2.1 The general analytical approach

The basic rule of Article III:4 is that internal regulations must treat imports no less favorably than like domestic products. Previously the Appellate Body's interpretation of ‘like products' attracted by far the most attention in the NT obligation. More recently, the focus has shifted to ‘treatment no less favorable’.Footnote ¹² The GATT panel in US–Section 337 Tariff Act stated that the term ‘treatment no less favorable’ does not require the identical treatment of imported and like domestic products, but the effective equality of competitive conditions between these like products in respect of the application of laws, regulations, and requirements. The Panel stated:

In Korea–Various Measures on Beef, the panel had wrongfully assumed that any regulatory differences based exclusively on the nationality or the origin of the products is inconsistent with Article III:4.Footnote ¹⁴ The Appellate Body rejected the panel's interpretation and made it clear that a formal difference in treatment between imported products 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.Footnote ¹⁵ 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 like domestic products.Footnote ¹⁶

The examination of whether imported products are afforded less favorable treatment cannot rest on a simple assertion, there must be further identification or elaboration of the implications of the measure for the conditions of competition in order properly to support a finding of less favorable treatment under Article III:4.Footnote ¹⁷ This is especially the case for origin-neutral measures.Footnote ¹⁸ On how to evaluate the implications of the contested measures for the equality of competitive conditions between imported and like domestic products, the Appellate Body has provided detailed guidance in Thailand–Cigarettes.

First, such an analysis must begin with careful scrutiny of the measure, including consideration of the design, structure, and expected operation of the measure.Footnote ¹⁹ 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.Footnote ²⁰ Second, if the regulation at issue indicates an origin-based, de jure discrimination, that is, the sole difference in regulatory treatment consists of requirements applied only to imported products, there is significant indication that imported products are accorded less favorable treatment.Footnote ²¹ In Thailand–Cigarettes, Thailand exempted three sets of VAT-related administrative requirements for resellers of domestic cigarettes, but imposed these administrative requirements on resellers of imported cigarettes. Without much difficulty, the Appellate Body ruled that less favorable treatment could be established in this case.Footnote ²² Third, 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 like domestic products.Footnote ²³ To identify the existence of a genuine relationship, the relevant question is whether it is the governmental measure at issue that affects the conditions under which like goods, domestic and imported, compete in the market within a Member's territory.Footnote ²⁴ In other words, the detrimental impact on competitive opportunities for like imported products must be attributable to the contested measure at issue.

Under Article III:4, less favorable treatment must affect the group of imported products, as compared to the group of domestic products. The NT obligation is breached only if imported products from the complaining party, on the whole, are treated less favorably than domestic like products. It is not enough that some like imported products from the complaining party receive worse treatment than some like domestic goods. This is because it is always possible to find a violation of Article III:4 as long as the disfavored type of product is imported and the favored type exists domestically.Footnote ²⁵ The Appellate Body report on US–Clove Cigarettes has provided the best illustration of the correct approach to identify the scope of ‘like products’ for ‘treatment no less favorable’ comparison in the NT provision. In this case, the Panel determined that the ‘treatment no less favorable’ requirement called for a comparison between treatment accorded to imported clove cigarettes and that accorded to domestically produced cigarettes that it had earlier found to be like products, menthol cigarettes.Footnote ²⁶ The Appellate Body reversed the panel and held that a panel is required to compare, on the one hand, the treatment accorded to all like products imported from the complaining Member with, on the other hand, that accorded to all like domestic products. In determining what the scope of like imported and domestic product is, a panel is not limited to those products specifically identified by the complaining Member.Footnote ²⁷ In addition, the NT obligation does not require Members to accord no less favorable treatment to each and every imported product as compared to each and every domestic like product. To put it another way, the ‘treatment no less favorable’ standard does not prohibit regulatory distinctions between products found to be like, provided that the group of like products imported from the complaining Member is treated no less favorably than the group of domestic like products.Footnote ²⁸

Based on this analytical framework, the Appellate Body in US–Clove Cigarettes held that the group of imported like products from Indonesia into the US included not only clove cigarettes but also non-clove cigarettes. Similarly, the group of like domestic products included not only menthol cigarettes but also domestically produced flavored cigarettes in the US.Footnote ²⁹ Nevertheless, only a small percentage of non-clove cigarettes was imported from Indonesia into the US and US domestic flavored cigarettes other than menthol cigarettes at best represented a relatively small market share prior to the ban. Consequently, the inclusion of non-clove cigarettes imported from Indonesia and domestically produced flavored cigarettes in the comparison would not have altered the Panel's conclusion that, essentially, the comparison was between imported clove cigarettes and domestic menthol cigarettes.Footnote ³⁰

2.2 The role of regulatory purpose in ‘treatment no less favorable’ analysis

Similar to the interpretation of ‘like products’, there has been a long-standing debate on whether the regulatory purpose in Article III:1 should be considered in the interpretation of ‘treatment no less favorable’ in Article III:4.Footnote ³¹ In EC‒Bananas III, the Panel considered it appropriate to discern the protective application of a measure from its design, architecture, and the revealing structure under Article III:4.Footnote ³² On appeal, the Appellate Body reversed the Panel and found that since Article III:4 does not specifically refer to Article III:1, a determination of whether there has been a violation of Article III:4 does not require a separate consideration of whether a measure ‘affords protection to domestic production’.Footnote ³³ In other words, a determination that the imported and domestic products in question are ‘like products’ and that the regulatory measure in dispute provides less favorable treatment to imported products is sufficient to establish a violation of Article III:4.Footnote ³⁴

In EC‒Asbestos, the Appellate Body implicitly retreated from EC–Bananas III and reiterated its earlier holding in Japan–Alcoholic Beverages II that Article III:1 ‘informs’ Article III and should act as a guide to understanding and interpreting the specific obligations contained in Article III. Footnote ³⁵ Setting aside the conventional wisdom of exploring the role of regulatory purpose in ‘like products’ analysis, the Appellate Body appeared to indicate that the regulatory purpose might be considered in connection with whether foreign products are accorded less favorable treatment compared with like domestic products. The Appellate Body stated:

Here the Appellate Body equated less favorable treatment in Article III:4 to protectionism in Article III:1. It is also clear from this paragraph that there is no separate and additional inquiry on the regulatory purpose of the measure under Article III:1. However, the Appellate Body did not address clearly the crucial question of how Article III:1 might inform the ‘treatment no less favorable’ analysis in Article III:4. As a result, after EC–Asbestos, there was a long-standing controversy of what message the Appellate Body meant to convey through the quoted paragraph above.Footnote ³⁷ For example, will a finding of disparate impact to the detriment of imported products automatically lead to a finding of less favorable treatment? Should consideration of regulatory purpose be part of the less favorable treatment analysis? What could the Appellate Body possibly have in mind when they use the tantalizing phrase whereby a Member may draw distinctions between like products without causing less favorable treatment? Is it because the purpose of the distinction is not trade protectionism?Footnote ³⁸

It was later clarified in Thailand–Cigarettes that the Appellate Body in EC‒Asbestos did not mean to say that the panel should try to explore the regulatory purpose of a disputed measure, nor that a non-protectionist explanation could render an otherwise discriminatory measure consistent with Article III:4. The correct understanding of the quoted paragraph in EC‒Asbestos above is that it confirmed what the Appellate Body had stated in Korea–Various Measures on Beef, that is, a formal regulatory distinction itself is not conclusive evidence of less favorable treatment. The investigative focus should be on whether competition conditions between like products were distorted to the detriment of imported products.Footnote ³⁹ The Appellate Body in EC–Seal Products reiterated this interpretation.Footnote ⁴⁰

On the critical question of whether a finding of detrimental impact on imported products vis-à-vis like domestic products, without more, is sufficient to conclude a violation of the NT obligation under Article III:4, the Appellate Body report on Dominican Republic‒Cigarettes caused even more confusion. In this case, Honduras argued that the Dominican Republic's requirement that importers and domestic producers post a bond of RD$5 million to ensure payment of taxes had a detrimental impact on the class of imported cigarettes compared with the class of domestic cigarettes. This was because importers sold fewer cigarettes in the Dominican Republic than domestic cigarette producers did. In the view of the Appellate Body, this detrimental effect on imported cigarettes was not enough to find a violation of national treatment:

After Dominican Republic‒Cigarettes, almost all panels and WTO experts understood the Appellate Body as saying that a modification of the conditions of competition to the disadvantage of imported products, without more, is not sufficient to find a violation of Article III:4. 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 imported products. For example, 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:

Similarly, in US–Tuna II, the Panel found that Mexico failed to demonstrate that the US ‘dolphin-safe’ labelling provisions had afforded less favorable treatment to Mexican tuna products. The Panel reasoned:

Clearly, in these cases, the panels interpreted the WTO Appellate Body report on Dominican Republic‒Cigarettes as requiring a separate and additional step of inquiry on 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. This is also the most sensible and straightforward reading of the Appellate Body's message.

In US–Clove Cigarettes, however, the Appellate Body has started to move away from its own interpretation of less favorable treatment in Dominican Republic‒Cigarettes. In footnote 372 of the Appellate Body report, the Appellate Body clarified that there is no additional inquiry of whether the detrimental impact is 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 Appellate Body report on Dominican Republic–Cigarettes simply stresses the fact that the sales of domestic cigarettes were greater than those of imported cigarettes on the Dominican Republic market. Consequently, per unit cost of the bond requirement for imported cigarettes was higher than domestic products.Footnote ⁴⁴ 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.Footnote ⁴⁵ It is now settled that what the Appellate Body 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 the challenged measure.Footnote ⁴⁶ 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 less favorable treatment.

After refuting the popular misinterpretation of its report on Dominican Republic‒Cigarettes, the Appellate Body in footnote 372 of its US–Clove Cigarettes report also hinted that it would not examine the rationale for the detrimental impact on imported products under Article III:4 in future disputes:

Undoubtedly, the Appellate Body's interpretation in footnote 372 of the US–Clove Cigarettes Appellate Body report represents a sharp deviation from its loud and clear message in Dominican Republic‒Cigarettes that a separate and additional step of inquiry on whether the detrimental effect is explained by factors or circumstances unrelated to the foreign origin of the product should be conducted before any conclusion on ‘treatment no less favorable’ may be drawn. Instead, the Appellate Body inserted an ‘attribution’ test, which later morphed into a ‘genuine relationship’ test between the measure at issue and the disparate impact in Thailand–Cigarettes, into the ‘treatment no less favorable’ analysis.Footnote ⁴⁸ The Appellate Body did not offer any explanation as to why it ditched the origin test outlined in Dominican Republic‒Cigarettes. In essence, what the Appellate Body stated in footnote 372 was that everybody misinterpreted its ruling in Dominican Republic‒Cigarettes. As I will explain in Section 2.3 below, the real reason for the Appellate Body's move is its reluctance to transpose an Article XX-style justification analysis to Article III:4 and its preference for a clear division of labor between Article III:4 and Article XX of the GATT 1994.

It was not until EC–Seal Products that the Appellate Body finally stated unequivocally its position on the role of regulatory purpose in interpreting ‘treatment no less favorable’. In this case, the EU argued that for the purpose of establishing a violation of Article III:4, a finding that a measure has a detrimental impact on competitive opportunities for imported products, compared to like domestic products, is not dispositive. The EU submitted that a panel must conduct an additional inquiry into whether the detrimental impact on competitive opportunities for like imported products stems exclusively from a legitimate regulatory distinction.Footnote ⁴⁹ In essence, the EU requested the Appellate Body to transpose the legal standard for the ‘treatment no less favorable’ under Article 2.1 of the TBT Agreement to Article III:4.Footnote ⁵⁰ Under Article 2.1 of the TBT Agreement, the analysis of ‘treatment no less favorable’ goes beyond consideration of the detrimental effects of the measure on the competitive opportunities for like imported products, and necessarily involves consideration of possible policy rationale for such a detrimental effect on trade.Footnote ⁵¹

The EU's arguments actually represent a very popular view among WTO commentators.Footnote ⁵² Indeed, from a regulatory perspective, the EU's position is rather appealing. Article 2.1 of the TBT and Article III:4 of the GATT overlap in their scope of application in respect of technical regulations. The NT obligation in both provisions is built around the same core terms ‘like products’ and ‘treatment no less favorable’.Footnote ⁵³ The TBT Agreement and the GATT 1994 also share the same objective and purpose: to strike a balance between trade liberalization and regulatory autonomy.Footnote ⁵⁴ One uniform approach to ‘treatment no less favorable’ will not only clarify the scope of the NT obligation in WTO law, but also help to make case law more coherent across WTO Agreements with regard to domestic regulations.

However, the Appellate Body categorically rejected the EU's position. To the disappointment of many WTO commentators, the Appellate Body took a strictly textualist approach to solve the long-standing controversy over the role of regulatory purpose in interpreting the NT obligation. Since Article III:4 does not explicitly refer to Article III:1, the Appellate Body considered that this omission of a textual reference to Article III:1 must be given meaning.Footnote ⁵⁵ Then the Appellate Body held that Article III:4 is, itself, an expression of the principle set forth in Article III:1. Consequently, if there is less favorable treatment of the group of like imported products, there is, conversely, protection of the group of like domestic products. In other words, the Appellate Body in EC–Seal Products supported the position that less favorable treatment is equal to a detrimental impact on the competitive opportunities for imported products. There is no need to consider the regulatory purpose of the measure in the ‘treatment no less favorable’ analysis, and a finding of a detrimental impact, without more, will automatically send the measure to be justified under Article XX.

2.3 An analysis of the Appellate Body's Interpretation of ‘treatment no less favorable’ in EC–Seal Products

As the Appellate Body has illustrated in US–Clove Cigarettes, the detrimental impact on imported products vis-à-vis domestic like products, by itself, is not sufficient to constitute less favorable treatment in Article 2.1 of the TBT Agreement. This is because a legitimate regulatory distinction can rectify the detrimental effects of the measure on trade.Footnote ⁵⁶ This is in sharp contrast to the Appellate Body's interpretation of the identical ‘treatment no less favorable’ phrase in Article III:4 of the GATT 1994.

The Appellate Body's interpretation of ‘treatment no less favorable’ in EC–Seal Products reminds the readers of its earlier position in EC‒Bananas III in 1997. Admittedly, this interpretation has the advantage of a clear division of labor between Article III:4 and Article XX. Although the TBT Agreement and the GATT 1994 share the same objective of striking a delicate balance between the pursuit of trade liberalization and member states' right to regulate, this balance is expressed in different ways. In the GATT 1994, this balance is expressed by the NT obligation in Article III:4 as qualified by the general exception clause in Article XX. The WTO Appellate Body in EC–Seal Products made it clear that only the trade impact of the contested measure will be considered under Article III:4. Any possible policy justifications for the detrimental trade impact, which delineate the scope of a Member's right to regulate, will be considered under Article XX only. After EC–Seal Products, the long-standing controversy on whether Article III:4 itself affords policy space for a WTO Member to consider Article XX-like policy rationale is over.

Indeed, one major difficulty with the EU's argument in EC–Seal Products with regard to Article III:4 is that it is likely to disrupt the delicate balance between Article III:4 and Article XX and risk rendering Article XX inutile. Like it or not, WTO Members have agreed on a rule-exception structure in the GATT 1994. All non-economic policy justifications are supposed to be considered under Article XX. Bearing in mind the contextual discrepancies between Article III:4 of the GATT 1994 and Article 2.1 of the TBT, it is challenging to introduce concepts such as ‘legitimate regulatory distinction’ and ‘even-handedness’ in Article 2.1 of the TBT to Article III:4,Footnote ⁵⁷ and neither is it possible to consider adequately all possible non-protectionist policy justifications under Article III:1. One may dismiss such concerns as misguided because if the jurisprudence under Article 2.1 of the TBT Agreement only applies to origin-neutral measures under Article III:4, then doing so could at most imply a less frequent recourse to Article XX. Article XX will still have a role to play in cases where a violation of Article III:4 is established by origin-based discrimination or in cases coming from Article XI or other Articles.Footnote ⁵⁸ But this argument represents a radical deviation from WTO case law precedents. In view of the Appellate Body's formalist approach to the interpretation of WTO texts, arguably out of legitimacy concerns,Footnote ⁵⁹ it is unlikely for the Appellate Body to adopt such an approach in practice.

Of course, the Appellate Body's interpretation of ‘treatment no less favorable’ in EC–Seal Products is not free from doubts and concerns. It is submitted that there are at least four problems with regard to the Appellate Body's interpretive approach. First, the Appellate Body's interpretation almost deprives Article III:1 of any meaning. It is well settled that the purpose of the NT obligation is to avoid protectionism and that Article III:1 informs Article III:4. The negotiating record of the NT obligation also suggests that the specific intention of incorporating Article III:1 was to mandate purpose inquiries in dealing with origin-neutral measures under Articles III:2 and III:4.Footnote ⁶⁰ In other words, Article III:4 of the GATT 1994 itself affords some policy space for WTO members to regulate for non-protectionist policy goals. If this is true, then origin-neutral internal measures that genuinely serve non-protectionist purposes with only incidental negative trade effects should not be found in violation of the NT obligation in the first place, let alone the necessity to seek justification under Article XX. That being said, the negotiating history of the role of Article III:1 in interpreting Article III:4 is not entirely unambiguous.Footnote ⁶¹ Indeed, if all negotiating parties have shared a common understanding on this issue, we would not see the rise and fall of the ‘aims and effects’ test in the first place.Footnote ⁶² In any case, the Appellate Body seldom resorts to travaux préparatoires because they are only a supplementary means of interpretation.Footnote ⁶³

Second, the Appellate Body's interpretation of ‘treatment no less favorable’ in EC–Seal Products may cause inconsistent rulings between Article III:2, the second sentence, and Article III:4, leading to a curious result that an internal fiscal measure may enjoy wider policy space than internal non-fiscal measures under the NT obligation. The second sentence of Article III:2 specifies an NT obligation for internal taxation on directly competitive or substitutable products (DCS products). The Appellate Body reasoned that since the second sentence specially refers to Article III:1, a separate finding that the ‘so as to afford protection’ requirement in Article III:1 has been violated is necessary to establish a violation under the second sentence of Article III:2.Footnote ⁶⁴ The Appellate Body further illustrated in Chile–Alcoholic Beverages that the regulatory purpose of the measure at issue and, more specifically, whether the structure of the measure relates to the asserted purpose is the key to determining whether the measure has the effect of ‘so as to afford protection’ to domestic products.Footnote ⁶⁵ Other than an explicit reference to Article III:1 in the second sentence of Article III.2, there does not seem to be a valid policy rationale to explain why the Appellate Body should reject any regulatory purpose analysis in Article III:4. That said, it is difficult to fault the Appellate Body for adopting a strict textualist approach to Article III:4. The Appellate Body has adopted the same strict textualist approach in its interpretation of Article III:2, the first sentence since the establishment of the WTO in 1995.Footnote ⁶⁶

Third, the Appellate Body's interpretation of ‘treatment no less favorable’ 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. It is questionable whether this closed list written 60 years ago is adequate to rescue various legitimate government regulations in the twenty-first century.Footnote ⁶⁷ Some suggested that Article XX may not be available in the realm of culture, consumer protection (beyond life and health), and socio-economic policies in general.Footnote ⁶⁸ By contrast, under Article 2.1 of the TBT Agreement, the list of possible legitimate objectives that may factor into the ‘treatment no less favorable’ analysis is open. Thus, a technical regulation that has a detrimental impact on imports would be permitted if such detrimental impact 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.Footnote ⁶⁹ Therefore, if WTO Panels were to confine purpose inquiries to Article XX, the policy space that WTO members would like 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.Footnote ⁷⁰

However, the Appellate Body does not share such concerns. The Appellate Body argued in EC–Seal Products:

As the defending party, the EU was understandably reluctant to point out any example in which a legitimate objective would not fall within the scope of Article XX because such a move would prevent the EU from invoking Article XX in certain disputes in the future.Footnote ⁷² On the other hand, the Appellate Body seemed to imply that WTO Members should not worry about the alleged inconsistency between Article III:4 of the GATT 1994 and Article 2.1 of the TBT Agreement going forward. To begin with, the Appellate Body 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.Footnote ⁷³ If so, then the fact that this balance is expressed in different forms should not result in inconsistent rulings. Furthermore, by challenging the EU to point out a concrete example, the Appellate Body seemed to hint that the scope of legitimate objectives that could be invoked under Article XX of the GATT 1994 is largely the same as the TBT Agreement. If this were true, then the Appellate Body would necessarily interpret the ten subparagraphs of Article XX more flexibly in order to cover all possible legitimate objectives that a WTO Member may pursue in future disputes. The Appellate Body's liberal interpretation of ‘public morals’ in Article XX (a) in EC–Seal Products may have already pointed in this direction.Footnote ⁷⁴

Finally, the Appellate Body's interpretation of ‘treatment no less favorable’ does not help to alleviate the rule-exception dichotomy embedded in the GATT 1994. The normative hierarchy whereby the default norm is liberalized trade, and non-economic competing values, such as human health and safety and protection of the environment, have to be justified is simply unacceptable to many people.Footnote ⁷⁵ However, to be fair, the WTO Appellate Body should not be blamed for the rule-exception dichotomy embedded in the WTO text. It is difficult to imagine an aggressive role for the Appellate Body to respond to such sensitivity. The role of the Appellate Body is to interpret the WTO text as it was written by the negotiators. The institutional constraints of the WTO also militate against the Appellate Body's ability to play such a progressive role.Footnote ⁷⁶

Ultimately, the debate on the merits of the Appellate Body's interpretive approach to ‘treatment no less favorable’ analysis in EC–Seal Products boils down to this question: where should a panel consider the regulatory purpose of the measure at issue, under the less favorable treatment analysis in Article III:4 itself or under Article XX? I submit that, with regard to the outcome of a dispute, there is not really much difference. This point could be illustrated by the Appellate Body's interpretation of Article 2.1 of the TBT Agreement. For Article 2.1, the first step in the Appellate Body's ‘treatment no less favorable’ analysis is the same disparate impact test in Article III:4; the second step of determining whether the detrimental impact stems exclusively from a legitimate regulatory distinction, sort of the ‘regulatory purpose’ test favored by many leading WTO commentators, is very similar to the chapeau test of Article XX of the GATT 1994.Footnote ⁷⁷

As the TBT case law shows, the second step of the analysis in Article 2.1 has been interpreted expansively by the WTO Appellate Body,Footnote ⁷⁸ and sometimes so expansively that, intentionally or unintentionally, the Appellate Body may do under Article 2.1 what it is supposed to do under the necessity test in Article 2.2.Footnote ⁷⁹ This is of course not surprising because no facts are per se excluded from the assessment of whether or not the measure is implemented even-handedly under the chapeau test in Article XX. Such facts may include the objective of the measure, whether it is legitimate, how is the measure applied, the extent to which the measure contributes to the objective at the chosen level, the existence of less trade restrictive alternatives, whether the restrictiveness of the measure is somewhat disproportionate, and even the consistent treatment of similar situations.Footnote ⁸⁰

Then, we may wonder what purpose does it really serve to transpose the essentially Article XX case law to the ‘treatment no less favorable’ analysis in Article III:4? As my analysis has shown, despite some reasonable concerns, the Appellate Body has valid reasons to choose a bright line ‘disparate impact’ test for the ‘treatment no less favorable’ analysis. It has the benefits of a clear division of labour between Article III and Article XX, acute awareness of the textural differences among WTO Agreements, and the avoidance of depriving Article XX of effect utile. Moreover, given that the WTO Appellate Body has established a long and increasingly stabilised jurisprudence under Article III:4 and Article XX, the cost of trying to reinvent the wheel is simply too heavy and the benefits do not obviously outweigh the costs incurred.

3.1 The ‘genuine relationship’ requirement

The first issue is the Appellate Body's interpretation of the ‘genuine relationship’ requirement between the measure at issue and its adverse impact on competitive opportunities for imported products vis-à-vis 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 nevertheless 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 Appellate Body required that in every case such a genuine relationship must exist.Footnote ⁸³ 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.Footnote ⁸⁴ In Dominican Republic–Cigarettes, a fixed expense, such as the annual fee for the bond, led to different per-unit costs among supplier firms, to the extent that these firms had different volumes of production or volumes of sales.Footnote ⁸⁵ Because imported cigarettes had a smaller market share, per unit costs of the bond requirement for imported cigarettes were higher than domestic cigarettes. The Appellate Body 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 in the Dominican Republic market. In EC–Seal Products, the Appellate Body identified a second possible scenario where the disparate market impact was caused not by the governmental measure but entirely by private choices.Footnote ⁸⁶ It is not clear how the Appellate Body could make such a determination. Governments routinely make normative regulations that affect private choices.Footnote ⁸⁷ In reality, a disparate market impact can hardly be caused entirely by private choice and the measure at issue does not play any role in causing such an adverse impact.Footnote ⁸⁸ It is quite possible that the Appellate Body may loosen this seemingly insurmountable test in future cases.

Interesting questions may arise as to how to determine the causal link between the governmental measure at issue and the detrimental impact on imported products. It is well known that the ‘but-for’ test is commonly used to determine a defendant's responsibility in tort and criminal law. It states that causation exists only when the result would not have occurred without the accused party's conduct. The ‘but-for’ test simply asks the question: ‘but for the existence of X, would Y have occurred?’ If the answer is yes, then factor X is an actual cause of result Y.Footnote ⁸⁹ Applying the ‘but-for’ test to Dominican Republic–Cigarette, the uniform 5M bond requirement is surely a causal factor because, without the bond requirement, the disparate impact would not arise in the first instance. This reasoning is of course not what the Appellate Body wanted as the Appellate Body looked at other contributing factors and made the determination that market share, rather than the measures at issue, was accountable for the disparate impact on trade. This means that the identification of a genuine relationship entails examining and dismissing other factors to which the detrimental impact might be attributable: climatic or geographical or demographic factors, changing consumer tastes or ethical values, policies of the governments of the exporting countries that could affect how the market is structured, etc.Footnote ⁹⁰ To establish the attribution, or a genuine relationship to the measure, a Panel must arguably be sure that the detrimental competitive impact in question is not due to the other factors, largely or entirely. In Dominican Republic–Cigarettes, the Appellate Body in essence held that the market share was a closer cause for the disparate impact on imports. Though there is no denying that the measure at issue was also a causal factor for the disparate impact, its link to the detrimental impact on trade was arguably not sufficient or as strong as the market share factor.

Furthermore, as the Appellate Body stated in Korea–Various Measures on Beef, a formally identical measure could lead to a disparate impact in the market place. Precisely because imported cigarettes have a smaller market share, the uniform 5M bond requirement had a detrimental impact on imported cigarettes on a per unit basis in Dominican Republic–Cigarettes. This heavier burden would likely make it more difficult for imported cigarettes to increase their market share. To ensure the equality of competitive conditions, it is reasonable not to levy the bond at a uniform amount of 5 million. Rather, the amount should vary in light of the smaller market shares held by imported cigarettes. Therefore, the denial of the causal relationship between the disputed measure and the detrimental impact by the Appellate Body in Dominican Republic–Cigarette is not entirely convincing.

A more appropriate interpretation of the Appellate Body's report on Dominican Republic–Cigarettes is that the Appellate Body recognized that to ensure absolute equality of treatment in all circumstances is too costly and that it would make the design and implementation of a bona fide regulation unnecessarily complicated. The uniform bond requirement in the case had only an incidental and unintended disparate impact on imported cigarettes. Another causal factor having little or nothing to do with the governmental measure, different market shares, better explains the disparate impact identified by the complaining party. Related, the calculation carried out by the panel showed that the bond requirement represented a very small cost for the importer ‒ equivalent to 0.2 per cent of the value of cigarette imports made by the importer in the year 2003.Footnote ⁹¹ Such a small cost had not affected the exportation of cigarettes from Honduras to the Dominican Republic. Indeed, the exports from Honduras had increased significantly over the years since the measure was introduced.Footnote ⁹² The Appellate Body thus concluded:

This analysis shows that the genuine relationship requirement offers some room for a clearly non-protectionist measure with only an incidental and negligible trade impact on imports to be consistent with the ‘treatment no less favorable' requirement. In future trade disputes, whether there is a genuine relationship between the measure at issue and its adverse impact on competitive opportunities for imported products vis-à-vis like domestic products will be a main battlefield for the defending party.

3.2 The role of ‘design, structure and expected operations of the measure’

The second issue is the role of ‘the design, structure and expected operations of the measure’ in the NT analysis. In EC–Seal Products, the Appellate Body explicitly denied the relevance of Article III:1 in interpreting ‘treatment no less favorable’. Even so, there is lingering doubt on whether the Appellate Body has implicitly considered the regulatory purpose of the contested measure. This point could be observed from the Appellate Body's emphasis on examining ‘the design, structure and expected operations of the measure’ when assessing ‘treatment no less favorable’ in Thailand–Cigarettes.Footnote ⁹⁴ This is precisely the same approach that the Appellate Body has taken when assessing whether the dissimilar taxation of imported and domestic like DCS products is applied ‘so as to afford protection’ in the second sentence of Article III:2.Footnote ⁹⁵

The second sentence of Article III:2 specifies a NT obligation for internal taxation on DCS products . It specifically refers to Article III:1. Article III:2 reads:

Article III:2, second sentence, has an Interpretative Note that reads:

In the Appellate Body's view, Article III:1 is relevant for the whole of Article III:2. However, it informs the two sentences of Article III:2 in different ways. The first sentence of Article III:2 does not refer specifically to Article III:1. The Appellate Body interpreted the omission as meaning that the presence of ‘so as to afford protection’ in Article III:1 need not be established separately.Footnote ⁹⁶ Unlike the first sentence of Article III:2, the language of the second sentence of Article III:2 specifically invokes Article III:1. The Appellate Body reasoned that:

On how to establish this element, the Appellate Body made it clear that this is not an issue of intent behind a measure, but is an issue of how the measure in question is applied. In practice, it is not necessary for a panel to sort through the many reasons legislators and regulators often have for what they do and weigh the relative significance of those reasons to establish legislative or regulatory intent. It is irrelevant that protectionism was not an intended objective.Footnote ⁹⁸ The Appellate Body further held that it is possible to examine objectively the underlying criteria used in a particular tax measure. The Appellate Body noted:

In Chile–Alcoholic Beverages, the Appellate Body found that the structure of the Chilean tax did not relate to the purposes stated by Chile, and concluded that this lack of correlation confirmed its finding that the Chilean measure was applied ‘so as to afford protection’.Footnote ¹⁰⁰ Moreover, in the course of agreeing with Chile that its measures need not be shown to be necessary to Chile's asserted purposes, the Appellate Body stated:

Here the Appellate Body seems to say that the inquiry was undertaken precisely to identify the measure's objective or purpose.Footnote ¹⁰² Revealingly, in Thailand–Cigarettes, the Appellate Body 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.Footnote ¹⁰³ By examining the same elements, the Appellate Body will be able to ascertain, even if implicitly, whether the measure in dispute is applied ‘so as to afford protection’ for domestic production. The Appellate Body'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 disputed 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 protectionism 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.Footnote ¹⁰⁴ Rather than being a separate step in Article III:2 analysis, it may be argued that consideration of Article III:1 is subsumed within the ‘treatment no less favorable’ analysis. As a matter of fact, many leading WTO experts hope that the Appellate Body might step back from its drastic decision to ignore all possible policy justifications under Article III:4 itself in future decisions.Footnote ¹⁰⁵

3.3 Consumer preferences and ‘like products’ determination

The third issue is how flexible the Appellate Body will be when using the different criteria for the determination of ‘like products’. The case law emphasized the importance of a competitive market relationship between products, from a consumer perspective, as a way to determine their likeness.Footnote ¹⁰⁶ A Panel is mandated to examine each of the criteria, and then weigh all of that evidence, along with any other relevant evidence, in making an overall determination on ‘like products’. In this weighing and balancing exercise, some evidence may reflect a competitive relationship between products, while other evidence may point to a different conclusion. The Appellate Body made it clear that not all products which are in some competitive relationship are ‘like products’ under Article III:4.Footnote ¹⁰⁷ What is at stake is a contextual and qualitative judgment about competitive relationships, not merely the economic analysis of cross-elasticity of demand between two groups of products. Such an assessment must be made on a case-by-case basis, informed by the general principle of anti-protectionism which informs all of Article III.Footnote ¹⁰⁸

In particular, it has been frequently suggested that some criterion such as consumer tastes and habits may be used to distinguish between products that would otherwise be seen as like, for example, sustainable versus unsustainable, or carbon-intensive versus low-carbon goods.Footnote ¹⁰⁹ In EC–Asbestos, the Appellate Body remedied the narrow scope given to the concept of likeness in prior case law by allowing non-economic interests and values, such as health, to be considered in the determination of ‘like products’.Footnote ¹¹⁰ In US–Clove Cigarettes, the Appellate Body confirmed that the regulatory concerns may play a role in determining ‘like products’.Footnote ¹¹¹ A more flexible and imaginative use of ‘like products’ analysis is surely a promising path to open up more regulatory space for a regulating WTO Member.

However, there are some stumbling blocks on this path. In US–Clove Cigarettes, the Appellate Body emphasized that the regulatory concerns are relevant to ‘like products’ determination only to the extent that they are reflected in the competitive relationship between and among the products concerned.Footnote ¹¹² In other words, it is one thing to argue that an increasing number of consumers are interested in, and sensitive to, the labor, environmental and other social concerns embodied in a product, it is quite another to demonstrate that such regulatory concerns indeed shape consumer preferences and guide consumer choices in the market place. If there is evidence showing that a significant segment of consumers perceive two groups of products differently, then a strong argument could be made that two products at issue are not ‘like products’. However, in reality, more often consumers are primarily guided by the price and quality of the products in their choice between products.Footnote ¹¹³ If so, it is unlikely to overturn the ‘like products’ determination between two otherwise identical products with only divergent environmental impact. Looking at the issue from this perspective, EC–Asbestos is a unique case because what is at stake is human health, the importance of which is undisputable.Footnote ¹¹⁴ International standards also support the classification of asbestos as a carcinogen, whose use should be eliminated. This made the interpretation of consumer tastes and habits in EC–Asbestos fairly unambiguous, even if there was not sufficient evidence supporting the claim.Footnote ¹¹⁵

In US–Tuna II, the preference by American consumers for dolphin-friendly tuna was so intense that all major American retailers no longer sell tuna caught by setting on dolphins.Footnote ¹¹⁶ The finding of the Panel that American tuna and Mexican tuna are ‘like products’ was not appealed by the parties. However, it could have been argued that in view of the intense consumer preferences, dolphin-friendly tuna and dolphin-unfriendly tuna were not ‘like products’, so that a different regulatory regime for dolphin-unfriendly tuna would have been deemed legitimate.Footnote ¹¹⁷ In any case, the US never raised a ‘like products’ argument and both parties agreed that the products at issue were simply tuna.Footnote ¹¹⁸

Another stumbling block is the relationship between government regulation and consumer preferences. In Japan–Alcoholic Beverages I, the panel warned that while the subjective factor of consumer taste should not be left out of, a tax should not be allowed to ‘crystalize’ consumer preferences for traditional domestic products.Footnote ¹¹⁹ This position was later repeated in other panels' reports.Footnote ¹²⁰ The underlying assumption, that government regulation should not influence consumer preferences, is not surprising, given the free market basis of the WTO.Footnote ¹²¹ However, when applying this logic to a public policy context, it has radical implications. Governments routinely make normative regulations such as sustainability criteria which affect consumer preference and influence consumers to adapt to shifting or emerging norms. These regulatory preferences likely reflect their national values and priorities. It is important that the WTO system not act as a braking mechanism for more progressive regulations that support emerging norms such as regulations that support environmental sustainability, or climate change mitigation.Footnote ¹²²