1. Introduction
Thailand – Customs and Fiscal Measures on Cigarettes from the Philippines (DS371),Footnote 1 hereafter labelled Thai–Cigarettes, has at its heart a number of questions surrounding national treatment in Article III:2 of GATT 1994 and the concept of ‘less favourable treatment’ in Article III:4. In particular, Thailand was charged with engaging in practices that resulted in excessive customs duties on imported cigarettes and excessive value-added taxes charged to imported cigarettes in comparison with domestic competing brands. Moreover, the structure of the Thai VAT placed more onerous administrative burdens on resellers of imported cigarettes than on resellers of domestic like products, suggesting that imports were treated less favourably. In principle, this system could alter the conditions of competition in favour of Thai's monopoly producer and restrain sales of imported brands.
The dispute arose from issues of customs valuation, with the Philippines claiming that Thai Customs had, in the period 2006–2007, systematically rejected the import values declared by Philip Morris Thailand (PMT) in favour of a deductive method that was outside its own procedures and improper under the rules of the WTO Customs Valuation Agreement (CVA). This method was alleged to establish higher cif (cost, insurance, and freight) import prices than would exist had PMT's declared values been accepted. The import price was the foundation for a constructed maximum retail-selling price (MRSP) for imported brands set by Thailand's Directorate General of Excise (DG Excise). The corresponding MRSP for domestic brands was built up from the ex-factory prices of the Thailand Tobacco Monopoly (TTM). The customs-valuation problem, along with added customs duties (for imports only) and a reserve for marketing costs (for both domestic and imported goods, but calculated differently), implied that the MRSPs for PMT's brands were systematically higher than those for TTM's brands.
The tax significance was that the MRSPs were the prices on which VAT collection was based. Because the MRSPs on imports were higher, they faced higher taxes in absolute terms. Thailand countered that its system was consistent with national treatment because the VAT rate was the same between domestic and imported cigarettes and the tax base was being established in the same manner for domestic and imported cigarettes in most instances.Footnote 2 However, a further significant issue was that resellers of domestic brands were fully exempted from collecting and paying VAT, while resellers of foreign brands were not. This imposed a stiffer administrative burden on the latter, the basis for a charge that the system offered less favourable treatment to imports.
After seeking consultations, in 2008 the Philippines requested a WTO Panel to investigate these issues. The Panel was established on 17 November 2008 and the Panel Report was circulated to WTO members two years later on 15 November 2010, much beyond the DSU's suggested time frame of nine months between Panel establishment and Report circulation. The Panel found that Thai Customs had acted inconsistently with its obligations under the CVA by: (i) rejecting PMT's declared values in 2006 and 2007 in violation of Articles 1.1 and 1.2; (ii) failing to communicate the grounds for this rejection in violation of Article 1.2(a); (iii) failing to provide an adequate explanation for how it determined customs values for imported cigarettes in violation of Article 16; (iv) assessing improperly the deductive value of imported cigarettes in violation of Article 7.1; (v) failing to properly inform PMT of the customs value determined under Article 7.1 and the method used to calculate that value in violation of Article 7.3; and (vi) revealing certain confidential business information to the media in violation of Article 10. These findings, though contested by Thailand in the Panel hearings, were not appealed to the Appellate Body.
The Panel also found Thailand had violated its national-treatment obligations in three ways. First, DG Excise violated GATT Article III:2, first sentence, by discriminating against imported cigarettes in the calculation of MRSPs, on which VAT was based, by using a less favourable method of calculating marketing costs for imported cigarettes than it used for calculating such costs for domestic cigarettes. Second, the VAT exemption for resellers of domestic cigarettes subjected imported brands to a VAT liability in excess of that facing like domestic cigarettes, a policy also inconsistent with Article III:2, first sentence. Third, the exemption subjected imported brands to less favourable treatment in violation of Article III:4 by imposing more burdensome administrative requirements on resellers of such brands as compared to resellers of domestic cigarettes. The Panel rejected Thailand's defense that the administrative burdens were necessary to safeguard against tax evasion, fraud, and counterfeiting of imported cigarettes. The defense was offered both as a claim that the discrimination did not violate Article III:4 and that it was permitted under the general exception found in Article XX(d). Thailand appealed all of these findings except for the first-described Article III:2 violation relating to the calculation of MRSPs.
The Panel also found that Thailand had violated a number of provisions of GATT Article X in connection with its imposition of customs duties and VAT on imported cigarettes. In particular, the Panel found that Thailand violated the publication requirements of Article X:1 by failing to publish (i) the methodology used to determine the price on which value-added tax would be levied on imported cigarettes and (ii) the general rules pertaining to the release of customs guarantees. The Panel also found that Thailand violated the reasonable administration requirement of Article X:3(a) in respect of delays in the [Customs] Board of Assessment process. In respect of Article X:3(b), which requires that WTO members must maintain judicial, arbitral, or administrative tribunals or procedures for the prompt review and correction of administrative action relating to customs matters, the Panel found that Thailand violated that obligation in respect of (i) customs valuation decisions and (ii) appeals of guarantee decisions. Only the latter finding regarding guarantee decisions was appealed.
Thailand appealed the various rulings indicated above in March 2011 and the oral hearing took place in April of that year, with both disputants and several third-party participants, including the United States, Australia, and the European Union, making representations about aspects of the case. In its Report, issued on 17 June 2011,Footnote 3 the Appellate Body upheld the Panel's finding that Thailand had acted inconsistently with the national-treatment requirement in Article III:2 by subjecting resellers of imported cigarettes to a higher tax liability. It also upheld the Panel ruling that the VAT exemption subjected imported cigarettes to less favourable treatment than domestic cigarettes, in violation of Article III:4. Further, while recognizing that tax systems may be defended under Article XX(d), the Appellate Body ruled that Thailand had not justified the necessity of its measure in that regard. Finally, it agreed with the Panel that Thailand had failed to provide the requisite independent review tribunals or processes to oversee decisions regarding guarantee charges.
On paper, this case seems rather straightforward and it is hard to fault either the Panel or the Appellate Body Reports. However, the case does raise a number of interesting legal and economic questions, which are the subject of the remainder of this paper. In Section 2, a brief review of trade data is offered to set an economic context. Section 3 presents the legal analysis, focused closely on the meaning of ‘less favourable treatment’. In Section 4, we turn to the economics of this situation, raising specific questions. For example, if tax evasion is a problem more endemic to imported than domestic cigarettes, as Thailand claims, is a strict application of national treatment actually optimal? Another question worth asking is how such differential administrative burdens might change competitive market conditions, particularly in the presence of a domestic monopoly supplier. A further wrinkle here arises from the interdependence of the importer with its Philippines supplier and the determination of customs value, when that determination directly affects relative VAT charges. Finally, there is the question of how such administrative uncertainty in tax collections may influence competitive behaviour. We wrap up in a final section.
2. The international trade context
Thailand is a reasonably large market for cigarette consumption. According to the World Health Organization, around 24% of the country's 68 million people smoked cigarettes in 2009, though this proportion has been declining for some time because of active attempts to discourage this behaviour.Footnote 4 Indeed, survey data suggests that reported consumption fell from 1.26 billion packs in 1999 to 862 million packs in 2009.Footnote 5 At the same time, there appear to be significant amounts of counterfeit or smuggled cigarettes in circulation.
Domestic manufacture of cigarettes is undertaken solely by the Thailand Tobacco Monopoly (TTM), which is a branch of the Ministry of Finance. Domestic production is monopolized, in principle, to help the government raise revenue from excise taxes, avoid tax evasion, control retail prices, and engage in deterrence campaigns. These objectives seem somewhat at odds with the mandate for TTM to maximize its profits.Footnote 6
While Thailand prevents entry of domestic and international firms into local production for the market, imports have been successively liberalized, largely as a result of tariff cuts under the ASEAN Free Trade Agreement (signed in 1992). Thus, the share of imported brands of total legal cigarette sales rose rapidly from 3% in 1996 to 22% in 2006 (Pavananunt, 2009). Over the same period, the tariff rate faced by ASEAN partners, such as the Philippines, fell from 20% to 5% in three discrete stages. In contrast, the excise tax on tobacco products, levied on both imports and domestic goods, rose from 68% to 79%, attesting to the relatively high taxation of this commodity.
The Philippines is, by far, the largest source of imports of cigarettes into Thailand. Table 1 shows the value and quantities in recent years of Thai imports from the Philippines in ‘cigarettes containing tobacco’, category 240220 in the 2002 HS system. It also shows the share of Philippine exports in this category going to Thailand. Thus, Thailand's total imports were stable at around $93 million in this period, though they fell off to $81 million in 2009 before rebounding to over $100 million in 2010. On average, Philippine exporters had about 67–75% of the imports in Thailand. Looking at export data, the Philippines has averaged around $90 million to $100 million in total exports, with Thailand taking about 65–70%. Clearly, the Thai market is quite important for cigarette exporters in the Philippines. In that context, there are certainly political-economy reasons for the Philippines to be concerned about Thai restrictions on trade in this sector.
Table 1. Trade data in cigarettes, Thailand and the Philippines
Source: United Nations COMTRADE database, category 240220.
However, the dispute in question surrounds actions taken against a particular importer, Philip Morris Thailand (PMT), in the years 2006 and 2007. As may be seen, Thailand's recorded imports from the Philippines rose in these years, especially in 2007. Thus, there seems little reason to think that the Philippines request for a Panel was motivated by a marked reduction in aggregate trade. Rather, it was aimed at correcting what it saw as improper customs actions raised specifically against PMT and the associated impacts on value-added tax liabilities under Thailand's internal tax regulations.
3. Legal issues: a role for the ‘less favourable treatment’ standard of Article III:4?
As described above, the Thai–Cigarettes case involved a broad-based challenge by the Philippines to the treatment of imported cigarettes by Thailand. At the Panel level, the case was noteworthy from the legal perspective because of the extensive consideration given to the Customs Valuation Agreement (CVA), which has not often been the subject of dispute settlement.Footnote 7 However, the Panel's findings of several violations of the CVA were not appealed, nor was its basic finding of tax discrimination arising through the calculation of MRSPs under Article III:2 appealed. This, in itself, is striking because it appears that the genesis of the case was a perception on the part of PMT that Thailand had decided to increase significantly the customs duties and taxes payable on imported cigarettes through manipulation of the customs value of those cigarettes and their MRSP on which value-added taxes were based. Instead, the appeal was limited to rather narrow legal issues related to certain of the Panel's findings under GATT Article III (and an Article XX defense) and GATT Article X. In particular, the Appellate Body Report is noteworthy, if at all, because (i) it confirmed the Appellate Body's practice of giving a quite broad reading to Article III:2 and III:4; (ii) it seemed to restrict the coverage of the Article XX(d) defense compared to one past precedent; and (iii) it gave an expansive reading to Article X's obligation to provide review of customs decisions. In addition, there were two procedural issues of note discussed by the Appellate Body. The most interesting issue in the case was probably the Appellate Body's consideration of the less favourable treatment issue that arises in GATT Article III:4 claims, where it continued its practice of requiring little to be shown by a complainant to establish such treatment.
3.1 Procedural issues
There were two interesting procedural issues that should be mentioned in passing that arose in Thai–Cigarettes.
1. Typographical errors
First, on appeal of the Panel's findings on Article XX, it appeared that the Panel may have made an error in a cross-reference that was key to its conclusion on the inapplicability of the Article XX(d) defense raised by Thailand.Footnote 8 In defending the Panel's rejection of that defense, the Philippines argued that the cross-reference was a mere clerical error and that the Appellate Body should correct it. Since only the Panel could definitively answer the question of whether there was such an error, it would seem that the Panel could have been asked to correct or confirm the error (if it agreed that there was an error), but the Panel was not asked. Instead, the Appellate Body concluded that while the cross-reference was an ‘obvious error’, it was not clear what the correct cross-reference would have been, given that the Panel's analysis of the Article XX(d) defense was ‘extremely brief’, suggesting perhaps that more than a mere typographical error may have been involved.Footnote 9 Accordingly, the Appellate Body reversed the Panel's finding that Thailand had failed to establish a valid Article XX(d) defense.Footnote 10 It then proceeded to ‘complete the analysis’ of the issue and concluded that the defense was not established for reasons that are discussed below.Footnote 11
As noted, it appears that the Panel was never asked to clarify the cross-reference. Assuming that the cross-reference was an error, it is unclear why the Panel was not asked. Perhaps it was felt that since the issue was first raised on appeal, the Panel was somehow incapable of responding to a request for clarification. It is not clear why this would be so. While it could be argued that a Panel ceases to exist, and therefore can no longer act on any issue, once it has issued its Report to the parties (or, at least, at the point when the Report is circulated to the WTO membership), it would seem that there is no inherent reason why a Panel could not correct typographical errors after its Report is issued and circulated. Indeed, corrigenda are routinely issued by the Secretariat, although not necessarily with the participation of the Panel. Moreover, unlike the days of GATT dispute settlement when the Panel had no further involvement with a case once the Report was issued, under the WTO Dispute Settlement Understanding the original Panel may be called upon to serve as a compliance Panel under DSU Article 21.5 or as the arbitrator of the level of suspension of concessions under DSU Article 22. Thus, Panels seem to have a continuing existence and could be asked by the Appellate Body to correct a typographical error if one is suspected.Footnote 12
In this regard, it is noteworthy that judicial and arbitral bodies typically have such authority. For example, under the US Federal Rules of Civil Procedure, a federal district court may correct a clerical mistake or a mistake arising from oversight or omission whenever one is found in a judgment, on motion or on its own, with or without notice.Footnote 13 Similar provisions are found in the UNCITRAL and ICSID arbitration rules.Footnote 14 Such a procedure should be available at the WTO, as it would seem to be the best and most efficient way to deal with the sort of typographical error that arose in this case. While the WTO does not have detailed procedural rules like those of UNCITRAL and ICSID to which an appropriate provision could be added, in our opinion the power to correct errors in findings should be viewed as an inherent power of a tribunal.
2. Right to respond to late-filed evidence
In response to a question from the Panel following the second meeting with the parties that is at the very end of the typical Panel proceeding, Thailand presented evidence on one aspect of the VAT issue. This evidence essentially rebutted evidence offered at the second meeting by the Philippines. As provided for by the Panel, the Philippines were allowed to comment on Thailand's answer. In its comment, the Philippines submitted an additional piece of rebuttal evidence. Thailand did not ask to comment and did not comment on the new piece of evidence, although at the interim-review stage, Thailand argued that the Panel should not have relied on the new piece of evidence. In its final Report, the Panel ruled that it could rely on the evidence and noted that, as rebuttal evidence, it was not filed in an untimely manner under the Panel's rules.Footnote 15 On appeal, Thailand argued that its due-process rights were violated by the Panel's acceptance of the evidence and that the Panel had accordingly violated DSU Article 11.Footnote 16 The Appellate Body rejected the appeal, noting that the piece of evidence in question was only one of several pieces on the VAT issue that the Panel considered and that the contested piece of evidence did not seem to be the principal basis for the Panel's ruling in any event.Footnote 17 It also noted that Thailand did not request an opportunity to respond.Footnote 18 This seems to be the correct result. One of the authors has served as a panelist on several occasions, and it would seem that Panels are generally open to allowing parties to comment on new evidence, whenever presented. However, the Panel's rules of procedure, which seem fairly standard, provided as follows:Footnote 19
The parties shall submit all factual evidence to the Panel no later than the first substantive meeting, except with respect to factual evidence necessary for the purposes of rebuttals, answers to questions or comments on answers provided by each other. Exceptions to this procedure will be granted where good cause is shown. In such cases, the other party shall be accorded a period of time for comment, as appropriate.
It would seem that the last sentence should be revised to make it clear that there is always a right to comment on evidence presented. For example, it could be rewritten to read:
In all cases, the other party, on request, shall be accorded a period of time for comment, as appropriate.
Normally, there would only be a need for such a request in the situation that arose here, since typically there will in fact be an opportunity for the other party to comment on any evidence that is presented to the Panel. While it is possible, under such a rule, that parties will continue to trade counterevidence and countercomments for some weeks after what was intended to be the end of the proceedings, the experience of the author who has been a panelist suggests that the parties run out of things to say rather quickly. Thus, adoption of such a rule should satisfy legitimate due-process concerns and should not in any way delay the issuance of the Panel Report.Footnote 20
3.2 Article III
Under the applicable Thai tax regime, resellers of domestic cigarettes, which must be obtained from the government-owned tobacco monopoly, were not subject to value-added tax on their sales of cigarettes.Footnote 21 As a consequence, they were also not subject to any of the applicable administrative requirements connected with the tax. Resellers of foreign cigarettes were subject to value-added tax on their sales of foreign cigarettes and the related administrative requirements. However, under Thai law, the amount of VAT collected on the sale of cigarettes at each stage of the resale chain from the manufacturer or importer to the ultimate seller to consumers is based on the maximum retail selling price set by the government for each brand, which means the tax due is the same on each sale. Thus, as is typical in a VAT system, since those subject to VAT are allowed to deduct the VAT paid by them to the seller of the cigarettes from the VAT they receive from their purchaser, the net VAT liability for those selling imported cigarettes is zero, assuming they comply with the applicable administrative requirements in respect of filing VAT returns and of keeping records. In other words, resellers of both domestic and imported cigarettes would normally have no liability for value-added tax on those products – the former because they were exempt from the tax and the latter because the net tax due would be zero given the use of a single price for calculating VAT liability at all stages of the distribution of cigarettes.
Given these circumstances, the Panel found a violation of Article III:2 since sellers of domestic cigarettes were exempt from the value-added tax, while resellers of imported cigarettes were subject to the tax. In the view of the Panel, while resellers of domestic cigarettes could never be subject to VAT on their sales of domestic cigarettes, resellers of foreign cigarettes were subject to VAT and might incur liability if certain conditions were not met (i.e., they failed to file the proper tax forms, retain the proper tax invoices, and meet other administrative requirements for obtaining a VAT refund).Footnote 22 Similarly, the Panel found a violation of Article III:4 in that resellers of foreign cigarettes were subject to various administrative requirements – filing tax returns, maintaining certain records, and filing certain reports, and penalties for failure to comply with these requirements – in connection with the VAT regime, while resellers of domestic cigarettes were not.Footnote 23
1. Article III:2
On appeal of the finding of a violation in respect of Article III:2, Thailand essentially argued that there was no difference in tax liability because resellers of imported cigarettes could, by filing the proper forms and obtaining and keeping the necessary records, avoid any tax liability. The Appellate Body rejected this argument:Footnote 24
[W]e do not consider that Thailand's measure precludes a finding of inconsistency with Article III:2 due to the fact that resellers of imported cigarettes may take action to avoid imposition of VAT liability. In our view, the availability of such a course of action does not alter the legal assessment of whether, under Thai law, imported cigarettes are subject to internal taxes or other internal charges in excess of those applied to domestic cigarettes. As we have explained, Thailand's measure provides for circumstances in which resellers of imported cigarettes will be subject to VAT liability, to which resellers of domestic cigarettes will never be subject.
… Imposing legal requirements that result in tax liability on imported products when resellers do not satisfy prescribed conditions necessary to avoid that liability, but which never result in tax liability on like domestic goods, is inconsistent with the requirements of Article III:2, first sentence.
In light of past precedents, this result is not at all surprising. For like products, no tax differential is permitted.Footnote 25 Here, such a differential could have occurred because of the structure of the Thai system. Ergo, that system is not consistent with Article III:2 under past case law. While one could question the significance of this differential treatment, there seemed to be no justification at all for treating resellers of domestic cigarettes and imported cigarettes differently. Requiring Thailand to treat both types of resellers the same way for tax purposes certainly does not seem to infringe in any meaningful way on Thailand's ability to tax and regulate cigarettes.
2. Article III:4
The appeal of the finding of a violation of Article III:4 presented a somewhat more difficult issue in that, unlike Article III:2, first sentence, which permits no differential treatment in respect of taxation, Article III:4 does permit different treatment of imported products, so long as the treatment is not less favourable than that accorded to domestic like products. Thus, origin-based discrimination is in theory permitted under Article III:4. Indeed, in Korea–Beef, where the Panel found that discrimination between domestic and imported products explicitly based on origin violated Article III:4, the Appellate Body reversed the Panel and stated:Footnote 26
A formal difference in treatment between imported and like domestic products is thus neither necessary, nor sufficient, to show a violation of Article III:4. Whether or not imported products are treated ‘less favourably’ than like domestic products should be assessed instead by examining whether a measure modifies the conditions of competition in the relevant market to the detriment of the imported products.
We conclude that the Panel erred in its general interpretation that ‘[a]ny regulatory distinction that is based exclusively on criteria relating to the nationality or the origin of the product is incompatible with Article III’.
Although the Appellate Body quoted the foregoing in part in the Thai–Cigarettes case, as we will see, it arguably did not require a very rigorous demonstration that the conditions of competition between Thai and imported cigarettes had been modified.
The Panel in the Thai–Cigarettes case had reasoned that the additional administrative requirements to which resellers of imported cigarettes were subject ‘could potentially affect’ the conditions of competition between domestic and imported cigarettes because the burden associated with the additional administrative requirements could increase the operating costs of resellers of imported cigarettes, which could cause some potential resellers of imported cigarettes to not sell them so as to avoid incurring those additional costs. The Panel found support for its position in evidence that showed that there was a degree of switching and cross-price elasticity between certain domestic and imported brands, which the Panel felt indicated that the market for those cigarettes was very competitive.Footnote 27 In such a market, the increased costs involved in selling imported cigarettes might well deter some resellers from handling such cigarettes. The Panel did not consider how much it would cost to comply with the additional administrative burdens, nor did it ascertain how prevalent this situation would be since the potential cost difference would largely apply only where the reseller of domestic cigarettes did not sell other products subject to VAT (since if it did sell such products, the added administrative structures would have already been in place). Moreover, the Panel did not consider the extent to which a reseller would view these minor costs to be worth incurring by the competitive imperative to offer a full range of cigarettes to its clients.Footnote 28
In its approach to the less favourable treatment issue in Thai–Cigarettes, the Appellate Body began by stressing that its resolution in specific cases requires a ‘careful examination “grounded in close scrutiny of the ‘fundamental thrust and effect of the measure’”’, although such scrutiny need not involve empirical evidenceFootnote 29 and that ‘[t]he implications of the contested measure for the equality of competitive conditions are, first and foremost, those discernible from the design, structure, and expected operation of the measure’.Footnote 30
It then considered Thailand's arguments that the ‘could potentially affect’ standard used by the Panel was insufficiently rigorous and that the Panel had failed to consider sufficiently the implications of the challenged measure in the marketplace. The Appellate Body rejected this appeal since in its view:Footnote 31
an analysis of less favourable treatment should not be anchored in an assessment of the degree of likelihood that an adverse impact on competitive conditions will materialize. Rather, an analysis under Article III:4 must begin with careful scrutiny of the measure, including consideration of the design, structure, and expected operation of the measure at issue. Such scrutiny may well involve – but does not require – an assessment of the contested measure in the light of evidence regarding the actual effects of that measure in the market. 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 to support a finding that imported products are treated less favourably.
The Appellate Body also noted that the Panel's use of the word ‘potentially’ did not suggest that the Panel had found less favourable treatment based on a ‘remote, unsubstantial or ‘theoretical possibility’’ that competitive conditions might be modified. Rather, it interpreted the Panel's use of the word to reflect the fact that the Panel believed that it was not required to enquire into the actual effects of the Thai measure in order to establish less favourable treatment.Footnote 32
Having said all that, the Appellate Body accepted the Panel's view that the additional administrative burdens imposed on resellers of imported cigarettes would affect their costs and would, in turn, modify the conditions of competition to the detriment of imported cigarettes. While the Appellate Body noted that the Panel could have enquired further into market implications of the Thai measure, ‘the mere fact the additional administrative requirements are imposed on imported cigarettes, and not on like domestic cigarettes, provides, in itself, a significant indication that the conditions of competition are adversely modified to the detriment of imported cigarettes.’Footnote 33 Thus, in essence, the case stands for the proposition that differential treatment is less favourable for purposes of Article III:4 if it seems to impose any conceivable (even if speculative) additional burden on imported products.
The Appellate Body's interpretation of the less favourable treatment component of GATT Article III:4 has always seemed incomplete and not completely consistent with its national-treatment case law overall. In its first Article III case – Japan–Alcoholic Beverages II – the Appellate Body emphasized that Article III:1, which provides that internal taxes and regulations should not be used ‘so as to afford protection to domestic production’, was the motivating principle of Article III.Footnote 34 Indeed, in its view any differential taxation of like products implied that the taxation was so as to afford protection, although it required that a separate enquiry be made under the second sentence of Article III:2, which governs directly competitive and substitutable products, to determine if the tax measure was applied so as to afford protection.Footnote 35 Nonetheless, after having emphasized the importance of the language of Article III:1 and required it as a separate component of establishing a violation of Article III:2, second sentence, the Appellate Body found, in EC–Bananas III, that in the case of Article III:4 in deciding whether there was ‘less favourable treatment’, it was not necessary to consider whether a measure was applied so as to afford protection.Footnote 36 This decision was always difficult to understand since it would seem that a natural way of interpreting ‘less favourable treatment’ in Article III would consider whether the treatment was protectionist, given the important contextual role played by Article III:1.
In the cases following Bananas, the Appellate Body seemed to suggest that the less favourable treatment requirement would be given meaningful content. The next Article III:4 case was Korea–Beef, where, in the passage quoted above, the Appellate Body emphasized that a mere distinction based on origin was not sufficient to establish less favourable treatment, but rather that an effect on the conditions of competition between the imported and domestic products had to be shown. The next occasion for the Appellate Body to discourse on the concept of less favourable treatment was in the EC–Asbestos case. Since the Appellate Body had found that the products involved in the case were not like, there was no need to consider the less favourable treatment issue. However, the Appellate Body had given a broad reading to the concept of like products under Article III:4 – it was held to cover like products as that term was used in Article III:2, first sentence, plus at least some portion of the directly competitive and substitutable products subject to Article III:2, second sentence. The principal justification for an expanded view of like products in Article III:4 was that otherwise the ban on the use of discriminatory taxes on directly competitive and substitutable products could easily be evaded if there was not a similar restraint on the use of discriminatory internal regulations in respect of such products.Footnote 37 However, in order to allay concerns that giving a broad scope to ‘like products’ under Article III:4 would lead to many findings of an Article III:4 violation, the Appellate Body recalled that a violation of Article III:4 also required a showing that the imported product received less favourable treatment, suggesting that likeness was only part of the relevant analysis of an Article III:4 violation. Indeed, it specifically stated that ‘[t]he term “less favourable treatment’ expresses the general principle, in Article III:1, that internal regulations ‘should not be applied … so as to afford protection to domestic production”’.Footnote 38
Since Korea–Beef and EC–Asbestos, however, Panels have generally not required much to establish less favourable treatment through modification of the conditions of competition. Where there has been origin-based discrimination, less favourable treatment has been found without much analysis. Any difference in treatment has been viewed as modifying the conditions of competition. Indeed, in China–Publications and Audio Visual Products, the Panel noted that:Footnote 39
the phrase ‘treatment no less favourable’ is not qualified by a de minimis standard. Accordingly, any less favourable treatment of imported products … is contrary to the obligation in Article III:4, provided such treatment modifies the conditions of competition to the detriment of imported products.
In the Thai–Cigarettes case, the Panel noted that ‘additional administrative requirements, albeit slight, imposed only on imported products can potentially have a negative impact on the competitive position of [those products]’.Footnote 40
The only exception to this tendency to require little to establish less favourable treatment is the 2005 Dominican Republic–Import and Sale of Cigarettes case, where there were two non-origin-based rules challenged – one required the affixing of tax stamps in the Dominican Republic; the other required the posting of a bond to ensure tax payments. In the case of the bond requirement, the Appellate Body upheld the Panel's rejection of the Article III:4 claim. In doing so, it noted that while imported cigarettes bore a higher cost on a per capita basis, that was due to the fact that they had a small market share (and thus the cost of the bond on a per cigarette basis was higher), not because the cigarettes were imported. Accordingly, it found that there was not less favourable treatment for purposes of Article III:4.Footnote 41 While the Appellate Body upheld the Panel decision, the case has not played a significant role in subsequent national-treatment decisions, such as the instant case.Footnote 42 Indeed, quite recently, the Appellate Body has specifically stated that the Dominican Republic–Import and Sale of Cigarettes case does not require an inquiry as to whether the detrimental impact of a measure is related to its foreign origin or is explained by other facts or circumstances, but rather only requires consideration of whether the conditions of competition have been modified to the detriment of imported products.Footnote 43 From this, one can speculate that the Appellate Body would have upheld a Panel conclusion that the bond requirement modified the conditions of competition to the detriment of importers if the Panel had found that importers are typically smaller firms than domestic suppliers, and that, accordingly, a bond requirement would impose heavier costs on them on a per-item basis.
Thus, the state of the WTO jurisprudence today on Article III:4 seems to be that any difference in treatment between imported and domestic goods – no matter how minor – will suffice to modify the conditions of competition between them and violate Article III:4. This result does not sit easily with the Appellate Body's Article III:2 jurisprudence, or its recent TBT Agreement jurisprudence.
The inconsistency in approach between Article III:2 and Article III:4 can be seen by slightly modifying the facts of Thai–Cigarettes. Suppose, in the instant case, that cigarettes were found to be directly competitive and substitutable products (not like products) for purposes of Article III:2.Footnote 44 (They, of course, would still be like products for purposes of Article III:4 and TBT Article 2.1.) With this change in facts, the Article III:2 issue would not be so clear since the Philippines would have to establish that the taxation was dissimilar. Since the difference would normally be zero, assuming a VAT return was filed, this might be difficult. Moreover, it would be necessary to establish that the dissimilar taxation – normally zero – was applied so as to afford protection to the domestic industry, which would probably be even more difficult to establish. Thus, it is quite possible that there would be a finding of no violation of Article III:2, second sentence. But the same analysis as occurred in the instant case would apply in respect of Article III:4. Thus, there would be no tax discrimination, but the related administrative requirements would violate Article III:4. This highlights the problem of having Article III:4 apply to products that would be treated as directly competitive or substitutable under Article III:2, when ‘so as to afford protection’ is an element of the Article III:2 claim but not the Article III:4 claim. This, of course, undermines the rationale in EC–Asbestos for giving a broad reading to like products for purposes of Article III:4 since it ultimately leads to disparity in – not harmonization of – the respective coverages of Article III:2 and III:4.
The Appellate Body's Article III:4 jurisprudence is also in tension with its 2012 Reports dealing with the TBT Agreement, where the Appellate Body has given considerable prominence to the ‘treatment no less favourable’ language of TBT Article 2.1. Article 2.1 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.
This language is obviously quite similar to Article III:4, although an important difference between GATT and the TBT Agreement is that, while the obligations of Article III:4 are moderated by the exceptions in Article XX, the coverage of TBT Article 2.1 is not subject to any exceptions (although the preamble to the TBT Agreement contains language to the effect that Members should not be prevented from taking certain actions for the protection, inter alia, of health, the environment, and consumers).
The Appellate Body has emphasized the similarities between the language of TBT Article 2.1 and GATT Article III, particularly insofar as determining like products. More significantly, in trying to reconcile the effective coverage of GATT Article III, with exceptions, and TBT Article 2.1, without exceptions, and to prevent Article 2.2 from being rendered inutile by Article 2.1, the Appellate Body in US–Clove Cigarettes has decided that for purposes of TBT Article 2.1:Footnote 45
Although we are mindful that the meaning of the term ‘treatment no less favourable’ in Article 2.1 of the TBT Agreement is to be interpreted in the light of the specific context provided by the TBT Agreement, we nonetheless consider these previous findings by the Appellate Body in the context of Article III:4 of the GATT 1994 to be instructive in assessing the meaning of ‘treatment no less favourable’, provided that the specific context in which the term appears in Article 2.1 of the TBT Agreement is taken into account. Similarly to Article III:4 of the GATT 1994, Article 2.1 of the TBT Agreement requires WTO Members to accord to the group of imported products treatment no less favourable than that accorded to the group of like domestic products. Article 2.1 prescribes such treatment specifically in respect of technical regulations. For this reason, a panel examining a claim of violation under Article 2.1 should seek to ascertain whether the technical regulation at issue modifies the conditions of competition in the market of the regulating Member to the detriment of the group of imported products vis-à-vis the group of like domestic products.
However, as noted earlier, the context and object and purpose of the TBT Agreement weigh in favour of interpreting the ‘treatment no less favourable’ requirement of Article 2.1 as not prohibiting detrimental impact on imports that stems exclusively from a legitimate regulatory distinction. Rather, for the aforementioned reasons, the ‘treatment no less favourable’ requirement of Article 2.1 only prohibits de jure and de facto discrimination against the group of imported products.
Accordingly, where the technical regulation at issue does not de jure discriminate against imports, the existence of a detrimental impact on competitive opportunities for the group of imported vis-à-vis the group of domestic like products is not dispositive of less favourable treatment under Article 2.1. Instead, a panel must further analyze whether the detrimental impact on imports stems exclusively from a legitimate regulatory distinction rather than reflecting discrimination against the group of imported products. In making this determination, a panel must carefully scrutinize the particular circumstances of the case, that is, the design, architecture, revealing structure, operation, and application of the technical regulation at issue, and, in particular, whether that technical regulation is even-handed, in order to determine whether it discriminates against the group of imported products.
Thus, on the one hand, the Appellate Body seems to suggest that the less favourable treatment language of GATT Article III:4 and TBT Article 2.1 should be interpreted similarly, while, on the other hand, it gives real content to the language in Article 2.1 by permitting a modification of the conditions of competition if it stems from a ‘legitimate regulatory distinction’ (at least in cases of de facto discriminationFootnote 46), even though there is no such consideration of that issue under Article III:4. Of course, the justification is that the context of TBT Article 2.1 requires this result, but it is hard to see why the context of Article III:4 – and specifically the context supplied by Article 3.1 – should not also require such a result.
It remains to be seen whether this distinction announced by the Appellate Body can survive. In a very recent case – US–Tuna II (Mexico) – the Appellate Body stated that the Panel, which found a violation of the TBT Agreement, should have also analysed the case under Article III:4.Footnote 47 The reason given by the Appellate Body for criticizing the Panel's exercise of judicial economy was that it was contested whether the measure at issue was a technical regulation and thus the Panel's findings on the TBT Agreement could have been nullified if the Appellate Body had ruled that the measure was not a technical regulation. It also noted that the scope of the two provisions was different. In the future, it is likely that Panels will feel compelled to consider both TBT and Article III:4 claims, at least so long as the respondent argues that its measure is not a technical regulation. One can easily imagine the situation where a measure will be found to distinguish de facto between imported and domestic like products and accord less favourable treatment to the imported products. The measure will not violate Article 2.1 of the TBT Agreement if a legitimate regulatory distinction is established, but it will violate Article III:4. Since the coverage of Article XX is not as broad as the ‘exception’ that the Appellate Body has read into the TBT Agreement, one could imagine a measure found to violate GATT Article III:4, but not TBT Article 2.1. Given that the TBT Agreement is the more specialized agreement and the one that would prevail in the event of a conflict with Article III, this result – which probably could not be called a conflict under existing jurisprudence – would seem incongruous. In any event, complainants may again decide to shy away from the TBT Agreement when a measure can be challenged under Article III:4 since the TBT Agreement may afford respondents more scope to argue that a measure is justified than does Article XX.
All in all, one can understand why the Appellate Body has taken the position it has in respect of Article 2.1 of the TBT Agreement. It needed to harmonize the TBT Agreement and GATT Articles III:4 and XX, and it needed to give effect somehow to the preamble of the TBT Agreement. Its approach in US–Clove Cigarettes, however, highlights the problems with the way it has largely read the less favourable treatment requirement out of Article III:4. As outlined in the foregoing paragraphs, there are potential inconsistencies between its jurisprudence under GATT Article III:4 and its jurisprudence under GATT Article III:2, second sentence, and TBT Article 2.1. Eventually, some reconciliation will be required. If the Appellate Body is unwilling to import the ‘so as to afford protection’ test into the less favourable treatment analysis under Article III:4, perhaps it will give more prominence to the discrimination involved, i.e. whether it is de jure or de facto, and require a more rigorous showing of a modification of the conditions of competition in de facto discrimination cases, such as arguably occurred in the Dominican Republic–Import and Sale of Cigarettes case.Footnote 48
3.3 Article XX(d)
The Article XX issue was the one that may have involved a typographical error. Article XX(d) provides an exception for measures that are ‘necessary to secure compliance with laws or regulations which are not inconsistent with [GATT]’. Thailand claimed that the administrative requirements found to violate Article III:4 were necessary to secure compliance with its VAT regime. The Panel ruled that Article XX(d) did not apply because it had already ruled that the laws or regulations with which compliance was sought were not GATT consistent and cross-referenced its finding that the administrative requirements violated Article III:4. The Philippines argued that the cross-reference was a typo, that the Panel intended to refer to its finding that the differential VAT treatment was inconsistent with Article III:2, and that the administrative requirements could therefore not be justified given that the underlying tax provision was invalid. However, the Panel's reasoning on the Article XX(d) issue was so brief it was not clear to the Appellate Body what the Panel had meant, so the Appellate Body itself examined whether Thailand had established the elements of an Article XX(d) defense.Footnote 49
In setting the elements of an Article XX(d) defense, the Appellate Body noted that:Footnote 50
when an Article XX(d) is invoked to justify an inconsistency with Article III:4, what must be shown to be ‘necessary’ is the treatment giving rise to the finding of less favourable treatment. [The Appellate Body cited the GATT Panel Report, US–Section 337 for this proposition, where it is clearly stated.] Thus, when less favourable treatment is found based on differences in the regulation of imports and of like domestic products, the analysis of an Article XX(d) defense should focus on whether those regulatory differences are ‘necessary’ to secure compliance with ‘laws or regulations’ that are not GATT-inconsistent.
Having set out this standard, the Appellate Body found four ‘critical flaws’ in Thailand's defense, the first of which was that ‘Thailand sought to justify administrative requirements relating to VAT liability generally, rather than to justify the differential treatment afforded to imported versus domestic cigarettes under its measure’.Footnote 51 The Appellate Body also recalled that Thailand had not discussed in any detail several other matters essential to a successful Article XX(d) defense.Footnote 52
As noted, the Appellate Body's approach is completely consistent with the approach of the US–Section 337 case, which the Appellate Body has often cited with approval in respect of other issues, especially involving Article XX. It does, however, appear to be inconsistent with the first Appellate Body case – US–Gasoline, where the Appellate Body stated in considering an Article XX(g) defense that the Panel had erred because:Footnote 53
the Panel asked itself whether the ‘less favourable treatment’ of imported gasoline was ‘primarily aimed at ‘the conservation of natural resources [i.e., whether the refusal to provide individualized baselines to foreign refiners was primarily aimed at the conservation of natural resources], rather than whether the ‘measure’, i.e. the baseline establishment rules, were ‘primarily aimed at’ conservation of clean air.
When it analyzed the Article XX(g) defense itself, the Appellate Body started by looking at ‘[t]he baseline establishment rules, taken as a whole (that is, the provisions relating to establishment of baselines for domestic refiners, along with the provisions relating to baselines for blenders and importers of gasoline).’Footnote 54 Had it followed the approach taken in the instant case, the Appellate Body would have only looked at the aspect of the baseline establishment rules found to violate Article III:4.
The Appellate Body's analysis of the Article XX(d) defense was quite cursory, and it gave no indication that it was changing the approach taken in Gasoline. Thus, it is not clear whether it intended to make a change or simply overlooked the Gasoline case. Hopefully, it intended the change since the Gasoline approach has always seemed to be the wrong approach. Indeed, when the Appellate Body did what it did in Gasoline, it made no reference to past GATT cases, such as the US–Section 337, which followed the approach it has now adopted in the instant case. In any event, by accident or design, it appears that the Appellate Body has reached the right result. It makes no sense to ask if a general law falls within the exceptions of Article XX. The issue is whether a GATT violation is excused, so the issue must be whether the violation is necessary (or, in the case of XX(g), is related to conservation). This can be seen in the extensive discussion of the issue in the Section 337 case, where the Panel noted that GATT parties must use more GATT-consistent alternatives where such alternatives are reasonably available. It should be noted that the Appellate Body's position in the Gasoline case in fact does not seem to have been discussed in other Article XX cases, so the instant case may simply be restating what the rule has always been considered to be (notwithstanding the Appellate Body's position in Gasoline).Footnote 55
3.4 Article X:3
The Philippines raised a number of GATT Article X claims in respect of the Thai regime for taxing imported cigarettes. In that regard, the Panel found that Thailand violated the publication requirements of Article X:1 by failing to publish (i) the methodology used to determine the price on which value-added tax would be levied on imported cigarettes and (ii) the general rules pertaining to the release of customs guarantees. The Panel also found that Thailand violated the reasonable-administration requirement of Article X:3(a) in respect of delays in the [Customs] Board of Assessment process. In respect of Article X:3(b), which requires that WTO members must maintain judicial, arbitral, or administrative tribunals or procedures for the prompt review and correction of administrative action relating to customs matters, the Panel found that Thailand violated that obligation in respect of (i) customs valuation decisions and (ii) guarantee decisions. Only the latter finding regarding guarantee decisions was appealed.
Under Thai law, when there is doubt as to the applicable duty owed on an import, Thai customs officials may undertake a more in-depth examination of the matter. The goods at issue may be released from customs pending this further examination if the importer pays the uncontested amount due and provides a guarantee that covers the maximum duty that might be payable. The guarantee may be in the form of a cash deposit or a bank (or some sort of government) guarantee. Thailand appealed two aspects of the Panel's decision: (i) whether the decision to require a guarantee constitutes administrative action relating to customs matters as used in Article X:3(b) and (ii) whether the right to appeal the guarantee decision upon the final assessment of duties satisfies Thailand's Article X:3(b) obligations.
On the first issue, the Appellate Body considered the meaning of the components of the phrase ‘administrative action relating to customs matters’. It seems clear that the decision to require a guarantee is an administrative action and that it relates to customs matters. The only real issue is whether the decision should be treated as a preliminary one that is reviewable at a later point of time, which raises the question of whether that later review would meet the promptness requirement of Article X:3(b). As to the preliminary nature of the decision, the Appellate Body noted that the Customs Valuation Agreement only specifically requires review of customs-valuation decisions and that the Anti-Dumping Agreement only requires review of ‘final’ anti-dumping action. However, the Appellate Body was of the view that this context was not all that relevant, particularly since Article X:3(b) was not qualified by any limitation to ‘final’ action.Footnote 56 The Appellate Body was of the view that the decision on the guarantee was a final decision with content of its own, as there was no further action contemplated insofar as securing payment of customs duties was concerned.Footnote 57
As to the question of whether the ‘prompt review’ requirement was satisfied if an appeal was permitted once the final assessment was issued, the Appellate Body noted that once a final assessment has been issued, the guarantee ceases to have any purpose since the amount due will then be collected. That implies that during the effective or useful life of the guarantee, there is no possibility of review, which means that the prompt-review requirement is not met.Footnote 58
On a first look, this decision seems rather intrusive. While the Appellate Body correctly parses Article X:3(b), it seems to require rather early review of an arguably minor, preliminary matter.Footnote 59 However, on reflection, it is worth noting that guarantee decisions can be abused, and customs authorities can delay decisions for a long time. In fact, there were examples of such behaviour being threatened by Thai customs presented to the Panel.Footnote 60 Thus, there may be a real problem here. Moreover, the actual requirements to implement this decision may be rather minor. A review of a guarantee decision would not necessarily require the correct valuation of the goods to be established, but rather only whether customs officials had reason to conduct a further investigation and whether the amount of the guarantee was reasonable. At this early stage of the valuation process, presumably Thailand could implement the decision without much effort and the new review process could afford considerable discretion to customs authorities. Thus, in the end, this broad reading of Article X:3(b) may make eminent sense to the extent that it requires little action by WTO members, but may prevent occasional abuses by individual customs officials. Indeed, the preferred course might be to have a general-review provision allowing the challenge of preliminary-type decisions on some sort of gross abuse of discretion grounds.
3.5 Conclusion
The Appellate Body Report in the Thai–Cigarettes case is consistent with past Appellate Body jurisprudence (except as noted in respect of Article XX) and does not raise significant issues, except insofar as one can question whether the Appellate Body has given adequate consideration to Article III:4's requirement that imported products must be shown to have been accorded treatment no less favourable than that accorded to like products of national origin.
4. Economic analysis
The circumstances underlying the complaint brought by the Philippines against Thailand were described above. However, it is useful to place them into an economic-policy context in order to offer some commentary about interesting issues they raise.
4.1 Market structure, customs, and taxes
As noted, there is one domestic manufacturer of tobacco products, including cigarettes, in Thailand. TTM is an agency of the Ministry of Finance, though it is enjoined to maximize profits. TTM has been the sole domestic supplier for decades and until the market was opened to imports, beginning in 1991, had a virtually complete monopoly. Currently, TTM commands about 75% of the market, with imports taking up the rest. According to the Panel Report, TTM produces 19 brands. TTM sells its brands at its chosen ex-factory prices to wholesalers and retailers (called resellers in the case). These brands come in essentially three quality classes, with retail prices (pre-VAT) ranging from 37 baht per pack at the low end to 58 baht per pack at the medium level and 75 baht per pack for deluxe brands. A key factor is that retail prices of TTM's cigarettes invariably equal the ‘maximum retail selling prices’ (MRSPs), which are set by the government based on a constructed formula reviewed below. TTM may petition the authorities to permit retailers to set lower prices on its brands but, as a branch of the government, never does so.
At the time of the Panel Report, there were 86 imported brands of cigarettes, coming from a large group of countries. As noted, however, the Philippines is the primary source of imports and a major producer is Philip Morris Philippines (PMP), which manufactures and sells them directly to an importing firm called Philip Morris Thailand (PMT), which declares the cost-inclusive cif price it pays and also must pay a 5% duty. Because these entities both exist under the same parent company, there is a risk that the declared price will not really reflect actual production and transactions costs and will instead be a transfer price. PMT then sells these imports to wholesalers and retailers. These resellers may also sell TTM's brands. Like those brands, each imported brand is assigned an MRSP, which is the basis for VAT calculations. However, because these imported products are procured from private firms it is more likely that those firms may wish to petition the authorities to sell at retail prices below the MRSPs.
Part of the dispute revolves around the Thai Department of Customs' treatment of two brands sold by PMT: Marlboro, which competes in the premium market, and L&M, a mid-level brand. On certain occasions in 2006 and 2007, Thai Customs refused to accept the declared customs value put forward by PMT for these cigarettes. Customs officials argued that, despite having accepted such valuations before, in these cases the agency had serious doubts about the legitimacy of those prices. The only grounds offered were that PMT and PMP were closely related and could not be trusted to declare a price approximating an arm's-length valuation. Moreover, Thai Customs argued that similar types of cigarettes were imported at those times at declared prices three to four times higher, a characterization disputed by PMT. The authorities issued a preliminary customs valuation based on a ‘deductive valuation’ method that resulted in a substantially higher cif value and also required PMT to post a fiscal guarantee against the final valuation. The Panel found these procedures to be inappropriate under terms of the CVA.
For our purposes, the interesting feature is that Thai Customs' decision to raise the cif valuation directly increased the VAT basis (MRSP) for PMT's cigarettes. The reason is that DG Excise uses a mechanical formula for determining MRSPs, as follows.
For TTM cigarettes:
$$\eqalign{{\rm MRSP} = \ & {\rm ex{\hbox-}factory}\;{\rm price} + {\rm excise}\;{\rm tax} + {\rm health}\;{\rm tax} + {\rm television}\;{\rm tax} \cr & + {\rm local}\;{\rm provincial}\;{\rm tax} + {\rm VAT} + {\rm marketing}\;{\rm costs}.} $$For imports:
$$\eqalign{{\rm MRSP} = \ & {\rm cif}\;{\rm value} + {\rm customs}\;{\rm duties} + {\rm excise}\;{\rm tax} + {\rm health}\;{\rm tax} + {\rm television}\;{\rm tax} \cr & + {\rm local}\;{\rm provincial}\;{\rm tax} + {\rm VAT} + {\rm marketing}\;{\rm costs}.} $In these equations, the excise tax is 79% of either the ex-factory price (for TTM) or of the sum of cif value + duties paid (for imports). The health tax is 2% of the excise tax, and the television tax is 1.5% of the excise tax. The local provincial tax is a fixed amount (1.86 baht per pack), and marketing costs could be computed by DG Excise either as a residual or from direct market information. VAT is the value-added tax collected from resellers on domestically produced or imported cigarettes, and is 7% of the associated price (ex-factory or cif, respectively).
It is evident from these expressions that the MRSP on domestic cigarettes is likely to be lower than that on imported brands in the same price category because the former calculation does not include customs duties paid. Further, when Customs assigns a high deductive value to the cif value the relative increase carries through via the excise tax and other tax components. There is a further degree of freedom for DG Excise in the designation of marketing costs for domestic and imported brands. Indeed, the Panel Report noted that the MRSPs were commonly higher for imports than for competing TTM goods, including during the period of deductive valuation. In turn, the absolute value of VAT paid on imports was higher, due to this higher tax basis, which the Panel Report found inappropriate.
Despite that fact, Thailand argued that there was no effective discrimination because the ad valorem VAT rates were the same on both imported and domestic goods. Moreover, the initial VAT payment could be deducted by resellers at each stage (wholesalers, retailers) as they collected VAT at those stages. Ultimately, final consumers would pay the tax, which was assessed at 7% on all cigarettes, despite the different tax bases. The problem here, as discussed above, was the significant difference in administrative burdens. Resellers of TTM products were exempt from collecting and claiming VAT and, therefore, did not have to keep the corresponding books and tax records. Resellers of imported cigarettes, however, had to engage in such accounting, with monthly reporting to the government, a burdensome element in terms of time costs at least.
Thailand's justification for this difference is that domestic cigarettes are produced by a government-owned monopoly and the initial VAT would automatically be posted as government revenue. There is no possibility of evading such a tax (in principle). However, because imports were sold by private firms across borders, it was thought that tax evasion was more likely. Thus, Thailand's Directorate General of Excise essentially argued that the difference in record keeping, despite its burdensome nature, was justifiable as a necessary policy to safeguard internal taxation under Article XX(d). As discussed earlier, both the Panel and Appellate Body rejected this claim.
4.2 Economic issues
We turn next to two interesting economic issues this situation raises about which more may be said.
1. Conditions of competition
First, consider the question of how the policy scenario described above would affect market competition. One way to think about this is to trace through impacts in retail markets, assuming that domestic brands and imported brands are close, but imperfect, substitutes within similar quality bands. For now, set aside how the domestic manufacturer (TTM) sets its ex-factory price and the importer (PMT) declares its cif value. As noted above, these prices, along with added taxes and costs, determine the MRSPs, which are the VAT tax bases. Retailers may sell domestic, imported, or both brands. Assume (strongly) that retailers are able to order the quantities that would clear demand at these MRSPs. (Recall that the MRSP for domestic brands are essentially equivalent to retail prices in all cases. This is generally true for imports as well, although retailers may choose to sell at a lower price, even if the MRSP remains fixed.) Finally, assume initially that resellers face the same (zero) reporting requirements for VAT on both goods.
In this simple framework, the following results may be readily shown in competition between an imported brand and a domestic brand. First, in any initial equilibrium the MRSP is higher for the import due to the differentially higher add-on factors noted above. At this point, suppose Customs denies the declared cif value for PMT and imposes a higher value. This will carry through into a higher MRSP and VAT liability for that product, essentially an increase in costs of retailing it. The quantity sold will fall at the higher MRSP, reducing PMT's profit. Further, this higher price for the import will raise demand for TTM's brand at the given MRSP, expanding sales and raising TTM's profit. Second, the extent of these impacts depends on market parameters, including own-price elasticities and the cross-price elasticity of the two brands. For example, if we assume a high cross-elasticity (significant demand shift) the derived demand increase for TTM would be relatively large. Third, welfare effects would depend on these elasticities, initial market shares, and the VAT tax rate.Footnote 61 Welfare impacts in Thailand would depend on the loss of consumer benefits in the import segment in comparison with the higher TTM profits and the net change in tax revenues, which could be positive or negative.
Now consider some complications to this basic story. First, although the changes in customs values (setting cif value above the declared value) ordered by Thai Customs happened only during the time period in question, in principle a Customs authority could act intermittently over time, generating uncertainty in the importer's cif value without similar uncertainty in the domestic ex-factory price, a possibility exacerbated by the lack of transparency in setting the higher import price. Similarly, decisions by DG Excise to use constructed marketing costs for imports but a residual balancing item for TTM raises more cost uncertainty. In this context, an intermittent uncertainty would appear in retail prices (MRSPs) for resellers of imported goods. However, the uncertainty would be one-way, in that it seems unlikely that Customs would choose to reduce cif value given its resort to deductive computations. To the extent that consumers prefer to avoid such uncertainty, we may anticipate a reduced average demand for the PMT product and, again, a rise in demand for the TTM product. Indeed, it is conceivable (depending on parameters) that for some imported brands this uncertainty would reduce demand sufficiently to deter some resellers from carrying them. An interesting feature here is that, in order to safeguard its tax collections, DG Excise is generally unwilling to permit a reduction in the MRSP (which is based on the cif value). Of course, PMT could attempt to offset the reduction in demand by encouraging its resellers to lower retail price, suggesting that the potential for ‘pass through’ is an interesting issue here and perhaps should be considered in any future competitive analysis of similar disputes regarding unequal treatment. Even if it did so, however, the MRSP would remain at a higher level, and consumers would pay VAT on an artificially high price. In this context, both the uncertainty in customs and the policy-based stickiness in the MRSPs seem anticompetitive.
Second, Thailand might still argue that the procedures considered to this point are not discriminatory in the law, as opposed to outcomes. However, the key element of national treatment in this case is the exemption from bookkeeping and reporting requirements for sellers of domestic brands. Sellers of imported brands are not given the same advantage and, as a result, face both these administrative costs and the potential for undergoing audits and paying fines if they report inadequately. Despite the fact that resellers pass along the VAT on imported cigarettes to consumers, these relatively higher costs and risks presumably diminish the willingness of some resellers to carry imported brands. In economic terms, the supply curves of import resellers would shift upward, raising price and cutting the derived demand for imports. It would also expand demand for the domestic competing brands. Note also that resellers may well transact in both domestic and imported goods. Thailand argued that, for these firms, the need to fill out forms for imported brands meant they were saddled with the same costs, implying they would not limit their foreign sales. This seems a dubious claim on its face. It is probably noteworthy that essentially all large retailers sell both imports and domestic products; it is only small retailers that opt out of the foreign brands.
Third, it is surely relevant for competition that the domestic manufacturer, TTM, is a government-owned monopoly. In that context, thinking of TTM and PMT as passive suppliers could be highly misleading. Rather, what matters is how they compete strategically against each other. There is little in the Panel Report to shed light on this issue, so we are left to speculate. The fact that TTM sets an ex-factory price and PMT reports a declared cif value might support a model of price competition. In that case, the cost-increasing factors on the import side discussed here would effectively shift upward the reaction curve of the importer in price space. The result would be higher prices and lower outputs of both goods. However, the impacts would be smaller for the domestic monopolist, raising its market share and profits. Profits of the importer would likely fall. At the retail level, both MRSPs (and retail prices) would be pushed up and consumers would be worse off. Competition would be diminished.
It is possible, however, that these firms compete in quantities. In that event, these policies would shift the foreign reaction curve downward, cutting import sales but expanding sales of the domestic monopolist. Welfare calculations would be more complicated but qualitatively like those in the basic analysis above.
In all of these cases, the competitive conditions would seem to be changed by the policies of Thai Customs and DG Excise in favour of TTM, the domestic monopoly. In that context, it is possible to agree in principle with the Panel and Appellate Body in their findings of inconsistency with Article III.2. Thailand's policies also do not comport with the requirement in Article III.4 that imported products shall be accorded treatment no less favourable than that accorded to like products of national origin.
All that said, return to the legal analysis offered earlier in the paper regarding the meaning of ‘no less favourable’ in Article III.4. That analysis pointed out that the Panel's rigorous interpretation of this principle is somewhat at odds with the finding in Korea–Beef that a measure must actually alter the conditions of competition to be a violation. Jurisprudence under the TBT agreement also points to the need for careful scrutiny of the circumstances of individual cases, at least where there is de facto discrimination, to identify modifications of these conditions. If these standards were to be applied, a careful economic and market analysis would be needed to score Thailand's policies, rather than having them rejected directly on grounds of differences in treatment. In this context, both the VAT calculations and the differential reporting requirements presumably generated both de jure and de facto discrimination. The latter could be disputed on grounds of zero net VAT liability for import resellers and the fact that many retailers sell both domestic and imported cigarettes. However, while the economic factors listed above point to the likelihood of favouritism on behalf of TTM's products, it is ultimately an empirical question.
2. The Article XX(d) claim
Working somewhat against the straightforward conclusion that Thailand's policies were discriminatory and ran afoul of Articles III.2 and III.4 is that country's claim that it imposed these differentially higher VAT reporting requirements on resellers of imported cigarettes in order to avoid tax evasion, and this policy was acceptable under the general-exceptions clause. It is worth reproducing XX(d):
Subject to the requirement that such measures are not applied in a manner which would constitute a means of arbitrary or unjustifiable discrimination between countries where the same conditions prevail, or a disguised restriction on international trade, nothing in this Agreement shall be construed to prevent the adoption or enforcement by any contracting party of measures:
(d) necessary to secure compliance with laws or regulations which are not inconsistent with the provisions of this Agreement, including those relating to customs enforcement, the enforcement of monopolies operated under paragraph 4 of Article II and Article XVII, the protection of patents, trademarks and copyrights, and the prevention of deceptive practices.
Thailand in essence argued that resellers of imported goods had to be required to account for VAT liability and reporting requirements in order to ensure compliance of imported cigarettes with the tax laws. In contrast, because TTM is a government agency (indeed, a component of the Ministry of Finance) there is no question about VAT collection on its sales.
The Panel and Appellate Body took a straightforward view of this matter. In this regard, it was not ‘necessary’ to have differential reporting requirements to safeguard tax compliance. One obvious alternative to achieve this would be a nondiscriminatory reporting requirement applied to resellers of all cigarettes. Further, as discussed earlier, since the policies in question were ruled inconsistent with GATT rules, they would seem to fail to qualify for XX(d) protection. It is hard to disagree with this interpretation.
At the same time, one might wonder whether a strict national-treatment principle serves as a straightjacket here. A literature has emerged recently on the economics of national treatment (NT). As Saggi and Sera point out, there are conditions under which NT can be harmful to at least one country if products are heterogeneous in quality and markets differ in size. Arguably, those factors are not much in play in the present case, though a careful study of quality differentiation in the cigarette market might find otherwise.Footnote 62
More relevant, however, is Horn's observation that NT constrains national-policy sovereignty in cases where imports generate a larger negative externality than does domestic production.Footnote 63 For example, if consumption of imported goods generates greater external costs, such as environmental damage or loss of biodiversity, a country might reasonably prefer to establish a higher consumption tax on it than on the domestic competing good. It is prevented from doing so and, indeed, the ‘no less favourable treatment’ provision permits only differentiating policy with a lower (not higher) tax on imports.
Arguably, this kind of situation may exist in Thai–Cigarettes, though one has to think through it carefully. There is no obvious reason why smoking imported cigarettes generates more health damage than does smoking domestic brands. In this case, the relatively higher VAT burden on imports in Thailand surely fails Horn's test for relief. However, suppose we conceive of tax evasion on imports as a negative externality, perhaps because it shifts the tax burden onto consumers of domestic goods. In the circumstances of this case, it seems evident that tax evasion is impossible for TTM but possible for imports, perhaps through an invalid declared customs valuation. In this context, a higher ‘tax’ (in the form of reporting requirements) on imports may, indeed, solve this problem, while a similar ‘tax’ on domestic brands would be just a nuisance.
In the present case, we agree with the Appellate Body that this situation does not satisfy the conditions of Article XX(d), because the differential administrative burden is not necessary to resolve the problem and could be construed as a disguised restriction on trade. Moreover, the CVA already offers a useful tool for dealing with evasion through customs valuation, if not other methods. At the same time, there is room for debate about whether future dispute-settlement cases might recast Article XX as a means of getting beyond the restrictions posed by NT in episodes where a differential tax or treatment of imports really would be necessary to correct a trade-related market externality. At the same time, a word of caution is in order because such an interpretation presumably would run the risk of inviting countries to label an unacceptably wide range of discriminatory policies as needed treatments for external costs.
5. Concluding remarks
On its face, Thai–Cigarettes is a straightforward case of discrimination against like imported products arising from differential means of calculating the basis for the VAT tax and the additional and burdensome reporting requirements imposed on resellers of imported cigarettes. Both the Panel and the Appellate Body took a strict approach to the meaning of discrimination in finding Thailand's VAT regime for this product a violation of both Articles III.2 and III.4. At the basic level, it is difficult to disagree with these findings. We registered the caution, however, that this rigour seems at odds with the more permissive approach to ‘less than favourable treatment’ underscored by other recent WTO jurisprudence. In this regard, the Panel might have been better advised to undertake a fuller economic and market analysis to determine whether, in fact, the differential rules really did have a noticeable impact on conditions of competition.
