparent; (2) consistent with other comparable regulations affecting similar products; and (3) performance-oriented in that it is in line with objective standards for the area in question, then the panel may infer that the measure is legitimate and should be considered in the determination of "like products."'421 In making these determinations, however, the WTO panels must defer to national democratic and transparent regulatory

processes to the extent they exist. Objective criteria should be used, but science may not always be the benchmark. Within this procedural framework, responding member states should have the burden of proving the legitimacy of their regulatory measures, and that no alternative means of accomplishing relevant public objectives exist. Finally, where a certain measure is best assessed at the regional level, such as in the case of Mexican sugar policies against the U.S., a WTO panel should defer to a regional tribunal, where possible, for final adjudication.

WTO panels may borrow from NAFTA tribunals in evaluating possible national treatment violations in their consideration of legitimate regulatory policy. Chapter 11 tribunals place the burden on the respondent to show that regulatory policies "not motivated by preference of domestic over foreign owned investments.'422 WTO panels, too, may place such a burden on respondent governments by making domestic legislatures accountable for policies in conflict with their obligations under GATT.423 It is within the domestic regulatory framework that the problems of special interest group bargaining and the effects of capture can best be addressed. Therefore, regional tribunals can better assess the costs and benefits of domestic regulatory policies concerning regional agreements in a Pope & Talbot-style balancing test.

Until states decide the role of the WTO vis-a-vis their own national sovereignty, the WTO should be a forum for bargaining among its members.425 In deciding which regulatory measures are not protectionist, the WTO panels should give member states the opportunity to prove their measures' legitimacy. While deference to member states may not be appropriate in every circumstance, a WTO panel performs best in asking the domestic institution closest to the implementation of such measure to "clarify" issues of legitimacy.426 While deferring, WTO panels do not necessarily abdicate their roles of adjudicator and facilitator of free trade. Rather, they embrace those roles within a global framework that incorporates multiple sovereign interests. For this reason, in adjudicating regulatory measures under Article III, WTO panels should shift the burden onto

respondents and their legislatures to prove the legitimacy of regulatory measures rather than ignore questions of legitimacy behind the strict formalism of the "like products" test. This should occur within the scope of Article III and should not be confined solely to the high threshold of "necessity" under Article XX exceptions. In this way, the WTO can encourage its members to develop regulatory processes that are more transparent.

Because of the binding nature of its determinations, the WTO dispute resolution body can exert quite a bit of influence on its members.427 Members may be deterred from ignoring panel determinations because of the retaliatory power of their counterparts under the GATT. Such retaliation may, however, lead to continued paralysis for the involved parties and frustration of any real resolution of the dispute at hand.428 In this way, the WTO panels may unwittingly undermine confidence in the legitimacy of the multilateral trade regime. The WTO should derive its legitimacy in adjudicating matters of domestic regulatory policies from standards established by domestic regulatory structures and not from higher norms the WTO purportedly creates.429 Those standards may evolve within the legal regimes of its member states, and the WTO may impact this evolution by using the decision-making power of its dispute resolution body to make domestic governments accountable from within their legal regimes. The WTO should not, however, alienate its members by adhering to strict formalism or ignoring domestic political processes.430 In inviting such dialogue, the possibility of legitimate regulatory measures that are independent of the domestic regulatory scheme and derive authority solely from the WTO.431 In adjudicating these measures under Article III, the WTO panels may best align the objectives of GATT with their role as facilitators of free trade.

the WTO

recognizes

C. Applying Reciprocal Deference

A reciprocal deference approach

to the WTO mimics the antidiscrimination model to the extent that it favors a procedural deference to decisions of domestic regulatory agencies when the WTO adjudicates domestic measures. This deference to the procedures of the domestic regulatory agencies is designed to promote democracy and transparency within

those systems.432 The reciprocal deference approach is nonetheless different in that it contemplates a multilateral system in which domestic regulatory structures and regional tribunals will occasionally defer to the WTO decisions.433 This approach would allow for fluidity and put pressure on both sides of the problem-the domestic and regional side as well as the international side-so as to sustain a bridge between the two. Domestic governments enforce WTO decisions, and regional tribunals may look to WTO jurisprudence for guidance in their own adjudicatory power. This is clear. However, in what ways should the WTO panels "defer"?

In applying Article III, the WTO panels should treat de jure and de facto discriminatory measures differently. For de jure discriminatory measures, the stricter "like products" test based on the competitive relationship of the products and their substitutability may suffice. This, for example, is the case for Mexico-Tax Measures. In larger bilateral disputes that involve the measure in question, such as the sugar dispute between the U.S. and Mexico in the case of Mexico-Tax Measures and Corn Products, the WTO panel should encourage the issue to be resolved at the regional level. The panel can state that the national treatment problem at hand is part of a larger dispute that cannot be resolved by the WTO and should be by the NAFTA tribunal.

With respect to de facto measures, WTO panels should distinguish between those having discriminatory effects and those that are non-discriminatory but still place "incidental burdens" on trade. With the latter, the panels should not be as concerned. With the former, the panels can apply a procedural mechanism that assesses (1) the transparency of the measure; (2) its consistency with comparable regulations affecting similar products; and (3) its consistency with objective standards regarding the public goal that the government is trying to achieve.434 If there is a positive assessment of these three factors, the inference should be that the measure is legitimate. To ensure that governments do not hide behind sophisticated regulatory structures that may pass seemingly goal-oriented measures that are really meant to be protectionist, the panels must still place the burden on the respondent to prove the legitimacy of these standards according to domestic law and demonstrate that no alternative means more in line with WTO law exists. Of course, less developed nations may argue that "requiring" transparency in this way is not fair to them because their regulatory institutions are less developed and may not have the information-gathering power necessary to conform to WTO standards.435 This is a fair objection to my claim; however, reciprocal defer-

ence is meant to encourage transparency, not punish nations for their inability to achieve it. Therefore, the international community, through the auspices of the WTO, should have a way of providing technical assistance in these situations.436

The EC-Asbestos case provides a good example of the application of reciprocal deference procedure. There, the WTO Appellate Body considered the legitimacy of the measure in assessing whether chrysotile asbestos fibers were "like" PCG fibers.437 It considered the reason for the French Decree and noted that the fact that the asbestos fibers were carcinogenic was an important factor in determining their "unlikeness.'438 In applying a reciprocal deference approach, the panel would first place this Decree in the category of a neutral measure with discriminatory effects. Then it would assess whether the Decree was (1) transparent (it was authorized through a transparent democratic process); (2) consistent with other measures on similar products (though the measure did not affect other fibers, the panel could look at analogous measures to protect health); and (3) passed according to objective standards (in this case scientific evidence of its carcinogenic aspects) to further legitimate social goals (namely, health). The panel could then infer that in fact the products are not "like" and that the Decree is legitimate. The French government would still have the burden, however, of proving that the measure is legitimate according to French law and was not passed for protectionist purposes. The outcome would thus have been the same as it was in the actual EC-Asbestos ruling. However, not all cases are as "clean" as EC-Asbestos (where one of the products is deadly). Nonetheless, the reciprocal deference approach would also help in dealing with more challenging facially neutral measures.

In this way, WTO panels would recognize that not all regulatory measures are protectionist and would impose some criteria for assessing the discriminatory effects of the measures on free trade. Reciprocal deference recognizes that domestic governments will pass regulatory measures for legitimate reasons, but it also guides member states as to their commitments under GATT. At the same time, it allows the WTO to take on the role of an adjudicatory body rather than a supranational institution imposing standards. Finally, it gives the WTO panels freedom to adjudicate only in areas under its jurisdiction without dismissing areas that must be resolved at the regional level. WTO panels can state which issues are under its jurisdiction, recognize the impact these decisions may have at the regional level, and push regional tribunals to decide matters within their own jurisdiction. In this way, sustainable bridges between multilateral and regional trade regimes may form, creating a tighter international network.

Conclusion

An apparent conflict exists within the WTO multilateral regime in the adjudication of domestic regulatory measures of its member states under Article III of GATT. The more formalistic application of the traditional "like products" test runs the risk of invalidating legitimate domestic regulatory measures, particularly those that are facially neutral (de facto). The less formalistic approach under the "aims and effects" test requires WTO panels to determine the legitimate purpose of the measure in question as a matter of substantive law. Though the latter approach is more deferential to the sovereign right of member states to regulate domestically, it places a high burden on WTO panels to decide issues of legitimacy concerning domestic law.

The current "like products" test applied by WTO panels to adjudicate domestic regulatory policy, while attempting to deal with regulatory measures such as in EC-Asbestos, fails to answer important questions regarding the legitimacy of domestic regulatory measures. The WTO may learn from similar discussions in other areas of WTO law as well as under the U.S. dormant Commerce Clause jurisprudence and the NAFTA decisions. Whereas the U.S. Supreme Court has the authority to decide issues of legitimacy with respect to state regulatory policy and balance the interests of the state with those of Congress, the WTO panels cannot claim this same authority. As the gatekeeper of the multilateral trade regime, WTO panels, nonetheless, do have the responsibility of enforcing compliance with the GATT objectives. In this way, the WTO takes the form of a regulatory model. The WTO dispute resolution bodies may make substantive decisions as to international law under WTO Covered Agreements; however, they should not do the same with respect to domestic law. WTO panels must encourage coordination and transparency within the domestic regulatory structures of its member states through procedural mechanisms that make member states accountable for their internal regulatory policies. As done by NAFTA tribunals, WTO panels should apply greater deference to transparent domestic regulatory processes when answering questions of legitimacy with respect to domestic regulatory measures and alleged violations of national treatment requirements under Article III. While this approaches an antidiscrimination model for the WTO, such deference should not be absolute. Rather, it should be in the form of inferences and burden-shifting; that is, even if the procedural mechanism raises an inference of no national treatment violation, WTO panels should require that respondent member states show that their domestic legislatures have proven "no alternative means" of furthering legitimate domestic goals not in violation of GATT. In this way, domestic democratic processes can allow for greater accountability, and the WTO can affect compliance at its source.

By deferring to domestic regulatory structures as a procedural matter, WTO panels may place the burden of proving legitimacy on those institutions closest to the implementation of the measure. In doing this, the WTO would also encourage active participation from domestic governments in

the multilateral trade regime. Proof of legitimacy should be within certain parameters set by the WTO panels. To borrow from the U.S. dormant Commerce Clause jurisprudence, WTO panels should require proof of "no alternative means" of furthering legitimate domestic policy that would be more closely aligned with commitments under GATT. By "remanding" such questions of legitimacy, the WTO is not taking on an autonomous role as the ultimate judge of domestic regulatory policy. Rather, it becomes an important internal player nudging domestic democratic political processes in the direction of compliance with commitments under GATT. the WTO panels also encourage domestic regulatory In this way, frameworks to be transparent in passing regulatory policy. WTO panels cannot ignore their impact on the execution of regional agreements and should recognize that some issues would be better settled there. In addition, WTO panels can exert a more positive influence at the domestic level without alienating member states when strict formalism is inappropriate. The WTO, through its adjudicatory power, can weave a tight web of influence at the domestic level, binding the multilateral regime together while maintaining its role as external gatekeeper of the regime.

Furthermore, when the WTO promotes coordination within the multilateral regime, WTO panels should recognize the impact of WTO decisions at the regional level. Regional tribunals such as the NAFTA Chapter 11 tribunals will defer to WTO decisions in applying provisions of regional agreements. Perhaps this deference allows for more adhesiveness within the multilateral regime as well. However, WTO panels cannot simply ignore this tendency; they must also recognize that some disputes may be better resolved at the regional level while still deciding matters where it clearly has jurisdiction. For this reason, WTO panels should exhibit procedural deference to regional agreements and domestic regulatory structures when examining regulatory measures. This deference will help solidify the desired coordination. A reciprocal deference approach will continue to encourage domestic regulatory institutions to follow commitments under the GATT agreements while allowing WTO panels to better influence those same institutions. For it is within its web of influence at the domestic level that the WTO can truly have an impact in pushing member states toward economic integration and globalization.

--- FOOTNOTES ---

T Associate Professor, Suffolk University Law School. Email: [email protected]. This paper was presented at Florida State University College of Law 2006 Faculty Enrichment Speaker Series in 2006 and the Law and Society Association Annual Meeting, Baltimore, Maryland, July 2006. I am especially grateful to my colleagues at Florida State University College of Law for their support and input in completion of a draft. Special thanks to Frederick M. Abbott, Florida State University College of Law; Peter Carstensen, University of Wisconsin Law School; David Gantz, University of Arizona James E. Rogers College of Law; Frank Garcia, Boston College School of Law; Robert L. Howse, University of Michigan Law School; and Douglas Kysar, Cornell Law School for their comments on a draft. Thanks tojamila G. Gooden, Blake Heraghty, and Nicole Stafford for their able research assistance. 40 CORNELL INT'L LJ. 201 (2007)

1. General Agreement on Tariffs and Trade, Oct. 30, 1947, 61 Stat. A-11, 55

U.N.T.S. 194 [hereinafter GATT].

2. See generally Understanding on Rules and Procedures Governing the Settlement of Disputes, Apr. 15, 1994, Marrakesh Agreement Establishing the World Trade Organization, Annex 2, Legal Instruments-Results of the Uruguay Round, app. 1, 33 l.L.M. 1125 (1994) [hereinafter DSUJ. WTO panels occasionally look to other sources of international law, such as Article 31 of the Vienna Convention when dealing with issues of treaty interpretation.

3. See generally GATT, supra note 1. The GATT establishes a wide range of provisions instructing parties to the agreement on what trade and tariff policies should be implemented and what types of policies are permissible. This level of power and control demonstrates the WTO's ability to direct matters of domestic regulatory policy.

4. See GATT, supra note 1, Art. III; id., Annex I, ad. art. III. 5. GATT, supra note 1, Art. II1. Paragraphs 1 and 2 read as follows: The contracting parties recognize that internal taxes and other internal charges, and laws, regulations and requirements affecting the internal sale, offering for sale, purchase, transportation, distribution or use of products, and internal quantitative regulations requiring the mixture, processing or use of products in specified amounts or proportions, should not be applied to imported or domestic products so as to afford protection to domestic production.

The products of the territory of any contracting party imported into the territory of any other contracting party shall not be subject, directly or indirectly, to inter-

nal taxes or internal charges of any kind in excess of those applied, directly or indirectly, to like domestic products. Moreover, no contracting party shall otherwise apply internal taxes or other internal charges to imported or domestic products in a manner contrary to the principles set forth in paragraph 1.

6. See Robert Howse & Elisabeth Tuerck, The WTO Impact on Internal Regulations: A Case Study of the Canada-EC Asbestos Dispute, in THE EU AND THE WTO: LEGAL AND CONSTITuTIONAL ISSUEs 283, 285 (Grainne de Buirca & Joanne Scott eds., 2001), available at http://www.worldtradelaw.net/articles/howseasbestos.pdf (describing that member states can "simply shift protectionism from explicit facially discriminatory measures, to regulatory schemes that [are] covertly or structurally discriminatory").

7. See GATT, supra note 1, Art. Ill. 8. Methanex Corp. v. United States, Final Award of the Tribunal on Jurisdiction and Merits, Part 1ll, Ch. B, 9 37 (NAFTA Ch. 11 Arb. Trib. Aug. 3, 2005) available at http:// www.state.gov/documents/organization/51052.pdf [hereinafter Methanex, Final Award].

9. See infra Part Ill.

10. See Methanex, Final Award., supra note 8, at Part II, Ch. B, ' 6; id. at Part IV, Ch.

B, 9 37.

11. Robert E. Hudec and Daniel Farber wrote an interesting comparison of these two doctrines in which they demonstrated ways that the laws under both the GATT and the dormant Commerce Clause both deal with regulatory measures enacted through legitimate local processes, but which in some way affect the competitive interests of those outside the local legal domain. 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, 1403 (1994) (stating that the GATT and the dormant Commerce Clause share common concerns). They continue to argue that "[t]he modern regulatory state inevitably produces burdens on trade, if only because of the unavoidable lack of regulatory uniformity. For various reasons, many of these burdens likely are unwarranted, and at least some are in fact due to protectionist efforts by local industries. Yet, tribunals have only a limited warrant to override the policy choices of local legislatures. Tribunals must accord respect to the democratic process as well as to the prerogatives (or sovereignty) of local governments." Id.

12. Request for Institution of Arbitration Proceedings, Corn Products International v. United Mexican States, October 28, 2003. The antidumping dispute was decided by the WTO in Panel Report, Mexico-Anti-Dumping Investigation of High Fructose Corn Syrup from the United States, WT/DS132/R (Jan. 28, 2000) (adopted Feb. 25, 2000) [hereinafter Mexico-HFCS] and a NAFTA Chapter 19 Tribunal in Review of the Final Determination of the Antidumping Investigation on Imports of High Fructose Corn Syrup, Originating from the United States of America, MEX-USA-98-1904-01 (Aug. 3, 2001). 13. Request for Institution of Arbitration Proceedings, Corn Products, '1 54. 14. Id. 15.

Id. 111 82-84. In addition, the claimant asserts that such a tax amounts to a performance requirement prohibited under 1106 and expropriates his investment under 1110. Id. P1 91-93, 98; see also NAFTA, infra note 33, art. 1102.

16. See Request for the Establishment of a Panel by the United States, Mexico-Tax Measures on Soft Drinks and other Beverages, WT/DS308/4,June 11, 2004. The claimant alleges violations under Articles 111:1, 111:2 and 111:4 of GATT. Id.

17. See Panel Report, Mexico-Tax Measures on Soft Drinks and other Beverages,

4.110, WT/DS308/R, Oct. 7, 2005 [hereinafter Mexico-Tax Measures].

18.

Id. cT 4.71-4.108. It is important to note that the Mexican sugar industry has

traditionally been regulated and protected by the Mexican government.

19. See id. 11 8.155-8.157.

20. See infra Parts III, IV. 21. The remedy under NAFTA Chapter 11 is in the form of monetary damages imposed against a guilty government by the private investor. On the other hand, a violation found by a WTO dispute settlement body allows the innocent government to retaliate against the guilty government. See NAFTA, infra note 33, art. 1135; see also DSU, supra note 2.

22. See GATT, supra note 1. 23. See id. at Part 11, art, Ill. 24. See id. at Part II, art. Il, ' 1.

25. See JOHN H. JACKSON, WORLD TRADE AND THE LAW OF GATT 5 (1969) (stating that one of the goals of GATT law is to ensure "efficient allocation of resources and the maximization of economic product").

26. See Marrakesh Agreement Establishing the World Trade Organization, Legal Instruments-Results of the Uruguay Round, 33 I.L.M. 1125 (1994) [hereinafter WTO Agreement]; see also DSU, supra note 2, at arts. 2, 17.

27. Appellate Body Report, European Comniunities-Measures Affecting Asbestos and 84-154, WT/DS135/AB/R (Mar. 12, 2001), reprinted

Asbestos-Containing Products, in 40 I.L.M. 1193 (2001) [hereinafter EC-Asbestos].

28. See id.; see infra Part L.A (discussing traditional application of the "like products" test according to the Border Tax Adjustments and to the competitive substitutability of the products).

29. WTO panels occasionally look to other sources of international law such as Article 31 of the Vienna Convention when dealing with issues of treaty interpretation. See Appellate Body Report, United States-Import Prohibition of Certain Shrimp and Shrimp Products, 157, WT/DS58/AB/R (Oct. 12, 1998).

30. See, e.g., Agreement on the Application of Sanitary and Phytosanitary Measures, including annexes, Apr. 15, 1994, Marrakesh Agreement Establishing the World Trade Organization, Annex 1A, Multilateral Agreements on Trade in Goods; Agreement on Technical Barriers to Trade, including annexes, Apr. 15, 1994, Marrakesh Agreement Establishing the World Trade Organization, Annex IA, Multilateral Agreements on Trade in Goods; Uruguay Round Agreement on Trade-Related Aspects of Intellectual Property Rights, Apr. 15, 1994, Marrakesh Agreement Establishing the World Trade Organization, Annex IA, Multilateral Agreements on Trade in Goods; Agreement on TradeRelated Investment Measures, Apr. 15, 1994, Marrakesh Agreement Establishing the World Trade Organization, Annex IA, Multilateral Agreements on Trade in Goods. Note that all of these agreements either specifically refer to Article III of GATT in ensuring that permissive measures do not amount to non-tariff trade barriers or state that national treatment obligations must be respected.

31. See generally H.C. Claus-Dieter Ehlerman, Dispute Settlement in the WTO (Oct.

2005), http://www.wilmerhale.com/publications/whPubsDetail.aspx?publication=2671 (discussing WTO standard of review).

32. NAFTA is a free trade agreement among the United States, Canada, and Mexico

[hereinafter Parties]. See NAFTA, infra note 33.

33. North American Free Trade Agreement, U.S.-Can.-Mex., Dec. 17, 1992, 32 I.L.M.

289, 296-456, 605-800 (1993) [hereinafter NAFTA].

34. See, e.g., id. art. 2005. 35. See infra Part III (discussing NAFTA chapter 11 cases in which the NAFTA chapter 11 tribunals, in defining "like circumstances" under NAFTA chapter 11, deferred to WTO determinations regarding Article III of GATT ). 36. See, e.g., EC-Asbestos, supra note 27,

'VT 96-97; Panel Report, United States-Taxes on Automobiles, WT/DS31/R (Sept. 29, 1994) [hereinafter U.S.-Taxes on Automobiles]; Panel Report, United States-Measures Affecting Alcoholic and Malt-Beverages, WT/DS23/R-39S/206 (Mar. 16, 1992) (adopted June 19, 1992) [hereinafter U.S. -Malt Beverages].

37. See infra Part II.C (defining "incidental burdens" in the context of the U.S. dor-

mant Commerce Clause jurisprudence).

38. See, e.g., Report of the Panel for Conciliation, Italian Discrimination Against Imported Agricultural Machinery, L/833-7S/60 (July 15, 1958) (recognizing that regulatory measures may foster domestic economic development and not be in violation of GATT). See infra Part I.B.ii for detailed discussion.

39. See, e.g., Panel Report, Mexico-Measures Affecting Telecommunications Services, WT/DS204/R, (June 1, 2004) (recognizing that telecommunications are partially regulated by the Mexican government but concluding that Mexico was in violation of its commitments under Reference Paper, April 24, 1996 and the GATS Annex by failing to prevent anti-competitive behavior by its largest service provider, Telmex) [hereinafter Mexico- Telecommunications].

40. See Notice of Intent to Submit a Claim to Arbitration under Section B of Chapter 11 of the North American Free Trade Agreement, Corn Products International, Inc. v. United Mexican States, '1 27-32, Jan. 28, 2003.

41. See Mexico-Telecommunications, supra note 39, 9 2.2. Some of the U.S. affiliated

carriers are Worldcom, through Avantel, and AT&T, through Alestra. Id.

42. See Joseph D. Kearney & Thomas W. Merrill, The Great Transformation of Regu-

lated Industries Law, 98 COLUM. L. REV. 1323, 1352-55 (1998).

43. See, e.g., Jim Rossi, The Electric Deregulation Fiasco: Looking to Regulatory Federalism to Promote a Balance Between Markets and the Provision of Public Goods, 100 MICH. L. REv. 1768 (2002); Elizabeth Trujillo, State Action Antitrust Exemption Collides with Deregulation: Rehabilitating the Foreseeability Doctrine, 11 FORDHAM J. CORP. & FIN. L. 349 (2006). But cf. John 0. McGinnis, The Political Economy of International Antitrust Harmonization 45 WM. & MARY L. REV. 549, 552 (2003) (proposing a "limited and modest antidiscrimination international antitrust regime located within the WTO" with a rationale "com[ing] principally from international trade law rather than antitrust law"). 44. See Mexico-Telecommunications, supra note 39, 9l 7.222-45 (defining "anticom-

petitive practices" as "actions that lessen rivalry or competition in the market").

45. See generally Richard H. Steinberg, Judicial Lawmaking at the WTO: Discursive, Constitutional, and Political Constraints, 98 AM. J. INT'L. L. 247 (2004) (arguing that the WTO dispute settlement system is constrained by international legal discourse and politics as well as constitutional structure). Professor Steinberg characterizes the present GATT dispute settlement system as one "more legalized and fundamentally more adjudicative" than diplomatic. See id. at 250. He also discusses various forces influencing the WTO's decision-making process through its dispute settlement system and questions the suggestion by many scholars that "expansive judicial lawmaking might undermine the political support of powerful states for the WTO." See id. at 257.

46. See John 0. McGinnis & Mark L. Movsesian, Commentary: The World Trade Constitution, 114 HARv. L. REV. 511, 517-19 (2001) (identifying deficiencies in the regulatory model).

47. See Sandeep Gopalan, The Creation of International Commercial Law: Sovereignty Felled?, 5 SAN DIEGO INT'L L. J. 267, 274-82 (2004) (discussing the definition of "harmonization").

48. See McGinnis & Movsesian, supra note 46, at 530-36. 49. See Steinberg, supra note 45, at 257 (stating that undermining political support by powerful member states would "be politically devastating, as it is well established that multilateral organizations that are not supported by the world's most powerful states will either collapse or become irrelevant.").

50. See id. at 267.

51. See Appellate Body Report, European Communities-Measures Concerning Meat

and Meat Products, '1 19, WT/DS26/AB/R (Jan. 16, 1998).

52. See EC-Asbestos, supra note 27, 'Hl 87-100, 120, 123. 53. See generally McGinnis & Movsesian, supra note 46. 54. See Alan Sykes, Regulatory Protectionism and the Law of International Trade, 66 U. CHI. L. REV. 1, 4-5, 21-22 (1999) (explaining measures that amount to the "least restrictive means" of furthering legitimate state regulatory goals).

55. See, e.g., Methanex Final Award. supra note 8,

12; see also S.D. Myers, Inc. v. Canada, Partial Award, 40 l.L.M. 1408, cl 245-46 (2001) [hereinafter S.D. Myers, Partial Award].

56. Closing Statement of the Panel, Mexico-Tax Measures on Soft Drinks and Other

Beverages, WT/DS 308 (Feb. 24, 2005).

Wines and Alcoholic Beverages, §§ 5.5-5.10 (Nov. 10, 1987).

59. See Sykes, supra note 54, at 7 (recognizing that governments "have an array of devices at their disposal for protecting domestic industries against outside competition"); see also JAGDISH BHAGWATI, PROTECTIONISM 43-54 (1988) (describing most nontariff barriers as deriving from legitimate regulatory institutions which administer measures impacting imports); ANNE 0. KREUGER, ECONOMIC POLICIES AT CROSS PURPOSES, 108-31 (1993) (describing several U.S. regulatory measures, including antidumping measures, that may be construed as protectionist).

60. See generally Paul Krugman, Toyota, Moving Northward, N.Y TIMES, July 25, 2005

at A19. 61. 62. See Michael E. Levine &Jennifer L. Forrence, Regulatory Capture, Public Interest, and the Public Agenda: Toward a Synthesis, 6J. L. ECON. & ORG. 167, 167-98 (1990). But see generally, Sykes, supra note 54. Professor Sykes demonstrates that "regulatory protectionism" is the least efficient kind of protectionist measure (as compared to tariffs, quotas, and subsidies for example) because of the additional deadweight losses it produces on society and its tendency to further the effects of capture and transfer of rents among self-interested politicians. See Sykes, supra note 54, at 5, 7-12.

63. I.R.C. § 4064 (2006). 64. US-Taxes on Automobiles, supra note 36. Note that this case also involved a lux-

ury tax placed on certain vehicles valued beyond a threshold amount. Id.

65. Id. 66. Id. 67. For an explanation of the "inherence" test, see generally Michael J. Trebilcock & Shiva K. Giri, The National Treatment Principle in International Trade Law, in HANDBOOK OF INT'L TRADE, VOL. 11: ECONOMIC AND LEGAL ANALYSES OF TRADE POLICY AND INSTITUTIONS 195-96 (E. Kwan Choi & James C. Hartigan eds., 2005).

68. Robert E. Hudec, GATT/WTO Constraints on National Regulation: Requiem for an "Air and Effects" Test, 32 INT'L LAW. 619, 628-29 (1998), available at http://www.world tradelaw.net/articles/hudecrequiem.pdf [hereinafter Hudec, Requiem].

69. Id. at 626-29. 70. See Daniel J. Gifford & Mitsuo Matsushita, Antitrust or Competition Laws Viewed in a Trading Context: Harmony or Dissonance?, in 2 LEGAL ANALYSIS, FAIR TRADE AND F[uMONIZATION: PREREQUISITES FOR FREE TRADE? 269, 282-83, (Jagdish Bhagwati & Robert E. Hudec eds., 1996) (stating that "the most frequent and widely experienced conflicts between industrial policy and competition policy occur in circumstances in which a domestic industry, having lost its competitiveness, is under heavy pressure from foreign competitors"). In this context, governments "take steps to encourage the modernization and/or restructuring of the domestic industry and to protect the industry temporarily from foreign competition." Id. The authors also clarify that this type of restructuring would normally come under the scope of the Emergency Action on Imports of Particular Products of GATT Art. XIX. Id. at 283.

71.

Id. at 285 (stating that "Illimitation of imports in connection with an industrial policy which seeks to rescue and/or restructure a distressed domestic industry is... an activity which the GATT itself contemplates."). The inherent challenge in Article XIX is that defining when an emergency action is applicable also turns on the definition of "like or directly competitive products." Id.

72. See id. at 5. 73. See id. at 283. The authors state that:

Viewed in the light of ensuring the survival of a domestic industry, crisis cartels and associated import restrictions seem more tolerable interferences in the free market .... Accordingly, measures which are designed as ad hoc efforts to rescue a troubled domestic industry with the ultimate objective of restoring marketplace competition cannot be seen as engendering substantial trade problems.

Id. at 287-88.

74. See Ian Wooten & Maurizio Zanardi, Antidumping versus Antitrust: Trade and Competition Policy in 2 HANDBOOK OF INTERNATIONAL TRADE: ECONOMIC AND LEGAL ANALYSES OF TRADE POLICY AND INSTI-TUTIONS 383 -84 (E. Kwan Choi & James C. Hartigan eds., 2005) (stating that in the context of government regulation, "the instruments that are available to a government to discipline a domestic firm are not accessible in the case of a foreign-based enterprise").

75.

Id. at 384 (asserting that "[tihe disadvantage of AD policies is that they do not fully address the problem of inappropriate use of market power and, indeed, often induce more distortions in the market than they resolve and are often captured by special interests").

76. See Joseph D. Kearney & Thomas W. Merrill, The Great Transformation of Regu-

lated Industries Law, 98 COLUM. L. REV. 1323, 1354-55 (1998).

77. See, e.g., Harry First, Regulated Deregulation: The New York Experience in Electric Utility Deregulation, 33 Loy. U. CHI. L.J. 911 (2002); Rossi, supra note 43, at 1770; Trujillo, supra note 43, at 353-54 (discussing the expansion of the state action immunity clause in the deregulated electricity industry as being a means of furthering the effects of capture in dominant suppliers already existing in those markets prior to implementation of deregulatory policies).

78. Hudec, Requiem, supra note 68, at 624-26, 634-35. 79. See Mexico-Tax Measures, supra note 17, 1 4.179. Id. 9 4.89. 80. 81. See id. ' 4.122. 82. See Methanex, Final Award, supra note 8, at Part II, c1 1-25. 83. See NAFTA, supra note 33, Chapter 6, arts. 601-09. Article 602 of NAFTA allows Mexico many reservations for cross-border trade in energy since the energy sector is highly regulated and constitutionally protected. See STEPHEN ZAMORA ET AL., MEXICAN LAW 285 (2004) (discussing economic and environmental regulation).

84. GATT distinguishes between tariff duties and other kinds of taxes such as internal taxes. Internal taxes are collected after the goods exit customs and enter into domestic commerce. However an Interpretative Note clarified that a tax collected at the time the product is in customs could still be construed to be an internal tax. Furthermore, exemptions from income taxes are not covered under Article Ill. See JACKSON, supra note 25, at 280; see also GATT, supra note 1, Annex 1, ad. art. Ill.

85. See RAj BHALA,

INTERNATIONAL TRADE LAW: THEORY AND PRACTICE 423 (2d ed. 2001) (explaining that paragraph one of Article Ill sets forth a general principle applicable not only to internal taxes but also to internal legislation and regulations more gener-

ally; on the other hand, paragraph specifically).

two of Article III deals with internal

taxes

86. GATT, supra note 1, art. 111:1. 87. See id., art. 111:2; see also Jackson, supra note 25, at 279. 88. The fourth paragraph of Article Ill of GATT states:

The products of the territory of any contracting party imported into the territory of any other contracting party shall be accorded treatment no less favourable than that accorded to like products of national origin in respect of all laws, regulations and requirements affecting their internal sale, offering for sale, purchase, transportation, distribution or use. The provisions of this paragraph shall not prevent the application of differential internal transportation charges which are based exclusively on the economic operation of the means of transport and not on the nationality of the product. GATT, supra note 1, art. 111:4.

Much of the discussion regarding the scope of "like products" under Article 111:4 concerns whether Article 111:4 deals only with "like products" but not necessarily "directly competitive or substitutable" products. Arguably, Article 111:1 does not apply to Article 111:4, thereby not requiring the determination of whether the non-fiscal measure was passed "so as to afford protection." See Trebilcock & Gibi, supra note 67, at 195-96.

89. See JACKSON, supra note 25, at 279-80. 90. The first sentence of Art. 111:2 specifically refers to "like domestic products" in order to prevent internal measures from discriminating between these products based on their origin. See id. at 281; see also GATT, supra note 1, art. 111. The second sentence of Art. 111:2 incorporates the more general obligation of Art. 111:1 that member states should not pass internal measures that "afford protection to domestic production." See GATT, supra note 1, art. 111:2; see also JACKSON, supra note 25, at 281.

91. See GATT, supra note 1, art. 111:1. 92. See Frieder Roessler, Diverging Domestic Policies and Multilateral Trade Integra-

tion, in FAIR TRADE AND HARMONIZATION: PREREQUISITES FOR FREE TRADE? 29 (Jagdish Bhagwati & Robert E. Hudec eds., 1996) (noting that the Panel's focus on the Border Tax Adjustments criteria may be "underinclusive," allowing some protectionist measures, and "overinclusive," invalidating legitimate domestic legislation that places no major burdens on free trade). See generally, Lothar Ehring, De Facto Discrimination in WTO Law: National and Most-Favored Nation Treatment- or Equal Treatment?, 36 J. of World Trade 921 (2002), available at http://www.jeanmonnetprogram.org/papers/01/013201. html (highlighting and exploring the question of whether national treatment violations under Article III include facially origin-neutral measures that have no or very little overall protective effect on domestic industries).

11ol1. 4 0

93. See JACKSON, supra note 25, at 281-82 (If "a measure complies with the first sentence is it possible that it can nevertheless contravene the second sentence? The answer appears to be clearly 'yes"'). The original language of Article III in the Havana Charter included the following:

[I]n cases in which there is no substantial domestic production of like-products of national origin, no contracting parties shall apply new or internal taxes on the products of the territories of other contracting parties for the purpose of affording protection to the production of directly competitive or substitutable products which are not similarly taxed; and existing internal taxes of this kind shall be subject to negotiation for their reduction or elimination.

Id. at 281. This language was later eliminated and the Interpretative Note specified that Article III obligation attached to products that were "directly competitive or substitutable products."

Id. at 282.

94. See Hudec, Requiem, supra note 68, at 632. 95. PETER SUTHERLAND ET AL., THE FUTURE OF THE WTO: ADDRESSING INSTITUTIONAL CHALLENGES IN THE NEW MILLENNIUM 15 (2004) (stating that the WTO does not "disallow market protection; rather it sets some strict disciplines under which governments may choose to respond to special interests").

96. See generally Robert E. Hudec, Differences in National Environmental Standards: The Level-Playing-Field Dimension, 5 MINN. J. GLOBAL TRADE 1 (1996) (discussing generally ideas behind the "'level-playing-field' complaint" and demonstrating it as a doctrine of "fairness" in trade matters); see also EC-Asbestos, supra note 27, 98, at 37 (stating that Article III eliminates disguised trade restrictions that "affect[ ] the competitive relationship, in the marketplace, between the domestic and imported products involved, 'so as to afford protection to domestic production'). 97. See JACKSON, supra note 25, at 289. 98. See Hudec, Requiem, supra note 68, at 624-26. 99. See, e.g., Panel Report, Argentina-Measures Affecting the Export of Bovine Hides

and the Import of Finished Leather, 9 8.230, WT/DS155/R (Dec. 19, 2000).

100. See Panel Report, Japan-Taxes on Alcoholic Beverages, WT/DS8/R, WT/DS10/R, WT/DS11/R, 1996 WL 406720, (July 11, 1996) [hereinafter 1996 Japan-Alcoholic Beverages] (asserting that complainant has the burden of proof in demonstrating that products are "like products" and that foreign products are taxed in excess of domestic ones); see also, Edward S. Tsai, "Like" is a Four-Letter Word- GATT Article Ill's "Like Product" Conundrum, 17 BERKELEYJ. INT'L LAW 26, 37-38 (1999) (stating that after the 1987 Alcoholic Beverages Panel decision, the test for national treatment consisted of three steps: "1) whether products are like; 2) whether the contested measure is an 'internal tax' or 'other internal charge' ...; and 3) whether the tax imposed on foreign products is in excess of the tax imposed on like domestic products"). Furthermore, in cases where it is not so clear if products are "like,"

[a] tax conforming to the requirements of the first sentence of paragraph 2 [of Art. ll] would be considered to be inconsistent with the provisions of the second sentence [of Art. 111:21 only in cases where there was competition between, on the one hand, the taxed product and, on the other hand, a directly competitive or substitutable product which was not similarly taxed.

WORLD TRADE ORGANIZATION, WTO ANALYTICAL INDEX GUIDE TO WTO LAW AND PRACTICE, ad. art. III 1 2 (2003).

Id. at 621-22.

101. See Hudec, Requiem, supra note 68, at 620. 102. 103. See Sykes, supra note 54, at 4-5 (recognizing that some regulations may have "protective effects" but may be justified because of a "nonprotectionist regulatory objective"). The author refers to these measures as the "'least restrictive means"' and purports to exclude those constituting "regulatory protectionism." Id.

types of measures as

these

104. See, e.g., Panel Report, Dominican Republic-Measures Affecting the Importation

and Internal Sale of Cigarettes, 17 5.67-68, WT/DS302/R (Nov. 26, 2004).

105. See EC-Asbestos, supra note 27, 'VI 96-97.

106. See GATT Dispute Panel Report, Japan-Customs, Duties, Taxes and Labeling Practices on Imported Wines and Alcoholic Beverages, L/6216-34S/83, 1987 WL 421964 (GATT) (Oct. 13, 1987) (adopted Nov. 10, 1987) [hereinafter 1987 Japan-Alcoholic Beverages].

107. See U.S.-Malt Beverages, supra note 36. 108. The claimants alleged that this tax scheme applied different tax rates to "like" products and therefore violated the national treatment obligations under Article Ill; the other alcoholic beverages included vodka, gin, brandy, etc. Id.

2.3.

109. In the first Panel decision, the WrO Panel rejected Japan's position and took a flexible approach in defining "like products." In supporting a case by case analysis for interpreting the terms, the Panel concluded that "like" in this case could refer not only to the same physical characteristics and end uses but also to "directly competitive or substitutable products" having common end-uses even if their physical characteristics differ. See 1987Japan-Alcoholic Beverages, supra note 106.

110. U.S.-Malt Beverages, supra note 36. 111. Also, among other state measures Canada alleged were in violation of Article Ill, the Federal Alcohol Administration Act required federal permits for all wholesalers. Id. c1 2.1. 112. 113. Id. 114. Id. '1 5.1.

Id. 1 5.23-26.

115. See generally Hudec, Requiem, supra note 68 (stating that the United States' fiscal measures were non-discriminatory and thus legal); see also Tsai, supra note 100, at 35-36 (citing U.S.-Malt Beverages' use of the "aim-and-effect" test in CH 5.12-15 and emphasizing that under this one-step test, "the sole determination [is] whether or not a measure was a protectionist measure").

116. Farber & Hudec, supra note 11, at 1423 (stating that the GATT panel "concluded that the 'likeness' issue must always turn on the purpose for which the product distinction is being tested").

117. U.S.-Malt Beverages, supra note 36; see also Farber & Hudec, supra note 11, at 1423 (stating that "the purpose of Article Ill 'like product"' standard [was] to distinguish between protectionism and good faith regulation"). Instead, the purpose of Article Ill, was as described under Article 111:1, not to allow internal measures that are applied "so as to afford protection to domestic production." U.S. -Malt Beverages, supra note 36, cl 5.25; see also GATT, supra note 1, art. 111:1.

118. U.S. -Malt Beverages, supra note 36 (holding that the purpose of Article IllI "like product" standard is to distinguish between protectionism and good faith regulation, the decision concluded that the "likeness" of the 2 products must depend on whether any non-trade regulatory purpose was served by distinguishing between the 2 types of wine").

119. U.S.-Malt Beverages, supra note 36. 120. U.S.-Malt Beverages and U.S.-Taxes on Automobiles seem to suggest that for there to be a violation of Article 111:2, the measure be motivated by a legislative intent to protect domestic industry. See generally Tsai, supra note 100.

121. U.S.-Malt Beverages, supra note 36, 1 5.26.

122. See Roessler, supra note 92, at 29 (stating that "the starting point of the analysis cannot be the concrete objects to which an internal tax or regulation is applied but only the abstract categories of products distinguished by the contracting party").

123. See id. Roessler also distinguishes between policies that discriminate between domestic and foreign products and policies that merely discriminate between different categories of products regardless of their origin. Id. 124. U.S.-Malt Beverages, supra note 36, 1 5.72. 125. See WTO Appellate Body Report, Japan-Taxes on Alcoholic Beverages, WT/DS8/

AB/R (Oct. 4, 1996) [hereinafter 1996 Alcoholic Beverages Appellate Body].

126. The Working Party Report on Border Tax Adjustments provides a set of criteria that Panels have used in interpreting the "likeness" of products. These criteria include the following:

1. physical characteristics of a product including its properties, nature, and quality; 2. the end-uses of a product in any given market; 3. the tastes and habits of consumers' tastes and habits, which may vary; and 4. the tariff classification of the products.

See Report of the Working Party on Border Tax Adjustments, '118, L/3464 (adopted Dec. 2, 1970). Whether the products fall under the same tariff classification (HTS) is an important factor in determining similar physical characteristics but it in itself is not conclusive. See 1987Japan-Alcoholic Beverages, supra note 106. 6.15.

127. 1996 Japan-Alcoholic Beverages, supra note 100, 128. Unlike the 1987 panel decision, the Appellate Body emphasized that Article 111:2 should be read in light of the purpose of Article 111:1: "the broad and fundamental purpose of Article III is to avoid protectionism in the application of internal tax and regulatory measures." 1996 Alcoholic Beverages Appellate Body, supra note 125, at 16.

129. 1958 Panel Report on Italian Discrimination against Imported Agricultural Machinery stated that "it was considered ... that the intention of the drafters of the GATT was clearly to treat the imported products in the same way as the like domestic

products once they had been cleared through customs; otherwise, indirect protection could be given." Italian Discrimination Against Imported Agricultural Machinery, supra note 38, ' 11.

130.

Id.; see also EC-Asbestos, supra note 27, H 95-97 (stating that the purpose of the Article III provision of GATT was to eliminate domestic measures that were in place "so as to afford protection to domestic production").

131. 1996 Japan-Alcoholic Beverages, supra note 100, '1 6.12-6.13. 132. Roessler, supra note 92, at 29 (1996) (stating that "the starting point of the analysis cannot be the concrete objects to which an internal tax or regulation is applied but only the abstract categories of products distinguished by the contracting party"). But see Andrew Green, Climate Change, Regulatory Policy and the WTO: How Constraining are Trade Rules?, 8 J. INT'L ECON. L. 143, 168-69 (concluding that after EC-Asbestos, "national treatment analysis of any regulatory categories would result either in a form of implicit balancing of the regulatory purpose and the trade restriction or in a test which captures both innocent and protectionist measures that disproportionately impact imports"). See also Appellate Body Report, Dominican Republic-Measures Affecting the Importation and Internal Sale of Cigarettes, 96, WT/DS302/AB/R (May 19, 2005) (approving a bond requirement for both domestic and foreign producers of cigarettes even though it gave an "extra burden" to foreign importers due to their smaller tax bracket).

133. Trebilcock & Giri, supra note 67, at 189, 196 (characterizing the two different approaches in the Article Ill decisions as either the "textual" approach taken in Japan-Alcoholic Beverages or the "contextual" "aim and effects" test applied in U.S.-MaltBeverages).

134. See generally EC-Asbestos, supra note 27.

135. See id.

1-2 (discussing French Decree No. 96-1133, decret no. 96-1133 relatif a l'interdiction de l'amiante, pris en application du code de travail et du code de la consummation).

Id. 'IN 4, 30. Id. '1 30. Id.

136. 137. 138. 139. EC-Asbestos, supra note 27, 140.

97-98.

32.

33. The opinion states that the European Community's interpretation of "like" concluded that, "instead of comparing the products prohibited by the French Decree at issue (all carcinogenic asbestos fibres), the Panel erroneously compared the allegedly 'like' products with an arbitrary third category of products, namely 'fibres with certain applications."' See id. 99.

141. 142. Id. The Appellate Body went on to state that not all products in some competi-

tive relationship are necessarily "like products." Id.

143. Id.

144.

Id. 1 98. See also, Lothar Ehring, supra note 92 (examining EC-Asbestos and the Appellate Body's broadening of "like product" into groups of "like products" and treatment" as the second condition of the Article 111:4 describing "less favorable analysis).

145. EC-Asbestos, supra note 27, '11 101-02 (stating that these criteria "provide a framework for analyzing the 'likeness' of particular products" and are "simply tools to assist in the task of sorting and examining the relevant evidence"); see also Howse & Tuerk, supra note 6, at 289 (noting that "a finding of 'likeness,' on market-based criteria, will not be dispositive of a finding of violation of Article 111:4.").

146. EC-Asbestos, supra note 27, l1 111-12. 147. Id. '1 114. 148. Id. cII 112-14. 149. Id. ' 103. 114. 150. Id. 151. Id. 1 103. 152. See Howse & Tuerk, supra note 6, at 288 (stating that the Appellate body in EC-Asbestos reaffirmed the "basic purpose of Article III as the discipline of protectionist measures" and not merely guaranteeing "market access" of like imported products).

153. EC-Asbestos, supra note 27, '1 114 (noting that the carcinogenic effects of asbes-

tos had been well-established).

154. Appellate Body Report, Dominican Republic-Measures Affecting the Importation

and Internal Sale of Cigarettes, supra note 132, IN 94-96.

155. Id. 96. 156. See id. 157. Id. 158. The Appellate Body found that the products were "like;" however, despite unequal treatment in applying the bond requirement, the effects were not sufficiently detrimental to amount to a national treatment violation. See id.

159. See Howse & Tuerk, supra note 6, at 284 (stating that "[i]f the WTO is to regain citizens' confidence, it has to prove its ability to balance the freedom of governments to pursue legitimate domestic objectives with the need to secure the benefits of trade liberalization"); see also Hudec, Requiem, supra note 68, at 620 (stating that "[tihe policing 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 member states").

160. See Jan Bohanes, Risk Regulation in WTO Law: A Procedure-Based Approach to the Precautionary Principle, 40 COLUM. J. TRANSNAT'L L. 323, 326 (2002) (stating that science "functions as a distinguishing criterion between legitimate and illegitimate trade barriers in the form of national health regulation"). The author explores the use of the precautionary principle WTO assessments of legitimate regulatory measures. See generally id.

Id. at 326.

161. 162. Italian Discrimination Against Imported Agricultural Machinery, supra note 38. 163. Id. c1 2. 164. Id. The panel considered whether there was a violation of Article 111:4. Signifi-

cantly, however, this report did not reach the Final Determination Stage.

3. 16.

165. Id. 166. Id. 167. Id. 168. Id.

[Tihe extension of the credit facilities in question to the purchasers of imported tractors as well as domestically produced tractors would detract from the attainment of the objectives of the Law, which aimed at stimulating the purchase of

tractors mainly by small farmers and co-operatives in the interests of economic development. If, on the other hand, the objective of the Law, although not stated in the text thereof, were to protect the Italian agricultural machinery industry, the Panel considered that such protection should be given in ways permissible under the General Agreement rather than by the extension of credit exclusively for purchases of domestically produced agricultural machinery.

Id. Ultimately, the panel reached its determination based on the adverse effects of the Italian law on imports of agricultural machinery according to Paragraph 2 of Article XXIII rather than on a determination of an Article III violation. See id.

25.

169. See, e.g., Mexico-Measures Affecting Telecommunications Services, supra note 39. 170. Id. 9[ 2.2. 171. Id. 2.1. 172. Id. 3.1(a). 173. Id. c1 3.1(b). 174. Id. 'H 7.222-269 (defining "anticompetitive practices" as "actions that lessen

rivalry or competition in the market").

175. Id. 1 7.235-237. 176. Id.

7.223.

177. 1996 Japan-Alcoholic Beverages, supra note 100, '1 6.17-6.19. 178. Id. at 11 6.21-6.23. 179. Farber & Hudec, supra note 11, at 1413. 180. See, e.g., City of Philadelphia v. N.J., 437 U.S. 617 (1978); C&A Carbon, Inc v. Clarkstown, N.Y., 511 U.S. 383 (1994). Note that with facially discriminatory measures, the presumption is that the state measure is unconstitutional. See e.g., Lewis v. BT Investments Managers, 447 U.S. 27 (1980) (declaring unconstitutional a state law that prevented out-of-state banks from owning investment advisory businesses within the state); Reynoldsville Casket Co. v. Hyde, 514 U.S. 749 (1995).

181. Such policy is permissible if it is an instrument of state police power. See generally Southern Pacific Co. v. Arizona, 325 U.S. 761 (1945); DiSanto v. Pennsylvania, 273 U.S. 34 (1927); Cooley v. Board of Wardens, 53 U.S. 299 (1851); Cooley v. Board of Wardens, 53 U.S. 299 (1851); Wilson v. Black-Bird Creek Marsh Co., 27 U.S. 245 (1829); Gibbons v. Ogden, 22 U.S. 1, 203 (1824). See also Maine v. Taylor, 477 U.S. 131, 138 (1986) (citing Hughes v. Okla., 441 U.S. 322, 336 (1979 . For further examples, see also Sporhase v. Nebraska ex rel. Douglas, 458 U.S. 941, 957 (1982); Hunt v. Wash. St. Apple Advertising Comm., 432 U.S. 333, 353 (1977); Dean Milk Co. v. Madison, 340 U.S. 349, 354 (1951). Although the Court usually discusses the need for a "legitimate" local purpose in the context of highly deferential rational basis review, in several of the dormant commerce clause cases the Court states that the justification for discrimination against interstate commerce must be subjected to the strictest scrutiny. Taylor, 477 U.S. at 141. These cases strongly suggest that the Court must find that the state has at the very least an important or possibly a compelling reason for discrimination before it will uphold a statute or regulation.

182. Farber & Hudec, supra note 11, at 1412. 183. See, e.g., Hunt v. Washington State Apple Advertising Commission, 432 U.S. 333, 354 (1977) (rejecting a state law requiring state-approved labels as creating discriminatory effects that could have been avoided with other nondiscriminatory measures). But see Pike v. Bruce Church, Inc., 397 U.S. 137, 142 (1970).

184. Taylor, 477 U.S. at 138. The state may also demonstrate that the measure is an

instrument of the state's general police power. Id.

185. Pike, 397 U.S. at 142 (citing Hughes, 441 U.S. at 336). See also Farber & Hudec, supra note 11, at 1413 (distinguishing between the burden of proving no alternative nondiscriminatory measures for cases involving discriminatory effects and the balancing test applied in cases involving measures with only "incidental burdens on interstate commerce"); Hunt, 432 U.S. at 353.

186. Farber & Hudec, supra note 11, at 1413. 187. Pike, 397 U.S. at 142 (stating that a "statute [that] regulates evenhandedly to effectuate a legitimate local public interest and its effects on interstate commerce are only incidental, it will be upheld unless the burden imposed on such commerce is clearly excessive in relation to the putative local benefits").

188. Factors U.S. courts consider in deciding the legality of facially neutral regula-

tions include:

1. the existence of a discriminatory intent to economically advantage local business; 2. the measure, while facially neutral, consists of a "proxy characteristic" that will advantage local business at the expense of outsiders; 3. the measure has the effect of embargoing outside products; 4. the measure provides a competitive advantage to local products over the same outside products; and 5. the measure may lack uniformity and consistency in its application, burdening outside products.

See Farber & Hudec, supra note 11, at 1416-17.

189.

Id. at 1412-13.

the interests of interstate commerce with those of intrastate commerce. '90 In dealing with these "incidental burdens," the Court seems to presume a non-protectionist intent.191 Furthermore, in the context of dormant Commerce Clause, the Court makes an exception when the state participates in the market. 1

In the dormant Commerce Clause context, U.S. federal courts avoid focusing on protectionist intent and tend to prefer a balancing approach approving of measures that are facially neutral and create only "incidental burdens" on interstate commerce. This approach, of course, places more responsibility on federal judges in deciding whether regulatory benefits are sufficient to justify burdens on outside states.193 On the other hand, in applying Article III, the WTO panels stray so far in the direction of free trade that they ignore the possibility of a legitimate domestic regulatory policy that (1) though discriminatory, places only "incidental burdens" on free trade; and (2) arises from domestic industries traditionally regulated or "partially" regulated. Furthermore, the "like products" test of JapanAlcoholic Beverages does not consider the possibility of an even-handed regulation having protectionist effects but intended to further a domestic purpose that is legitimate and non-protectionist.'94

The fact that U.S. and GATT law handle regulatory measures in similar but distinct ways invites us to consider whether they address different sets of concerns. Legal doctrines dealing with the dormant Commerce Clause are arguably concerned with the preservation of a "political union." In other words, regulatory measures that amount to state protectionism are hostile to preserving the larger interests of a centralized government and

190. Id. at 1413. 191. See id. (stating that the difference in treatment of measures with discriminatory effects and those with only incidental burdens "suggests an absence of protectionist purpose").

192. This is referred to as the "market participant exception." See Hughes v. Alexandria Scrap Corp., 426 U.S. 794 (1976); White v. Massachusetts Council of Construction Employers, Inc., 460 U.S. 204 (1983). See also Dan T. Coenen, Untangling the Market Participant Exemption to the Dormant Commerce Clause, 88 MICH. L. REV. 395, 399-406 (1989) (discussing the development of the market-participant rule).

193. See Farber & Hudec, supra note 11, at 1414-17; see also Donald H. Regan, The Supreme Court and State Protectionism: Making Sense of the Dormant Commerce Clause, 84 MicH. L. REV. 1091, 1093 (1986) (arguing for a judicial approach that stresses a finding of protectionist intent rather than a balancing test, especially in the context of movement of goods cases).

194. EC-Asbestos may be a step in this direction. See supra Part I.C. For example, in the context of the U.S. dormant Commerce Clause, America Can. Co. v. Oregon Liquor Control Commission, 517 P.2d 691 (1973), involved an Oregon law intended to protect the environment by discouraging the use of non-returnable, non-reusable beverage containers by placing a refundable deposit on the containers. The deposit was repayable to consumers and retailers upon the return of the container. The effect of the statute was to encourage use of glass bottles instead of metal cans. This would increase transportation costs (because glass is heavier than cans), which tended to favor Oregon bottlers over out of state bottlers. The Appellate Court found no protectionist purpose and the statute was upheld. See generally id.

195. See Regan, supra note 193, at 1112-13 (discussing the "concept-of-union" objec-

tion to state protectionism).

196. U.S. CoNsT. art. I § 8. The Commerce Clause authorizes Congress to "regulate Commerce with foreign Nations, and among the several States, and with the Indian Tribes." Id.

197. Amy M. Petragnani, The Dormant Commerce Clause: On its Last Leg, 57 ALB. L. REv. 1215, 1215 (1994). But see generally, Jim Chen, A Vision Softly Creeping: Congressional Acquiescence and the Dormant Commerce Clause, 88 MINN. LAw REv. 1764 (2004) (arguing that the power to overturn the dormant Commerce Clause rests in the legislative branch and that Congress's silence on the issue is significant).

198. See Jim Chen, supra note 197, at 1796 (stating that "the shameless judicial formulas as that comprise the dormant Commerce Clause doctrine continue to uphold the core federal interest in a nationwide common market").

199. See Farber & Hudec, supra note 11, at 1405-06 (stating that the "postwar initiative that created GATT also rested on a collective perception that ruthless treatment of the economic interests of outsiders is inconsistent with the conditions of peaceful international society").

200. See id. at 1404. 201. See infra Part II.B. 202. Farber & Hudec, supra note 11, at 1403 (stating that the GATT and the dormant

Commerce Clause share common concerns). The authors argue,

The modern regulatory state inevitably produces burdens on trade, if only because of the unavoidable lack of regulatory uniformity. For various reasons, many of these burdens likely are unwarranted, and at least some are in fact due to protectionist efforts by local industries. Yet, tribunals have only a limited

warrant to override the policy choices of local legislatures. Tribunals must accord respect to the democratic process as well as to the prerogatives (or sovereignty) of local governments.

See id.

203. See id. (stating that "GATT agreements also contain a number of DCC-like prescriptions, prohibiting national regulatory measures that constitute 'unnecessary obstacles to international trade' or a 'disguised restriction on international trade"').

204. See id. at 1404 (stating that in the context of the dormant commerce clause, "we accord legislatures the freedom to choose economically harmful results, including the freedom to choose the least efficient way to accomplish a particular social objective").

205. GATT, supra note 1, art. xx. 206. Hudec, Requiem, supra note 68, at 620, 626-29. 207. Id.; see also GATT, supra note 1, art. xx. 208. See supra Part L.C for discussion on EC-Asbestos and showing that EC-Asbestos

case settled the issue somewhat.

209. Hudec, Requiem, supra note 68, at 621-22. 210. Most of the cases involving decisions on Article XX have dealt with dejure discriminatory measures. See id. See generally ROBERT E. HUDEC, ENFORCING INTERNATIONAL TRADE LAW 373-585 (1993) (providing a list of GATT legal complaints from 19481989).

211. See FREDERICK M. ABBOTT, LAW AND POLICY OF REGIONAL INTEGRATION: THE NAFTA AND WESTERN HEMISPHERIC INTEGRATION IN THE WORLD TRADE ORGANIZATION SYSTEM 107-108 (1995).

212. See Sutherland Report, supra note 95, at 11 (stating that in reaping the benefits of free trade, "[m]uch remains in the hands of individual governments"); see also id. at 29-34 (discussing issues of sovereignty and free trade).

213. See JOHN H. JACKSON, THE JURISPRUDENCE OF GATT AND THE WTO: INSIGHTS ON TREATY LAW AND ECONOMIC RELATIONS 21 (2000) (stating that the goals behind the GATT agreement included "the prevention of war and the establishment of a just system of economic relations" as well as "the economic benefits that might derive from international trade and economic stability") [hereinafter JACKSON, JURISPRUDENCE OF GATT]; see also Uruguay Round Agreement Establishing the World Trade Organization, Apr. 15, 1994, pmbl.

214. See McGinnis & Movsesian, supra note 46, at 590-96.

215.

Id. at 556-58. Id. at 550-51. Id. at 530-31.

Id. at 566 (describing the WTO's adjudication of national regulations as one following an "antidiscrimination model" and describing the difficulties in the regulatory model). The authors describe the "antidiscrimination model" as one which can promote the development of democratic and transparent regulatory frameworks. See id. at 566-72. 216. 217. 218. 219. Id. 220.

Id. at 572. The authors define "covert protectionism" as having have two main characteristics: 1) "the measure must be one that would not have been enacted but for the benefits it gives domestic industries by restraining imports; 2) the measure must lack a public interest foundation." Id. The authors do not clearly define measures with a "public interest foundation" but seem to presume it if in fact the determinate rules are met. See id. at 572-73 (stating that the "second criterion may be subsumed as a practical matter by the first"). 221. Id. at 572-73. 222. Id. at 572. 223. Professors McGinnis and Movsesian emphasize that transparency is one of the key principles necessary for building true democratic institutions. See id. at 574 (also stating that "[tiransparency levels the playing field by ensuring that regulations are publicized and the steps for compliance are clear").

224. Id. at 568, 573-76. 225. Id. at 577-78.

226. Id. 227. Id. at 576. 228. See id. at 577-78. 229. Id. at 579. 230. See JACKSON, supra note 25, at 3 (describing the GATT as the wheel of a larger

machine that was the International Trade Organization).

231. See id. at 43. 232. See Thomas J. Dillon, Jr., The World Trade Organization: A New Legal Order for World Trade?, 16 MICH. J. INT'L L. 349, 351-52 (1995); see also KENNETH W. DAM, THE GATT: LAw AND INTERNATIONAL ECONOMIC ORGANIZATION 10-16 (1970); JACKSON, supra note 25, at 35-57 (providing a detailed history of the GATT agreement).

233. See Dillon, supra note 232, at 352-55. 234. See id. at 355. 235. See JACKSON, JURISPRUDENCE OF GATT, supra note 213, at 134; see also Robert E. Hudec, Comment by Robert E. Hudec, in EFFICIENCY, EQUITY, AND LEGITIMACY: THE MULTILATERAL TRADING SYSTEM AT THE MILLENIUM 297 (Roger B. Porter et al. eds., 2001) (stating that the WTO may be viewed as "a freestanding institution" as well as "an ingredient of the domestic decision-making process of national governments") [hereinafter Hudec Comment].

236. See Agreement Establishing the World Trade Organization, supra note 213, art. III (providing that the WTO "shall facilitate the implementation, administration and operation, and further the objectives" of the WTO and "shall provide the forum for negotiations" among member states).

237. See Rufus H. Yerxa, Comment by Rufus H. Yerxa, in EFFICIENCY, EQUITY, AND LEGITIMACY: THE MULTILATERAL TRADING SYSTEM AT THE MILLENNIUM 381, 382 (Roger B. Porter et al. eds., 2001).

238. See JACKSON, JURISPRUDENCE OF GATT, supra note 213, at 155 (describing members of WTO panels as "delegates" of member states, entrusted with authority "to vindicate the political decisions made by the members as a whole"); see also Agreement Establishing the World Trade Organization, supra note 213, pmbl.

239. See Steinberg, supra note 45, at 250-51. 240. See id. at 250. 241. See Robert Howse & Kalypso Nicolaidis, Legitimacy and Global Governance: Why Constitutionalizing the WTO is a Step Too Far, in EFFICIENCY, EQUITY, AND LEGITIMACY: THE MULTILATERAL TRADING SYSTEM AT THE MILLENIUM 227 (Roger B. Porter et al., 2001) (distinguishing between the GATT as a forum for bargaining among its self-interested member states and the later WTO, which can be viewed as "performing constitutional functions or to be an incipient global economic constitution").

242. See Steinberg, supra note 45, at 250 (describing the "fundamental purpose" of the original GATT dispute settlement system as "facilitat[ing] the diplomatic settlement of trade disputes between contracting parties"). The author goes on to describe the current WTO system as one "far more legalized and fundamentally adjudicative." Id. at 250-51. But see JACKSON, JURISPRUDENCE OF GATT, supra note 213, at 155 (describing members of WTO panels as "delegates" of member states, entrusted with authority "to vindicate the political decisions made by the members as a whole").

243. Examples of such regional and bilateral trade agreements include the U.S. Tariff Act of 1974; the European Union; Mexico's SECOFI; and Canada's Court of International Trade.

244. See generally Kent Jones, Contrary Opinion: Who's Afraid of the WTO?, 41 CHALLENGE 105, 105-19 (1998) (deconstructing anxieties accompanying the reality of economic integration among world countries).

245. See U.S. Tariff Act, 19 U.S.C.A. §1677 (1974) (referring to the WTO). See Hudec Comment, supra note 235, at 297 (stating that the WTO may be viewed as "a freestanding institution" as well as "an ingredient of the domestic decision-making process of national governments").

246. See David Gantz, A Post-Uruguay Round Introduction to International Trade Law in the United States, 12 ARiz. J. INT'L & COMP. LAw 7, 9-10 (1995) (stating that U.S. trade law results from a political process that includes various special interests).

247. See Bohanes, supra note 160, at 346 (stating that "science is clearly linked to the

notion of standards of review in the adjudicative setting").

248. Id. at 356 (discussing differences between "scientific risk assessment and public perceptions of risk"). Specifically, the author explains that public assessments of risk made through a legitimate democratic process should be respected despite the risk of inconsistency with scientific conclusions. See id. at 356-62.

249. See generally Deborah Cass, The "Constitutionalizaton" of International Trade Law: Judicial Norm-Generation as the Engine of Constitutional Development in International Trade, 12 EUR. J. INT'L L. 39, 53-57 (2001) (arguing that the case law of the WTO Appellate Body reflects a gradual "constitutionalization" analogous to adjudication under international law).

250. GATT Appellate Body Report, EC-Measures Concerning Meat and Meat Products

(Hormones), GATT Doc. WT/DS26/AB/USA Appellate Body-Beef Hormones].

[ 124-25 (Jan. 16, 1998) [hereinafter

251. Bohanes, supra note 160, at 354 (emphasizing that there was not a finding of a disguised trade restriction but rather that "the EC ban was simply bad lawmaking because the results of the scientific risk assessments did not sufficiently and reasonably warrant the European ban").

252. See generally Mark L. Movsesian, Essay, Enforcement of WTO Rulings: An Interest Group Analysis, 32 HOFSTRA L. REV. 1 (2004) (arguing that the retaliation remedy of the Dispute Settlement Understanding, in which a prevailing party before the WTO Appellate Body can suspend trade concessions that it owes to the losing party, is necessary to counter the influence of domestic interest group upon trade regulations).

253. See id. at 4 (explaining that a retaliation remedy can "create incentives for ...that favor free trade [and] can

another set of domestic interest group-exportersneutralize the effect of the domestic groups that oppose it").

254. 1996 Japan-Alcoholic Beverages, supra note 100, EEC pleadings. 255. See JACKSON, JURISPRUDENCE OF GATT, supra note 213, at 115-17 (describing GATT dispute resolution as having "meager resources" to deal with many questions of domestic policy accurately).

256. DSU, supra note 2, art. 1.1 (stipulating that the cause of action of WTO dispute

mist be found in the covered agreements).

257. See SUTHERLAND, supra note 95, at 15. 258. See infra Part Ill (explaining how NAFTA tribunals tend to defer to WTO deci-

sions when applicable).

259. Many thanks to Frank Garcia for his comments in this respect. 260. See Sykes, supra note 54, at 4. 261. See Movsesian, supra note 252, at 4 (discussing the positive effects on free trade

that interest groups may have at the domestic level).

262. See Arvind Subramanian & Jayshree Watal, Can TRIPS Serve as an Enforcement

Device for Developing Countries in the WTO?, 3 J. OF INT'L ECON. L. 403, 406 (2000).

263. See G. Richard Shell, Trade Legalism and International Relations Theory: An Analy-

sis of the World Trade Organization, 44 DUKE L. J. 901-04 (1995).

264. See id. at 895-901. 265. See Peter M. Gerhart, The World Trade Organization and Participatory Democracy,

37 VAND. J. TRANSNAT'L L. 897, 908 (2004).

266. See FREDERICK M. ABBOTT, LAW AND POLICY OF REGIONAL INTEGRATION: THE NAFTA AND WESTERN HEMISPHERE INTEGRATION IN THE WORLD TRADE ORGANIZATION SYSTEM 41 (NAFTA Law and Policy Series eds., 1995).

Id. art. 102, § 2.

267. See id. at 107-108. 268. See NAFTA, supra note 33, art. 103. 269. 270. See id. art. 30; see also Gantz, supra note 246, at 15-16. 271. See e.g., Pope & Talbot, Inc. v. Canada, Award on Merits, Phase Two, 11 73-104 (NAFTA Ch. 11 Arb. Trib. Apr. 10, 2001), available at http://www.dfait-maeci.gc.ca/tnanac/documents/AwardMerits-e.pdf [hereinafter Pope & Talbot, Award on the Merits].

272. NAFTA, supra note 33, art. 1102. 273.

Id. (emphasis added). Article 1102 also protects "investments of investors of

another Party" in the same way. Id.

274. NAFTA tribunals also look to OECD agreements where the term "like situations" is defined. See S.D. Myers, Partial Award, supra note 55, 91 245-46; Pope & Talbot, Inc. 78 (NAFTA Ch. 11 Arb. Trib. June 26, 2000), available at v. Canada, Interim Award, [hereinafter Pope & http://www.international.gc.ca/tna-nac/documents/pubdoc7.pdf Talbot, Interim Award].

275. See NAFTA, supra note 33, art. 1136(1) (providing that arbitral awards are bind-

ing only on the parties and only in respect to the particular case).

276. See Daniel M. Price & P. Bryan Christy, 1II, An Overview of the NAFTA Investment Chapter: Substantive Rules and Investor-State Dispute Settlement, in THE NORTH AMERICAN FREE TRADE AGREEMENT: A NEW FRONTIER IN INTERNATIONAL TRADE AND INVESTMENT IN THE AMERICAS 165, 165 (Judith H. Bello, Alan F. Holmer, & Joseph J. Norton eds., 1994).

277. See NAFTA, supra note 33, art. 1120 (allowing investors to submit a claim to

arbitration under ICSID or UNCITRAL arbitration rules).

278. See generally Surya P. Subedi, The Challenge of Reconciling the Competing Principles within the Law of Foreign Investment with Special Reference to the Recent Trend in the Interpretation of the Term "Expropriation", 49 INT'L LAw. 121 (2006).

279. S.D. Myers, Partial Award, supra note 274. These restrictions were in accordance

with Canadian commitments under the Basel Convention.

id. 310. Id. 1 131.

280. 281. 282. Id. 283. Id. cj 252. 284. Id. cj 194.

255-56.

the objectives of NAFTA.285

285. Id. ' 195. The Tribunal noted that the "indirect environmental objective" was "to keep the Canadian industry strong in order to assure a continued disposal capability." Id.

286. Id. c9 243-44. 287. Id.

246. Much like the WTO panels, the NAFTA Tribunal failed to recognize the difference in an Article XX justification for regulatory measures that for all other purposes was a violation of Article III, and the possibility that under the auspices of Article Ill, the measure may not necessarily be a violation at all.

288. S.D. Myers, Partial Award, supra note 279, ' 248. OECD does consider the relevant domestic policy objectives in its determination of whether the investments are in "like circumstances." See id. '11 248, 250. Note that the tribunal took an expansive view of "sector." Id.

289. Id. 1 250. 290. Id. 291. But see generally EC-Asbestos, supra note 27. 292. S.D. Myers, supra note 279, 9 252. It is important to note that in fact there was a comparable Canadian industry in the disposal of PCBs. Dividing the 1102 analysis along the lines of discriminatory measures between nationals and non-nationals was a natural result.

293. See id. at 11 255. 294. See id. 295. Id.

255; see Sykes, supra note 54 (identifying some measures as the "least restrictive means" of achieving a legitimate regulatory objective). Interestingly, this conclusion is also reminiscent of cases along the lines of Hunt v. Washington State Apple Advertising Commission in the dormant Commerce Clause scenario, where facially nondiscriminatory measures with discriminatory effects on outsiders would be struck down unless there was proof of no alternative means of obtaining the legitimate state goal. See Hunt v. Washington State Apple Advertising Com'n, 432 U.S. 333, 353-54 (1977).

296. S.D. Myers, Partial Award, supra note 279, c1 95-102. 297. See id. 298. Pope & Talbot, Interim Award, supra note 274. 299. See id. 300. Other allegations included violations of articles 1106 and 1105. The investor was Pope & Talbot Ltd., a corporation organized under the laws of the Province of British Columbia, which manufactured and sold standardized and specialty wood lumber.

Harvest in the province of BC and operated three sawmills and two forestry divisions in that same province. See Softwood Lumber Agreement, Can.-U.S., art. It, 9 1, Apr. 1, 1996, available at http://www.dbtrade.com/casework/softwood/175976w.htm#art10.

301.

In June of 1996, Canada issued SLP Export Permit Fees Regulations, introducing an administrative fee to be paid by an exporter for the issuance of a permit in respect of exports of softwood lumber products in the EB and for export fees to be paid for permits in the LFB of $50 and in the UFB of $100 per thousand board feet. See id.

302. See id. 303. Pope & Talbot, Interim Award, supra note 298, 304. 305. 306. 307. Pope & Talbot, Award on the Merits, supra note 271, 99 93-95. Claims under

Id. 6. Id. 30. Id. 9 47-48.

30.

1105 were also dismissed. Id.

308. 309. 310. 311.

Id. 44. Id. 99 45-72. Id. 9 56. Id.

It becomes a matter of "rights." The Disproportionate Disadvantage Test focuses on whether the measure disproportionately disadvantages the foreign investor as compared to the domestic investor. It requires that there be a domestic investor "in like

circumstances." It does not take into account any disparate representation that the domestic investor has in that economic sector-that is, unless it is demonstrated that the disadvantaged Canadian group of investors is smaller than the advantaged group, no discrimination could exist. Canada derived this test from WTO/GATT Panel decisions, but the tribunal shows how it is used differently in those cases. The tribunal points out that these cases first decided if the services/goods were "in like circumstances" or dealing with like products. See id. 42.

312. Pope & Talbot, Award on the Merits, supra note 271, c 65. 313. 314.

Id. Id. ("That is, the situation at that point was that two Canadian companies were

66.

free to operate, while their American competitor was effectively out of business.").

315. See id.

77 (agreeing with Canada that "like circumstance" should incorporate "the entire background of its disputes with the United States concerning softwood lumber trade between the two countries").

316. The tribunal looked to the OECD's understanding of "like situations" in defining 78, n.73.

"like circumstances" in this way. Id.

317. Pope & Talbot, Award on the Merits, supra note 271, 318. Id. 319. See id. T1 78-79.

78.

320. Id. l 78. 321. See id. '99 78-79. 322. Id. 323. Pope & Talbot, Award on the Merits, supra note 271.,

79.

79 ("A formulation focusing on the like circumstances question, on the other hand, will require addressing any difference in treatment, demanding that it be justified by showing that it bears a reasonable relationship to rational policies not motivated by the preference of domestic over foreign owned investments.").

324. See EC-Asbestos, supra note 27. 325. GAMI Investments, Inc. v. Mexico, Final Award, 9 24 (Nov. 15, 2004) [hereinafter GAMI Investments, Final Award]. The claimant also alleged that his investment violated the fair and equitable treatment standard under 1105. Id.

326. See id. 99 45-48. 327. See id. c 9 44-64.

328. Id. 1 51-53. 329. See id. T9 51-58. 330. See id. 99 57-64. 331.

Id., 99 73-76 (Nov. 15, 2004) (considering whether the role of the tribunal was to determine the legitimacy and legality of the Mexican government in its implementation, or lack thereof, of relevant regulatory policy).

332. See id. 91 78. See also Loewen Group, Inc. & Raymond L. Loewen v. United States, 42 I.L.M. 811, 830 Uune 26, 2003) (stating that the relevant question regarding the Loewen trial in a Mississippi court was whether the whole trial satisfied minimum standards of international law). In concluding in the affirmative, the tribunal in Loewen deferred to the U.S. judicial process itself, and whether the claimant had access to the same opportunities as nationals to benefit from the judicial process. See id. at 831-33. The tribunal found that it did. See id.

Id. 1 110.

333. See GAMI Investments, Final Award, surpa note 325, 9 76. 334. 335. See id. 1 111. 336. Id. 99cl 111-15. 337. Id. 1 114.

338. Id. 339. GAMI Investments, Final Award, supra note 325, e 114. 340. Pope & Talbot, Interim Award, supra note 298, e 73. 341. 342. See, e.g., Appellate Body Report, Canada-Certain Measures Affecting the Automo-

Id. J 72.

tive Industry, 91 106, WT/DS139/AB/R, WT/DS142/AB/R (June 19, 2000).

343. See U.S. Trucking Services, USA-MEX-98-2008-01 ' 295 (Feb. 6, 2001). The allegations included breaches of Articles 1202 (national treatment for cross-border services), 1203 (most-favored nation treatment for cross-border services), and 1102 (national treatment for investment), and 1103 (most-favored nation treatment for investment as well). See id. 1 1.

344. Id. l 271. Chapter 9 grants parties the right to maintain their own technical and licensing standards as long as in applying them, the parties comply with the obligations of national treatment and most-favored nation status. Also, the standards themselves must be pursuant to legitimate domestic objectives and any discriminatory treatment must be justifiable and non-arbitrary. See NAFTA, supra note 33, art. 904.

345. See U.S. Trucking Services, supra note 343, ' 247. Canada was not subjected to

the same blanket refusal to consider applications. See id.

346. See, e.g., Joost Pauwelyn, Recent Books on Trade and Environment: GATT Phantoms Still Haunt the WTO, 15 EUR. J. INT'L L. 575, 580 (2004) (stating that origin-neutral yet discriminatory measures can be justified under certain circumstances).

347. See, e.g., Joost Pauwelyn, Recent Books on Trade and Environment: GATT Phantoms Still Haunt the WTO, 15 EUR. J. INT'L L. 575, 580 (2004) (stating that under WTO law origin-neutral yet discriminatory measures can be justified under Article XX and XXI of GATT).

348. See supra Part II.C for a discussion of WTO panel analysis and decisions. 349. Arguably, EC-Asbestos took a similar position. However, in EC-Asbestos, the primary focus for the Appellate Body was the products themselves, and whether the measure in question was connected to an inherent component of the product. The Appellate Body was not as concerned with the legitimacy of the measure. See EC-Asbestos, supra note 27, 84-185 (focusing primarily on the "like products" and then determining whether this product fell within the scope of Article XX).

350. Cf. Pope & Talbot, Interim Award, supra note 298,

66-67 (pointing out that according to the Vienna Convention, which is reflective of customary international law, a governmental treaty or regulation must be looked at against relevant rules of international law applicable to the relations between the parties).

351. But cf. id.

73 (noting factors that must be considered when determining whether a party violates NAFTA, as opposed to adopting a presumption of a national treatment violation).

352. See ORTnNO, supra note 347, at 24. 353. Pope & Talbot, Award on the Merits, supra note 271,

76, n.69.

354. See Metalclad Corp. v. United Mexican States, Case No. ARB(AF)/97/1, Ili 50, 65 ICSID (W.Bank) (NAFTA Ch. 11 Arb. Trib. Aug. 30, 2000), available at http:// naftaclaims.com/Disputes/Mexico/Metalclad/MetalcladFinalAward.pdf [hereinafter Metalclad I]. The claimant Metalclad alleged violations of Articles 1105 and 1110. United Mexican States v. Metalclad Corp., [2001] 81 B.C.L.R. (3d) ' 20 (Can.), available at http://naftaclaims.com/Disputes/Mexico/Metalclad/MetalcladJudgement.pdf [hereinafter Metalclad II].

355. See Metalclad 1, 1 33-36. 356. The tribunal interpreted Article 1105 as incorporating a transparency requirement. See id. 9 76; see also Metalclad II, 39[ 23, 27 (summarizing the tribunal's findings). The Supreme Court of British Columbia later found that, with respect to 1105, the Tribunal erred in determining that transparency was a requirement of 1105. See Metalclad I, 19[ 68, 70-72.

357. See id. 9 66. 358. See Metalclad 1, 9 71, 81-84, 99. 359. See Metalclad I, 360. See Request for Institution of Arbitration Proceedings by Corn Prod. Int'l, submitted pursuant to Chapter 11 of the North American Free Trade Agreement, Case No. ARB(AF)/04/1, 9 2, 23-30 ICSID (W.Bank) (Oct. 21, 2003), http://naftaclaims.com/

71-72.

Disputes/Mexico/CPI/CPI%20-%20Claim.pdf [hereinafter Request for Institution of Arbitration Proceedings]. HFCS-55 is the grade of HFCS primarily used in soft drinks and Arancia, the wholly-owned subsidiary of Corn Products International is the only producer in Mexico. Id. 3 27.

361. See Notice of Intent to Submit a Claim to Arbitration Under Section B of Chapter 11 of the North American Free Trade Agreement by Corn Prod. Int'l, Inc., C3 27-29, available at http://naftactaims.com/Disputes/Mexico/CPI/CPI%/ 20-%20Intent.pdf [hereinafter Notice of Intent to Submit a Claim]; see also Request for Institution of Arbitration Proceedings, supra note 360, 53 (describing the tax). The soft drink industry had been taxed in various periods of Mexican history and even prior to IEPS, there was a tax on soft drinks. See Notice of Intent to Submit a Claim, supra note 361, 3 27. Specifically, the IEPS applies to the following: "Gasified or mineral waters; soft drinks; hydrating or rehydrating beverages; concentrates, powders, syrups, flavor essences and extracts, that when diluted allow to obtain soft drinks, hydrating or rehydrating beverages that use sweeteners different from sugar from cane." Id. 3 28. It also defines soft drinks as "beverages that are not fermented, made with water, carbonated water, fruit extracts or essences, flavoring or any other raw material, gasified or without gas, that may contain citric acid, benzoic acid or ascorbic acid or their salts as preservatives, as long as they contain fructose." Id. 3 28, n.13 (emphasis omitted). In 2002, Mexican Congress decided to maintain the HFCS tax and expand its application. See id. 3 32.

362. Request for Institution of Arbitration Proceedings, supra note 360, 1 29; see also Mexico-Tax Measures, supra note 17 (discussing the impact of tax measures on bottlers). 363. Request for Institution of Arbitration Proceedings, supra note 360, 331 28-29. 364. The Mexican market for soft drinks is dominated by multinational corporations such as Coca-Cola (71.9% of market) and Pepsi Cola (15.1% of market). See Mexico-Tax Measures, supra note 17, 3 2.6.

365. See Notice of Intent to Submit a Claim, supra note 361, 3 10. 366. The tax measure was part of Law on the Special Tax on Production and Services (Ley del Impuesto Especial Sobre Producci6n y Servicios) and related regulations and resolutions. See Mexico-Tax Measures, supra note 17, 3 2.3.

367. Id. 3l 2.2.

See id. ' 8.3.

369. NAFTA, supra note 33, art. 1102. The claimants argued that the tax was "tantamount to expropriation." Article 1110 of Chapter 11 NAFTA prohibits Parties from "directly or indirectly nationaliz[ing] or expropriat[ing] an investment of an investor of another Party or take[ing] a measure tantamount to nationalization or expropriation of such investment" except for a specific public purpose. Furthermore, claimants contended that the tax illegally imposed "a performance requirement on soft drink and sweetener producers to procure and produce sweeteners made exclusively from Mexican inputs, more particularly sugar." Article 1106 of Chapter 11 of NAFTA prevents Parties from imposing performance requirements. Notice of Intent to Submit a Claim, supra note 361, 'IH 35-44.

370. See generally Mexico-Tax Measures, supra note 17. 371. See generally Methanex, Final Award, supra note 8.

372. Methanex did not originally bring the claim for breach of 1102 NAFTA, but introduced it in the draft Amended Statement of Claim of February 12, 2001. See id. Part IV, ch. C, '1 3.

27.

373. Id. Part 11, ch. D, 374. Id. cl 28. 375. Id. Part IV, ch. B, ' 7-8. 376. Methanex, Final Award, supra note 8, Part IV, ch. B, c 5. 377. Id. c1c 6-7. 378. Id. 1 9. Methanex explained that the analysis under 1102 required a three-step analysis: 1) a determination of "like circumstances;" 2) a determination of less favorable treatment; 3) if there is less favorable treatment, the respondent must prove legitimacy in the regulatory measure. The U.S. argued that the 1102 analysis should focus on discrimination on "the basis of nationality of ownership of an investment." See id. 111 13-14. Id. 9l 21. 379. Id. 380. Id. 1 28. 381. 382. Id. 91 24. 383. Id. 91 25.

384. Id. IN 33-35. 385.

Id. 1 37 (concluding that "Article 1102 ... [should] be read on its own terms and not as if the words 'any like, directly competitive or substitutable goods' appeared in it").

Id. 91 4.113.

386. See Mexico-Tax Measures, supra note 17, 91 4.148. 387. 388. Id. 9191 4.115-4.122. Mexico also claimed that the tax measure came under an exception under Article XX(d) which provides an exception for discriminatory measures that are "necessary to secure compliance with laws or regulations which are not inconsistent with the provisions of [GATT]." See GATT, supra note 1, art. XX(d).

389. Mexico-Tax Measures, supra note 17, 1 4.135. 390. Id. 91 4.150-4.157. 391.

Id. 91 4.57-4.58. Border tax adjustments relate to several provisions of GATT, including Article Ill, in attempting to require domestic governments to implement tax schemes that treat domestically consumed products in the same manner whether the products are produced domestically or imported. See generally JACKSON, WORLD TRADE AND THE LAW OF GATT, supra note 25, at 295. Whether the products fall under the same tariff classification (HTS) is an important factor in determining similar physical characteristics but it in itself is not conclusive. See Report of the Working Party on Border Tax Adjustments, supra note 126, 91 18; see also 1987 Japan-Alcoholic Beverages, supra note 106, 91 3.3.

392. See Mexico-Tax Measures, supra note 17,11918.37-8.59; see also GATT, supra note

1, art. I1,

2:1.

393. See Mexico-Tax Measures, supra note 17, c9 8.66. 394. Id. 11 8.37-8.59. 395. Id. 9° 8.66-8.96. In deciding whether the tax was applied "so as to afford protection," the panel, though cautioning against giving too much importance to the "subjective legislative intent of legislators and regulators in the drafting of a particular measure," focused on comments made during a meeting of the Mexican Chambers of Deputies annulling the tax exemption. Ironically, the reasons given for such annulment appear to be because the tax did not achieve the fiscal and non-fiscal objectives, including the protection of the cane sugar industry. See id. 11 8.91-8.95.

396. Id. 9 8.102-8.106. 397. 398. 399. See Roessler, supra note 92, at 24-25 (stating that the "no less favorable" of

Id. l 8.123. Id. q 8.105.

Article 111:4 is to be applied to Article 111:2 in determining discriminatory treatment).

400. See Mexico-Tax Measures, supra note 17, 11 8.42-8.45 (recognizing that the Mexican tax measure was an "indirect" tax on HFCS and that it can still be in violation of Article III due to high burdens it places on HFCS, a component of the finished product that is subject to the tax).

401. See id.

402. Id. ' 8.85. 403. See generally id.; see also GAMI Investments, Final Award, supra note 325, ' 45. 404. See Mexico-Tax Measures, supra note 17, c 4.71-4.97. 405. See id. 9 7.4. 406. See id. ' 8.232. 407. See Methanex, Final Award, supra note 8, Part IV, ch. B, c 15. 408. Id.

409. See S.D. Myers, Partial Award, supra note 274, 1 244. 410. See id. 411. See generally EC-Asbestos, supra note 27. 412. See Sykes, supra note 54, at 31, 42. 413. Id. at 4-5 (stating that trade agreements do not generally incorporate a balancing test). Professor Sykes explains that the lack of a "cost-benefit" analysis in trade agreements is primarily because it does not properly deal with political special interest groups that may reap most of the benefits of certain allowances under a trade agreement. He states:

[I]f rigorous cost-benefit analysis were the touchstone of policy, virtually all tariffs, quotas, and subsidies would fail to pass muster because of the deadweight losses evident .... Those measures, and others like them, survive because political officials care not only about the magnitude of costs and benefits, but also about their incidence. Modest benefits to a well-organized interest group can readily outweigh larger costs to a diffuse and poorly organized interest group in the political calculus.

Id. at 31.

414. Id. at 31-32. 415. See id. at 32-33.

416. See McGinnis & Movsesian, supra note 46, at 572-83. 417. This deferential approach would be more in line with the "antidiscrimination model" of the WTO than the regulatory model. However, it does not eliminate the possibility of a regulatory model approach in other instances. See generally McGinnis & Movsesian, supra note 46.

418. See id. at 566-72. 419. See Mexico-Tax Measures, supra note 17, ' 9. 420. See McGinnis & Movsesian, supra note 46, at 590. 421. See discussion regarding the three requirements proposed by Professors McGinnis and Movsesian of transparency, performance orientation, and consistency. Id. at 573-77.

422. Pope & Talbot, Award on the Merits, supra note 271, 1 79. 423. See McGinnis & Movsesian, supra note 46, at 566. 424. See id. at 567. 425. See generally Steinberg, supra note 45, at 263-69 (arguing that the WTO Appellate Body's decision are influenced by politics-both internally through the selection of Appellate Body members and externally through political bargaining among the most powerful member states, which in turn must deal with domestic special interest groups in deciding trade matters); see also McGinnis & Movsesian, supra note 46, at 581 (stating that their proposed procedural test for covert protectionist measures "would permit nations to make their own trade-offs between labor, environmental, health, and safety goals on the one hand, and economic growth on the other, so long as the nations do not discriminate against imports").

426. See Robert Hudec,Judicialization of GATT Dispute Settlement, in IN WHOSE INTEREST? DUE PROCESS AND TRANSPARENCY IN INTERNATIONAL TRADE 9 (Michael M. Hart & Debra P. Steger eds., 1992) (discussing European Communities-Measures Concerning Meat and Meat Products (Hormones) and considering the preferred option of the WTO Appellate Body deferring to its member states in that case).

427. See Dillon, supra note 232, at 387-88 (describing the remedies against members

found to be in violation of GATT, including retaliation).

428. See Steinberg, supra note 45, at 266 (explaining that some members may attempt to delegitimize WTO decisions through noncompliance which would then lead to "authorized retaliation" among member states). The author goes on to explain that noncompliance, despite the threat of retaliation, may be preferable to compliance for certain member states. See id.

429. Hudec Comment, supra note 235, at 297 (stating that as a part of domestic policymaking, the WTO must meet the "standards of legitimacy that national governments must meet when they exercise power" and these standards may be complex).

430. See id. at 298 (stating that the WTO "is a member-driven organization, and the force of its orders is the product of a process in which governments agree to participate and which they ultimately control").

431. See id.

432. See McGinnis & Movsesian, supra note 46, at 566. 433. See id. (discussing the deference in an antidiscrimination model to regional and

national laws rather than to a multilateral body's regulations).

434. See id. at 573-74 (discussing in the context of the antidiscrimination model the three requirements of transparency, performance orientation, and consistency). See also supra, Part lI.B.

435. See generally David Parker & Colin Kirkpatrick, Researching Economic Regulation in Developing Countries: Developing a Methodology for Critical Analysis, in CENTRE ON REGULATION AND COMPETITION: WORKING PAPER SERIES PAPER #34 (Fiona Wilson ed.,

2002) (discussing the various problems with developing nation's regulatory systems and the wide range of progress in development).

436. Many thanks to Rob Howse for his suggestions regarding technical assistance in

this context.

437. See EC-Asbestos, supra note 27, at 438. Id. 11 27, 30, 32.

30.