Taking Stock: What Do We Know, and Do not Know, about the National Treatment Obligation in the GATT/WTO Legal System?

Even though widely recognized as one of the core disciplines of international economic law, the interpretation of national treatment (“NT”) obligation has been long marked by legal indeterminacy. More recently, a series of landmark cases, including US—Clove Cigarettes, US—Tuna II (Mexico), US—COOL and EC—Seal Products, have fundamentally reshaped our collective understanding of the NT obligation in the GATT/WTO system. The objective of this article is to take stock of what we have already known about the NT obligation in the WTO law, identify the lingering uncertainties and discuss the options for the WTO Appellate Body to bring more clarity to the NT obligation in future dispute settlements.


Introduction
It is conventional wisdom that one fundamental tension in the world trading system is the need to strike a delicate balance between the pursuit of trade liberalization and member states' right to regulate. ulation about to what extent the AB may transpose its new approach to the NT obligation in Article 2.1 of the TBT Agreement to Article III:4 of the GATT.5 The speculation is now over after the AB issued its landmark EC-Seal Products ruling in June 2014.6 The objective of this article is to take stock of what we have already known about the NT obligation in the GATT/WTO system, identify the lingering uncertainties and discuss how the WTO Panels and the AB may bring more clarity to this core legal principle after EC-Seal Products in future dispute settlements. The paper is organized as follows. Section two reviews the WTO AB's evolving interpretation of the NT obligation in Article III:4 of the GATT 1994 before EC-Seal Products. Section three examines critically how the AB dealt with the NT obligation in Article 2.1 of the TBT Agreement in the three recent TBT disputes. Section four introduces the most recent EC-Seal Products case and reflects on how the AB has reshaped our collective understanding of the NT obligation in Article III:4 of the GATT 1994. Also explored in part four are some of the remaining uncertainties to be clarified in future dispute settlements. Section five concludes the article.

2
The NT Obligation in Article III:4 of the GATT 1994 before

EC-Seal Products
With regard to the domestic non-fiscal regulation of goods, the NT obligation is contained in Article III:4 of the GATT 1994. Article III:4 provides: The products of . . . any contracting party imported into . . . any other contracting party shall be accorded treatment no less favorable than that accorded to like products of national origin in respect of all laws, regulations and requirements . . .

Du
The Chinese Journal of Global Governance 1 (2015) 67-95 The essence of Article III:4 is that imported products must be offered "treatment no less favorable" than that accorded to domestic like products. In addition, Article III:4 should be read in light of Article III:1, which serves as a guiding principle of the whole Article III.7 Article III:1 reads: The Members recognize that . . . internal laws, regulations and requirements . . . should not be applied to imported or domestic products so as to afford protection to domestic production.
As will be discussed below, the interpretation of key terms of the NT obligation, such as "like products" and "treatment no less favorable" is not an easy exercise.8 In particular, what has persistently puzzled WTO scholars is how exactly Article III:1 shall inform the interpretation of "like products" and "treatment no less favorable" in Article III:4.9 Should a panel consider the regulatory purpose of a measure when determining whether or not there is a violation of the NT obligation? If yes, where should this consideration take place, in "like products" analysis, in "treatment no less favorable" analysis or in a separate and additional step after the "like products" and "treatment no less favorable treatment" analyses? To what extent could the consideration of Article III:1 render a measure with detrimental impact on imported products nevertheless consistent with Article III:4?

"Like Products"
In EC-Asbestos, the AB found that the determination of "like products" under Article III:4 is fundamentally a determination of the nature and extent of a competitive relationship between and among products.10 In assessing this competitive relationship, the AB usually examines (i) properties, nature and quality of products; (ii) the extent to which products are capable of serving the same or similar uses; (iii) the extent to which consumers perceive and treat the products as alternative means of performing particular functions in order 7  to satisfy a particular want or demand; and (iv) the international classification of the products for tariff purposes.11 All other evidence revealing the competitive relationship between products should also be considered.12 Ever since EC-Asbestos, the AB has taken the stance that a thorough examination of the competitive relationship between the two products in the marketplace, mainly through the traditional four-factor analysis, is sufficient to decide whether they are "like products". Even if regulatory purpose in Article III:1 needs to be considered when interpreting the NT obligation in Article III:4, it should not be considered as part of the "like products" inquiry.
However, this market-based "like products" test has been heavily criticized for technical, structural and normative reasons over the years. In EC-Asbestos, the AB pondered the technical difficulties in interpreting "like products" under Article III:4: First, the dictionary definition of "like" does not indicate which characteristics or qualities are important in assessing the "likeness" of products under Article III:4, since most products will have many qualities and characteristics . . . Second, it provides no guidance in determining the degree or extent to which products must share quality or characteristics . . . as products may share only a few characteristics or qualities, or they may share many . . . Third, it does not indicate from whose perspective "likeness" should be judged. Ultimate consumers may have a view about "likeness" of two products that is very different from that of producers of those products.13 (emphasis added) Structurally, if a domestic regulatory measure is found to discriminate against imports in violation of Article III, the regulating government can seek to justify that discrimination by proving that it is "necessary" or "related to" the achievement of some legitimate regulatory purposes enshrined in Article XX of the GATT 1994. However, this is not a desirable route as Article XX was traditionally interpreted stringently.14 In addition, Article XX provides a closed list of Du The Chinese Journal of Global Governance 1 (2015) 67-95 legitimate objectives for government intervention. It is doubtful whether this closed list written sixty years ago is adequate to justify all legitimate government intervention in the 21st century.
Normatively, the market-based "like products" test raises a deeper and troubling issue of the very symbolism of political identity, the way a society wants to understand its internal hierarchy of values. In this regard, the market-based test establishes a normative hierarchy whereby the default norm is liberalized trade, and for competing values, such as human health and safety and protection of the environment, to prevail, they have to be justified.15 Such concerns gave rise to an alternative approach to "like products" analysis, generally known as the "aims and effects" test, in the GATT case law in the early 1990s. The aims and effects test can be viewed as an effort to relax the stark dichotomy of Article III and Article XX, especially with regard to origin-neutral regulatory measures. According to this test, "like products" are not defined by reference to prevailing perceptions about the pair of products in the market place, but by reference to the regulatory aim pursued by the intervening government. 16 In 'aims and effects' test, the panel inferred from the text of Article III:1 that the purpose of Article III is not to prevent contracting parties from using their fiscal and regulatory powers by differentiating between different product categories for policy purposes unrelated to the protection of domestic protection. The panel considered that this limited purpose of Article III had to be taken into account in interpreting the term "like products". In the panel's view, the determination of the relevant features common to the domestic and imported products had to include an examination of the aims and effects of the particular measure: A measure could be said to have the aim of affording protection if an analysis of the circumstances demonstrated that a change in competitive opportunities in favor of domestic products was a desired outcome and not merely an incidental consequence of the pursuit of a legitimate policy goal. A measure could be said to have the effect of affording protection to domestic production if it accorded greater competitive opportunities to domestic products than to imported products.17 Although the aims and effects test has much to be commended, certain problems mitigated against its application. The panel report of Japan-Alcoholic Beverages II explained these problems in the context of Article III:2 and the same reasons also apply to Article III:4.18 To begin with, similar to Article III:2 the first sentence, Article III:4 contains no textual reference to Article III:1. Second, the adoption of the aims and effects approach has important repercussions for the burden of proof imposed on the complainant in the dispute settlement processes. Under such an approach, the complainant would bear the burden of showing not only discriminatory effects, but also that the aim of the measure is to afford protection to domestic production. However, the aim of a measure may be sometimes indiscernible. In the case of a multiplicity of aims, it would be a difficult exercise to determine which aim or aims should be determinative for applying the test. Finally, the aims and effects test may render Article XX virtually redundant. If panels were required to consider the regulatory purpose of a measure when deciding violation under Article III, all regulatory justifications provided in Article XX would have already been considered in the first stage determination of violation, leaving no reason to conduct the same inquiry again under Article XX.19 By contrast, the traditional objective "like products" test does not involve any subjective value judgments. Arguably, this protects the panel and the AB and may increase the perception of legitimacy of the AB rulings.20 Because of these defects and legitimacy concerns, the AB rejected the aims and effects test in Japan-Alcoholic Beverages II shortly after the establishment of the WTO in 1995.

"Treatment No Less Favorable"
The panel in US-Section 337 Tariff Act (1989) stated, "The words 'treatment no less favorable' calls for effective equality of opportunities for imported products in respect of the application of laws, regulations and requirements 17

Du
The Chinese Journal of Global Governance 1 (2015) 67-95 compared to the domestic products".21 In other words, the "treatment no less favorable" standard prohibits WTO Members from modifying the conditions of competition in the market place to the detriment of imported products vis-à-vis domestic like products. 22 Previously the AB's interpretation of "like products" received by far the most attention. More recently, the focus has shifted to "treatment no less favorable".23 In Korea-Beef (2001), the AB made it clear that a formal difference in treatment between imported and like domestic products, even if based exclusively on the origin of the products, is neither necessary, nor sufficient, to show a violation of Article III:4. Rather, what is relevant is whether such regulatory differences modified the conditions of competition to the detriment of imported products.24 In addition, the less favorable treatment, i.e., negative economic effects, must affect the group of imported products, as compared to the group of domestic products. In other words, the NT obligation will be breached only if imported products, on the whole, are treated less favorably than domestic products. It is not enough that one imported product is or may be treated less favorably than one domestic like product.25 Similar to the interpretation of "like products", there has been a long-term debate on how should Article III:1 inform the interpretation of "treatment no less favorable". In EC-Bananas III, the AB concluded that, as there is no specific reference to Article III:1 in Article III:4, there is no requirement to determine separately whether a measure is applied "so as to afford protection to domestic production".26 In EC-Asbestos, the AB stated that Article III:1 Taking Stock The Chinese Journal of Global Governance 1 (2015) 67-95 "informs" Article III:4 and should act as a guide to understanding and interpreting Article III:4.27 The AB states: The term "less favorable treatment" expresses the general principle in Article III:1: if there is "less favorable treatment" of the group of like imported products, there is, conversely, "protection" of the group of "like" domestic products. However, a Member may draw distinctions between products which have been found to be "like", without, for this reason alone, according to the group of "like" imported products less favorable treatment than that accorded to the group of "like" domestic products.28 Here the AB explicitly recognized that Article III:1 "informs" Article III:4. Recall that the "like products" inquiry does not consider the regulatory purpose of the measure, the AB seemed to indicate that the regulatory purpose might be considered when determining whether foreign products are accorded "treatment less favorable" than domestic "like products". It is also clear from the paragraph quoted above that there is no separate and additional inquiry on the regulatory purpose of the measure under Article III:1.
However, the AB did not make it clear how the regulatory purpose might inform "less favorable treatment" in EC-Asbestos.29 Will a finding of disparate impact to the detriment of imported products automatically lead to "less favorable treatment", and thus a violation of the NT obligation? If not, should the consideration of regulatory purpose be part of the "less favorable treatment" analysis? What could the AB possibly have in mind when they use the tantalizing phrase whereby "a Member may draw distinctions between 'like' products" without resulting in less favorable treatment? Is it because the purpose of the distinction is not trade protectionism?
As some prominent WTO commentators speculated, the ambiguous statement that a Member may draw distinctions between "like products" without resulting in less favorable treatment could certainly be a platform from which to embrace an intent analysis, reviving to some extent the aims and effects test.30 27  However, it was later clarified in Thailand-Cigarettes that in EC-Asbestos, the AB did not mean that the panel should try to explore the regulatory purpose of a disputed measure, nor may a non-protectionist explanation render an otherwise discriminatory measure consistent with the NT obligation. The correct understanding of the quoted paragraph above is that it restates what the AB stated in Korea-Beef, i.e., a formal regulatory distinction itself is not conclusive evidence of less favorable treatment. Rather, the investigative focus should be whether conditions for competition between and among like products were modified to the detriment of imported products.31 On the critical question of whether a finding of detrimental impact on imported products vis-à-vis domestic like products, without more, is sufficient to conclude a violation of the NT obligation under Article III:4, the AB report of Dominican Republic-Cigarettes caused even more confusion. In this case, Honduras argued that Dominican Republic's requirement that importers and domestic producers post a bond of RD$5 million to ensure payment of taxes has a detrimental impact on the class of imported cigarettes compared with the class of domestic cigarettes. In the view of the AB, this detrimental effect on imported cigarettes was not enough to find a violation of the NT obligation: [T]he existence of a detrimental effect on a given imported product resulting from a measure does not necessarily imply that this measure accords less favorable treatment to imports if the detrimental effect is explained by factors or circumstances unrelated to the foreign origin of the product . . .32 (emphasis added) After Dominican Republic-Cigarettes, almost all panels and WTO experts understood the AB as saying that it is not sufficient to find inconsistency with Article III:4 solely on the basis that the measure at issue modifies the conditions of competition to the disadvantage of imported products. The complainant must also show that those adverse effects are related to the foreign origin of the products. In other words, a non-protectionist explanation, that is, an explanation unrelated to the foreign origin of the product, may be evidence that there is no less favorable treatment to imports.33 In EC-Biotech Products, Argentina, the US, and Canada complained that the EC had accorded less favorable treatment to biotech products than to non-biotech products, despite the fact that they are "like products". The Panel stated: Argentina is not alleging that the treatment of products has differed depending on their origin. In these circumstances, it is not self-evident that the alleged less favorable treatment of imported biotech products is explained by the foreign origin of these products rather than, for instance, a perceived difference between biotech products and non-biotech products in terms of their safety, etc. In our view, Argentina has not adduced argument and evidence sufficient to raise a presumption that the alleged less favorable treatment is explained by the foreign origin of the relevant biotech products.34 (emphasis added) In US-Tuna II, the panel found that Mexico had failed to demonstrate that the US "dolphin-safe" labelling provisions afford less favorable treatment to Mexican tuna products. The panel reasoned: The impact of the US dolphin-safe provisions on different operators on the market and on tuna products of various origins depends on a number of factors that are not related to the nationality of the product, but to the fishing and purchasing practices, geographical location, relative integration of different segments of production, and economic and marketing choices.35 Clearly, in these cases, the panels interpreted the AB report of Dominican Republic-Cigarettes as requiring a separate and additional step of inquiring whether the detrimental effect is explained by factors or circumstances unrelated to the foreign origin of the product, before any conclusion on "treatment no less favorable" may be drawn.

Du
The Chinese Journal of Global Governance 1 (2015) 67-95 In US-Clove Cigarettes, the AB refuted this popular understanding. In footnote 372 of the report, the AB clarified that there is no additional inquiry of whether the detrimental impact was related to the foreign origin of the products or whether there are any non-protectionist policy justifications for such a disparate impact under Article III:4. The intriguing statement in the AB report of Dominican Republic-Cigarettes simply stresses the fact that the sales of domestic cigarettes are greater than those of imported cigarettes on the Dominican Republic market. Consequently, per unit cost of the bond requirement for imported cigarettes is higher than domestic products.36 In other words, the higher per unit costs of the bond requirement for imported cigarettes was not attributable to the specific measure at issue but was a function of sales volumes.37 It is now clear that what the AB meant to establish in Dominican Republic-Cigarettes was a test of "causation" under which a panel needs to consider whether any alleged disparate impact is caused by a challenged measure.38 If some other factors, rather than the disputed measure, are accountable to the disparate impact, then the measure in dispute has not modified conditions of competition, hence no "treatment less favorable".
In Thailand-Cigarettes, the AB has provided more guidance on how to approach "treatment no less favorable" in Article III.4. First, the AB stated that the analysis of the implications of the contested measure for the equality of competitive conditions must begin with careful scrutiny of the measure, including consideration of the design, structure, and expected operation of the measure.39 Such an analysis may involve, but need not be based on, the actual effects of the contested measure in the market place, nor should the Panel anchor the analysis of less favorable treatment in an assessment of the degree of likelihood that an adverse impact on competitive conditions will materialize.40 Second, if the regulation at issue indicates an origin-based, de jure discrimination, i.e., the sole difference in regulatory treatment consists of requirements applied only to imported products; there is a significant indication that imported products are accorded less favorable treatment.41 In Thailand-Cigarettes, Thailand exempted three sets of VATrelated administrative requirements for resellers of domestic cigarettes, but imposed these administrative requirements on resellers of imported cigarettes. 36  Without much difficulty, the AB ruled that less favorable treatment could be established.42 Finally, in any event, there must be in every case a genuine relationship between the measure at issue and its adverse impact on competitive opportunities for imported versus domestic like products.43

3
The NT Obligation in Article 2.1 of the TBT Agreement Article 2.1 of the TBT Agreement provides: Members shall ensure that in respect of technical regulations, products imported from the territory of any Member shall be accorded treatment no less favourable than that accorded to like products of national origin and to like products originating in any other country.
According to the AB, for a violation of the NT obligation in Article 2.1 to be established, three elements must be satisfied: (i) technical regulation, (ii) like products, and (iii) no less favorable treatment.44 For the purpose of this article, only the last two elements will be discussed.

"Like products"
In US-Clove Cigarettes, the panel questioned the conventional wisdom of interpreting like products primarily from a competition-oriented perspective. On top of the traditional four factors, the panel attempted to add an additional consideration, the regulatory purpose of the legitimate objective pursued. 45 The panel held that the determination of like products in Article 2.1 should be influenced by the fact that the measure in question is a technical regulation having the immediate purpose of regulating cigarettes for public health reasons.46 In essence, the panel attempted to revive at least part of the aims and effects test in determining like products under the TBT Agreement.
The AB unequivocally rejected such an attempt for two reasons. First, measures often pursue a multiplicity of objectives. It would not always be possible for a panel to identify all the objectives and be in a position to determine which objectives are relevant to the determination of like products. If a panel were to 42   focus only on one of the multiple objectives, it may reach a somewhat arbitrary result in the determination of what are the "like products" at issue.47 Second, the concept of like products serve to define the scope of products that should be compared to establish whether less favorable treatment is being accorded to imported products. If products that are in a sufficiently strong competitive relationship to be considered like are excluded from the group of like products on the basis of a measure's regulatory purpose, this would inevitably distort the less favorable treatment comparison.48 After US-Clove Cigarettes, it is settled that the determination of "like products" under Article 2.1 of the TBT Agreement adopts the same competitionoriented approach as Article III:4 of the GATT 1994. It is a determination about the nature and extent of a competitive relationship between and among products at issue.49 In such a determination, a panel should discount any distortive effects that the measure at issue may itself have on the competitive relationship, and reserve the consideration of such effects for the analysis of "less favorable treatment".50 Similarly, a panel should not consider the regulatory purpose of the measures in question when determining "like products". 51 In this connection, a distinction should be drawn between the regulatory purpose of the measures and the regulatory concerns underlying the measure. While the regulatory purpose should not be considered in determining "like products", the regulatory concerns may play a role in the determination of likeness to the extent that they are relevant to the examination of certain likeness criteria and are reflected in the competitive relationship between and among the products concerned.52 In EC-Asbestos, for example, the health risks associated with chrysotile asbestos are considered relevant in determining whether chrysotile asbestos and PCG fibres are like products. This is because health risks affect physical properties as well as consumer tastes and preferences.53

"Treatment no less favorable"
The interpretation of "treatment no less favorable" is a tricky one in the TBT Agreement. The TBT Agreement does not contain a "general exceptions" clause similar to Article XX of the GATT 1994. If "treatment no less favorable" 47  were interpreted in the same way as Article III:4, any technical regulation having a detrimental impact on the group of imported products vis-à-vis the group of domestic like products would violate Article 2.1 of the TBT Agreement. This will be the case even if the technical regulation in question has a legitimate regulatory objective and is executed even-handedly. Clearly, this is not an acceptable outcome, as it would prioritize trade liberalization with no regard to other non-economic social values.
In US-Clove Cigarettes, the AB first argues that the purpose of Article 2.1 is not to prohibit a priori any obstacles to international trade. Instead, a WTO Member has the right to pursue legitimate regulatory objectives. The AB points to the sixth recital of the preamble of the TBT Agreement to support this interpretation. The sixth recital expressly acknowledges that Members' right to regulate should not be constrained if the measures taken are necessary to fulfill certain legitimate policy objectives, and provided that they are not applied in a manner that would constitute a means of arbitrary or unjustifiable discrimination or a disguised restriction on international trade.54 After US-Clove Cigarettes, it is settled that a finding of "treatment no less favorable" under Article 2.1 of the TBT Agreement requires a two-step analysis. First, a panel must find that the measure at issue modifies the conditions of competition in the relevant market to the detriment of the group of imported products vis-à-vis the group of domestic like products. The existence of such a detrimental effect however is not sufficient to demonstrate less favorable treatment under Article 2.1. In the second step, a panel must further analyze whether the detrimental impact on imports stems exclusively from a legitimate regulatory distinction.55 If a regulatory distinction is not designed and applied in an even-handed manner because, for example, it is designed or applied in a manner that constitutes a means of arbitrary or unjustifiable discrimination, that distinction cannot be considered "legitimate".56 In assessing even-handedness, a panel must scrutinize the design, architecture, revealing structure, operation, and application of the technical regulation at issue.57 In US-Clove Cigarettes, because clove cigarettes and menthol cigarettes are "like products" and that clove cigarettes are banned while menthol cigarettes are permitted in the US, there was little difficulty for the AB to conclude that such differential treatment is to the detriment of the imported products. 54  In the second step, the AB found that the cigarettes that are prohibited by the US measure consist primarily of clove cigarettes imported from Indonesia. By contrast, the "like products" that are actually permitted consisted primarily of US domestically produced menthol cigarettes, accounting for about 26% of the total US cigarette market.58 This strongly suggests that the detrimental impact on competitive opportunities for clove cigarettes imported from Indonesia reflects discrimination. Moreover, the AB found that the reasons presented by the US for the exemption of menthol cigarettes from the ban do not demonstrate that the detrimental impact on imported clove cigarettes stems from a legitimate regulatory distinction. The US argued that the exemption of menthol cigarettes from the ban is necessary because of the potential impact of a menthol ban on US health care system due to withdrawal treatment, and the potential development of a black market. The AB found that US arguments do not really stick because it is not clear if these risks would materialize if menthol cigarettes were to be banned, insofar as regular cigarettes would remain in the market.59 In US-Tuna II, Mexico challenged a US measure that monitored and enforced the "dolphin-safe" label on tuna. The AB found that the "dolphin-safe" label has significant commercial value for tuna products on the US market. The US measure prohibited the use of the label on tuna marketed in the US in any instance where the tuna was caught by a method involving encircling or setting upon dolphins. Because Mexican vessels usually catch tuna by setting on dolphins in the Eastern Tropical Pacific (ETP), most tuna caught by Mexican vessels would not be eligible for the label. By comparison, most tuna from the US and other countries that are sold in the US market were caught by other fishing methods outside the ETP. Therefore, most tuna caught by US vessels are eligible for a "dolphin-safe" label. The lack of access to the "dolphin-safe" label has a detrimental impact on the competitive opportunities of Mexican tuna products in the US market.60 Further, the AB concluded that the detrimental impact of the US labelling requirement did not stem exclusively from a legitimate regulatory distinction. Under the US measure, tuna harvested in the ETP may be labelled "dolphin-safe" if the captain and an observer certify that no dolphins were killed or seriously injured and that there was no setting on dolphins during the same fishing trip.61 However, the US measure does not address mortality (observed or unobserved) arising from fishing methods 58  other than setting on dolphins outside the ETP. As a result, the tuna caught in a non-ETP fishery would be eligible for the US "dolphin-safe" label even if dolphins had in fact been killed or seriously injured during the fishing trip.62 The evidence shows that the US labelling requirements are not calibrated to the risks to dolphins arising from different fishing methods in different areas of the ocean, and less favorable treatment was afforded to imported Mexican tuna. The AB believed that it is feasible for the US to address the deficiency of the labelling requirements.63 In US-Cool, the US government imposed on retailers an obligation to provide US country of origin labelling (COOL) information on a range of meat products. In examining various possible methods of compliance with the COOL measure, the AB found that the design of the COOL measure and its operations within the US market, particularly the recordkeeping and verification requirements, meant that the least costly way is to rely exclusively on domestic livestock.64 In addition, the costs of compliance cannot fully be passed on to consumers. The AB then concluded that the COOL measure creates an incentive for US market participants to process exclusively domestic livestock and reduce the competitive opportunities of imported livestock.65 The COOL measure thus modifies the conditions of competition in the US market to the detriment of imported livestock. The AB then proceeded to consider whether the regulatory distinctions, i.e., different labels to reflect the different production steps of birth, raising and slaughter, embodied in the COOL measure are designed and applied in an even-handed manner. Looking at the design and application of the COOL measure, the AB found that the recordkeeping and verification requirements in the COOL measure impose a disproportionate burden on upstream producers and processors.66 This disproportionality cannot be explained by the need to convey to consumers through labels information on country of origin because the COOL measure as applied is unlikely to be successful in conveying accurate information to consumers for a number of reasons. First, the level of information conveyed to consumers through the mandatory labelling requirements is far less detailed and accurate than the information required to be tracked and transmitted by these producers and processors. As a result, consumers are not able to comprehend the meaning of the labelling programmer accurately. Second, some 62

Du
The Chinese Journal of Global Governance 1 (2015) 67-95 labels lead to confusing or inaccurate origin information conveyed to consumers. Third, upstream producers may be subject to the COOL measure's onerous recordkeeping and verification requirements even when the meat is ultimately exempt from the labelling requirements altogether.67 The AB concluded that the regulatory distinctions imposed by the COOL measure amount to arbitrary and unjustifiable discrimination against imported livestock, such that they cannot be said to be applied in an even-handed manner.
It is immediately clear that the first step in the AB's "treatment no less favorable" analysis mirrors much of Article III:4 case law. Moreover, the second step in the AB's "treatment no less favorable" analysis resembles the chapeau test of Article XX of the GATT 1994.68 This is not a surprise because the sixth recital of the preamble of the TBT Agreement closely mirrors the chapeau. The chapeau of Article XX sets forth a non-discrimination requirement that prohibits a measure from being "applied in a manner which would constitute a means of arbitrary or unjustifiable discrimination between countries where the same conditions prevail". In Brazil-Retreaded Tyres, the AB held that the determination of "arbitrary or unjustifiable discrimination" is essentially an analysis that relates primarily to the cause or the rationale of the discrimination put forward to explain its existence.69 As the AB stated: There is arbitrary or unjustifiable discrimination . . . when the reasons given for this discrimination bear no rational connection to the objective falling within the purview of a paragraph of Article XX, or would go against that objective.70 After Brazil-Retreaded Tyres, it is settled that the determination of arbitrary or unjustifiable discrimination will be examined in the light of Article XX's policy goals.71 In Brazil-Retreaded Tyres, Brazil exempted remolded tyres originating in MERCOSUR countries from the ban after a MERCOSUR tribunal found that Brazil's import ban constituted a restriction on trade prohibited 67  under MERCOSUR. The AB dismissed the argument because it bears no relationship to the objective of human health protection pursued by the import ban and even goes against this objective.72 It is also well settled that discrimination can also be arbitrary or unjustifiable where alternative measures exist that would have avoided or at least diminished the discriminatory treatment (less restrictive alternative) and would enable the regulating Member to achieve its legitimate policy goals with the same degree of efficiency and efficacy.73 In US-Gasoline, the AB found that the US could employ a couple of alternative measures to avoid the discrimination resulting from the application of different baseline establishment methods. 74 The US cited administration difficulties and financial costs against the use of alternative measures. The AB found that they were not insurmountable difficulties. Precisely because of the failure on the part of the US to seek cooperation with the two foreign governments and to take into account the costs for foreign refiners, the AB ruled that US different baseline establishment methods to domestic and foreign refiners constitute a means of arbitrary or unjustifiable discrimination.
In US-Shrimp, the AB found several aspects of the US regulation constitute arbitrary or unjustifiable discrimination. For example, the actual application of the US measure, in effect, requires other WTO Members to adopt essentially the same regulatory programs and enforcement practices without taking into account the different situations of other nations. Shrimp harvested using methods comparable in effectiveness to those employed in the US have been excluded from the US market solely because they have not been certified by the US. In the view of the AB, the measure is difficult to reconcile with the declared policy objective of protecting and conserving sea turtles. 75 The AB also pointed out the successful Inter-American Convention provides convincing demonstration that an alternative course of action was reasonably open to the US for securing the legitimate policy goal of its measure a course of action other than the unilateral and non-consensual procedures of the import prohibition. 76 In the recent three TBT cases the AB followed the same approach as the chapeau of Article XX in determining whether the detrimental impact on imports Du The Chinese Journal of Global Governance 1 (2015) 67-95 stems exclusively from a legitimate regulatory distinction. In its analysis, the AB is essentially scrutinizing the cause or the rationale of the regulatory distinction put forward by the defending party to explain its existence. At the same time, the AB examines whether alternative measures are available that would have avoided or at least diminished the discriminatory treatment (less restrictive alternative) and would enable the regulating Member to achieve its legitimate policy goals with the same degree of efficiency and efficacy. In US-Clove Cigarettes, the US argues that the regulatory distinction between menthol cigarettes and clove cigarettes is necessary considering the impact on the US health care system associated with treating millions of menthol cigarettes smokers affected by withdrawal symptoms and the risk of development of a black market if menthol cigarettes are banned. However, the AB found that it was not clear if these risks would materialize if menthol cigarettes were to be banned, insofar as regular cigarettes would remain in the market.77 In other words, the reasons given for the discrimination bear no rational connection to the regulatory objectives that the US government claimed to pursue. Also, the alternative measure such as a ban on menthol cigarettes would avoid the discriminatory treatment.
In US-Tuna, the US measure failed to address mortality arising from fishing methods other than setting on dolphins outside the ETP. As a result, the tuna caught in a non-ETP fishery would be eligible for the "dolphin-safe" label even if dolphins have in fact been killed or seriously injured.78 The regulatory distinction clearly cannot be reconciled with the regulatory objectives of dolphin conservation and provision of reliable information to consumers. The AB also examined the US assertion that alternative measures such as the appointment of an independent observer to monitor tuna-fishing are infeasible due to administrative and financial burdens.79 The AB dismissed the US assertion and suggested that requesting a captain to monitor tuna-fishing activities and issue certificates could be a less restrictive alternative. Similarly, in US-Cool, the origin labelling program in operation fails to achieve the regulatory objectives they are designed to pursue. The lack of rational relationship between the origin labelling program and the regulatory objective constitutes an arbitrary and unjustifiable discrimination.80 Given the close relationship between the TBT Agreement and the GATT 1994, including the similarities in their texts, the panels and the AB have 77  repeatedly found it useful, in examining the measure's consistency with the TBT Agreement, to refer to the AB's guidance in previous disputes concerning the obligations under the GATT 1994.81 In the three recent TBT cases, the AB has developed a clear-cut and consistent interpretation of the NT obligation embodied in Article 2.1. In particular, the interpretation of "like products" has adopted the same competition-oriented approach as "like products" in Article III:4 of the GATT 1994. It is a determination about the nature and extent of a competitive relationship between and among products at issue.82 By contrast, compared to Article III:4 of the GATT 1994, the AB has adopted a different approach in interpreting "less favorable treatment" in Article 2.1 of the TBT Agreement. Under Article III:4, the "less favorable treatment" represents a modification of the conditions of competition in the market place to the detriment of the group of imported products vis-à-vis the group of domestic like products. It is controversial how and to what extent Article III:1 informs the interpretation of less favorable treatment in Article III:4. In Article 2.1 of the TBT Agreement, similar to Article III:4, a panel must first find that the measure at issue modifies the conditions of competition in the relevant market to the detriment of the group of imported products vis-à-vis the group of domestic like products. Different from Article III:4, the AB made it clear that the existence of such a detrimental effect is not sufficient to demonstrate less favorable treatment. A panel must further analyze whether the detrimental impact on imports stems exclusively from a legitimate regulatory distinction.83 In examining whether the regulatory distinction is "legitimate", the AB has adopted an interpretative approach similar to the determination of arbitrary or unjustifiable discrimination in Article XX of the GATT 1994. It seems safe to conclude that the NT obligation in Article 2.1 of the TBT Agreement is largely an integration of Article III:4 and the chapeau of Article XX of the GATT 1994.
Arguably, Article 2.1 of the TBT Agreement helps to reduce the pro-trade bias embedded in the GATT 1994.84 Under the GATT 1994, Article XX is structured as "a general exception" even if they are designed to protect some legitimate public policies. If a non-fiscal internal measure is challenged, it will be scrutinized under Article III.4 first. Only after the measure is found in violation of Article III:4 will legitimate non-economic regulatory objectives be considered. This structure reflects a "hierarchy" between substantive trade obligations and domestic regulation priorities as legitimate regulatory objectives are subordinated as exceptions.85 This "obligation-exception" dichotomy tends in the first place to take a very suspicious attitude towards domestic regulatory measures. Even worse, the exceptions under Article XX are interpreted stringently. Consequently, this approach is criticized as likely to lead to a WTO Member losing or sacrificing its own regulatory autonomy in various areas of social policies solely for efficient resource allocation at the global level.86 Under the TBT Agreement, there is no such dichotomy. Legitimate regulatory objectives are not given an inferior status as "exceptions" to a general free trade obligation. Instead, Article 2.1 overcomes the pro-trade bias in the GATT 1994 by elevating such exceptional social values to the status of inherent rights of WTO Members. This unitary structure means that trade and non-trade values will be considered at the same time. If the regulatory distinction is legitimate, i.e., there is a rational relationship between the regulatory distinction and the regulatory objectives, there is no less favorable treatment and no violation of the NT obligation.
Of course, the significance of this structure difference should not be overstated. Besides the symbolic abandon of the "obligation-exception" dichotomy, the principle implication of this structure difference for the WTO jurisprudence is the allocation of burden of proof in practice. Under the GATT 1994, Article XX is an affirmative defence, with both the burdens of persuasion and proof on the defendant. In the context of Article 2.1 of the TBT Agreement, the complainant must prove its claim by showing that the treatment accorded to imported products is less favorable than that accorded to like domestic products.87

The EC-Seal Products Case and the Future of NT Obligation in GATT 1994
As can be seen in Part II of this article, prior to EC-Seal Products, the AB was rather vague on the role of regulatory purpose in GATT NT obligation suggests that the specific intention of incorporating Article III:1 was to mandate purpose inquiries in dealing with origin-neutral measure under Articles III:2 and III:4.94 The AB's approach denies any possibility for the panel to undertake an analysis of the regulatory purpose when determining "treatment no less favorable". Second, the AB's interpretation of less favorable treatment may create inconsistent rulings between the GATT 1994 and the TBT Agreement in the future. Article XX provides a closed list of legitimate objectives for government intervention.95 By contrast, under Article 2.1 of the TBT Agreement, the list of possible legitimate objectives that may factor into an analysis is open. Thus, a technical regulation that has a detrimental impact on imports would be permitted if such detrimental stems from a legitimate regulatory distinction, while, under Article III:4 of the GATT 1994, the same technical regulation would be prohibited if the objective that it pursues does not fall within the subparagraphs of Article XX.96 Therefore, if WTO panels were to confine purpose inquiries to Article XX, then the policy space that members want to retain with regard to the use of domestic measures in GATT 1994 may be severely restricted and the legitimacy of the multilateral trading system may be undermined. 97 Clearly, the AB did not share such concerns.98 The AB emphasized that the balance between trade liberalization and Members' right to regulate in the TBT Agreement is not, in principle, different from the balance set out in the GATT 1994.99 If so, then the fact that this balance is expressed in different forms in different Agreement should not result in inconsistent rulings. This may imply that the AB may interpretation the legitimate regulatory objectives in the subparagraphs of Article XX more liberally in future disputes. The AB's expansive interpretation of "public morals" in Article XX(a) is an example in point.100 The full implication of the AB's ruling remains to be seen. The AB seems to have imposed a "strict liability" on importing countries. As Howse pointed Du The Chinese Journal of Global Governance 1 (2015) 67-95 out, very few legislative or regulatory distinctions between products would not fail this test. National safety, environmental and health rules, for example, are quite likely to have a different impact on goods manufactured in different places. The logical implication is that a large universe of laws and regulations is now prima facie illegal under WTO law. The outcome seems extreme and hard to reconcile with the intent and text of the GATT. 101 To be fair, there are still a few possible "buffers" in Article III:4 after the AB decision in EC-Seal Products. Interpreted properly, these buffers may suggest some regulatory space for innocent regulations to pass muster under Article III:4. The first one is the AB's interpretation of the "genuine relationship" requirement between the measure at issue and its adverse impact on competitive opportunities for imported versus like domestic products.
To be clear, after EC-Seal Products, there is still room to argue that even if the measure at issue has a disparate impact on imported products compared to like domestic products, there is no "less favorable treatment". The key argument will be that a "genuine relationship" between the measure and the disparate impact on competitive opportunities does not exist. In Thailand-Cigarettes, the AB requires that in every case such a genuine relationship must exist.102 A genuine relationship means that it is the governmental measure at issue that affects the conditions under which like products, domestic and imported, compete in the market within a Member's territory.103 In Dominican Republic-Cigarettes, a fixed expense, such as the annual fee for the bond, leads necessarily to different per-unit costs among supplier firms, to the extent that these firms have different volumes of production or volumes of sales.104 Because imported cigarettes have a smaller market share, per unit costs of the bond requirement for imported cigarettes are higher than domestic cigarettes. The AB held that there was no "less favorable treatment" essentially because the disparate impact on imported products was not attributable to the measures at issue, but because of greater market shares of domestic cigarettes than those of imported cigarettes on the Dominican Republic market. In EC-Seal Products, the AB also identified a second possible scenario where the disparate market impact was caused not by the governmental measure but entirely by private choices.105 The "genuine relationship" requirement offers Du The Chinese Journal of Global Governance 1 (2015) 67-95 sentence only in cases where competition was involved between . . . the taxed product and . . . a directly competitive or substitutable product which was not similarly taxed. Therefore, to find violation of the second sentence of Article III:2, it must be established that the dissimilar taxation between the directly competitive or substitutable imported and like domestic products is applied "so as to afford protection".108 On how to establish this element, the AB noted: We believe [it] requires a comprehensive and objective analysis of the structure and application of the measure in question on domestic as compared to imported products . . . Although it is true that the aim of a measure may not be easier ascertained, nevertheless its protective application can most often be ascertained from the design, the architecture, and the revealing structure of a measure.109 In Thailand-Cigarettes, the AB similarly mandated the panel to examine "the design, structure and expected operations" of the measure when assessing whether the regulation at issue has modified the conditions of competition between imported and domestic like products.110 By examining the same elements, the AB will be able to ascertain, even if implicitly, whether the measure in dispute is applied "so as to afford protection" for domestic production.
The AB's blunt rejection of the need to consider the possible policy rationale in EC-Seal Products raises the question of the purpose of examining "the design, structure and expected operations" of the measure under Article III:4 in future disputes. One plausible interpretation may be that in Article III:4, the purpose of examining "the design, structure and expected operations" of the measure is to ascertain the disparate impact on imported products, whilst in the second sentence of Article III:2, the purpose is to ascertain the protectionist intent. But this is clearly an artificial distinction. In practice, a panel member will instinctively want to know if the measure has a bona fide regulatory purpose and to what extent its market effects are protective when they are called to decide whether the measure in question is in violation of Article III.111 Even though the AB has explicitly denied the need to consider Article III:1 it was explicitly denied by the AB in EC-Seal Products. Rather