an analysis on Japan's tax on alcohol beverages, Essays (university) of International Law

Download an analysis on Japan's tax on alcohol beverages and more Essays (university) International Law in PDF only on Docsity! STRICTED WORLD TRADE WT/DSSR 'WT/DS10/R WT/DS11/R ORGANIZATION 11 July 1996 (96-2651) Original: English JAPAN - TAXES ON ALCOHOLIC BEVERAGES Report of the Panel  L INTRODUCTION 11 On21 June 1995, the European Communities (“the Community”) requested consultations with Japan under Article XXII of the General Agreement on Tariffs and Trade 1994 (“GATT”) concerning the internal taxes levied by Japan on certain alcoholic beverages pursuant to the Japan’s Liquor Tax Law (WT/DS8/1). Cn 7 July 1995, pursuant to Article 4.11 of the Understanding on Rules and Procedures Governing the Settlement of Disputes (“DSU”), the United States (WT/DS8/2) and Canada {WT/DS8/3) requested to be joined in these consultations. Japan accepted these requests on 19July 1995 {WT/DS8/4). 1.2 On7 July 1995, Canada requested consultations with Fapan under Article XXII of GATT 1994 concerning certain Japanese liquor taxation laws (WT/DS10/1), On 17 July 1995, pursuant to Article 4.11 of the DSU, the United States (WT/DS10/2) and the Community (WT/DS10/3) requested to be joined in these consultations. Japan accepted these requests on 19 July 1995 (WT/DS10/4), 13 On 7 July 1995, the United States requested consultations with Japan under Article XXIII of GATT 1994 regarding internal taxes imposed by Japan on certain alcoholic beverages pursuant to the Liquor Tax Law (WT/DS11/1). 14 On20July 1995, the Community, Canada and the United States jointly held consultations with Japan with a view to reaching a mutually satisfactory resolution of the matter, but they were unable to reach such a resolution. On 21 July 1995, the United States and Japan consulted under Article XXHI:1, but they did not reach a mutually acceptable resolution of the matter. 1.5 On 14 September 1995, pursuant to Article XXIII:2 of GATT 1994 and Article 6 of the DSU, the Community requested the Dispute Settlement Body (“ DSB”) to establish a panel with standard terms of reference (WT/DS8/5). The Community claimed that: “a) Japan had acted inconsistently with Article III:2, first sentence, of GATT 1994 by applying a higher tax rate on the category of ‘spirits’ than on each of the two sub- categories of shochu, thereby nullifying or impairing the benefits accrued to the European Communities under GATT 1994; and that b) Japan has acted inconsistently with Article ILI:2, second sentence, of GATT 1994 by applying a higher tax rate on the category of ‘whisky/brandy’! and on the category of ‘liqueurs’ than on each of the two sub-categories of shochu, thereby nullifying or impairing the benefits accrued to the European Communities under GATT 1994. In the event that the liquors falling within the category of ‘spirits’ were found by the Panel not to de ‘like products’ to shochu within the meaning of the first sentence of Article I1I:2, the [Community] further claimed that: ¢) Japan has acted inconsistently with Article III:2, second sentence, of GATT 1994 by applying a higher tax rate on the category of ‘spirits’ than on each of the two sub-categories of shochu, thereby nullifying or impairing the benefits accrued to the European Communities under GATT 1994”. 1.6 On 14 September 1995, pursuant to Article XXIII of GATT 1994 and Articles 4 and 6 of the DSU, Canada requested the DSB to establish a panel with standard terms of reference (WT/DS10/5). Canada claimed that: ‘In the present Panel report the use of the term “whisky” includes also the term “whiskey” as used in the case of Irish whiskey and Tennessee whiskey. WT/DS3/R. WT/DS1O/R WT/DSLIU/R Page 2 “',., the higher rates of taxation on imported alcoholic beverages including whiskies, brandies, other distilled alcoholic beverages and liqueurs than on Japanese shochu imposed pursuant to the Japanese Liquor Tax Law are: a) inconsistent with Article IIE:1 and III:2 of GATT 1994; b) nullifying and impairing the benefits accruing to Canada pursuant to the WTO”. 1.7 On 14 September 1995, pursuant to Article XXIII:2 of GATT 1994 and Articles 4 and 6 of the DSU, the United States requested the DSB to establish a panel with standard terms of reference (WT/DS11/2). The United States claimed that: “ ,., the internal taxes imposed by Japan [pursuant to the Liquor Tax Law] on these beverages, and in particular the preferential tax treatment accorded to shochu, are inconsistent with Article III of GATT 1994, and otherwise nullify and impair benefits accruing to the United States under the GATT 1994”, 1.8 At its meeting of 27 September 1995, pursuant to the first request of the three complaining parties and with Japan’s acceptance, the DSB established a single panel with the mandate to examine the requests of the Community, Canada and the United States, all of which related to the same matter, in accordance with Article 9 of the DSU (WT/DSB/M/7). 19 During the 27 September 1995 meeting of the DSB, Norway reserved its right as a third party to the present dispute. However, on 7 November 1995, Norway informed the Panel of the withdrawal of its request to participate as a third party in the dispute (WT/DS8/7, D$10/7 and DS11/4). 1.10 At the same meeting of the DSB on 27 September 1995, the parties agreed that the Panel should have standard terms of reference as follows: “To examine, in the light of the relevant provisions of the covered agreements cited by the EC, Canada and US in documents WT/DS8/5, WT/DS10/5, WT/DS11/2, the matters referred to the DSB by the EC, Canada and the United States in those documents and to make such findings as will assist the DSB in making the recommendations or in giving the rulings provided for in those agreements”. 1.11 On 30 October 1995, the Panel was constituted with the following composition: Chairman: Mr. Hardeep Puri Panelists: Mr, Luzius Wasescha Mr. Hugh McPhail Il. FACTUAL ASPECTS A. The Japanese Liquor Tax Law 2.1 This dispute concerns the Japanese Liquot Tax Law (Shuzeiho}, Law No.6 of 1953 as amended (“Liquor Tax Law”), which lays down a system of internal taxes applicable to all liquors, which are defined as domestically produced or imported beverages having an alcohol content of not less than one degree and which are intended for consumption in Japan. WT/DS8/R WT/DS10/R WT/DS11/R Page 3 2.2 The Liquor Tax Law currently classifies the various types of alcoholic beverages into ten categories and additional sub-categories: sake, sake compound, shochu (group A, group B), mirin, beer, wine (wine, sweet wine), whisky/brandy, spirits, liqueurs, miscellaneous (various sub-categories), 1. Terminology and Definitions The Liquor Tax Law defines liquors involved in the present disputes - shochu, whisky/brandy, spirits and liqueurs - as follows:? “Article 3 Paragraph 5 ‘Shochu’ shall mean liquors produced by the distillation of alcohol containing substances. Included in this definition are those produced by adding water, sugar or other substances stipulated in government ordinances to the above-mentioned liquors. They must have an alcoholic strength of 45% vol or less. The liquor must be less than 36% vol in case distilled by a ‘continuous still", the definition of which is as follows: a machine that removes fusel oil, aldehyde and other impurities during the process of continuous distillation. The definition of the type of sugar which can be added is given by government ordinances. In case produced by adding substances other than water, the extract of the product ought to be less than 2g/100 ml. Note that chose enumerated below from (a) through (d) do not fall under the definition of ‘shochu’. ka) Liquors produced in whole or in part from malted cereals or fruit (including dried fruit or boiled-down or concentrated must, but excluding dates or other fruit as stipulated in government ordinances. The same shall apply hereafter). (b) Liquors produced by filtering it through white birch charcoal or other substances specified in government ordinances. {c) Liquors produced in whole or in part from saccharized substances (e.g. molasses, sugar, syrup and honey; excluding sugar as defined by government ordinances) and by the distillation at less than 95% vol. @ Liquors produced by flavouring alcohol by way of steeping ingredients of other substances during distillation. Paragraph 9 “Whisky/Brandy’ shall mean the following tiquors on condition that those listed in (a}, (b) and (d) be excluded in case covered by (b) through (d) of Paragraph 5: (a) Liquors produced by distillation of alcohol containing substance derived by first saccharifying malted cereals and water and then fermenting them.The above- mentioned liquors must be distilled at less than 95% vol. {b) Liquors produced by the distillation of alcohol containing substance derived by first saccharifying unmalted cereals with malted cereals and water and then fermenting them. The above mentioned liquors must be distilled at less than 95% vol. *These definitions (wanslations from the Liquor Tax Law) were submitted by Japan. WT/DS8/R WT/DS10/R WT/DS11/R Page 6 Spirits Alcoholic Strength Tax Rate (per | kitolitre) (1) below 38 degrees ¥367,300 (2) 38 degrees and above ¥367,300 plus ¥9,930 for each degree above 37 Liqueurs Alcoholic Strength Tax Rate (per 1 kilolitre) (1) below 13 degrees ¥98,600 (2) 13 degrees and above ¥98,600 plus ¥8,220 for each degree over 12 2.4 A special formula is applied to determine the rate applicable to beverages having an alcohol content below 13 per cent or, in the case of “liqueurs”, below 12 per cent (as a general rule, pre-mixes combining a liquor with water or with other non-alcoholic beverages). This formula yields the result that the tax rate per litre of pure alcohol levied on these beverages is the same as the tax per litre of pure alcohol that would be borne by a liquor of the same category at the legal standard strength. B. The 1987 Panel Report on Japan - Customs Duties, Taxes and Labelling Practices on Imported Wines and Alcoholic Beverages (1987 Panel Report”) 2.5 In 1986, the Community requested consultations with Japan in respect of Japan’s Liquor Tax Law, as it then existed. The consultations failed to resolve the matter and in 1987 a panel was established to consider, among others, the Community’s claim that the Liquor Tax Law violated Article II[:2. 2.6 As of 1987, the Liquor Tax Law divided the whisky/brandy category into whisky and brandy, and subdivided whisky into three grades, i.c., Special Grade, First Grade and Second Grade. The category shochu was sub-divided into Groups A and B. Specific tax rates were provided for each category and sub-category of alcoholic beverages. In addition, an ad valorem tax was applicable to inter alia, Special, First and Second Grade whiskies where the price exceeded a certain threshold. This tax was not applicable to either shochu group. 2.7 The 1987 Panel Report concluded that some aspects of the Liquor Tax Law were inconsistent with Article III:2, first and second sentences, and suggested that the CONTRACTING PARTIES recommend that Japan bring its taxes on whiskies, brandies, other distilled spirits (such as gin and vodka), liqueurs, still wines and sparkling wines into conformity with its obligations under the General Agreement. In particular, the Panel reached the following conclusions: “5.5... The Panel concluded that the ordinary meaning of Article III:2 in its context and in the light of its object and purpose supported the past GATT practice of examining the conformity of internal taxes with Article III:2 by determining, firstly, whether the taxed imported and domestic products are ‘like’ or ‘directly competitive or substitutable’ and, secondly whether the taxation is discriminatory (first sentence) or WT/DS8/R WT/DS10/R WT/DS11/R Page 7 protective (second sentence of Article III:2), The Panel decided to proceed accordingly also in this case. 5.6 .-. The Panel found that the following alcoholic beverages should be considered as “like products” in terms of Article II:2 in view of their similar properties, end-uses and usually uniform classification in tariff nomenclatures: imported and Japanese-made gin; imported and Japanese-made vodka; imported and Japanese-made whisky (including all grades classified as ‘whisky’ in the Japanese Liquor Tax Law) and ‘spirits similar to whisky in colour, flavour and other properties’ as described in the Japanese Liquor Tax Law; imported and Japanese-made grape brandy (including all grades classified as ‘brandy’ in the Japanese Liquor Tax Law); imported and Japanese-made fruit brandy Gncluding all grades classified as ‘brandy’ in the Japanese Liquor Tax Law); imported and Japanese-made ‘classic’ liqueurs (not including, for instance, medicinal liqueurs); imported and Japanese-made unsweetened still wine; imported and Japanese-made Sparkling wines. 5.7 The Panel did not exclude that also other alcoholic beverages could be considered as ‘like’ products. Thus, even though the Panel was of the view that the ‘fikeness’ of products must be examined taking into account not only objective criteria {such as composition and manufacturing processes of products) but also the more subjective consumers’ viewpoint (such as consumption and use by consumers), the Panel agreed with the arguments submitted to it by the European Communities, Finland and the United States that Japanese shochu (Group A) and vodka could be considered as ‘like’ products in terms of Article III:2 because they were both white/clean spirits, made of similar raw materials, and their end-uses were virtually identical (either as straight ‘schnaps’ type of drinks or in various mixtures), Since consumer habits are variable in time and space and the aim of Article [II:2 of ensuring neutrality of internal taxation as regards competition between imported and domestic like products could not be achieved if differential taxes could be used to crystallize consumer preferences for traditional domestic products, the Panel found that the traditional Japanese consumer habits with regard to shochu provided no reason for not considering vodka to bea “like” product. The Panel decided not to examine the ‘likeness’ of alcoholic beverages beyond the requests specified in the complaint by the European Communities (see ... ). The Panel felt justified in doing so also for the following reasons: Alcoholic drinks might be drunk straight, with water, or as mixes. Even if imported alcoholic beverages (e.g. vodka) were not considered to be ‘tike’ to Japanese alcoholic beverages (e.g. shochu Group A), the flexibility in the use of alcoholic drinks and their common characteristics often offered an alternative choice for consumers leading to a competitive relationship. In the view of the Panel there existed - even if not necessarily in respect of all the economic uses to which the product may be put - direct competition or substitutability among the various distilled liquors, among various liqueurs, among unsweetened and sweetened wines, and among sparkling wines. The increasing imports of “Western-style” alcoholic beverages into Japan bore witness to this lasting competitive relationship and to the potential products substitution through trade among various alcoholic beverages. Since consumer habits vis-a-vis these products varied in response to their respective prices, their availability through trade and their other competitive intez-relationships, the Panel concluded that the following alcoholic beverages could be considered to be ‘directly competitive or substitutable products’ in terms of Article III:2, second sentence: WT/DS8/R WT/DS10/—R WT/DS11R Page 8 - imported and Japanese-made distilled liquors, including all grades of whiskies/brandies, vodka and shochu Groups A and B, among each other; - imported and Japanese-made liqueurs among each other; - imported and Japanese-made unsweetened and sweetened wines among each other; and - imported and Japanese-made sparkling wines among each other. 5.9 a)... The Panel concluded ... that (special and first grade) whiskies/brandies imported from the EEC were subject to internal Japanese taxes ‘in excess of those applied ... to like domestic products’ (i.e. first and second grade whiskies/brandies) in the sense of Article III:2, first sentence. b) .... The Panel concluded ... that ... the imposition of ad valorem taxes on wines, spirits and liqueurs imported from the EEC, which are considerably higher than the specific taxes on ‘like’ domestic wines, spirits and liqueurs, was inconsistent with Article III:2, first sentence. d ... The Panel concluded that this imposition of higher taxes on ‘classic’ liqueurs and sparkling wines with higher raw material content was inconsistent with Article III:2, first sentence. 5.11 The Panel recalled its findings that distilled liquors, including all grades of shochu types A and B, were ‘directly competitive or substitutable products’ in terms of the interpretative note to Article IIL:2 (see above paragraph 5,7). The Panel noted that shochu was not subject to ad valorem taxes and that the specific tax rates on shochu were many times lower than the specific tax rates on whiskies, brandies and other spirits. The Panel noted that, whereas under the first sentence of Article III:2 the tax on the imported product and the tax on the like domestic product had to be equal in effect, Article III:1 and 2, second sentence, prohibited only the application of internal taxes to imported or domestic products in a manner ‘so as to afford protection to domestic production’. The Panel was of the view that also small tax differences could influence the competitive relationship between directly competing distilled liquors, but the existence of protective taxation could be established only in the light of the particular circumstances of each case and there could be a de minimis level below which a tax difference ceased to have the protective effect prohibited by Article MII:2, second sentence. The Panel found that the following factors were sufficient evidence of fiscal distortions of the competitive relationship between imported distilled liquors and domestic shochu affording protection to the domestic production of shochu: - the considerably lower specific tax rates on shochu than on imported whiskies, brandies and other spirits ... ; WT/DS3/R WT/DS10/R WT/DSIUR Page 11 3.3 The United States claimed that the Japanese tax system applicable to distilled spirits has been devised so as to afford protection to production of shochu. For this reason and because “white spirits” and “brown spirits” have similar physical characteristics and end-uses, the United States claimed that “white spirits” and “brown spirits” are “like products” in the sense of the first sentence of Article III:2, and therefore the difference in tax treatment between shochu and vodka, rum, gin, other “white spirits”, whisky/brandy and other “brown spirits” is inconsistent with Article III:2, first sentence. If the Panel were not able to make such a finding, the United States requested, in the alternative, that the Panel find that all “white spirits” are “like products” in terms of Article III:2 first sentence, and that all distilled spirits are “directly competitive and substitutable” in terms of Article III:2, second sentence for the same reasons. The United States concluded that irrespective of the legal analysis the Panel adopts, the Liquor Tax Law should be found to be inconsistent with Article III:2. 3.4 — Thedefending party, Japan, responded to the claims from the three complaining parties. Japan claimed that the purpose of the tax classification under the Liquor Tax Law is not to afford protection and does not have the effect of protecting domestic production. Therefore, Japan argued that the Liquor Tax Law does not violate Article HI:2. According to Japan, spirits, whisky/brandy and liqueurs are not “like products” to either category of shochu, within the meaning of Article II:2, first sentence, nor are they “directly competitive and substitutable products” to shochu, within the meaning of Article I1:2, second sentence. Consequently, Japan claimed that the Liquor Tax Law cannot violate Article II:2. Iv. ARGUMENTS OF THE PARTIES A. Preliminary Objection of Japan 41 The United States requested the Panel to declare that the reduction in excise taxes for small- volume producers, contained in the 1989 legislation®, discriminates on its face against imported shochu, sake and wine and therefore violates Article III:2, first sentence. The operative provision of the 1989 legislation is phrased in terms of “shipment from the factory”, terms which, under the Liquor Tax Law, refer exclusively to factories in Japan. These measures apply to producers of such alcoholic beverages whose taxable shipments from the factory do not exceed 1300 kilolitres in a given fiscal year; in the following fiscal year, such producers are entitled to a 30 per cent reduction in the excise tax otherwise due on their first 200 kilolitre shipped from the factory. This provision, in the US view, was provided as compensation to small domestic producers for the 1988 law reducing the other protection afforded by the tax differential between these alcoholic beverages and competing imports. In the 1994 amendment of the Liquor Tax Law, the tax relief measures for small producers of shochu A, shochu B, Tefined sake and wines were extended, so that the entire time period covered by the tax retief measures is 1 April 1988 through 31 March 1997, 4.2 In support of its claim, the United States referred the Panel to the Panel Report on “United States - Measures Affecting Alcoholic and Malt Beverages” (“1992 Malt Beverages”),° whichexamined a similar provision which provided for an excise tax on beer of $18 per barrel in general, but for a reduced tax of $7 per barrel for the first 60,000 barrels produced by US breweries with annual production of not more than 2 million barrels. Inthe 1992 Malt Beverages case, the panel found that the application of a lower rate of excise tax on beer, which was not available in the case of imported beer, constituted SArticle 87 of the Taxation Special Measure Law, Law No. 26 of 1957, as amended; provision applying to producers of refined sake, shochu A, shochu B or certain types of wine. “Panel report adopted on 19 June 1992, BISD 39S/208, para. 2.7. WT/DS8/R WT/DS10/R WT/DS11/R Page 12 less favourable treatment of the imported product in respect of internal taxes and was therefore inconsistent with Article IH:2, first sentence. The United States concluded that even if the present Panel were not to find that the tax reduction programme was inconsistent with the first sentence of Article I[I:2 because of the preliminary objection of Japan, this measure was a fact and must be taken into account in the Panel’s evaluation of tax/price ratios and in its evaluation of protectionist purpose and effect. This measure was a significant element of the Japanese tax policy with regard to shochu and provided important evidence of the essentially protectionist nature and effect of that policy. 43 Japan responded that WTO Members who request consultations in accordance with the provisions of Article 4.4 of the DSU must identify the “measure(s)” at issue. However, the request of the United States did not mention the specific measure for small-volume producers described in paragraph 4.1 above. Nor did the United States raise this issue with Japan in the course of the bilateral consultations. The lack of identification of this measure in the request for consultations, as well as the failure to raise the issue in the bilateral consultations, constituted a case of non-compliance with Article 4 of the DSU. Japan submitted further that Article 6.2 of the DSU requires that the request for the establishment of a pane! must identify the specific measures at issue. The lack of identification of this measure in the US request for the establishment of the present Panel also constituted failure to abide by the DSU. Japan noted also that it is a GATT practice that a panel would not render judgment on matters not taised during consultations or not included in the request to establish the panel. This practice, inJapan’s view, should not be ignored because it is intended to encourage the parties to a dispute to attempt to obtain satisfactory adjustment before having recourse to the panel procedure and because it is important also for the interests that third parties may have in the dispute. Japan, therefore, requested the Panel to consider the interim measure under the Special Taxation Measures Law as being beyond the terms of reference of the present Panel. 4.4 Incase the Panel decides to rule on this issue, Japan submitted that the Panel should find the measure to be a temporary measure to ensure compliance with the recommendations of the 1987 Panel Report, and therefore consistent with Japan’s obligations under GATT 1994, Historically, this interim measure was introduced as part of the overall package for compliance with the recommendations of the 1987 Panel Report. It has served to alleviate adjustment costs felt by small scale manufacturers. The measure was extended in 1994 in order to facilitate the amendment which removed distortions arising out of price changes since the 1989 amendment. In other words, Japan suggested that this interim measure is part of the process to ensure and maintain compliance with Article IIT, and has no distortional effect or protective intent. In response to the reference by the United States to the 1992 Malt Beverages report, Japan responded that in that case, the federal and state measures found inconsistent with Article III were permanent measures, inherently distortional to trade. Japan noted, however, that most of these measures are still in place. Similarly, Japan noted that the Community has a permanent mechanism to reduce the tax rate applicable to small breweries. By virtue of this mechanism, Austria, Belgium, Denmark, Finland, Germany, Luxembourg and the Netherlands apply a reduced tax rate to small breweries, and Austria, Germany and Spain give a tax benefit to small manufacturers of distilled liquors. The United States noted in response that Japan’s measure was hardly “interim” at this point, and had been routinely renewed in 1994: there were no guarantees that it would not be renewed repeatedly into the indefinite future. B. The Legai Value of the 1987 Panel Report 4.5 For the Community, since the specific tax rates applied to shochu have remained at a much lower level than the rates applicable to other “like” or “directly competitive or substitutable” products, Japan has failed to implement in full the recommendations of the 1987 Panel Report. Consequently, WTIDS8/R WT/DS10/—R WT/DS11/R Page 13 the Community claimed that the Liquor Tax Law is inconsistent with Article III:2, first and second sentences, of GATT 1994 (worded identically to Article III:2 of GATT 1947). In view of the time already elapsed since the adoption of the 1987 Panel Report and of the regulatory changes implemented in the meantime, the Community argued that it was more appropriate to request the establishment of a new panel to examine and find that the Liquor Tax Law, as it stands now, is inconsistent with the obligations of Japan under GATT 1994 rather than requesting the full implementation of the 1987 Panel Report under GATT 1947. Nonetheless, to the extent that the claims contained in the Community Tequest reiterate some of the claims already examined by the 1987 Panel Report, the findings of that panel report should be accorded a special precedential value. According to the Community, the following findings of the 1987 Panel Report, in particular, should provide decisive guidance to this Panel: - vodka and shochu A are “like products” within the meaning of Article HI:2, first sentence. Other distilled spirits may also be “like products”; - all distilled spirits are “directly competitive and substitutable” among each other within the meaning of the Note to the second sentence of Article III:2; - the differences in taxation between shochu and the other distilled spirits afford protection to the Japanese production of shochy, 46 For Canada, Article 3.2 of the DSU makes clear that a panel “serves ... to clarify existing provisions of [covered] agreements”. In so clarifying these provisions, Canada considers that Article XVI:1 of the Agreement Establishing the WTO (“WTO Agreement”) provides clear guidance to apanel and the DSB respecting the legal value of reports adopted by the CONTRACTING PARTIES under GATT 1947. Article XVI:1 of the WTO Agreement provides: “Except as otherwise provided under this Agreement or the Multilateral Trade Agreements, the WTO shall be guided by the decisions, procedures and customary practices followed by the CONTRACTING PARTIES to GATT 1947 and the bodies established in the framework of GATT 1947.” For Canada, inthis case, neither the WTO Agreement nor the Multilateral Trade Agreements “otherwise provide”. Canada considers that there is a strong factual nexus between the 1987 Panel Report and the current dispute. Accordingly, for Canada, in determining the consistency of the Liquor Tax Law with Article III:2, second sentence, the Panel and the DSB are, at a minimum, to be directed by the 1987 Panel Report. In Canada’s view, the 1987 Panel Report is particularly authoritative in determining the consistency of the Liquor Tax Law with Article III:2 second sentence. In addition, in Canada’s view, it can be argued that since the 1987 Panel Report was adopted by the CONTRACTING PARTIES, it forms now an integral part of the WTO Agreement through Article 1(b)(iv) of GATT 1994.’ TArticle 1 of GATT 1994 reads as follows: “Ll. The General Agreement on Tariffs and Trade 1994 (‘GATT 1994’) shall consist of: fa) the provisions in the General Agreement on Tariffs and Trade, dated 30 October 1947, ... : (b) the provisions of the legal instruments set forth below that have entered into force under GATT 1947 before the date of entry into force of the WTO Agreement: @ protocols and certifications relating to tariff concessions; di Protocols of accession .... ; ii) decisions on waivers granted under Article XXV of GATT 1947 and still in force on the date of entry into force of the WTO Agreement; {iv) other decisions of the CONTRACTING PARTIES to GATT 1947; «} the Understandings set forth below: ...; and, @ the Marrakesh Protocol to GATT 1994.” WT/DS8/R WT/DS10/R WT/DS11/R Page 16 by the consumer use of shochu blended in various proportions with whisky, brandy or other drink” .”° Japan submitted that the first and second factors no longer exist in the current law and that the statements of the Panel in the third and fourth points were not based on facts. First, it was not true that shochu was produced almost exclusively in Japan. Nor is Japan the largest manufacturer of shochu. Shochu is widely produced in Southeast and East Asia. Second, consumers do not mix shochu with whisky or brandy under normal circumstances. Moreover, the increase in imports of Western-style liquors does not by itself prove the existence of substitutability, or the cross-price elasticity of demand, between shochu and other distilled liquors. For all these reasons, Japan argued that the present Panel cannot tely on the reasoning and conclusions of the 1987 Panel Report. 4.11 The Community submitted that it agreed with Japan’s views on the precedential value of panel reports. However, the Community argued that Japan had taken the position that subsequent panel reports should be analyzed on the basis of the fex posterior principle because panel reports are adopted by the CONTRACTING PARTIES. In the Community's view, this was a thoroughly misguided view of such decisions. For the Community, panel reports do not jose their character of decisions on individual cases because they have become part of the “acquis gattien” under Section | (b) (iv) of GATT 1994. Nor should panel reports lose this trait because they have been adopted by the CONTRACTING PARTIES. Moreover, as there is no rule of stare decisis as between panel reports, the present panel would not be bound by the 1992 Malt Beverages report in any case, Cc General Presentation of the Arguments of the Parties on Article HI:2 4.12 The Community argued that assessing compliance with Article II:2 imposes two different tests for the first and second sentences. Under the first sentence of Article III:2, the examination of the conformity of a system of internal taxation involves a two-step analysis: first, it must be determined whether the domestic product and the imported product are “like”, having regard to their physical characteristics and end-uses; and second, it must be ascertained whether the imported product is subject to internal taxes in excess of those applied to the domestic product. For the Community, once it has been established that the two products are “like” and that the imported product is subject to higher taxes, a finding that Article III:2, first sentence, has been infringed is automatic. It is not necessary to show that the difference in taxation affords protection to the domestic product; under the first sentence, unlike the second sentence of Article III:2, such protective effect is irrefutably presumed in all cases. For the Community this was in accordance with the 1987 Panel Report. Whether or not the tax differential has a protectionist aim is also irrelevant. The examination of the aim of the measure only becomes relevant in order to determine whether the infringement of Article If:2, first sentence, may be justified under one of the general exceptions of GATT Article XX. 4.13 In this context and in response to the suggestion that an aim-and-effect test is to be favoured under Article III, the Community responded that the aim-and-effect test is inconsistent with the ordinary meaning of the words and the specific purpose of Article III:2 and with the general objectives of Article III. Nor does the aim-and-effect test incorporate any proportionality requirement. Moreover, the burden of proof, which in GATT Article XX and the WTO Agreement on Technical Barriers to Trade (“TBT Agreement”) lies with the respondent, is subtly but effectively shifted to the complainant, Given the extreme difficulty of positively proving a protectionist purpose (as opposed to refuting the 61987 Panel Report, para. 5.11. WT/DS3/R WT/DS10/R WT/DS11/—R Page 17 proof that a measure does not have a specific non-protectionist purpose), the mere invocation by a defendant of a non protectionist purpose may suffice in practice to exclude the application of Article TI:2. In this context, and unlike situations which would be dealt with through exceptions under Article XX, the aim-and-effect test would confront panels with the situation of having to adjudicate a dispute without a clearly defined standard of review. In response to the allegation that such a strict reading of the first sentence of Article I:2 would result in legitimate, non-protectionist policies being found inconsistent with Article ITE:2, the Community argued that it may be possible to introduce two kinds of flexibility into the interpretation of Article HI:2. The first flexibility is formed by the notion of discrimination in the case of graduated systems of taxation of like products. This form of flexibility seems to have been envisaged by the 1987 Panel Report. The second flexibility is formed by the dividing line between like products and directly competitive and substitutable products, These two flexibilities, together with the general exceptions of Article XX, may offer sufficient scope to deal adequately with all situations covered by Article [11:2, first sentence. In the present case, no such reference to any of the two flexibilities is necessary. The Community concluded that spirits are like products to shochu A and B, and since the Liquor Tax Law applies a higher tax rate on the category of spirits than on each of the two like products, it automatically violates GATT Article III:2, first sentence. 4.14 For the Community, the application of the second sentence of Article III:2 involves a different test: first, if must be determined whether the imported product and the domestic product are “substitutable or directly competitive” in the light not only of their physical characteristics and end-uses, but also of other criteria such as their cross-price elasticity, their availability in the same trade channels, etc., determined in an ex-post manner {i.e., by taking into account not only actual competition but also potential competition); second, it must be established whether an internal tax is applied “so as to afford protection to domestic production”, which does not involve any application of the so-called aim-and-effect test. The Community considers that the aim-and-effect test is not warranted by the ordinary meaning of the words “so as to afford protection”, and is contrary to the purpose of Article [11:2 and, more generally, to the basic objective of Article III. For the Community, the existence of a protectionist purpose is never a necessary condition for a finding that Article III:2 has been infringed. The aim of the measure only becomes relevant under the first sentence of Article III:2, in order to determine whether it may be justified under Article XX. For the Community, the protectionist effect of ameasure is the only relevant criteria for assessing whether the measure is “so as to afford protection”. A protectionist purpose only becomes relevant to the extent that it may provide an indication of protectionist effect. The Community therefore concluded that (1) shochu and other distilled spirits are directly competitive and substitutable products, and (2) since Japan does not dispute that the tax rates on shochu Aand shochu B are much lower than the rates on spirits, whisky/brandy and liqueur, in terms of taxation per litre of beverage and of taxation per litre of pure alcohol, and since a comparison of tax/price ratios is irrelevant and not a proof of tax neutrality, then the Liquor Tax Law has the effect of affording protection to shochu and therefore is inconsistent with the second sentence of Article HI:2. 4.15 Canada considers that at its core, Article 1I:2, second sentence, is designed to protect the competitive relationship between imported and domestic products. In Canada’s view, this is made clear in the “United States - Taxes on Petroleum and Certain Imported Substances” (“Superfund”)” report, the 1987 Panel Report and the 1992 Malt Beverages report. Canada considers that these reports make clear that an organizing principle in considering whether a measure affords protection to domestic production is whether the internal fiscal measure at issue distorts the competitive relationship between imported and domestic products. For Canada, Article IIL:2 second sentence, together with Article III:1 and the Interpretative Note ad Article Ill, Paragraph 2, set out four tests that must be met for a panel “Panel report adopted on 17 June 1987, BISD 348/136. WT/DS8/R WT/DS10/R WT/DSI1/R Page 18 to determine that in its application to Canadian whisky and shochu, the Liquor Tax Law is inconsistent with Article III:2, secondsentence. These four criteria are: (1) whether whisky is a “directly competitive or substitutable product” with shochu; (2) whether the taxes levied pursuant to the Liquor Tax Law are “internal taxes or other internal charges”; (3) whether whisky and shochu are “not similarly taxed”, even in using the tax/price ratio yardstick suggested by Japan; and (4) whether the taxes levied pursuant to the Liquor Tax Law afford protection to domestic production of shochu. 4,16 The United States began its discussion of Article III:2 with the observation that GATT panels had addressed taxes or regulations like those at issue in the present case, which are origin-neutral on their face, only on a very few occasions. In each such case, the analysis of whether a violation had occurred, had turned not on the tax’s trade effects, but, on whether the tax was targeting imports as such. GATT panel findings over the years had confirmed that the first sentence of Article III:2 requires that imports be provided equivalent opportunities; the second sentence addressed situations where the targeting of imports is more subtle. The United States emphasized that WTO Members retain the right to enact and maintain taxes and regulations that do not, on their face, discriminate against imports, and which classify similar products into distinct categories, but only if the categories are objective and based on neutral fiscal or regulatory policies. The central concern of Article III was the targeting of imports as such for special treatment. Therefore, in determining whether two products subject to different treatment are “like products”, it was necessary to consider whether such product differentiation is being made “so as to afford protection to domestic production” as provided in Article III:1. If the aim and effect, or purpose and effect, of the distinction was to target imports as such, the imported and domestic products should be deemed to be “like products” and a tax measure which treats the imported product less favourably than the domestic product would be inconsistent with Article III:2, first sentence. The United States further emphasized that this aim-and-effect test would apply only in the smali subset of Article III cases that involve origin-neutral legislation. The aim-and-effect test would have no application in cases that invoive measures that discriminate on the basis of origin. 4.17 The United States suggested that an examination of a measure’s aim or purpose should focus, inter alia, on the legislation’s wording and stated rationale, its preparatory work if any, statements by legislators, structural incentives, the treatment of products distinguished, the arbitrariness of distinctions drawn and ex ante knowledge that the distinction would discriminate between domestic and imported products. The United States argued that a measure could be said to have the aim of affording protection if analysis of the circumstances in which it was adopted, in particular an analysis of the instruments available to achieve the declared policy goal, demonstrated that a change in competitive opportunities in favour of domestic products was a desired outcome and not merely an incidental consequence of the pursuit of a legitimate policy goal. The examination of the effects should focus on the qualitative alteration of the conditions of competition, such as targeting of imports and evidence of cress-clasticity of demand between the favoured and disfavoured categories. The United States pointed out that Article IIE was one element of an interlocking system of obligations under GATT, the implications of which had been developed with great consistency in interpretations in panel reports of the previous ten years. This system included a number of basic propositions: a) the equivalence imparted to national treatment obligations under Article III:2 and Article III:4, which meant that a Member could not accomplish with regulation what it could not legally accomplish with taxation, or vice versa; 6) the equivalence between “like product” in Article III and in Article I:1; c) the ban on border adjustment of taxes (such as direct taxes) or regulations not imposed on a product as such; and d) the restrictive interpretation given to the exceptions contained in Article XX. Within the context of this system, the aim-and-effect test offered the only acceptable approach to the legal analysis of origin-neutral taxation or regulation. The present case, which was limited to Article III:2 and included no claims under Article LII:4 or Article XX, was the wrong occasion for a Panel to reopen and reconsider the aim-and-effect test and the complex of interlocking legal interpretations in this area. WT/DS8/R WT/DS1OR. WT/DS1I1R Page 21 4.22 The Community submitted that a system of taxation is “discriminatory” within the meaning of the first sentence of Article III:2 if, inter alia, the tax rate applied to the domestic product is lower than the tax rate applied to the like imported product. Any difference, however small, between the tax applied to the imported product and the tax applied to a domestic product violates Article III:2, first sentence, regardless of its effects on the volume of trade. Once it has been established that che two products are like and that the imported product is subject to higher taxes, a finding that Article III:2, first sentence, has been infringed is automatic. Unlike under the second sentence of Article III:2, it is not necessary to show that the difference in taxation affords protection to the domestic product because such protective effect is irrefutably presumed in all cases. In the Community’s view, whether the tax differential has a protectionist aim is also irrelevant. The first sentence of Article HI:2 thus lays down the evidentiary rule that tax discrimination between like products constitutes automatically a violation of Article II:2, first sentence. The examination of the aim of the measure only becomes relevant in order to determine whether the infringement of Article IH:2, first sentence, may be justified under any of the general exceptions of Article XX. As further argued hereafter in paragraph 4.36 and following, in the Community’s view the aim-and-effect test is inconsistent with the ordinary meaning of Article III:2, first sentence, and is contrary to the specific purpose of this provision as well as with the general objectives of Article III. Moreover, the aim-and-effect test effaces the clear textual difference between the first and the second sentences of Article III:2. The ordinary meaning of Article II1:2 is that the first and the second sentences of this provision set out different legal requirements and that only the second sentence refers to the first paragraph of Article III. This results clearly from the use of the word moreover at the beginning of the second sentence as well as from the Note ad Article IIT:2, which provides that: “A tax conforming to the requirements of the first sentence of paragraph 2 would be considered to be iticonsistent with the provisions of the second sentence only in cases where competition was involved between, on the one hand, the taxed product and, ‘on the other hand, a directly competitive or substitutable product which was not similarly taxed”. Por the Community, this does not mean that the first sentence of Article III:2 has a purpose contrary to the general principle set forth in Article III:1. The reason why the first sentence of Article III:2 does not refer to Article III:1, is because the imposition of taxes on imported products in excess of those imposed on domestic products is presumed inherently protective and therefore contrary in all cases to the general principle set forth in Article IH:1:; in the case of tax discrimination between like products a demonsiration of protectionist effect is not required. The aim-and-effect test nullifies this tule through the subterfuge of forcing the requirement “so as to afford protection” into the definition of “like product”. b) The Test Suggested by Canada 4.23 Canada did not raise any claim under Article III:2, first sentence. However, in response to questions from the Panel on the legal test arising out of Article III:2, first sentence, Canada stated that the negotiating history, the first Working Party Report on Brazilian Internal Taxes,” and a plain reading of the wording of Article I11:2, and the Note ad Article IIL, Paragraph 2, makes clear that unlike Article 1:2, second sentence, Article III:2 first sentence does not refer to Article IH: 1 and is thus a self-contained obligation. Canada thus considers that Article I-:1 should not be “read into” 4GATT/CP.3/42, adopted 30 June 1949, BISD 1/181. WT/DS8/R WT/DS10/R WT/DS11/R Page 22 Article IIl:2, first sentence. For Canada, to interpret Article III:2, first sentence, by adding as a separate Jegal requirement the provisions of Article III:1 does not clarify a Member’s rights and obligations but rather goes against the explicit wording of Article III:1 and Article 11:2, first sentence, of GATT 1994, Canada further argued that a fair reading of the first sentence of Article III:2 as well as the object and purpose of Article IM:2, make it clear that the sentence does not authorize permissible regulatory distinctions respecting discriminatory taxation on imported products that are “otherwise” “like” products, Canada thus considers that the words “aim and effect” should not ultimately be read into an analysis of Article III:2, first sentence. Article III:2, first sentence, is singularly designed to ensure that internal taxes are not applied discriminatorily on “like” products. Therefore it does not authorize permissible regulatory distinctions on products that are otherwise like to justify discriminatory internal taxation. The structure of GATT 1994 is equally useful in considering the context of Article I1I:2 in relation to the WTO Agreement. This structure shows, in Canada’s view, that derogations from particular obligations for domestic policy reasons are permissible only when expressly stated. Thus, for example, Article 1:2, Article III:3 and Article XI:2 of GATT each set out explicit permissible regulatory derogations from stated obligations. ©) The Test Suggested by the United States 4.24 The United States began by pointing out that GATT panel reports over the years have confirmed that the first sentence of Article III:2 requires that imports be provided equivalent opportunities. The central concern of Article II:2 is the targeting of imports as such for a special treatment. For the United States, in examining whether two products subjected to different regulatory or tax treatment are “like” or “directly competitive or substitutable”, it is necessary to determine first whether the product differentiation is being made “so as to afford protection to domestic production”. This determination requires a determination whether the distinction has the “aim and effect” of affording such protection. Physical characteristics are only a small sub-set of the legitimate distinctions that exist. However, in the US view, when a panel merely goes by its own perceptions of physical resemblance, uses, etc., it is implicitly, and non-transparently, judging the appropriateness of the tax or regulatory policies embodied in distinctions between products. Thus, the notion of “likeness” cannot be separated from the purpose with respect to which products are “like”, and the objectives of the regulatory scheme that draws a distinction between two otherwise similar products. 4.25 The United States noted that the WTO Agreement recognizes that origin-neutral regulatory distinctions can be compatible with WTO principles even if these distinctions do not appear in Article XX of GATT 1994. Under the WTO Agreement on Technical Barriers to Trade (“TBT Agreement”), technical regulations are permitted where they fulfil “legitimate objectives”. Technical regulations by their nature differentiate among otherwise like products. “Legitimate objectives” are not exhaustively defined in the TBT Agreement, but Article 2.2 of the TBT Agreement provides a partial list (i.e., national security requirements, prevention of deceptive practices, protection of human health or safety, animal or plant life or health, or the environment). There are many other legitimate objectives, and even those specifically listed in the TBT Agreement are not all found in Article XX. The TBT Agreement was about measures that make regulatory distinctions among what may otherwise be like products. In the US view, no one could read the TBT Agreement as prohibiting technical regulations, or find that they are per se inconsistent with Article III of GATT 1994 whenever imports are disproportionately affected, simply because the regulations in question are based on criteria other than those in Article XX. The United States further noted that the TBT Agreement is not an exception to GATT 1994. 4.26 As a result, argued the United States, WTO Members are permitted to make regulatory distinctions among products that might otherwise be considered “like”, in pursuance of a legitimate objective other than trade protection. This is what the 1992 Malt Beverages report recognized in WT/DS8/R WT/DS10/R WT/DS11/R Page 23 analyzing the regulatory regime for low alcchol and high alcohol beer. In the US view, that panel sought to avoid a result that would make even an unintentional coincidence between domestic regulation and the presence or absence of foreign competition in the market amount to a violation of Article IIE:2. Such a result would force policy harmonization and encroach on the policy options available to legislators and regulators to an extent unanticipated when GATT was drafted. The United States also pointed out that the European Court of Justice (ECJ) had reached similar conclusions in interpreting Article 95 of the EC Treaty. In its cases on fiscal incentives, the EC] has determined that “ ,. at the present stage of the development of Community law and in the absence of any unification or harmonization of the relevant provisions, Community law does not prevent Member States from granting tax advantages, in the form of exemption from or reduction of duties, to certain products or to certain classes of producers. Indeed, tax advantages of this kind may serve legitimate economic or social purposes, such as the use of certain raw materials by the distilling industry, the continued production. of spirits of high quality, or the continuation of certain undertakings such as agricultural distilleries” .” The Court has limited permissible differentiation to cases in which the distinction drawn “... pursues objectives of economic policy which are themselves compatible with the requirements of the Treaty and its secondary legislation, {and] the detailed rules are such as to avoid any form of discrimination, direct or indirect, in regard to imports from other member States or any form of protection of competing domestic products. 4.27 The United States saw this analysis by the EC] as remarkably similar to that in the 1992 Malt Beverages panel report. In specific cases, the ECJ had applied the rule permitting preferential tax treatment for the pursuit of “legitimate economic and social aims” to permit, for instance, tax exemptions for beer from small-volume breweries,” preferential treatment of fruit-based spirits distilled by small cooperatives that use their own raw materials,” higher taxes imposed on luxury goods,” higher taxes on cars with larger engine displacement,” lower taxes on natural sweet wines the production of which is traditional and customary*’ and lower taxes on game machines typically used by children. The ECT has in this way been able to draw the line between Community integration and member State fiscal sovereignty. It is therefore possible for an international tribunal to examine origin-neutral legislation and determine on the facts of each case, whether the measures in question are “such as to avoid ... protection of competing products”. 4.28 On this comparison between the ECJ case law and GATT Articte III, the Community argued that it is important to begin by pointing to a basic difference between GATT Article III:2 and the corresponding tax discrimination provision of the EC Treaty (Article 95). Article III:2 is covered by the general exception of GATT Article XX; Article 95 is not subject to any exception. The latter 5H. Hansen jun. and O.C, Baile GmbH & Co. v. Hauptzollamt Flensburg, Case 148/77, {1978] ECR 1806; cited, c.g., in Commission v. Italian Republic (regenerated petroleum products), Case 21/79, [1980] ECR 1, 12. *John Walker & Sons Ltd v. Ministeriet for Skatter og Afgifter, Case 243/84, [1986] ECR 875. "Bobie Getrinkevertrieb GmbH y. Hauptzollamt Aachen-Nord, Case 127/75, [1976] ECR 1079. Hansen & Balle, note 22 above. Commission v. Italy, Case 319/81, [1983] ECR 601, paras. 14, 21. *Commission v. Italian Republic, Case 200/85, [1986] ECR 3953. “Commission v. France (natural sweet wines), Case 196/85, [1988] CMLR 851. “Gabriel Bergandi v. Directeur général des impéts, Case 252/86, [1988] ECR 1343. WT/DS8/R WT/DS10/R WT/DS11/R Page 26 production if it accorded greater competitive opportunities to domestic products than to imported products. ... [the] central objective of the analysis remained the determination of whether the regulatory distinction was made ‘so as to afford protection todomestic production.’ The analysis of aims and effects of the measure were elements that contributed to that determination.” In examining whether distinctions have the aim of affording protection, the United States noted that the US Auto Taxes panel report looked at the stated policy objective for the tax or legislative measure in question, the statements by legislators, preparatory work and the wording of legislation as a whole. Italso looked at the treatment of products on either side of the regulatory distinction drawn, and whether it was known at the time the legislation was enacted that it would draw a line between one group of products that would be foreign and another group that would be domestic. The panel also examined the incentives created by the legislation, and whether these incentives would lead to a result consistent with the stated policy behind the legislation. In connection with its examination whether the measure had the effect of affording protection, the panel found that the distinction drawn for the luxury tax (between automobiles above and below $30,000 in value) “did not appear arbitrary or contrived in the context of the policies pursued”.*' The US Auto Taxes panel report then examined for each of the measures at issue whether the distinctions drawn had the effect of affording protection to domestic production, in terms of conditions of competition. The panel examined data on sales and trade flows for evidence of a change in the conditions of competition favouring domestic products. It also examined other data, including whether the characteristics distinguished were inherent to domestic products or foreign products, and whether there was a large difference in tax rates at the threshold. 4.31 The United States went on to argue that the same analysis that was developed in the US Auto Taxes panel report should be applied to the Japanese taxes at issue here. The United States argued that Japan’s tax system applicable to distilled spirits has the aim and effect of affording protection to domestic production of shochu. For this reason, and because white and brown spirits have similar physical characteristics and end-uses, the United States requested the Panel to find that white and brown spirits are “like products” in the sense of the first sentence of Article HI:2, and therefore that the difference in tax treatment between shochu and vodka, rum, gin, other white spirits , and whisky, brandy and other brown spirits is inconsistent with Article IIT:2, first sentence. 4.32 On the issue of the burden of proof, the United States argued that it is up to the complainant to produce a prima facie case that an origin-neutral measure has both the aim and effect of affording protection to domestic production. Once the complainant has demonstrated that this is the case, then it would be up to the defending party to present evidence to rebut the claim. The panel would decide whether it were more likely than not that the measure is applied so as to afford protection. As in the usual panel proceeding, only the respondent would have an interest in showing that one of its Corigin- neutral) domestic measures did not have protective aim or effect. @d) The Test Suggested by Japan 4.33 Japan, argued that the consistency with Article III:2, both the first and the second sentences, of a different treatment of products should be judged in light of paragraph 1 of the Article, in particular the language “not be applied ... so as to afford protection to domestic production”, and that whether or not the tax at issue is designed “so as to afford protection to domestic production” should then be judged by whether it had the “aim” and “effect” of affording protection. Japan submitted that the “Idem. para. 5.10, “Idem, para. 5.14. WT/DS8/R WT/DSIO/R WT/DSIUR Page 27 more recent criteria of “like products” can be found in the 1992 Malt Beverages panel report which stated that “the Panel considered that the limited purpose of Article [II has to be taken into account in interpreting the term ‘like products’ in Article IIT, It established that Article I]J:2 dees not prohibit classification of products for a legitimate policy goal: “The purpose of Article Ili is thus not to prevent contracting parties from using their fiscal and regulatory powers for purposes other than to afford protection to domestic production. Specifically, the purpose of Article II is not to prevent contracting parties from differentiating between different product categories for policy purposes unrelated to the protection of domestic production” .” That panel found under this criteria that beer with high alcoholic content is not like beer with low alcoholic content. 4.34 Japan continued in stating that these more recent criteria are in contrast to the 1987 Panel Report which found “likeness” on the basis of physical similarities, identical end-uses and customs classifications, and did not allow difference in treatment between like products. Japan also referred to the US Auto Taxes report which also applied the aim-and-effect test in respect of taxation: “The Panel noted that the purpose of Article III is set out in paragraph i of the article ... The Panel considered that paragraphs 2 and 4 of Article III had to be read in the light of this central purpose. The Panel reasoned therefore that Article III serves only to prohibit regulatory distinctions between products applied so as to afford protection to domestic production” .* “The Panel noted that the term ‘so as to’ suggested both aim and effect. Thus the phrase ‘so as to afford protection’ called for an analysis of elements including the aim of the measure and the resulting effects.” For Japan, in determining whether two products subject to different treatment are like products, it is necessary to consider whether such product differentiation is being made “so as to afford protection to domestic production”. In Japan’s view, the 1987 Panel Report would not allow differentiated treatment for socially legitimate policy goals (e.g., unleaded gasoline and leaded gasoline) between physically like products. The subsequent overruling of this approach proves the weakness of the reasoning of the 1987 Panel Report. Japan also emphasized that the 1987 Panel Report failed to deliver a clear-cut conclusion on the issue of “likeness” between shochu A and vodka. Although the 1987 Panel Report noted that these “could be considered as ‘like’ products”, the analysis ended in a discussion of directly competitive or substitutable products. Nor do these products appear on the list of pairs of like products inthat report. Japan concluded that the Liquor Tax Law does not apply taxes “so as to afford protection to domestic production” , and under the new criteria (US Auto Taxes and 1992 Malt Beverages), shochu A and shochu B are not “like” products to whisky, brandy, liqueur or spirits. 4.35 Developing further the criteria to be used under this aim-and-effect test, Japan argued that the aim of the Liquor Tax Law is not so as to afford protection cr protectionist. The legislation pertains to neutrality and horizontal equality in equalizing the tax/price ratio burden across tax categories. Nor “1992 Malt Beverages panel report, op. cit. 5, para. 5.25. “US Auto Taxes panel report, op.cit., para. 5.7. “Idem, para. 5.10 WT/DS3/R WT/DS10/R WT/DS11/R Page 28 does the Liquor Tax Law have the effect of protecting domestic production since it does not distort the competitive relationships between imported and domestic products based on the three following cumulative criteria: 1} the neutrality of the tax burden among categories of the legislation under examination, 2} the production of the allegedly protected products outside the imported country and of the allegedly “imported” products in the country, and 3) the absence of a directly competitive or substitutable relationship (cross-price elasticity) between the imported and domestically-produced products. Japan argued that if there is no difference in tax burden, the system does not distort trade; if directly competitive and substitutable relationships do not exist, differences in tax burden do not matter; if the products at issue are produced in and out of the country, the tax differentiation should not be construed to afford protection to domestic production. Protective distortion can be shown only when all of the three requirements are met. Japan further argued that as one examines the relative tax burden between the products in question, the tax burden should be measured by the tax/price ratio, a yardstick which best captures the impact on consumers’ behaviour. In examining whether or not the category in question is almost exclusively domestic, what needs to be examined is not import ratios, but whether or not anallegedly “domestic” category is produced in other countries and whether or not “imported” products in question are also domestically produced. 4.36 The Community responded that the aim-and-effect test is inconsistent with the ordinary meaning of Article III:2, first sentence, and contrary to the specific purpose of this provision. Previous panels have taken the view that the interpretation of the term “like product” should be made on a case-by-case basis as this term may have a different scope in each of GATT provisions in which it is used. For instance, the notion of “like product” has traditionally been given a broader interpretation in the context of Article 1:1 or Article III:2 than in the context of Article VI. Nonetheless, it remains that, in conformity with the ordinary meaning of the word “like”, the notion of “like product” is in all cases an objective one, exclusively related to the characteristics of the product. The “purpose (or aims) and the effect” of a regulatory measure are totally alien to that notion. An apple does not cease to be an apple only because the legislator does not pursue evil intentions when decreeing that it is an orange. Moreover, the aim-and-effect test effaces the clear textual difference between the first and the second sentences of Article III:2. For the Community, the ordinary meaning of Article III:2 is that the first and the second sentences of this provision set out different legal requirements and that only the second sentence refers to the first paragraph of Article III. This results clearly from the use of the word “moreover” at the beginning of the second sentence as well as from the Note ad Article II:2. This does not mean that the first sentence of Article III:2 has a purpose contrary to the general principle set forth in Article III:1. The Community argued that the reason why the first sentence of Article III:2 does not refer to Article III:1 is because the imposition of taxes on imported products in excess of those imposed on domestic products is presumed inherently protective and therefore contrary in all cases to the general principle set forth in Article HI:1; and this presumption is an irrefutable one. The first sentence of Articie II1:2 thus lays down the evidentiary rule that in the case of tax discrimination between like products, a demonstration of protectionist effect is not required. In the Community’s view, the aim-and-effect test nullifies this rule through the subterfuge of forcing the requirement “so as to afford protection” into the definition of “like product”. 4.37 Moreover, the Community argued that the interpretation of “so as to afford protection” as requiring both a protectionist purpose and a protectionist effect is by no means mandated by the ordinary meaning of these terms. For instance, the Shorter Oxford Dictionary defines the conjunction “so as” in the following terms: “followed by an infinitive, denoting result or consequence” . In turn, the French words “de mani@re 4” are defined by the Dictionnaire Petit Robert as “propre A obtenir telle conséquence”. Moreover, this interpretation is incompatible with the specific purpose of Article III:2 and, more generally, with the basic objective of Article III, which, in the Community’s view, has been variedly but consistently defined as: ensuring “that internal taxes on goods should not be used as means WT/DS8/R WT/DS10/R WT/DS11/R Page 31 paragraph 4.25 above. The United States further submitted that if national treatment obligations in GATT and the General Agreement on Trade and Services (“GATS”) were to be interpreted consistently, a discrimination test linked only to enumerated general exceptions would lead to difficult conclusions in the case of GATS which has a narrower exceptions list than GATT’s. Many laws and regulations governing the behaviour of services providers draw distinctions between categories of otherwise “like or competitive” providers based on objectives wholly unrelated to the generat exceptions set out in GATS Article XIV. The United States concluded that WTO Members are permitted to make regulatory distinctions among products that might otherwise be considered “like” , but they must have a legitimate objective for doing so. 4.41 For the Community, the parallelism drawn by the United States to the application by panels of Article XX or the TBT Agreement was misleading. Article XX lays down a limited list of exceptions. Moreover, the application of the exceptions is subject to a mumber of requirements defined inthe chapeau. of Article XX, as well as in each of the specific grounds of justification. Likewise, the TBT Agreement contains a list of grounds of justification. Even if the list is open, it provides panels with some guidance in order to determine what may constitute a “legitimate policy objective”. Further, the TBT Agreement lays down strict requirements concerning, inter alia, proportionality, risk assessment, duration, compatibility withexisting international standards, transparency and recognition of equivalent standards. For the Community, all these safeguards are absent in the aim-and-effect test. Under this approach, any non-protectionist ground may provide a valid justification for discriminatory taxation. Moreover, the aim-and-effect test does not incorporate any proportionality requirement. Last but not least, under this approach public policy objectives are not exceptions to an obligation but the criteria for defining the scope of the obligation. As a result, the burden of proof, which in Article XX and the TBT Agreement lies with the respondent, is subtly but effectively shifted to the complainant. Given the extreme difficulty of positively proving a protectionist purpose (as opposed to refuting the proof that a measure does not have a specific non-protectionist purpose), the mere invocation by a defendant of a non-protectionist purpose may suffice in practice to exclude the application of Article TI:2. For the above reasons, the aim-and-effect test could have the perverse effect of rendering Article XX redundant with respect to Article III:2. The grounds of justification listed in Article XX can also be invoked under the aim-and-effect test (why should legitimate policy objectives already recognized by GATT be deemed less worthy than non-recognised ones?). Since the conditions for the application of Article XX are more restrictive, no Member would bother to invoke Article XX. Indeed, why would any Member go through the pain of arguing that a discriminatory measure is “necessary” to protect human health, if it could simply argue that because the measure has no protectionist purpose, the products concerned are not ‘like’ and the prohibition of discrimination does not apply? Concerning the reference toGATS made by the United States, the Community argued that GATS Article XVI differs substantially from GATT Article III:2, first sentence. While a finding that Article III:2, first sentence, has been infringed does not require proof of protectionist effect, GATS Article XVI provides that formally identical or formally different treatment shall be deemed to be less favourable only if it modifies the conditions of competition in favour of like domestic services or service suppliers. Moreover, GATS Article XVII only applies to the sectors inscribed inthe Member’s schedule and subject toany conditions and qualifications set out therein. In contrast, the national treatment obligation in GATT Article IIT applies in respect of all sectors and may not be subject to any conditions or qualifications. 4.42 The Community also pointed out that the Panel should be advised of the risks involved in the aim-and-effect approach. For example, the aim-and-effect test could open the door to claims that the extraterritorial application of environmental regulations concerning non-product related processes WT/DS8/R WT/DS10R WT/DSIUR Page 32 and production methods is not contrary to Article III. Likewise, under the aim-and-effect test, one could argue that the imposition of a higher sales tax rate on products which have been manufactured by workers whose wages are below a certain level or who are required to work on Sundays does not infringe Article III:2 (even if the taxes have a disproportionate impact on imported products) because the tax differential is based on non-protectionist social considerations.” Moreover, there is a risk that the aim-and-effect test could contaminate other GATT provisions and, more generally, the entire WTO system by substituting for some of the hard-and-fast rules at the heart of the system the unpredictable balancing of ill defined “legitimate policy objectives”. The Community concluded that if the general opinion were that the exceptions provided for in Article XX are not sufficient in relation to Article IN:2, first sentence, the only approach consistent with the WTO Agreement would be to amend Article XX in order to add new grounds of justification and/or to relax the conditions for their application. Articles 3.2 and 19.2 of the DSU make clear that panels cannot add to or diminish the rights and the obligations of the Members under GATT. Thus, panels are precluded from engaging in the creation of new exceptions to existing obligations even when this appears to be necessary in order to fill GATT lacunae. For all these reasons, the Community requested the Panel not to consider the aim-and-effect test for the application of the first sentence of Article IIL:2. 4.43 Japan argued that in its view even the Community has come to agree that the analysis of the phrase “so as to afford protection to domestic production” requires the analysis of both purpose (i.e. aim) and effect despite its argument to the contrary, and that, therefore, the Community should accept that the aim of the measure should not be irrelevant if and when the phrase “so as to afford protection to domestic production” needs to be examined in regard to the first sentence of Article I1:2. Japan further argued that there is no reason to require the “aim” of the tax difference in question to fall under one of the exceptions in Article XX. The 1992 Malt Beverages panel report concluded that the regulatory distinction was not so as to afford protection and reached this conclusion as a matter of the interpretation of Article II; the panel did not apply Article XX to a measure which is otherwise inconsistent with Article HI. There is no reason, therefore, to restrict justifying “purposes” to those listed under Article XX. This issue ought to be interpreted on a case-by-case basis in light of the purpose of Article II. Moreover, for Japan, the Community’s concern over the burden of proof of the “aim” is unfounded. Japan does not believe that a mere claim of “legitimate purpose” can, or should, automatically override obligations under Article III. The practice of GATT dispute settlement is that both parties play the alternate role of claim and rebuttal. Under such practice, the defending party will be expected to present evidence to prove that there is legitimate rationale other than protection, which the complaining party or parties will rebut. The complaining party or parties, in turn, will present evidence of protectiveness, to which the defending party will make its rebuttal. On the basis of evidence thus available, panels can and should render judgment on protectiveness in the “aim” of a disputed measure. Therefore, Japan does not believe that the aim-and-effect test would lead to loss of discipline under Article II1:2, “According to the Community, the generally accepted view is that an imported product and a domestic product manufactured in accordance with different non-product related PPMs (processesand production methods) are still “like” and cannot therefore be treated differently under internal regulations. This interpretation is supported by the evo unadopted panel reports on “US - Restrictions on Imports of Tuna” (see Tuna I, panel report dated 3 September 1991, not adopted but published in BISD 398/155, and Tuna If, panel report dated 10 June 1994, DS29/R, not adopted). Under the aim-and-effect test one could argue that oana harvested with a high rate of incidental dolphin killing is not like to other tuna because the distinction does not have a protectionist purpose. Under the traditional interpretation, this would be contrary to Article II] because the more taxed products are “like” products manufactured by workers who are paid above the minimum wage level and do not work on Sundays. See panel report on “Belgium - Family Allowances”, adopted on 7 November 1952, BISD 18/59. WT/DS8/R WT/DS10/R WT/DS1I/R Page 33 4.44 The Community replied that in the end panels would have to put the burden of proof on one of the parties. If the legitimate purpose is integrated into Article III of GATT as it is in Article 2.2 of the TBT Agreement, the final assessment of a panel would be along the following lines: The complaining party has offered no convincing proof that the legitimate purpose mentioned by the defendant was not the real purpose but protectionism. Whereas in an Article XX situation, the final assessment would be: The defendant party has not been able to show that the measures was necessary for, e.g., animal health. In the final analysis, the difference is important and the shift of burden of proof is important. In Article 2.2 of the TBT Agreement this was clearly desired by the Uruguay Round negotiators; in respect of GATT Article I1i:2 it was not. 4.45 The Community went on to argue that, without abandoning the traditional interpretation of Article III:2, first sentence, two kinds of flexibility could be read into its wording. The first flexibility is found in the interpretation of the notion of discrimination in the case of graduated systems of taxation of like products. For example, if all cars are considered like products, it may be possible to accept that a system of graduated taxes based on engine displacement or weight is not discriminatory as long as the tax increases proportionally to engine displacement or weight and applies equally to domestic cars with the same engine displacement or weight. This requires a review of the tax system as a whole as applied to a broad category of like products (here: cars). The Community argued that this form of flexibility seems to have been envisaged by the 1987 Panel Report which, after noting that the specific tax rates on special grade whisky/brandy were considerably higher than the specific tax rates on first and second grade whisky/brandy (ail of which had been previously found to be like products), observed that “,.. [it) was unable to find that these tax differentials corresponded to objective differences of the various distilled liquors, for instance that they could be explained as a non-discriminatory taxation of their respective alcohol contents”. The second flexibility is found in the definition of the dividing line between like products and directly competitive and substitutable products. In the above-mentioned example of cars, one can take the view that cars with different engine displacement are not like, but rather directly competitive and substitutable products. In that case it is possible to let the “so as to afford protection” criterion play a role. Por the Community, differences in taxation based on alcohol content might be covered by the first flexibility; differences in taxation between leaded and unleaded gasoline or between recyclable and non-recyclable containers might be justified under the second flexibility; regulatory distinctions between cups made from a material producing toxic gas when incinerated and cups made from other materials might be covered by the second flexibility or, alternatively, by Article XX (b); the preservation of historical buildings is covered by Article XX(; Sunday closing laws could be analyzed by panels in the same way as the ECJ, i.e., by laying down a distinction between those requirements that affect directly the distribution and sale of products as such and those which concern the regulation of commercial activities and have only an incidental impact on the sale of goods. The Community concluded that these two flexibilities, together with the general exceptions of Article XX, may offer sufficient scope to deal adequately with the examples of worthy regulatory distinctions between products cited by Japan and the United States. 4.46 The United States reiterated that the plain language of Articte III:2, first sentence, condemns measures that explicitly target foreign products and accord them less favourable treatment. This, in the US view, makes sense because the discriminatory aim of such measures is apparent. However, when a measure is origin-neutral and therefore such an aim cannot be presumed, it does not make sense “41987 Panel Report, para. 5.%a). WT/DS8/R WT/DS10/R WT/DS11/R Page 36 An examination of the effects can tell if a measure is targeting imports, and if there is cross-price elasticity of demand between the relevant products. Moreover, as the United States had argued in relation to the Community’s discussion of “effects”, an “effects only” test is also a poor idea and unimplementable. For the United States, all taxation or regulation causes some degree of market distortion, and often it is not possible to know in advance what the effect of a tax or regulation on the market or the economy will be. Domestic products and imported products are different groupings of goods with differing characteristics. In the US view, the “effects only” test would imply that all internal measures maintained by the Community and its Member States should be reviewed and judged solely in relation to whether they happen to disadvantage imports, a position which would have sweeping effects. Since it is generally accepted that imported products have an income elasticity greater than one, and domestic goods have an income elasticity less than one, adoption of an “effects test” as posited by the Community would mean that every time the Bundesbank, or any other central bank, takes actions that reduce economic growth, these measures would be inconsistent with Article DI. 2. Application to the Present Case of the Legal Analysis Suggested by the Community for Article ITI:2, First Sentence a) The First Step of the Test: Like Products 4.51 In referring to the first step of the legal test it suggested for the first sentence of Article III:2 ~- the like product assessment, the Conamunity argued that the physical characteristics and manufacturing process of spirits and shochu A and B are similar; The two categories of shochu and most of the liquors falling within the category “spirits” are white/clear beverages with a relatively high alcoholic content made by distillation from the same large variety of raw materials (¢.g., grains, potatoes ...). A comparison of the legal definitions of shochu and of the category of “spirits” contained in articles 3.5 and 3.10 of the Liquor Tax Law demonstrates that the only differences between these two categories are that shochu cannot (1) be made from sugar cane and distilled at less than 95 per cent of alcohol {such as rum); (2) have other ingredients added at the time of distillation (such as gin); (3) be filtered with charcoal of white birch (such as vodka); (4) have an alcoholic content in excess of 45 per cent, in the case of shochu B, or 36 per cent, in the case of shochu A. In practice, as mentioned above, both types of shochu typically have an alcoholic strength of 20 per cent to 35 per cent, with 25 per cent being the most common strength. The legal definition of the category of “spirits” does not provide for a maximum alcohol content but in practice, the average alcohol content of the liquors falling within this category is 40 per cent, For the Community, the above differences between shochu and each of the main types of “spirits” are clearly minor and do not prevent all of them from qualifying as like products, Similar differences (if not more significant ones) exist also among the various types of western- style distilled spirits, despite of which ali of them have been included into a single category of “spirits” and taxed at a uniform rate. The Community submitted that the differences in alcoholic strength are moreover rendered irrelevant by the drinking habits of the Japanese consumers: both shochu and the liquors falling within the category of “spirits” tend to be drunk heavily diluted with water or other non-alcoholic beverages and end up at roughly the same strength. 4.52 Insupport ofits allegation, that shochu and spirits are like products, the Community also argued that shochu and spirits have essentially the same consumers’ uses and customs classification. Shochu and “spirits” have essentially the same end-uses. All of them are drunk “straight”, “on the rocks” or, more frequently, diluted with water or other non-alcoholic beverages. Moreover, both shochu and “spirits” are widely drunk by all categories of consumers, regardless of age, sex or occupation. WT/DS8/R WT/DS10/—R WT/DS11R Page 37 In support of its argument, the Community submitted two market studies.” Moreover, shochu and all “spirits” other than gin and rum fall within the same HS sub-heading (HS 2208.90). This confirms that the differences between shochu and the category of “spirits” may be less significant than the differences among the various types of liquors falling within the category of “spirits”. 4.53 The Community then submitted that a striking illustration of the “likeness” between shochu and “spirits” and, at che same time, of the arbitrariness and artificiality which are inherent to the criteria ‘on the basis of which the Liquor Tax Law attempts to distinguish them, has been recently provided by the change in the tax categorization of the brand “Juhyo”. This brand had been traditionally sold by the local manufacturer Suntory as vodka and accounted for almost half of the Japanese production of that liquor. However, as from June 1993, Suntory started to market the same product as “Juhyo shochu”, All that was required in order to obtain this change in tax category was to discontinue the use of charcoal of white birch as a filtering material.*' The Community argued that the change was Inade with the aim of escaping the higher taxes levied on “spirits” and was followed by an immediate and substantial reduction in the retail prices of “Juhyo”. In support of its argument, the Community submitted an article from the Teiin Shkuryo Shinbun. The Community concluded by teferring the Panel to the findings of the 1987 Panel Report where it was stated that “Japanese shochu (Group A) and vodka could be considered as like products in terms of Article II:2 because they were both white/clear spirits, made of similar raw materials, and their end-uses were virtually identical (either as straight schnaps type of drinks or in various mixtures)”*? and that other types of spirits, in addition to shochu A and vodka, could also be like products. For the Community therefore, the liquors falling within the category “spirits” and the two sub-categories of shochu are, in light of all the criteria that have been identified above as relevant, “like products” within the meaning of the first sentence of Article 1:2. 4.54 Japan argued that in its view the Community acknowledged that the differences in physical characteristics between whisky/brandy and shochu are sufficiently large to prevent the two categories from qualifying as like products. It also noted that the Community’s claim of likeness applies only between the category of “spirits” and shochu A and B. Japan argued that, in examining the “likeness” of “spirits” and shochu, the Community looked at the following four criteria: (i) the product’s properties, nature and quality, (ii) its end-uses, (iii) consumers’ tastes and habits, and (iv) the HS classification. Japan argued that if the Community’s four criteria were correctly applied to the facts, “spirits” and shochu A and B would not be “like products”, because: as to (i) the product’s properties, nature and quality: - The alcoholic strength of shochu (mostly 20 to 25 per cent) is closer to wine and sake (12 to 15 per cent) than to “spirits” (around 40 per cent). ~ Most shochu does not undergo a post-distillation value-adding process (over 99 per cent is not aged in wooden casks) while “spirits” are characterized by value-addition through flavouring, purification with white birch charcoal or aging; picking a few examples from the vast array of shochu brands should not cloud the overall picture. - Bulky plastic, glass and paper bottles over 1.8 litres are the most popular containers for shochu. while 0.7 litre glass bottles are common for “spirits”. A market survey conducted by the Japan Market Research Bureau in December 1994 and a market survey conducted by an ependent research company in May 1994. ‘lIn addition, barley and rice were added as raw materials in order to alter the taste of the product. Nevertheless, this change was not required by the Liquot Tax Law in order to make “Juhyo” qualify as shochu. 1987 Panel Report, para. 5.7. WT/DS8/R WT/DS10/R WT/DS11/R Page 38 as to {i) end-uses and (ii} consumers’ tastes and habits: - 60 per cent of consumers drink shochu during meals but 63 per cent drink “spirits” after meals. - 42 per cent of shochu consumers, but only 4 per cent, 1 per cent, and none of vodka, gin, and rum consumers, respectively, drink the product in question with hot water; and none of shochu consumers but 26 per cent, 32 per cent, and 15 per cent of vodka, gin and rum consumers, respectively, drink the product in question with tonic water, according to the data submitted by the Community. - The study by ASI Market Research Inc. submitted by the complaining parties concludes that “(s)hochu is not seen as so much of a competitor (i.e., substitutable product) in the eyes of the consumers”. - According to a study, only 6 per cent of shochu consumers responded that they would drink “spirits” if shochu is not available. - Contrary to the Community’s allegation, the evidence submitted by the Community shows that shochu consumers are only as often (not more often) found in the “regular consumers” of premium brands of spirits and liqueurs as are found in all respondents. and as to iv) classification in the HS: - The 1996 version of the HS gives separate headings for rum (2208.40), gin (2208.50) and vodka (2208.60), as opposed to shochu (2208.90, “other”}. Japan submitted that the HS is established for purposes other than internal taxation and does not offer appropriate criteria by which to judge “likeness” in terms of Article III, but even if “likeness” should be examined onthe basis of identity of the HS heading, as the Community and the 1987 Panel Report suggest, shochu and vodka would not be “like” under the 1996 version of the HS. 4.55 The Community responded that as far as shochu and “spirits” are concerned, Japan had been able to identify only two main differences in physical characteristics: the alcohol content and the packaging. According to the Community, the differences in alcohol content between shochu and “spirits” are not reflected in their respective legal definitions and, therefore, cannot provide a valid justification for applying different tax rates. There is nothing in the Liquot Tax Law preventing the manufacture of vodka of 25 per cent. In practice, some brands of vodka do have an alcohol strength of 25 per cent as illustrated by the case of Juhyo. On the other hand, shochu B may have an alcoholic strength of up to 45 per cent, whilst the maximum alcohol content of shochu A is set at 36 per cent, i.e., only four degrees below the average strength for spirits. High alcohol shochu is by no means a rarity. In 1994, the sales volume of shochu of 35 per cent was larger than the total sales volume of all types of “spirits”. The alleged differences in packaging are irrelevant for a like product determination. The physical properties of shochu remain the same irrespective of the size and the material of the packages in which it is sold. 4.56 Japamsubmitted that such commonality of sales outlets and advertising styles between “spirits” and shochu as pointed out by the Community are also observed among all alcoholic and non-alcoholic beverages and thus fails to demonstrate that products are “like”, Though the Community points out similarity in the shochu-based pre-mixes and the pre-mixes made from other liquors, for Japan, such would not be evidence of “likeness” of shochu and “spirits”, just as the similarity among tequila-based, wine-based and beer-based “margaritas” in the United States would not render tequila, wine and beer “like”. For Japan, “Juhyo Vodka” and “Juhyo Shochu” are two distinct products with different raw materials and different production methods sold under the same established brand name, and are not “like products”. Japan also noted that the 1987 Panel Report failed to deliver a clear-cut conclusion on the issue of “likeness” between shochu A and vodka. Although the panel noted that these “could WT/DS8/R WT/DS10/R WT/DSIDR Page 41 E. Article IEI:2, Second Sentence 1. The Different Legal Analyses Suggested by the Parties for the Interpretation of Article III:2, Second Sentence a) The Test Suggested by the Community 4.62 The Community argued that the 1987 Panel Report, Article I1I:2, second sentence, the first paragraph of Article III to which reference is made in the second sentence of Article III:2; and the Note ad Article III:2, all establish a two-step test for the examination of the conformity of a system of internal taxation with Article IIE:2, second sentence. First, it must be determined whether the taxed imported and domestic products are “directly competitive or substitutable”; and secondly, it must be established whether the taxation is “protective”. For the Community, all liquors falling within the categories of “whisky/brandy” and of “liqueurs” and the two sub-categories of “shochu” are “directly competitive or substitutable” among each other. Should any of the liquors falling within the category of “spirits” be found by the Panel not to be a “like product” to shochu, the Community submitted that shochu and “spirits” are, at the very least, “directly competitive or substitutable” products. By applying higher tax rates to the categories of “whisky/brandy”, “liqueurs” and “spirits” than to each of the two shochu sub-categories, the Liquor Tax Law affords protection to the domestic production of shochu, thereby violating Article IIT:2, second sentence. 4.63 For the Community, the concept of “directly competitive or substitutable product” - the first step of the test - is wider than the concept of “like product” and may include products with different physical characteristics but substitutable in terms of uses such as, for instance, skimmed milk powder and vegetable proteins;* apples and oranges;* butter and oleomargarine;® tung oil and linseed oi1;°” or natural rubber and synthetic rubber. In order to determine whether two products are directly competitive or substitutable, the following criteria may be relevant: the aptitude of the two products to serve the same uses (it is not necessary, however, that the two products are substitutable in respect of all their potential uses); the extent and the form in which the two products are available to the public; the respective prices of the products and the responsiveness of the dernand for one of the products to the changes in the price of the other. 4.64 As for the second step of the test it suggested for the second sentence of Article [II:2, the Community argued that the following criteria may be relevant inorder to determine whether a difference in taxation is “protective” of domestic production. (1) The level of the tax differential, but contrary to the first sentence of Article III:2, a tax difference does not lead automatically to a violation of the second sentence of Article III:2. On the other hand, even small tax differences may be protective. Nonetheless a de rninimis differential may, in certain cases, be found not fo afford protection. (2) The degree of substitutability and competition between the two products. Logically, the protective effect of a system of taxation increases with the degree of substitutability and competition. (3) Whether the less taxed product is produced in other countries. A system of taxation is protective of the domestic production if the less taxed category is almost exclusively produced in the country imposing the taxes. In contrast, the fact that the more taxed product is produced also in the country applying the internal “Panel report op. cit note 16 (EEC - Measures on Animal Feed Proteins), adopted on 14 March 1978, BISD 258/49, para. 4.3. SEPCT/A/PV/9, p.7. “Reports of Committees and Principal Subcommittees, UN Conference on Trade and Employment, 1948, p. 61. SBICONF.2/C.3/SR.11, p- 1 and Corr.2, S*B/CONF.2/C.3/SR.11, p. 3. WT/DS8/R WT/DS10/R WT/DS1II/R Page 42 taxes is irrelevant. The Community also submitted that since Article III:2 protects trade expectations on the competitive relationship between imported and domestic products rather than expectations on trade volumes, it is not necessary, in order to establish a violation of Article III:2, second sentence, to show that the difference in taxation has had an actual effect on the volume of trade. b) The Test Suggested by Canada 4.65 Canada’s claim was limited to the second sentence of Article III:2. Canada submitted that taxes imposed on Canadian whisky as compared to those imposed on domestically-produced shochu are inconsistent with the provisions of Article III:2, second sentence, For Canada, the Liquor Tax Law that in 1987 was determined to be inconsistent with Article III:2, second sentence, continues, even as amended, to be inconsistent with Article III:2, second sentence. 4.66 For Canada, Article 3.2 of the DSU makes clear that it is the express wording of the WTO Agreement that ultimately defines the rights and obligations of Members and thus, whether the Liquor Tax Law is inconsistent with Article III:2. To the same effect, Canada cited Professor Lauterpacht regarding the role of treaties in defining the rights and obligations of States: “The rights and duties of States are determined in the first instance by their agreement as expressed in treaties - just as in the case of individuals their rights are specifically determined by any contract which is binding upon them. When a controversy arises between two or more States with regard to a matter regulated by a treaty, it is natural that the parties should invoke and that the adjudicating agency should apply, in the first instance, the provisions of the treaty in question”? Canada submitted that the relevant “rights and obligations of Members under the covered agreements”, ie., the “provisions of the treaty in question”, respecting Japan’s Liquor Tax Law are set out in Article IH:1; Article III:2, second sentence, and Note ad Article ITI, paragraph 2. 4.67 Canada referred the Panel to the exact wording of Article III:2, second sentence, which provides that “No Member shall otherwise apply internal taxes or other internal charges to imported or domestic products in a manner contrary to the principles set forth in paragraph III: 1”. The “principles set forth in Article If]:1” provide that “Members recognize that internal taxes and other internal charges ... should not be applied to imported or domestic products so as to afford protection to domestic production”. Canada recalled that the Note ad Article III, Paragraph 2 to Article IIT:2, second sentence, ‘was added pursuant to the Havana Conference Report of Sub-Committee A of the Third Committee on Tariff Negotiations, Internal Taxation and Regulation “so that it would be easier for Members to ascertain the precise scope of their obligations under this Article”.° The report stated that “A tax conforming to the requirements of the first sentence of paragraph 2 would be considered to be inconsistent with the provisions of the second sentence only in cases where competition was involved between, ‘on the one hand, the taxed product and, on the other hand, a directly competitive or substitutable product which was not “similarly taxed”. Canada argued, therefore, that a clear reading of these provisions establishes that under Article III:2 second sentence, four criteria must be satisfied for Japan’s Liquor Tax Law to be found to be inconsistent with this provision: (1) the taxes levied pursuant to the Liquor Tax Law are internal taxes or other internal charges; *Lauterpacht, International Law: Collected Papers, 86-87 (1970). Havana Reports, at p. 61, para. 36. WT/DS8/R WT/DS10OR WT/DS1I1/R Page 43 (2) whisky is a directly competitive or substitutable product with shochu Group A and shochu Group B; (By whisky and shochu Group A and shochu Group B are not similarly taxed; and (4) the taxes levied pursuant to the Liquor Tax Law afford protection to domestic production of shochu Group A and shochu Group B. For Canada, these criteria are based on the general principle, enunciated in the Havana Reports and confirmed in the 1989 panel report on “United States - Section 337 of the Tariff Act of 1930”, that internal taxes should not be applied in such a manner so as to afford protection to domestic production. The Working Party Report on Border Tax Adjustments” and the 1987 Panel Report confirmed that Article III:2 gives effect to this general principle by ensuring the trade “neutrality” of internal taxation measures that are applied to imported and domestic products. Internal taxation measures that are not trade neutral distort the conditions of competition between imported and domestic products thereby affording protection to domestic production. To assess whether an internal taxation system affords protection, Canada noted that the 1987 Panel Report sets out three variables that are applicable in the Present case: (i) whether there is a considerably lower tax rate on shochu than on imported whisky, (ii) whether the shochu consumed is almost exclusively produced in Japan, and (iii) whether shochu and whisky are mutually substitutable. ©) The Test Suggested by the United States 4.68 The United States submitted that since Japan's tax system applicable to distilled spirits has been devised so as to protect domestic production of shochu and because all distilled spirits have similar physical characteristics and end-uses, they are “directly competitive and substitutable” in terms of Article 11:2, second sentence. Therefore, the United States considered that the difference in taxation between distilled spirits exceeds any de minimis level because that difference materially alters the conditions of competition between domestic and imported products. In the present case, the United States submitted that the change in conditions of competition is illustrated by factors such as the demonstrated effect on consumption choices and the cross-price elasticity of demand discussed further below, 4.69 The United States reiterated that the plain language of Article III:2, first sentence, condemns measures that explicitly target foreign products and accord less favourable treatment. This, in the US view, makes sense because the discriminatory aim of such measures is apparent. However, when a measure is origin-neutral and therefore such an aim cannot be presumed, it does not make sense to say that the purpose of the measure becomes irrelevant. The United States noted that Article HI is designed to protect against discrimination, not to create a per se rule of absolute liability for any greater burden or restriction on international trade. All direct and indirect regulation of goods has domestic and international trade-restricting effects, because by its nature regufation imposes burdens, For the United States, the rule proposed by the Community would mean that a government could not adopt any measure, irrespective of its purpose, if the measure had the effect at some point of burdening foreign more than domestic products. Such a “pure effects” test would give no guidance or certainty to legislators or to their legal advisers, because in any situation its application could change from day to day based on international and domestic factors that could not be anticipated at the time a measure is adopted. [hid at p. 41, para. 7. Panel report adopted on 7 November 1989, BISD 368/345, para. 5.10. “Panel report adopted on 2. December 1970, BISD 185/97, para. 9. WT/DS8/R WT/DS10/R WT/DS1U/R Page 46 “whisky/brandy” are spirits obtained by distillation and with a relatively high alcoholic content. The main differences between the two categories are thus restricted to the fact that neither malted grains nor grapes can be used in the production of shochu. For the Community, this difference is only relative, as most shochu is made, like whisky, from different types of grain, albeit not malted. Other differences are that shochu is, as a general mule, a white/clear spirit, while whisky and brandy are brown-coloured; whisky and brandy are matured/aged and, as a general rule, blended, while shochu is not. These last two differences are becoming irrelevant as an increasing number of shochu brands claim to be blended and aged in barrels and are brown coloured. For the Community, the absence of any fundamental differences between shochu and “whisky/brandy” is attested by the fact that the advertising of many shochu brands tends to emphasize their similarities with whisky and/or brandy in terms of raw materials, ingredients, manufacturing process and tradition. In some cases, this policy has been pursued to the extreme of modifying the traditional manufacturing methods of shochu in a deliberate attempt to confer upon it a whisky-like appearance and taste. Concerning “liqueurs”, this category is comprised of a very heterogeneous variety of liquors which haveas their only common characteristic an extract content in excess of two per cent. The 1987 Panel Report found that differences concerning the level of extract content were minor and did not prevent two products from being like products. A fortiori, differences in the extract content are not sufficient in themselves to prevent liquors falling within the category of “liqueurs” from being considered as “directly substitutable and competitive” with “shochu”, “spirits” and “whisky/brandy”. Moreover, it must be recalled that a major portion of the sales in this category consists of bottled or canned pre-mixes made from “shochu”, “spirits” or “whisky/brandy” which are, therefore, identical to home-made mixed beverages from the same liquors. 4.75 The Community argued that the fact that all distilled spirits and liqueurs have the same basic properties and are objectively apt to serve the same end-uses is confirmed by the consumption patterns observed in the Japanese market. In support of its allegation, the Community submitted the results of a research conducted by the Japanese whisky industry on the presence of shochu and whisky in “snack bars”. All of them are drunk “straight”, “on the rocks” or, more frequently, diluted with water or other non-alcoholic beverages. The drinking styles of the various distilled spirits (including shochu) and liqueurs are virtually the same. Furthermore, the advertising of the different types of spirits and liquors tends to promote the same drinking styles. Both shochu and the other types of distilled spirits and liqueurs are widely drunk across all categories of consumers, regardless of their age, sex and occupation. A very high proportion of shochu consumers are also regular consumers of whisky and other spirits and liqueurs and this proportion is higher than among consumers of alcoholic beverages in general. The same pattern has been observed with respect to the consumption of premium brands of western-style spirits and liqueurs sold at the highest prices. In the last few years, a new market has emerged, especially among young consumers, for bottled or canned pre-mixed drinks combining spirits and soft drinks. As shown by the advertising materials submitted by the Community, the executional style, target market and drinking style of the shochu based pre-mixes and the pre-mixes made from other liquors are identical. 4.76 The Community further argued that shochu and the other types of spirits and liqueurs are directly competitive since they are available in the same trade channels and are promoted and advertised in asimilar way. All of them are sold at the same outlets, both for on-premise consumption and for home “Thus, in May 1988 (i.c., shortly after the adoption of the 1987 Pane! Report), the Japanese manufacturer Takara started marketing “Jun Legend”, a light amber coloured brand of shochu produced by blending two types of alcohol distilled from barley and com and maturing them in charred white oak barrels for one to five years. According to Takara, “the most noticeable characteristic of this brand is a flavour and taste similat to whisky”. When the new brand was launched, Takara announced its expectations that the new product would appeal to former consumers of second grade whisky which, as a result of the 1987 Panel Report, was expected to become subject to much higher tax rates as from 1989. WT/DS8/R WT/DS10/R WT/DS1UR Page 47 consumption. Although in the past, there may have been a certain specialization among on-premise outlets, in recent years this specialisation has disappeared. The Community submitted that the results of a recent survey shows that in the Tokyo area 71 per cent of the “snack bars” (a category of outlet where western-style liquors were traditionally predominant and which represents approximately 40 per cent of the total on-premise market) now serve both whisky and shochu. Similarly, an increasing number of “izakayas” (a once “traditional” shochu/sake/beer style of outlet which accounts for approximately 20 per cent of the on-premise market) are now serving whisky and other western-style spirits and liqueurs. Both at on-premise outlets and at outlets selling for home consumption, shochu is positioned and promoted side-by-side with the other types of spirits and liqueurs, thus evidencing that both the retailers and the public regard them as substitutable and competitive products. The level of advertising spending on shochu brands is comparable to the advertising spending on the brands of other spirits and liqueurs. The advertising of shochu and of other spirits and liqueurs is very similar in executional style, is targeted towards the same categories of customers (young consumers and “salarymen”) and aims at projecting similar images, regardless of whether the products are of “traditional Japanese origin”, like shochu, or western-style. The distribution of advertising spending among the different media is similar for shochu and other spirits and liqueurs, a sign that similar markets are being targeted. 4.77 For Canada, the evidence makes abundantly clear that in Japan, shochu and whisky continue to be directly competitive or substitutable products in that shochu and whisky have many common characteristics and are commonly consumed at similar diluted alcoholic strengths and manufacturers of shochu in Japan capitalize on these common characteristics by marketing some shochu products on the strength of their similarity to whisky and on the basis that they can be consumed in the same manner as whisky. Canada suggested that the processes and raw materials used in the production of shochu and Canadian whisky are very similar. Canada submitted that both Canadian whisky and shochu are produced from a variety of grain sources such as wheat, barley, rye and corn, although a slightly broader range of agricultural raw materials, such as rice, can be used to manufacture shochu. Many varieties of shochu use exactly the same raw grains that are used in whisky production. The enzymes and yeasts used in the production of whisky and shochu are also similar. With respect to production methods, the processes of milling, cooking and conversion are common to both whisky and shochu, The same equipment is used and the fermentation processes are similar. Whisky and shochu are both produced using continuous and pot distillation methods, or through a blend of the two methods. Up to the end of the distillation process, Canadian whisky and shochu can be identical. The main distinction between Canadian whisky and shochu is that Canadian whisky must be aged. This is not so for shochu, although some shochu is currently being aged. 4.78 Canada also argued that in Japan, both whisky and shochu can be consumed in common styles, i.e., “straight”, with water or “on ice”, and submitted evidence of current liquor advertising in Japan. As distilled liquors, whisky and shochu are both sold to the public at alcohol strengths significantly above the strengths common for other Liquors such as beer and wine. Canadian whisky is sold at the retail level in Japan at an alcohol strength of at least 40 per cent. Shochu can be sold at the retail levet up to an alcohol strength of 36 per cent for shochu Group A and 45 per cent for shochu Group B, {although the most common retail strength is 25 per cent for both shochu groups). Distilled liquors such as whisky and shochu are commonty consumed ina diluted style, with the resulting alcohol content being similar for both products. Pre-mixed, i.e., diluted, products sold in Japan contain whisky at “tn support of its allegation Canada submitted a letter from Hiram Walker & Sons Lid which confirmed: “The process of milling, cooking and conversion are common to both spirits. The same equipment, grains, and enzymes are used in the production of both shochu and Canadian whisky. The details of these processes will vary by distiller and formula but the basic process is the same for both types of spirits. The same similarities hold true for the fermentation process”. WT/DS8/—R WT/DS10/R WT/DS11/R Page 48 an alcoholic strength of between five per cent and eight per cent and contain shochu at an alcoholic strength of between four per cent and six per cent. Advertisements for the sale of shochu refer variously to the similarity in the raw materials used to produce whisky and the advertised shochu, the similarity in the production processes that produce whisky and the advertised shochu product and the similarity in the physical appearance (¢.g., coloration) of whisky and the advertised shochu product. In addition, Canada argued that recently, recognizing consumer perceptions that whisky and shochu are alternative choices, the Japan Spirits and Liquor Maker’s Association stated that “the drinking patterns of whisky/brandy and shochu [are] becoming alike and [that] they are competing with each other [in the] tmarket”. 4.79 Japan argued that “spirits” and “shochu” differ in physical characteristics, end-use, and in tariff lines as is described in paragraph 4.54 above. Japan also argued that whisky/brandy and shochu differ in materials (with malts versus without malts; Bourbon, Tennessee, and Canadian whiskies without malts are classified as “spirits” under the Liquor Tax Law), in the post-distillation processing (aged in wooden casks versus over 99 per cent not aged in wooden casks), in alcoholic strength (around 40 per cent versus 20 to 25 per cent), in colour (0.2 to 0.8 of optical density versus 0.08 of optical density) and in containers (0.7 litre glass bottles versus bulky plastic, glass and paper bottles over 1.8 litres). For Japan, they also differ in end-uses: according to a study in Japan, 60 per cent of shochu consumers drink shochu during meals, while 72 per cent of whisky consumers drink whisky after meals; and according to a study submitted by the Community, only eight per cent of consumers of shochu drink the beverage “on the rocks” while 68 per cent of bourbon whisky consumers do. None of bourbon whisky consumers mix such whisky with hot water or juice, while 42 per cent and 37 pet cent of shochtt consumers do respectively. They also differ in tariff lines: whisky is classified as “2208.30 whisky” while shochu is classified as “2208.90 Other”. Japan also argued that the commonality in availability to the public mentioned by the Community exists only to the extent applicable to all alcoholic and non-alcoholic beverages: the menus and promotion leaflets submitted by the Community list not only whisky(ies) and shochu but also sake, wine, beer, juice, coffee and tea side by side. As to the Canada’s comment that pre-mixed shochu and pre-mixed whisky are taxed at the same rate in Japan, Japan argued that the fact that pre-mixed wine and pre-mixed spirits are taxed at the same rate in Canada does not imply a directly competitive or substitutable relationship between wine and spirits, the two products taxed at completely different rates in Canada. 4.80 Japan also noted that the aptitude of the two products to serve the same uses raises the issue of the extent of the sameness. Since the use for quenching the thirst, for example, is common to ali beverages, and since the use for enjoying alcohol is common to all alcoholic beverages, the concept of “sameness” should be understood in a narrower sense. According to Japan, the Community argues that sameness in drinking habits between shochu and other distilled liquors is sufficient to meet the criteria. However, Japan’s evidence shows a good degree of divergence in drinking habits not only between shochu and spirits but between shochu and Bourbon whisky as well. The aptitude to serve the same uses does not seem to exist beyond what would apply to all alcoholic beverages. 4.81 The Community argued that Japan’s criticisms were unfounded. The study on drinking styles submitted to the panel showed that the end-uses of shochu and bourbon whisky were the same except that shochu is not drunk with tonic water and bourbon is not mixed with warm water or juices. Moreover, the study showed that three out of the five most frequent end-uses of shochu were also found among the five most frequent end-uses of bourbon. WT/DS8/R WT/DSIOR WT/DSIUR Page 51 commissioned by the European Commission in 1994," to Japanese consumption data for the past 20 years, based on household surveys by the Bureau of Statistics of the Japanese Management and Coordination Agency. Using prices of shochu, whisky, beer, wine and sake, the household consumption expenditures, and the trend factor as seven explanatory variables, 16 equations were developed in order to explain the shochu consumption and the whisky consumption, respectively. The result was striking. Neither the impact of the whisky price on the shochu consumption nor the impact of the shochu price onthe whisky consumption was proven significant: the whisky price cannot logically explain the shochu demand; equations containing this variable lead to either the relationship that the higher is the shochu price, the larger is the shochu consumption or that of the higher is the whisky price, the lower is the shochu consumption. In contrast, prices of shochu and beer explain the shochu demand in a highly significant manner. Whisky consumption equations containing the shochu price as an explanatory variable either lead to the non-logical relationship that the higher the shochu price, the lower the whisky consumption, or result in a lower absolute value of t-statistics, or the lack of significance. In other words, shochu and whisky are not competing with each other in the Japanese market. Thus, the distinction between shochu and whisky in Japan should be less distortional to the market than the distinction between beer, wine and distilled liquors in the European markets. 4.86 The United States criticised the Japanese econometric study. The first point related to the conclusions of the Japanese study. Using the volume of shochu consumed as the dependent variable, the Japanese model found that shochu’s own price elasticity is positive and the cross-price elasticity with whisky is negative; it suggested that when the price of shochu goes up the volume consumed proportionately increases by two to three times as much, and when the price of whisky increases the consumption of shochu decreases. The t-statistics corresponding to the model’s estimated coefficients were greater than 2 or less than -2, meaning that the model as specified attaches a high degree of statistical significance to these results. Yet a finding that consumers react to an increase in the price of a product by increasing consumption is contrary to one of the fundamental tenets of microeconomic theory -- a downward sloping demand curve. No credence could be given to a model that states, with a high degree of statistical robustness, that demand for a good increases because its price increases; at a minimum such a counter-intuitive result casts serious doubts on the validity of the regression procedure. Secondly, looking further at the regression results in the Japanese study, the United States noted that although the model generally produced elasticities of the expected sign (i.c., negative own elasticity and positive cross-elasticity), the low t-statistics for the cross-elasticity estimates indicated that the variables, as specified, do not significantly explain movements in the consumption of whisky. The United States concluded that in no way does the model support the conclusions cited by Japan. To the contrary, in the US view, the results of the regression analysis only showed that the underlying model is mis-specified and the methodology is flawed. Thirdly, the United States pointed out the Japanese model’s failure to correct for basic problems in estimation of time-series models, such as serial correlation and auto-correlation. Because serial correlation biases the standard error of the tegression, a naive analysis will draw two erroneous conclusions: (1) the conclusion that the parameter estimates are more precise than they actually are -- the t-statistics will be higher than they should be in a correctly specified model, and (2) a high R? statistic that gives an overly optimistic picture of the success of the regression model in estimating relationships between variables. Japan had failed to use On the question of the cross-price elasticity of demand between alcoholic beverages, Japan had referred the Panel to a statistical analysis made by Bossard Consultants, Competition Between Difference Categories of Alcoholic Drinks (1994) which was commissioned by the European Commission in 1994. This study by Bossard Consultants resulted in a series of findings for the European markets: 1) When the price of wine rises 1%, the consumption of distilled liquors will increase 1.4%, 0.55-0.9% and 0.4%, respectively, in Spain, the United Kingdom and West Germany. 2) One percentage rise in the beer price will lead co the expansion of the distilled liquors by 1.3%, 1.2% and 0.9% respectively, in West Germany, the Netherlands and Denmark. WT/DS3/R WT/DS10/R WT/DS11/R Page 52 well-recognized techniques for correcting these problems such as “first-differencing” (using the change ina variable from the preceding period) or the Cochrane-Orcutt estimation procedure (used in the Bossard study referred to by Japan). In a correctly specified model, a trend variable would not be needed because the effect of increased consumption due to increased income would be picked up by the coefficient of the consumption function variable. An additional benefit of specifying the model in this form is that the coefficients of the price variables would be the actual own- and cross-price elasticities. Finally, the United States raised concerns with the Japanese data itself. The model utilized annual data since 1974. The implicit assumption that no other factors have affected the consumption of shochu or whisky (such as a change in consumer tastes) was unwarranted. For the United States, it would be preferable touse monthly or quarterly data over a shorter time period. Finally, Japan had failed to use real (i.e., inflation-adjusted) prices to avoid an inflationary bias. The United States concluded that the results of the regression study cited by Japan do not support the Japanese Government’s contention that whisky and shochu are not directly competitive or substitutable products. The methodology used in the study was flawed, differing from both the Bossard study and accepted econometric practice. The model should be re-estimated as outlined above, preferably utilizing more detailed consumption and pricing data. 4.87 Japan responded that this first criticism is equal to saying that it is improbable that shochu consumption is not explained by prices of shochu and whisky but by prices of shochu and beer. Similarly, in response to the Community’s criticism that it was improbable that many regressions did not yield a valid result whilst one variable less or more suddenly produced very strong results in one or two regressions, Japan argued that if the beer price is a real factor and the whisky price is a mere noise, “one variable less or more” should totally alter the result. According to Japan, what the Community and the United States are requesting is the result “any hypothesis can explain shochu consumption”, a result which is truly improbable. Japan submitted that it is a basic econometrics principle that the level of significance depends on the combination of explanatory variables. In the present analysis, the level of significance attained by an identical set of explanatory variables did not fluctuate widely depending on methods of conversion or estimation. For Japan, the aim of an econometric analysis is to examine the validity of hypotheses. If a hypothesis led to a finding contrary to accepted economic theory under any of the standard models and methods — linear, log-inverse, log-log, Cochrane-Orcutt, maximum likelihood -- and another led toa meaningful result, what should be rejected is not the models but the former hypothesis. Japan explained that it rejected the hypothesis that “the consumption of shochu is affected by the shochu price and the whisky price” because it produced a result contrary to accepted economic theory. It chose instead the “shochu price and the beer price” hypothesis, because the result accorded with general economic principles. In response to the US criticism that the Japanese study failed to follow the basic point of econometrics technique by failing to use log-linear transformation, which the Bossard study used, Japan noted that there is no reference in the Bossard study to the use of log-linear models, although its use of the linear models, log-log models and log-inverse models is noted. The Japanese study also used linear models, log-log models and log- inverse models, but did not use log-linear models. This is because log-linear models are regarded as more appropriate at the initial stage of introduction of new products into a given market, and are not effective for the analysis of established products such as liquor. Contrary to the US criticism that the Bossard study used the Cochrane-Orcutt method of estimation to correct for autocorrelation and that the Japanese study did not, Japan responded that the Japanese study used not only the ordinary least squares but the Cochrane-Orcutt method as well. Additionally, the maximum likelihood method, which is believed to be a better corrector for autocorrelation, was used. In response to the US criticism against Japan’s use of nominal prices, Japan submitted that the Bossard study used nominal price indices. Moreover, Japan submitted that reliability of conversion of nominal prices into real prices by the use of the Consumer Price index (CPI), for example, hinges on whether or not a change in the CPI affects consumption to the same extent as a change in the nominal price of the product at issue. If consumers tend to be influenced more heavily by a change in the price of the product than a change in the price WT/DS3/R WT/DS10/R WT/DSIL/R Page 53 level in general, nominal price indices are more reliable. Moreover, introduction of the CPI could add noise to the analysis because of the issues of weight given to product categories or the reference yeat. Since prices were generally stable during the years used for the study, it is more appropriate to explain consumer behaviour on the basis of nominal prices. Japan additionally submitted an analysis using real (deflated) prices, which, according to Japan, ted to similar results as the analysis using nominal prices. Japan also noted that according to the analysis submitted by the United States, which utilized real (deflated) prices and the technique of “first-differentiating”, “the annual price indices of whisky and shochu and annual household expenditures do not account for movements in the quantity consumed of these products”. Japan argued that this result also supports Japan’s rejection of the hypothesis that the prices of shochu and whisky affect the shochy consumption. Japan thus concluded that the criticism raised by the United States was off the mark, 4.88 The expert for the Community further challenged the Japanese econometric study. According to this expert, time series which are used in Japan study, are inherently difficult to analyze as they tend to be beset by at least three types of problems that make a statistical analysis difficult. Concerning the trends, this means that the development of the different variables is driven by factors that cannot be explained by an econometric model, but happen due to autonomous factors. For instance, consumption of a product may change simply due to fashion, economic growth, population growth, etc. When a large part of the changes in consumption is influenced by factors of this nature, it will be difficult to separate statistically the influence of price movements from that of trends. Concerning autocorrelation, this means that extraordinary influences on the variables in one year are likely to be present in the following year. An advertising campaign in one year that pushed up consumption of the product will also have an impact on consumption of the following year. If it is not corrected, autocorrelation decreases the precision of the econometric estimates, if it is corrected (e.g. by using the Cochrane-Orcutt method), it will reduce the number of available data points. Concerning multicollinearity, this means that for the case at hand, changes in the consumption of one type of liquor are related to changes in the consumption of another liquor. A hot summer, for instance, will increase the consumption of all beverages. Again, the results show how difficult itis to statistically separate the influence of one variable from that of another. According to the Community expert, the data set used for the Japanese statistical analysis is fraught with all three problems, which would make it difficult to prove any statistical connection between shochu and whisky consumption. In addition to these problems that occur naturally in time series analysis, the date set suffers from two further limitations: first, there are breaks in the data series. During the time period analyzed there have been several tax reforms that affected the structure of liquor prices and consumption, Such breaks in the data series have, as a consequence, that what is measured at the beginning of the time period is not identical with what is measured at the end of atime period. Second, there is illegitimate aggregation of product groups. There are major shifts within a product group such as whisky. Although domestic and foreign whisky are inherently close products, the aggregation into one category hides the fact that up to 1989 the tax system treated them differently. Because the tax reform has affected domestic and foreign whisky in an opposite ditection, an aggregation of the two types will lead to misleading statistical results. The Community expert submitted also that the importance of the tax reform of 1989 can be seen graphically. As the reform made domestic whisky more expensive and imported whisky less expensive, for 1989 and 1990 the statistics show a major increase in imported and a decrease in domestic whisky. After this period, the two whisky varieties start moving in line with each other. This is to be expected as they are identical products and are affected by the same external factors. For the Community, the inspection of the graph therefore leads to two evident conclusions: Firstly, it can be seen very clearly that a reduction in tax for imported whisky has a substantial impact on its consumption. Secondly, any statistical analysis that fails to take account of the structural break in the time series will come to misleading conclusions, because what is measured with the aggregate whisky consumption before and after 1989 is evidently not thesame. Spurious statistical results are the natural outcome. According to the Community expert, WT/DS8/R WT/DSIO/R WT/DS1UR Page 56 These large tax differences have no option for whisky/brandy but to set its sale price relatively higher, and this leads not only to decrease in consumers’ support to whisky/brandy , but also te narrow consumers’ freedom of choice. As a result, coupled with the present consumers’ price-oriented attitude, the consumption of whisky/brandy has remarkably declined” Thus, Canada argued, the lower the price of imported whisky the greater its competitiveness with shochu an alternative choice and consequently the greater its domestic consumption. Canada concluded that consumer choice between shochu and whisky is price responsive. 4.93 In addition to their arguments and counter arguments suggested in the present section, the Community, Canada, the United States and Japan submitted further evidence on cross-price elasticity of shochu and other imported distilled spirits in their discussions of the application of the aim-and-effect test, when arguing, more specifically, whether the Liquor Tax Law has the effect of distorting the competitive relationship between shochu and other imported liquors so as to afford protection (see Sections F and G below). bd) The Second Step of the Test suggested by the Community for Article III:2, Second Sentence: “ ... So as to Afford Protection” 4.94 Astothe second step of the legal test it suggested for the second sentence of GATT Article II:2 in assessing whether a measure imposed on substitutable or directly competitive products is “so as to afford protection”, the Community reiterated that the following criteria may be relevant in order to determine whether a difference in taxation is “so as to afford protection” to domestic production: 1) The level of the tax differential (but contrary to the first sentence of Article [II:2, a tax difference does not lead automatically to a violation of the second sentence of Article IJI:2); 2) The degree of substitutability and competition between the two products; 3} Whether the less taxed product is produced in other countries. In this context the Community recalled the conclusion of the 1987 Panel Report which found that the following factors were sufficient evidence of fiscal distortions between imported distilled liquors and domestic shochu affording protection to the domestic production of shochu: - the considerably lower specific tax rates on shochu than on imported whiskies, brandies and other spirits ...; - the imposition of high ad valorem taxes on imported whiskies, brandies and other spirits and the absence of ad valorem taxes on shochu; ~ the fact that shochu was almost exclusively produced in Japan and that the lower taxation of shochu “did afford protection to domestic production” (Article III: 1) rather than to the production of a product produced in many countries (say butter) in relation to another product (say oleo margarine as in the example referred to by Japan ...); - the nmitual substitutability of these distilled liquors, as illustrated by the increasing imports into Japan of Western-style distilled liquors and by the consumer use of shochu blended in various proportions with whisky, brandy or other drinks. 4.95 For the Community, the above factors are still present and, therefore, continue to warrant the conclusion that the Liquor Tax Law affords protection to the Japanese domestic production of shochu: WT/DS8/R WT/DS10/R WT/DS11/R Page 57 (i) Despite the 1989 and the 1994 tax reforms, the tax rates on shochu A and shochu B are still much lower than the rates on “spirits”, “whisky/brandy” and “liqueurs”. The taxes on shochu are from 2.45 to 9.6 times lower in terms of rates per litre of beverage and from 2 to 6 times lower in terms of rates per litre of pure alcohol and these differences can thus hardly be considered as de minimis. Even though the tax differentials have been reduced in absolute terms since the adoption of the 1987 Panel Report, their protectionist effect has actually become more acute in the context of the current recessionary economy which has made Japanese consumers much more price sensitive. @Q Shochu continues to be produced almost exclusively in Japan. In 1994 imports of shochu represented 1.7 per cent of the total sales of shochu and barely 1 per cent of the total sales of distilled spirits and “authentic liqueurs”. In contrast, during the same year, imports from third countries accounted for 27 per cent of the total sales of whisky, 29 per cent of the total sales of brandy, 18 per cent of the total sales of “spirits” and 78 per cent of the total sales of “authentic liqueurs”. Sales of domestically produced shochu account for almost 80 per cent of the total sales of domestically produced distilled spirits and “authentic liqueurs”. Thus, by affording protection to shochu, Japan is in fact affording protection to the majority of its domestic production of spirits and liqueurs. (3) Shochu and other imported liquors are mutually substitutable as evidenced by their cross-price elasticity, argued in paragraphs 4.82 and following above in the Community’s discussion of the first step of the legal test it suggested for the second sentence of Article IIT:2. The Community also recalled that since Article I1I:2 protects trade expectations on the competitive relationship between imported and domestic products rather than expectations on trade volumes, it is not necessary, in order to establish a violation of Article IIL:2, second sentence, to show that the difference in taxation has had an actual effect on the volume of trade. 4.96 Japan responded to the Community’s arguments on the three criteria. First, concerning the potential protective effect, Japan submitted that the tax differential should be measured on the basis of the tax/price ratio, as it is a criterion to judge whether or not a tax affords protection, and for Japan, there is no differential in the tax/price ratios. Secondly, for Japan, shochu and other distilled liquors do not show the aptitude of the two products to serve the same uses, and differ in the extent and the form in which the two products are available to the public, beyond what would apply to all alcoholic beverages. Cross-price elasticity of demand does not, therefore, exist. If a directly competitive or substitutable relationship were to be found in this case, it would have to be found between all alcoholic beverages, and, consequently, any liquor taxation currently in force would become inconsistent with Article III, unless all products show the same tax/price ratio. The degree of substitutability and competition between the products is minimal at best. Third, shochu is widely produced in Asian countries, and the third criterion is not met. Thus, Japan concluded that if the Community’s interpretation is applied to the facts, one inevitably zeaches a conclusion that Japan’s liquor tax is consistent with Article III:2, second sentence. 4.97 Japan argued that the Community is criticizing Japan’s tax distinction among distilled liquors while dividing wine into six categories in its liquor tax directive and legitimizing Germany’s application of four completely different rates to categories of wines. For Japan, a position which holds that champagne and sherry may be distinguished from other wine while shochu and whisky should be treated alike, is equal to turning Article III into an instrument of harmonization of internal taxes with a system of a particular group of countries. Japan reiterated that the purpose of Article III is not to require Members to adopt a particular system of taxes or regulations, nor to harmonize taxation systems. Japan argued that only a small number of WTO Members apply a flat rate to all categories of distilled liquors and a larger number of Members apply more than one rate in one way or another. In Japan’s view, WT/DS38/R WT/DSLOR WT/DSLIR Page 58 the conclusion advocated by the Community in the present case would substantially affect other countries as well. Japan referred the Panel to its chart on worldwide “Taxation on Distilled Liquors” (see Annex HD. 4.98 The Community agreed that Article III does not require Members to adopt a particular system of taxes or regulations, or to harmonize with a system of a particular Member. Nonetheless, the Community believed that the fiscal autonomy of Members is limited by their obligation under GATT to afford imported products equal conditions of competition, inter alia, with respect to internal taxes. The Community was not requesting Japan to introduce any particular system of taxation or to set the tax rates for distilled spirits at any particular level. All the Community was asking from Japan was that shochu and ail other like and substitutable or directly competitive distilled spirits not be taxed in a discriminatory or protectionist manner. For the rest, the Community submitted that Japan retains full autonomy to choose its own tax system. Following the adoption of a panel report upholding the Community’s claims, Japan would enjoy complete discretion to decide whether to maintain the current system of specific taxes or to replace it by, for example, a system of ad valorem excise taxes or by a system of ad valorem consumption taxes or by a mixed system combining specific and ad valorem taxes. The Community reiterated that Japan also retained complete freedom to decide not to impose any tax at all on distilled spirits. Japan would also be able to choose the level of the tax rates. If, for example, Japan decided to maintain a system of specific excise taxes, the rates for all distilled spirits could be set at or above the current level for “whisky/brandy” (the highest) or at or below the current level for shochu B (the lowest), as well as at any level in between. As it stands, the Liquor Tax Law is inconsistent with the second sentence of Article III:2. The Community also added that the taxation systems of countries other than Japan are not covered by the terms of reference of the present dispute. e) Application of the Second, Third and Fourth Criteria of the Legal Test Suggested by Canada for Article III:2, Second Sentence 4.99 In its discussion of its legal test for the second sentence of Article III:2, Camada also referred the Panel to the 1987 Panel Report which set out four factors that established “sufficient evidence of fiscal distortions of the competitive relationship between imported distilled liquors and domestic shochu” so as to “afford protection to the domestic production of shochu”: (1) the considerably lower specific tax rates on shochu than on imported whisky; (2) the almost exclusive production in Japan of shochu; (3) the mutual substitutability of distilled liquors like whisky and shochu as illustrated by the increasing imports into Japan of “Western-style” distilled liquors; and (4) the imposition of an ad valorem tax on imported. whisky but not on shochu. 4.100 For Canada, the application of the three relevant criteria (omitting the fourth one) mentioned above to the current Liquor Tax Law confirms its inconsistency with the second sentence of Article III:2, in distorting the competitive relationship between Canadian whisky and domestically produced shochu: aw Canada argued that even a cursory examination of the Liquor Tax Law shows that whisky and shochu are not similarly taxed. The reference tax rate for whisky is set 6.3 times higher than the reference tax rate for shochu A and 9.6 times that for shochu B. And even the Lowest tax rate on whisky is still 2.5 times higher than the maximum tax rate on shochu. More importantly, Canada argued that even using Japan’s suggested basis for analysis -- the tax/price ratio which is the proportion of the retail price of whisky and shochu represented by the liquor tax levied under the Liquor Tax Law -- the evidence does not support Japan’s claim that the tax differential between distilled liquors under the Liquor Tax Law yields “roughly” equivalent tax/price ratios. Moreover, the evidence submitted by Japan is based on manufacturers’ suggested retail prices, not actual retail prices. Yet in the Japanese distilled liquor market, suggested retail prices are merely notional figures that bear little relationship WT/DS8/R WT/DS10/R WT/DS11/R Page 61 Of the presumption of “an adverse impact”, the panel report in the Superfund” case and the 1987 Panel Report”! enunciate the principle that overall increases in market share for imports of the products in issue does not constitute a rebuttal. Indeed, the Superfund case articulates the principle that a finding of fiscal distortion in the competitive relationship between imported and domestic products constitutes an “ircefutable presumption” of nullification and impairment of benefits. The 1987 Panel Report determined that the factors set out in paragraph 4.94 above ipso facto constitute “sufficient evidence of fiscal distortions of the competitive relationship between imported distilled liquors and domestic shochu”. In view of the conclusive evidence described in paragraphs 4.72 to 4.93, including, more specifically , the positive evidence of cross-price elasticity between shochu and other imported distilled liquors, it necessarily follows, in Canada’s view, that the Liquor Tax Law distorts the competitive relationship between imported distilled liquors and domestically produced shochu, and therefore is inconsistent with the second sentence of Article III:2. Canada noted that indeed, Japan’s Deregulation Subcommittee of the Administrative Reform Council, an independent advisory body whose members are appointed by Japan’s Prime Minister and approved by the Diet, stated that the tax rates under the Liquor Tax Law constitute “virtual restrictions on the buyer's activities” and are not “neutral in relation to consumer choice”. 4.105 Japan responded that the criterion unique to Canada’s test is distortion of the competitive relationship and that it is noteworthy that Canada’s argument rests on distortion of relative prices. Accurate data show, however, that the Liquor Tax Law does not distort relative prices between whisky and shochu, and, accordingly, this criterion of distortive effects is not met. In order to reach a finding of inconsistency with Article III, all of Canada’s criteria would have to be met, but it was not the case, since Canada had not proven that shochu and Canadian whisky are directly competitive, that shochu and Canadian whisky are not similarly taxed, or that the Liquor Tax Law affords protection to domestic production. For Japan, Canada had, therefore, not proven that the Liquor Tax Law is inconsistent with Article III:2, even under Canada’s interpretation of the provision. F. Application to the Present Case of the Legal Analysis Suggested by the United States for the Interpretation of Article III:2 4.106 Asargued in paragraphs 4.24 to 4.32 above, the United States submitted that the central concern of Article III is to prohibit the targeting of imports and suggested that application to the Liquor Tax Law of the aim-and-effect test of earlier panel reports would confirm the inconsistency of that measure with the provisions of Article III:2, second sentence, in that the regulatory distinctions made by the legislation are so as to afford protection. 1. The Aim of the Legislation 4.107 The United States argued that the protective aim of the Liquor Tax Law structure is apparent from (1) the stated policy objective and whether it was known at the time the legislation was enacted that it would draw a line between one group of products that would be foreign and another group that *Thus, in the Superfund case, para. 5.19, the panel stated: “A demonstration that a measure inconsistent with Article 1:2, first sentence, has no or insignificant effects would therefore in the view of the Panel not be a sufficient demonstration that the benefits accruing under the provision had not been nullified or impaired even if such rebuttal were in principle permitted”. "The 1987 Panel Report stated in para. 5.16: “ [AIn increase in imports could not refute the presumption that discriminatory ‘or protective taxes inconsistent with Article [1:2 had impaired the competitive benefits protected under Article III:2 because, inter alia, an increase in imports did say nothing about what the trade might have been in the absence of the inconsistent trade restrictions”. WT/DS8/R WT/DS10/R WT/DS11/R Page 62 would be domestic (ex-ante knowledge), (2) the internal inconsistencies of the legislation and its structural incentives, (3) legislative statements and the preparatory work, as well as from (4) the arbitrary and irrational categories of the legislation under scrutiny. The United States continued by stating that: (ly During the consultations, the Japanese Government asserted that the policy objective of the Liquor Tax Law system was to maximize tax revenue while ensuring that the tax is distributed among consumers in accordance with their “tax-bearing ability”. However, this objective is nowhere stated in the law, which has no general statement of purpose other than “Taxes shall be imposed on alcoholic beverages in accordance with this law”. The taxes provided for by the Liquor Tax Law are specific taxes, with no link between the tax rate and the actual price of the alcoholic beverage in question; their structure does not support the claim that they are designed to effectuate equity between categories of spirits. To base tax rates on consumers’ tax-bearing ability assumes that some products are consumed by the masses and should be low-priced, and other products are exotic luxuries consumed by the rich who can afford to be taxed heavily; this proposition was specifically rejected by the 1987 Panel Report. For the United States, statements connected with the 1994 revision of the Liquor Tax Law also offer a sample of the motivations behind enactment of this legislation. The official records of deliberations inthe Finance Committee of the Diet in March 1994 show that Ministry of Finance Tax Bureau Director Ogawa testified that the reason for the difference in tax treatment was “out of consideration for the higher materia! costs etc” of shochu B. He also testified that particular attention had been made to coordinate the tax increases with the increased costs of raw materials associated with factors such as the poor rice harvest in the case of refined sake and shochu, especially shochu B. The legislation raising taxes included as well an extension of tax reductions for small-volume producers of shochu A and B, and provision for a subsidy fund for shochu producers. The package in context demonstrates that the operative consideration in passing the legislation was the economic well-being of domestic shochu producers, not a neutral tax policy. 2) According to an article in a Ministry of Finance publication written by one of the Ministry drafters explaining the 1962 revisions,” the definitions were changed at that time in order to clarify and reinforce the distinction between shochu, whisky, brandy and spirits. The purpose of the change and the related exception was (a) to exclude certain products which would be classified as whisky, brandy, and spirits, but since dates were already being used as a raw material for shochu in Japan, these would be permitted as a fruit raw material for shochu; (6) to exclude vodka; (c} to exclude rum from the category of shochu, but permit Okinawan awamori made with barrel molasses to remain as shochu; (d) to exclude gin and similar genever-type drinks. Q) The lack of any policy rationale other than protection is apparent from the otherwise-arbitrary distinctions drawn in the product categories. The only difference between vodka and shochu A is that according to the definition in the Liquor Tax Law, shochu A cannot be filtered with white birch charcoal, although it can be filtered with any other material. Yet the tax rate on vodka is 2.55 times higher than the tax rate on shochu A, The Japanese government has never claimed that the ban on the use of white birch charcoal in filtering shochu was based on health reasons or any other policy, Thus the distinction cannot have any purpose other than excluding imported vodka from the tax benefits granted to the producers of shochu. (4) It is also arbitrary to set the maximum alcchol content for shochu made by continuous distillation methods (shochu A) at 36 per cent and the maximum alcohol content for shochu distilled otherwise 72Tan Hirosho, “Shuzeiho to no ichibu o kaisei suru horitsu” (The Law Partially Revising the Liquor Tax Law), in Zeisei Tsushin (Tax Policy News), June:1962, p. 23ff. The article identifies the author as the Deputy Director of the Ministry of Finance, Second Tax Policy Division. WT/DS8/R WT/DS10/R WT/DSIIR Page 63 (shochu B) at 45 per cent. All alcoholic beverages falling within the categories of “shochu”, “whisky/brandy” and “spirits” are classified as “liqueurs” and taxed at a uniform tate whenever they are pre-mixed with a sugared non-alcoholic beverage. However, the same alcoholic beverages, when sold undiluted, are classified within different tax categories and taxed at widely differing rates, even though they are often consumed in home-made mixes made with similar non-alcoholic beverages. Again, in the US view, the Japanese government has claimed no policy justification for this difference in taxation. The only rational explanation for it is that pre-mixes, unlike undiluted alcoholic beverages, are produced almost exclusively in Japan. For the United States, the arbitrariness of the distinction drawn between “spirits” and shochu can be seen in the recent move by Suntory, the producer of “Juhyo” brand vodka, to re-characterize it as shochu A. Before June 1993, Juhyo was sold as vodka, and accounted for almost half of Japanese vodka production. After June 1993, Suntory ceased using birch charcoal as a filtering material, and began selling Juhyo as shochu A, simply in order to reduce the tax burden on the product. Suntory was then able to, and did, reduce the retail price of Juhyo. Of course, because of the substantial tariffs on shochu, it is not possible for foreign vodka producers to do the same. Thus, forthe United States, the distinction drawn by the system of Japanese liquor taxation between shochu and all other distilled spirits is arbitrary and contrived. 4.108 Japan responded that the complaining parties seemed to confuse the present Japanese policy, as explained in the bilateral consultation, with that of 1987. Japan argued that it had not referred to the notion of the tax-bearing ability in bilateral consultation. The essence of the policy in 1987 was: “Since whisky consumers have a greater tax-bearing ability than shochu consumers, the tax/price ratio ought to be higher for whisky than for shochu.” In contrast, the present tax policy since the 1989 amendment is: “Tax/price ratio should be roughly constant between whisky and shochu for the sake of ensuring neutrality to consumers’ choice and of equity in between consumers of these products”. Examining excise taxes in view of the three criteria of neutrality, horizontal equity and vertical equity is common practice among tax authorities in the world, though which of the three is prioritized may differ according to prevailing socio-economic conditions: for example, the report on excise taxes issued by the United States’ Congressional Budget Office in 1990 starts its discussion with the examination of the three criteria. The lack of statements of policy goals in the Liquor Tax Law is only a standard practice of tax legislation in Japan, and, for example, introducing the 1989 amendment before the National Diet, the Minister of Finance stated, “The fundamental principles of the present amendment are to ensure equity in distribution of the tax burden and to maintain neutrality toward economic activities”. Japan further argued that the evidence cited by the United States is part of the record of Diet deliberations of the 1994 amendment. The amendments cannot conceivably have had a protective intent, however; it raised the tax rate for shochu A by 30 per cent, that for shochu B by 44 per cent, while raising the tax on “spirits” by a mere 11 per cent, but maintained the tax on whisky/brandy at the same level. The 1962 material written by the person who prepared the Liquor Tax Law merely referred to the problem common to any product classification and so did other alleged evidence of arbitrariness. The amendments of 1989 and 1994 which substantially raised the tax on shochu B refute, ipso facto, the allegation of policy distortion by local political forces. This led Japan to conclude that speculative inside stories are not appropriate as the basis for panel findings. Most whisky, brandy and spirits consumed in Japan are manufactured focally. For example, the rate of domestic production of whisky is 75 per cent, that of brandy, 72 per cent, and that of “spirits”, 82 per cent. These categories cannot be equated with imports as such. Categorization of these products, therefore, is not the targeting of imports as the United States claims. The tariff on shochu (currently £7.9 per cent, same as that on vodka and lower than that on mim) is irrelevant to the issues of Article III. Moreover, for Japan, the categories are not exceptional or arbitrary. It is arbitrary, according to the US submission, to distinguish vodka from shochu on the basis of filtration with white birch charcoal, but Japanresponded that any legal definition of a product encounters similar difficulty in translating a socially accepted concept. It is no more indicative of protective intent than the Community definition of sparkling wine WT/DS3/R WT/DS10/R WT/DS1U/R Page 66 substitutability of distilled liquors, as demonstrated by the increasing imports into Japan of distilled spirits and the consumer use of shochu in mixed drinks. For the United States an examination of the effect of the Liquor Tax Law in this present case should focus on qualitative alteration of conditions of competition such as targeting of imports, and evidence of cross-elasticity of demand between the favoured and disfavoured categories. 4.114 The United States pointed out that shochu consumed in Japan continues to be made almost exclusively in Japan. In 1994, imports of shochu were 1.7 per cent of total sales and 1 per cent of total sales of distilled spirits and “authentic distilled spirits”. Also, in 1994, imports from third countries accounted for 27 per cent of the total sales of whisky, 29 per cent of the total sales of brandy, 18 per cent of the total sales of spirits and 78 per cent of the total sales of “authentic liqueurs”. At the same time, domestically-made shochu accounted for over 80 per cent of all domestic sales of distilled spirits and authentic liqueurs. Thus, the protection given to shochu has had the effect of protection for domestic production. 4.115 On the market shares of shochu and the price-cress elasticity of shochu, the United States, in addition to the arguments detailed in paragraphs 4.82 to 4.93 above, noted that there were clear indications that the demand for shochu is largely influenced by fluctuations in demand for other distilled spirits and liqueurs. This could be seen in the rearrangement of the market place for distilled spirits after the 1989 tax reform. The 1989 reform unified tax rates on whisky, abolished the classification of whisky into three classes, and consequently more than tripled the tax rate on second-class whisky while lowering the taxes on other whisky, authentic liqueurs and spirits. The 1989 law also raised the tax on shochu by a small amount. In particular the United States submitted that: . Retail prices for second-class whisky almost doubled, and the market share for domestic whisky declined from 27 per cent in 1988 to 19.6 per cent in 1990. This trend has continued: in 1994 the market share of domestic whisky sank further, to only 13.2 per cent. Shochu makers were able to move into the place in the market formerly held by second-class whisky. Sales of shochu have steadily increased and reached 74.2 per cent of distilled spirits in 1994. - The prices of imported whisky, liqueurs and spirits declined and their sales rose. However, Japan entered a recession in 1992. The highest-taxed categories, whisky/brandy, authentic liqueurs and spirits, were hit worst and have lost sales both relatively and absolutely since 1992, while the market share of shochu continues to grow at their expense. - Because the prices of shochu and other distilled spirits have partially converged, their cross- elasticity of demand has risen. - Shochu continues to be made almost exclusively in Japan. In 1994, imports of shochu were 1.7 per cent of total sales and 1 per cent of total sales of distilled spirits and “authentic distilled spirits”. Also in 1994, imports from third countries accounted for 27 per cent of the total sales of whisky, 29 per cent of the total sales of brandy, 18 per cent of the total sales of spirits and 78 per cent of the total sales of “authentic liqueurs”. At the same time domestically-made shochu accounted for over 80 per cent of all domestic sales of distilled spirits and authentic liqueurs. Thus, the protection given to shochu has had the effect of protection for domestic production. 4.116 In support of its allegation of the protective effect of the Liquor Tax Law, the United States argued that there is a sudden or dramatic difference in rates at the margin. The rates on shochu A and B are still much lower than the rates on other distilled spirits. The United States noted that there WT/DS8/R WT/DS10R WT/DS1IR Page 67 is still a 9.6 to 1 tax differential between whisky and shochu B, a 6.3 to 1 differential between whisky and shochu A, and a 2.4 to 1 differential between spirits and shochu A, at the respective reference values for each. The United States added that the Liquor Tax Law targets inherent characteristics of the product. Foreign manufacturers of gin or rum cannot make a product that has access to the lower tax rate without changing the nature of the product.” 4.117 The United States further argued that the protective effect of the tax distinction can also be seen in the attempts of Japanese shochu producers to make their products resemble whisky or to emphasize the points of similarity between the raw materials, ingredients, manufacturing process, appearance and tradition of their shochu and whisky or brandy. In May 1988, the shochu manufacturer Takara began to market “Jun Legend”, a light amber coloured brand of shochu produced by blending two types of alcohol distilled from barley and corn and aging them in charred oak barrels for one to five years. Takara’s claim was that “the most noticeable characteristic of this brand is a flavour and taste similar to whisky”. When the new brand was launched, Takara announced its expectation that the new brand would appeal to former consumers of second-class whisky, the tax rates on which were expected to increase with unification of tax rates on all classes of whisky. The favourable treatment of shochu is a classic instance of use of a tax system to perpetuate existing consumer preferences. Foreign manufacturers of gin or rum cannot make a product that has access to the lower tax rate accorded. to shochu unless they change the nature of the product. Therefore, the United States concluded that the regulatory tax distinction made by the Liquor Tax Law between shochu and other imported liquor targets imports and has the aim and effect so as to afford protection, in contravention of the second. sentence of Article III:2 (as well as the first sentence of Article III:2 for which the United States suggested the same test). 4.118 Japan responded to the US claim that the Liquor Tax Law had a protective effect. For Japan, among the four reasons why the 1987 Panel Report found Japan’s liquor tax protective, the first one (large specific tax rate differential) is non-existent now under the appropriate yardstick of comparison, the second one (ad valorem tax on imported liquors) is already abolished, the third one (exclusive production of shochu in Japan) is incorrect, and the last one (increased imports of western style liquors, and consumer use of shochu blended with whisky etc.) is a combination of an irrelevant fact and an incorrect assumption. As argued with the Community, for Japan, “directly competitive or substitutable relationship” between domestic shochu and imported distilled liquors of other categories is not proven. Alleged “sudden or dramatic difference in rates at the margin” is common to many tax systems including taxes on wine in the US and in the Community, and is not evidence of protectiveness. The product “Jun Legend” is a case of experiment at the margin which, Japan argued, accompanies any product categories with tax differentials. Shochu, for which Japan is only the second or third largest producer in the world, is not an “inherently domestic product”, and whisky, for which Japan is the fifth largest producer in the world, is not an “inherently foreign product”. For Japan, none of the US criteria are met, The United States added that this is fundamentally different from, for example, a tax incentive for adding catalytic converters to automobiles: an automobile can qualify for the incentive without changing its nature. WT/DS8/R WT/DS10/R WT/DSIU/R Page 68 G. Application to the Present Case of the Legal Analysis suggested by Japan for the Interpretation of Article If1:2 4.119 Japan submitted that the Liquor Tax Law generally and the very nature of its tax classification system do not have the aim or the effect “so as to afford protection”. For Japan, the absence of Protective aims and effect of the regulatory distinction contained in the Liquor Tax Law, confirms that shochu and other imported liquors are not like products and that the legislation is not inconsistent with Article IIE:2, first and second sentences. Japan submitted three criteria to demonstrate that the regulatory distinction of the Liquor Tax Law is consistent with Article II:2: a) the nature of the categorization, b) the aims of the legislation, and more particularly the new policies of horizontal equity and neutrality, and c) the absence of protective effect evidenced by the fact that there is no competitive relationship between shochu and other imported liquors (no cross-price elasticity), the fact that shochu is produced abroad and the very neutrality of the Liquor Tax Law. 1. Categorization of the Liquor Tax Law 4.120 Japan submitted that shochu is readily distinguishable from the rest of the distilled liquors and these differences have resulted in different net-of-tax prices in relation te which tax rates are adjusted according to the tax categories of the Liquor Tax Law. Indeed, the categorization of distilled liquors under the Liquor Tax Law is not protectionist because it is based on three objective criteria: (i) the cost of the raw materials; (ii) the alcoholic strength; and (iii) the value added through the post distillation process. More specifically, the distinctions between the categories are the following: - shochu A and B have a low alcohol content, are produced from inexpensive materials and are “normally consumed without post distillation processing” ; + “spirits” are also produced ftom inexpensive raw materials but have a high alcohol content as well as a “higher value added in post-distillation process” (in the case of vodka, through a “specialized filtering process”; in the case of rum, through aging; and in the case of gin “through the flavour-adding process”); - “whisky/brandy” has a high alcohol content and is produced from expensive raw materials. Moreover it has a “higher added value in post-distillation” (through “wooden cask aging”); - “liqueur” , as described by the Community: “[T]his category includes two well differentiated groups of products: on the one hand, so-called single item liqueurs or authentic liqueurs with a relatively high alcohol content (often 40 per cent, although in some cases it may be only 16 per cent - 24 per cent) such as brandy liqueurs, orange liqueurs, anisette, cream liqueurs, emulsion liqueurs and certain bitters; and, on the other hand, cocktails and sparkling pre-mixes combining one or more liquors with non-alcoholic beverages and with an overall alcohol content between 4 per cent and 12 per cent”. The two groups of products, which previously were two separate tax categories, were merged into a single tax category of “liqueur” by the 1989 amendment because of the recommendation of the 1987 Panel Report. WT/DS8/R WT/DS1I0/R WT/DSIUR Page 71 as in the Liquor Tax Directive of the Community. For Japan, the Community’s argument against the Japanese criteria does not indicate arbitrariness inherent in the Liquor Tax Law, but rather the difficulty common to all product definitions. For Japan, despite allegations by other parties to the present dispute, there is nothing arbitrary in identifying these popular categories according to socially accepted concepts. 4.122 Japan then responded to another alleged arbitrariness in the Liquor Tax Law which is the ceiling on the alcoholic strength of shochu A (35 per cent) and B (45 per cent). The perceived arbitrariness, however, reflects the difficulty inherent in defining a historically developed concept. As an illustration of such common difficulty, Japan noted the following minimum alcoholic strengths for various products under the European definitions: 40% whisky, pastis 37.5 % rum, gin, vodka, ouzo, Kornbrand 36 % brandy 35 % grain spirit 32% Korn 30 % caraway-flavoured spirit drinks 25 % fruit spirit drinks 15% aniseed-flavoured spirit drinks For Japan, these floor rates are by no means arbitrary; they merely capture basic features of historically developed and socially accepted concepts. As stated earlier, the relatively low alcoholic ceiling for shochu reflects the fundamental features of this type of low-cost alcoholic beverage. The Liquor Tax Law does leave room for experimentation at the margin, as demonstrated by examples of Juhyo vodka or Jun Legend. Japan insisted that this is not inherent in the Japanese taxation system. As long as there is a tax differential between product categories, the opportunity for experimentation arises. This is an issue of optimal tax policy and not of GATT rules. Japan argued, for example, that Juhyo vodka contained only 20 per cent alcohol and would not meet the criteria of alcoholic strength for vodka in either Europe (37.5 per cent) or the United States (40 per cent). Although the filtering method of vodka was used, it was shochu in terms of alcoholic strength. This borderline product was consumed in much the same way as shochu and was subsequently reformulated into genuine shochu with alteration of raw materials to give a typical shochu taste. 4,123 For the Community, neither the Liquor Tax Law nor economic reality bear out Japan’s exercise of ex-post facto rationalization since - alcoholic strength is nota classification criterion under the Liquor Tax Law. Alcoholic strength is not an element of the legal definitions of either “whisky” or “spirits”. These categories may be manufactured and sold at any strength. On the other hand, the maximum legal strength of shochu A and of shochu B is 36 per cent and 45 per cent which can hardly be described as low alcohol content. Therefore, in the Community’s view, Japan cannot claim that differences in tax rates are based in differences in alcohol content. : the cost of a particular raw material may vary considerably from one country to the other as well as seasonally. Grapes, for instance, may be an expensive product in Japan, where there is hardly any production, but not in the Mediterranean region, where they are commonly used to produce industrial alcohol. Dates (one of the products that may be legally used in the manufacture of shochu) are not necessarily less expensive than malt. The Community argued that the comparison of the raw material costs is totally irrelevant since the price of tice fluctuates according to other tariff and non tariff barriers. In any event, there is no indication that the WT/DS&/R WT/DS1O/R WT/DS1U/R Page 72 rates for shochu are regularly adjusted so as to take account of the movements in the price of rice in the Chicago futures market. The minutes of the Diet’s debate on the 1994 tax reform rather suggest that the rates on shochu are only adjusted to take into account the results of the tice harvest in Japan. - aging, whether in wooden casks or in other containers, is not an element of the legal definition of either whisky or ram. On the other hand, the definition of shochu does not exclude the aging of this product. Therefore, Japan cannot claim that aging is a valid criterion for applying different tax rates. As regards other white spirits, one may wonder what is the value added by filtering vodka through charcoal of white birch, instead of any other material, or by adding to gin some flavouring substances. Do any of these manufacturing processes have the effect of multiplying the value of the liquor concerned by 3.22 (the current difference in taxation between “spirits” and shochu B)? The Community recalled that according tc Japan, shochu A and shochu B share the same characteristics: both have a low alcohol content; both are made from inexpensive raw materials; and both have “a lower value added in post-distillation process”. However, the tax rate on shochu A is 1.52 times higher than on shochu B. 4.124 In response to the Community’s argument, Japan pointed out that similar differentials in net of tax prices between categories do exist in countries which apply a flat-rate tax on distilled liquors as well: For Japan, the Community claimed similarity of prices of brandy and gin based on the promotional leaflet of a Belgian supermarket, picking the least expensive item out of five brandy brands advertised in the leaflet (armagnac, calvados and cognac) and the most expensive gin brand from the six on the leaflet (gin and geneva, most expensive when adjusted for differences in bottle size). Japan called it a classical case of a selective reference, saying that calculation of the average net-of-tax prices of all items appearing on the leaflet results in a significant price differential. Against the Community’s argument that grapes are very inexpensive and are used to produce industrial alcohol in the Mediterranean area, Japan questioned why is it then that brandy brands are more expensive than other distilled liquors in the Belgian supermarket. Moreover, according to a survey done by Business International, Inc., an affiliate of The Economist magazine, brandy is more expensive than whisky and whisky is more expensive than gin in 32 cities of the world where a flat tax rate is applied. Against the Community’s claim that no category of distilled spirits is inherently more expensive than others, Japan questioned why is it then that such price differentials prevail across the world. For Japan, the Community tends to ignore the fact that differences in the market value of distilled liquors correspond to subtle product differences. 4.125 The Community reiterated that the prices and taxation systems of whisky in countries other than Japan are outside the terms of reference of the present dispute. 4.126 The Community also referred the Panel to a recent recommendation of the Administrative Reform Council, a public law advisory body attached to Japan’s Prime Minister’s office, which passed the following judgement on the alleged rationality of the Liquor Tax Law: “The current liquor tax law divides spirits into 16 products, each with a different tax rate. Even if there were reasons for setting different tax rates for each individual liquor in the past, it may not be possible to provide a consistent explanation to justify those Teasons and the liquor tax structure as a whole at the present time. WT/DS3/R WT/DSLO/R WT/DS11/R Page 73 ... Therefore, it would be advisable, upon determining the categories and rates, to ensure that the rationale behind the decisions is logical and can be easily understood by consumers as well”. 4.127 Japan argued that the above translation provided by the Community and Canada, which says “the Liquor Tax Law classifies spirits into 16 products” deviates sharply from the original which says “The current liquor tax law divides liquors into 16 categories”. The Administrative Reform Committee’s comment concerns the taxation of alcoholic beverages as a whole, and contains no specific reference to distilled liquors, Japan admitted that there is still room for improvement in the tax as a whole, including taxation of brewed beverages. For example, sake and wine are both brewed and have asimilar strength. Yet the specific tax rate, and the tax/price ratio, of sake is more than twice the rate or ratio of wine. Japan argued that the Committee's critique of shortcomings in Japan’s taxation of alcoholic beverages as a whole would be in fact valid for any liquor tax in the world, citing the United States Congressionat Budget Office’s report on the United States excise taxes and the European Commission’s report on taxes on alcoholic beverages in Europe. It is true that there is room for improvement; but no tax is perfect. 4,128 The Community and Canada responded that the prices and taxation systems of liquors in countries other than Japan were outside the terms of reference of the present dispute. 4.129 Japan continued its explanation of the neutrality of the categorization in stating that the tax rates are set corresponding to the average net-of-tax price of each category. While net-of-tax prices vary from one category to another, the ratio of the tax over the retail price stays roughly constant between categories. Looking at the figures contained in Annex FV, one can see that on the basis of the weighted average of the suggested retail prices of some of the 20 best selling brands, per quantity containing the same amount of alcohol as a 750 mi, 40 per cent bottle, the amount of tax burden on each category varies substantially. The ratio of the tax burden over the retail price, on the other hand, is roughly the same; as measured in December 1995 vis-a-vis average suggested retail prices, at around 20 per cent. According to the figures, a consumer of any category, on average, paid roughly 20 per cent of the price for tax. Japan referred the Panel to 20 individual prices, the tax/price ratio of which is distributed within a similar range. If these categories were subject to the same tax rate per alcoholic content as currently applied to whisky, any shochu would bear a heavier tax burden than any imported whisky. 4.130 In support of its argument that the categorization of the Liquor Tax Law is reasonable, Japan submitted to the Panel that EU members, Canada and the United States apply significantly different tax rates between distilled liquors, wine, beer and intermediate products, Moreover, in eight Community Member States and the United States, the tax applicable to still wine is different from that applied to sparkling wine. In Japan’s view, this treatment of different categories is apparently not based on the degree of competitive relationship between products. For example, there seems to be little competition, if any, between cream liqueur and vodka, while the study by Bossard Consultants found a strong competitive relationship between wine and distilled liquors, and between beer and distilled liquors. In Japan’s view, the other parties to the dispute reject tax distinctions between whisky and shochu, a pair of products which have not been demonstrated to be competing against each other. On the other hand, Japan submitted that the complainants take for granted tax differentiation among beer, wine and distilled liquors, categories which the Bossard study found to be competitive products. Japan suggested to the Panel that if Japan shouid not apply different rates to distilled liquors by virtue of GATT rules, wine-producing France should not tax distilled liquors more heavily, nor should Germany impose a higher tax on distilled liquors than on beer, Japan added that seven Community Member States apply different rates to distilled liquors, Japan argued that not all of these practices are inconsistent with GATT rales and that Article III is not an instrument for harmonization;of internal taxes. WT/DS8/R WT/DSIO/R WT/DS11/R Page 76 concludes otherwise, and considers that as currently structured, the Liquor Tax Law distorts neutrality in relation to consumer choice. 4.137 Japan argued that what Canada read from the Subcommittee’s recommendation differed from the language of the Subcommittee report. Japan then submitted a chart which compares the relative tax burden between shochu A and imported vodka and whisky under the three yardsticks. For example, the left column indicates that the liquor tax per litre of vodka is 2.7 times higher than that of shochu A, and that of whisky, 7 times higher than that of shochu. Figure: Comparison of “Tax Discrimination Indices” Per Litre of Per Litre of Pure Tax/Price Ratio Beverage Alcohol Liquor Tax | VAT Liquor Tax | VAT Liquor Tax | VAT Shochu A | 1.0 10 1.0 10 1.0 1.0 2.7 3.4 1.6 2.0 0.8 1.0 8.0 4.0 47 0.9 1.0 Note: Calculated on the basis of weighted average of 20 most selling brands. 4.138 Japan argued that this figure demonstrates that i) the liquor tax is similar to VAT in terms of “tax discrimination indices”, and that ii) VAT would be regarded more “trade-distortive” than the liquor tax as long as a comparison is made on the basis of the taxes by the tax amount per litre of beverage or of pure alcohol. Japan submitted that VAT is regarded as one of the most trade-neutral indirect taxes and its introduction is one of the conditions to join the European Union. On the other hand, a comparison made with the amount of tax per litre of beverage or of pure alcohol would find such VAT as trade-distortive. In fact it is the use of those two yardsticks as tools of comparing taxes which is problematic, rather than the tax itself. Japan emphasized that a consumer usually does not buy a product exclusively on the basis of the size of the bottle or on the basis of the alcoholic strength. Consumers choose products by comparing the price and the overall value of a product, which depends upon the taste, flavour and other features and is not confined to the volume and strength. Japan argued that this is why the tax/price ratio is a better criterion to evaluate the effects of taxes on competitive conditions, and neutrality is achieved when the tax/price ratio is equalized, as is the case with the Japanese tax. 4.139 The Community submitted that the system established by the Liquor Tax Law was radically different from an ad valorem tax system. In a true ad valorem system, taxes are proportional to the actual sales value of each shipment and, therefore, neutral, provided that the rates are the same for all the categories. In contrast, under the Liquor Tax Law system, the tax amount is unrelated to the actual sales price. Instead, the assumption is made that certain categories of liquors are a priori more expensive than others. The tax rates are then set so as to reflect this “assumed value” and uniformly applied to all shipments, regardless of their actual price. The outcome of this system is not “neutrality” but arbitrariness because no category of distilled spirits is inherently more expensive than others. The WT/DS8/—R WT/DS10/R WT/DS11/R Page 77 evidence provided to the Panel shows that, even under the current tax system, the pre-tax price of some brands of whisky is lower than the equivalent price for some brands of shochu. 4.140 Japan argued that the lack of plausible alternatives further testifies to the lack of protective intent. Conceivable alternatives to ensure neutrality and equity are: (i) to raise the ad valorem value- added tax to a level comparable to that of the European Union, which applies not only to liquor consumption but to almost all consumption or (ii) to alter the liquor tax into an ad valorem tax. However, for Japan, neither of these is practical. First, the decision to raise the ad valoremconsumption tax from the present three per cent to five per cent beginning April 1997 was made in 1994 only after a prolonged, heated debate. It is not very likely that the rate would be raised to the Community level in the near future. Second, an ad valorem excise tax could easily invite tax evasion by way of transfer- pricing, particularly if applied at the shipping stage. Canada’s Federal Manufacturers Sales Tax suffered from the same difficulty and was abolished in 1991. On the other hand, enforcement cost of an ad valorem tax would be very substantial if applied at the retail level. 4.141 The Community responded that the reasons given by Japan for not pursuing the alleged objective of neutrality through the application of an ad valorem tax system are groundless. In the Community’s view, Japan’s definition of neutrality assumes that specific excise taxes are passed on in full to consumers. Thus, if the current tax system was truly neutral, there should be no reason for the Diet to object to its replacement by an ad valorem consumption tax. The Diet’s oppesition merely reflects the fact that under the current system shochu is much less taxed than it would under a truly neutral ad valorem consumption tax. In any event, internal political difficulties may not provide a valid justification for infringing GATT rules. For the Community, the tax evasion problems invoked by Japan are common to the application of all ad valorem systems. Similar issues arise, for instance, in connection with the application of ad valorem customs duties to import transactions between related parties. It is therefore suggested that any tax evasion problems related to transfer pricing could be appropriately tackled by using any of the alternative methods for the calculation of transaction values provided for in the Agreement on Implementation of Article VII of GATT 1994 (the Customs Valuation Code). In this respect, it is worth noting that Japan currently applies ad valorem customs duties on imports of a fair number of liquors, apparently without this giving rise to any major duty evasion problems. Moreover, until 1989, Japan applied ad valorem excise duties to certain liquors. These duties were abolished because they were found inconsistent with Article III:2 by the 1987 Panel Report and not because they were an invitation to tax evasion. Ad valorem excise duties on alcoholic beverages are currently applied by other WTO Members (e.g., in Denmark). Contrary to Japan’s assertions, the application of a flat rate tax to all distilled spirits would not necessarily render the Japanese system less “neutral” (according to Japan’s own definition of neutrality). For the Community, the scenario depicted by Japan is unrealistic because it does not take into account the likely dynamic effects of a tax readjustment. If shochu was taxed at the same rate as whisky, the cheapest brands of shochu would be expelled from the market, just like former second grade whisky was wiped out of the market by the 1989 tax reform. This would have the consequence of lowering the average tax/price ratio for the remaining brands of shochu. On the other hand, an increase of the rates on shochu would allow the entry into the market of cheaper brands of the other categories of liquors, thus driving up their respective average tax/price ratios. Moreover, there is no reason why a flat tax rate should be applied at the level of the current rate for whisky. It could as well be set at the level of the current rate for shochu or at any intermediate level between the current rates for shochu and “whisky/brandy”. The tax systems of other WTO Members which apply a flat tax rate to all distilled spirits are not less “neutral” in terms of tax/price ratios than the current Japanese system but rather the opposite. 4.142 Inresponse to the Community’s argument that the evasion problems are manageable since there are other cases of the ad vaforem tax, Japan made the following counterarguments. First, the Danish WT/DS8/R WT/DS10/R WT/DSIUR Page 78 ad valorem tax is expected to be abolished shortly. According to Japan, the abolition was prompted. by the fact that, as prices are determined on a case-by-case basis, the system became enormously confusing, and that, in particular, it was easy to manipulate prices while such manoeuvres were hard todetect. The Community now prohibits members from adopting an ad valorem tax. Second, although it is true that an ad valorem tax used to be part of the Japanese liquor tax before the 1989 amendment, detection of transfer pricing was possible then because retail prices were stable at the suggested retail Prices and the margin rates for retailers and wholesalers were broadly constant. However, liquors are now traded, in some cases, at prices substantially different from suggested retail prices and it has become virtually impossible to distinguish manipulation from normal trading. In fact there was another __ad valorem excise tax before the 1989 amendment which applied to consumer goods such as automobiles, cameras, watches and audio equipments. However, since these items were often sold at discount, there were a series of complaints over taxable prices, and the integrity of the tax was put into question. Third, for Japan, an ad valorem excise tax is fundamentally different from an ad valorem customs duty; and the success of an ad valorem customs duty does not mean that an ad valorem excise tax is feasible. The total tax burden under Japan’s liquor tax is 20 billion US dollars, and is more than 50 times greater than that of customs duties on iiquor. The aggregate magnitude of incentives for tax evasion is therefore far greater for the liquor tax than for the customs duty. Nevertheless, it is far more difficult to secure compliance with the excise tax. Customs duties are charged before the goods are withdrawn, while the liquor tax is levied after they are shipped. Importers must declare the prices to the customs authorities and pay applicable duties before they are authorized to withdraw the goods. In contrast, the liquor tax is levied after the goods have been shipped, on the basis of declaration for the preceding one month. Japan argued that careful examination of the examples of ad valorem duties put forward by the Community demonstrates the impracticality of an ad valorem liquor tax. In response to the Community’s claim concerning “the likely dynamic effect of a tax adjustment”, Japan raised the following points. First, since 1987, the tax rate on whisky has been halved. However, the share of premium whisky in the imported Scotch whisky market expanded from 33 per cent in 1987 to 51 per cent in 1994. Japan argued that, contrary to the allegation of the Community, the decrease of tax promoted premium whisky sales. Also, during the same period, in which the shochu tax rate increased twofold, containers of shochu A shifted from a medium size toa large size. Again, contrary to the Community’s allegation, higher tax has led to an increase in the share of low-cost brands. The Japanese market configuration shifted in a direction opposite to the Community’s “likely dynamic effect of tax adjustment”. Second, according to Japan, differentials in net of tax prices between categories similar to those in the Japanese market do exist in countries which apply a flat-rate tax on distilled liquors as well. According to a survey done by Business International, Inc., an affiliate of The Economist Magazine, brandy is more expensive than whisky and whisky is more expensive than gin in 32 cities of the world where a flat tax rate is applied. The Community argued that the application of a flat rate tax to all distilled spirits would not necessarily render the Japanese system less neutral. However, Japan calculated the tax burden in Belgium on the basis of average prices quoted in the promotion leaflet of the supermarket in Brussels submitted by the Community and found that any bottle of vodka or any bottle of gin is burdened with a higher tax/price ratio than any of the bottles of brandy in the shop (left hand side of the exhibit in Annex VY). On the other hand, under the Japanese system, tax/price ratios are broadly equal across categories (right hand side of the exhibit in Annex V). Japan concluded from the exhibit that a flat-rate tax was far inferior to the current Japanese tax in terms of neutrality, horizontal equity and vertical equity. Japan submitted two tables on “Tax/Price Ratio in Belgium and in Japan” (see Annex V). 4.143 Japan also submitted to the Panel charts which identified the “roughly equal” tax/price ratios, namely a chart showing the Average Retail Prices and Taxes of Liquor and another chart on the Percentage of Taxes in Retail Prices (see Annex VI). According to Japan, the Liquor Tax Law’s tax rates are set corresponding to the average net-of-tax price of each category. While net-of-tax prices WT/DS8/R WT/DS10/R WT/DS11/R Page 81 by the vigorous opposition of some Japanese legislators to the very modest increase in the tax on shochu proposed by the Government as part of the 1994 tax reform bill, According to a member of the Diet, that increase would have been “a blow to the more than 30% of shochu makers who show a deficit”. In order to appease this criticism, the Japanese Government was forced to admit that the tax increases had been kept “to the minimum level possible”. Contrary to the Japanese Government’ s claim that” the only public policy objective pursued by applying lower tax rates to shochu is to maximize the tax revenue while ensuring that the tax burden is distributed among the different categories of consumers in accordance with their respective tax bearing ability, the amount of the tax is unrelated to the actual price of the liquors: a liquor falling within a certain category and having a certain alcoholic content is always taxed at the same rate, regardless of its sales price. Instead, the Japanese authorities make the a priori assumption that certain categories of liquors are more expensive than others and are consumed by more affluent consumers. As a result, it is the amount of the tax that the Japanese authorities arbitrarily decide to impose on each category of liquors which determines its sales price and ultimately its consumption pattern, and not the opposite. 4.148 Against the Community’s allegation of political influences based on media articles, Japan referred the panel to a newspaper article conveying a contrary view and submitted that speculative inside stories were not appropriate as the basis of Panel findings. For Japan, the record of Diet deliberation of the 1994 amendment cannot be evidence of a protective intent, as the amendment raised the tax rate on shochy A by 30 per cent and that for shochu B by 44 per cent, while raising the tax on "spirits” by mere 11 per cent and maintaining the tax on whisky/brandy at the same fevel. Japan also argued that the complaining parties seem to confuse the present Japanese policy with that of 1987. Japan stressed that its current distilled liquor taxation prioritizes neutrality and horizontal equity over vertical equity, and that Japan is not saying that a pursuit of vertical equity can justify a non-trade neutral tax. For Japan, what it is saying is that in so far as a tax system satisfies the trade-neutral requirement, the less regressive it is, the better. In explaining the notion of vertical equity, Japan argued that tax inequity among consumers of different liquors can have a distributional impact when different income groups prefer different categories of liquor. Japan submitted evidence that compares the income, the consumption tax payments, the liquor tax burden in respect of distilled liquor under the current Liquor Tax Law and the hypothetical liquor tax burden under a flat specific rate, of an average household in income quintiles. Under the current Liquor Tax Law, the consumption tax burden or the present liquor tax burden track roughly the income level and are fairly proportional. However, the tax becomes highly regressive under the flat-rate regime. 4.149 For Japan, one of the possible factors contributing to the introduction of the flat-rate tax in Europe and North America seems to be the importance in the tax policy attached to the prevention of alcoholic dependence. Community documents prepared in the course of drafting the Community Council Directives of 1992 indicate that the impact on health has been an important factor. Similarly, the local liquor tax laws of the United States were introduced in the wake of the Prohibition, and some of them earmark the tax revenue specifically for fighting alcoholism. Another possible factor is the high ad valorem consumption tax. The proportionality of the ad valorem consumption tax (e.g., VAT, the state sales tax) could mitigate the overall regressiveness of taxes on liquor. For example in the Community member States’ VAT rates range from 15 per cent to 25 per cent. The Canadian federal government imposes a VAT of 7 per cent, and Ontario levies a 12 per cent sales tax. The rate of State and local sales tax in New York is 8.25 per cent. These taxes serve to ensure equity, enabling the liquor tax to focus on alcohol. These factors are not present in Japan. At the rate of 3 per cent, the ad valorem consumption tax alone cannot ensure the equitable distribution of tax burden among liquor consumers under a flat rate liquor tax system. Japan submitted that the National Diet demands annually an estimate of the ratio of the liquor tax burden relative to the income level for each income group. Japan argued that it was against this background of acute concern over the issue of distributional equity WT/DS8/R WT/DSLO/R WT/DSLU/R Page 82 that the Liquor Tax Law incorporated an element of distributional equity in its structure. Different specific rates apply to a variety of products, depending on the degree of value-added. However, Japan argued, the present policy pursues the goal of distributional equity only to the extent compatible with neutrality. For Japan, while the pre-1989 policy assumed that a consumer of imported whisky had a greater tax-bearing ability, and therefore should bear a greater tax burden, Japan’ s taxation of distilled liquors made a clear departure from that policy with the 1989 amendment following the 1987 Panel Report. Contrary to the claims by other parties to the dispute, the present policy requires an equal burden to be shared among consumers of imported whisky and shochu. For Japan, distortional effects of a tax on consumer choice are most pronounced in the case of flat-rate specific taxes, as currently applied in some countries of Europe and North America. For example, different categories of liquors are produced by different manufacturing processes; some are stored in wooden casks for many years, and others are consumed immediately after distillation. Their value or net-of-tax price differs accordingly. However, the flat-rate tax imposes the same amount of tax across various categories of distilled liquors if the amount of alcohol contained is the same, regardless of product differences. Under the flat-rate taxation, where the amount of tax is the same for all categories of distilled spirits, a bottle of gin the value of which is 4 pounds sterling will cost 12 pounds to purchase after tax. In contrast, a bottle of whisky the value of which is 20 pounds sterling will be priced at 30 pounds. In other wards, the tax makes the price of gin three times higher than its value, while making the whisky price only 50 per cent higher. This kind of taxation prejudices consumers’ choices against gin in favour of whisky because consumers choose products by comparing their price and their overall value. 4,150 Japan further argued that in order to ensure neutrality to consumers’ choice, a liquor tax should not be based exclusively on one element of the overall value: alcoholic strength. The Liquor Tax Law, on the other hand, captures other elements of value by classifying liquors on the basis of the difference in value added by post-distillation processing. Indeed, this results in a broadly constant value/price tatio across categories (80 per cent, as measured in December 1995 vis-a-vis average suggested retail prices) and a tax/price ratio (20 per cent as measured in the same manner). For example, prices of a bottle of whisky (¥3,000) and of a bottle of shochu (41,000) increase by a similar percentage after tax. Categorization of distilled liquors under the Liquor Tax Law thus serves the goal of neutrality, a legitimate tax policy objective. In fact, this role of varied specific rates is not peculiar to Japan. Eight countries out of 15 Member States of the Community apply a tax rate to sparkling wine several times higher than the one levied on still wine. Japan also submitted that the US tax on sparkling wine per litre of alcohol is 3.2 times higher than that on still wine, compared with the 3.9-10-1 tax differential per litre of alcohol between whisky and shochu A in Japan. Also, the import ratio of still wine in the United States is lower than that of sparkling wine, as is the impert ratio of shochu, which is lower than that for whisky. Moreover, in Japan, wine is wine and the same tax rate is applied to both sparkling and still wines, whereas in the United States, alcohol is alcohol and the same tax rate is applied to all distilled liquors. 4.151 The Community, Canada and the United States reiterated that the taxation systems of the countries other than Japan were outside the terms of reference of the present Panel. 4.152 The Community stated that although Japan argued that “equity” requires a “fair distribution of the tax burden among consumers”, it did not provide any clear explanation of what is considered as a “fair distribution” of the tax burden. In the Community’s view, Japan argues that “equity” means that tax rates on distilled spirits should be progressive or at least proportional to the consumers’ income Jevel: “Tax inequity among consumers of different categories can have a distributional impact when different income groups prefer different categories of liquors”. The Community submitted that different income groups do not have fixed and inherently different tastes. Consumers with lower incomes drink less than consumers with higher incomes simply because whisky is more expensive. In turn, it is more WT/DS8/R WT/DS10/R WT/DS11/R Page 83 expensive because it bears higher taxes. Thus, through the application of higher tax rates on the categories of liquors deemed a priori to be preferred by the rich, a groundless assumption is turned into a self- fulfilling prophecy. The assumption that certain income groups prefer certain liquors was also at the heart of the “tax bearing ability” principle invoked by Japan as a justification before 1987. This justification was rejected by the 1987 Panel Report. “Equity” is but another tag name for “tax bearing ability” and should be condemned on identical grounds. The Community concluded by stating that the above considerations led the 1987 Panel Report to dismiss in categorical terms a similar justification advanced by the Japanese Government: “The Panel was of the view that the use of product and tax differentiations with the view of maintaining or promoting certain production and consumption patterns could easily distort price-competition among like or directly competing products by creating price differences and price-related consumer preferences which would not exist in case of non-discriminatory internal taxation consistent with Article III:2. The Panel noted that the General Agreement did not make provision for such a far-reaching exception to Article III:2 and that the concept of taxation according to tax-bearing ability of prospective consumers of a product did not offer an objective criterion because it relied onnecessarily subjective assumptions about future competition and inevitably uncertain consumer responses” .” 3. The Effect of the Legislation 4.153 Japan reiterated that the “effect” of “so as to afford protection” must be judged by whether the tax distorts the competitive relationship between imported and domestic products. For Japan, the Liquor Tax Law does not distort the competitive relation between imported and domestic products for the following reasons. 1) The tax/price ratios of all tax categories are roughly the same. In terms of the examination of the tax burden, the tax/price ratio is the superior yardstick and better indicates the impact on consumers’ choice than the ratio of tax over product volume or alcohol content and it is common practice to employ a tax/price ratio in comparing the burden of an excise tax. 2) Shochu is produced outside Japan; indeed in examining whether or not the category in question is almost exclusively domestic, what needs to be examined is not import ratios but rather whether the “domestic” product is produced in other countries, and whether the “imported product” is also domestically produced. 3) There is no directly competitive or substitutable relationship between domestic products and imports, precluding therefore any possibility of protective effects. Since in Japan’s view, protective distortion exists only when the three above-mentioned cumulative requirements are met, it is clear that the Liquor Tax Law does not distort any competitive relationship and is therefore consistent with Article Til:2. a) The Tax/Price Ratio Rates under the Legislation are Neutral 4.154 For Japan the lack of protective effect of the Law is demonstrated by the neutrality of the tax. The application of vary ing rates dependent on average net-of-tax prices of distilled liquor categories Tesults in a fairly stable tax/price ratio. Thus the liquor tax is far less distortional than taxes of other countries, and far less likely to alter consumer choice. 4.155 The Community responded more specifically that the Liquor Tax Law is not, de facto, “neutral” even according to the standard of “neutrality” defined by Japan. The tax/price ratios, as calculated 1987 Panel Report, pata. 5.13. WT/DS8/R WT/DS10/R WT/DS11/R Page 86 tax base cannot be the only yardstick to measure the relative tax burden. Japan reiterated that the purpose of comparing the tax burden here is to judge whether or not the tax has a distortive effect on competitive conditions. The terms of comparison ought to be, therefore, such that would correctly capture the tax’s impact on consumer behaviour. Japan submitted the example of a bottle of imported whisky and a bottle of domestic shochu both priced at ¥2,000 before tax. The whisky is contained in a 0.7 litre bottle at 40 per cent alcohol, while the shochu is marketed in a four litre plastic bottle at a 25 per cent strength. If the amount of tax per volume of beverage is equalized, the tax levied on the bottle of shochu must be 5.7 times the tax on the bottle of whisky. Thus, if the whisky is to be taxed ¥700, the shochu must be taxed ¥4,000. Alternatively, if the tax per a quantity of alcohol contained is equalized, the tax on the shochu will be ¥2,500, or 3.6 times the ¥700 tax. In either case, a potential whisky customer is likely to remain undeterred. However, a shochu customer would be forced to alter its choice toward other beverages such as beer or sake. For Japan, this tax would distort consumers’ behaviour; in fact the least distortive method is to levy the identical amount if products’ before-tax prices are equal; ¥700 tax both on ¥2,000 shochu and ¥2,000 whisky. 4.159 Canada recalled that Japan repeatedly asserts that the tax/price ratio between all alcoholic beverages is “roughly constant” and that, accordingly, the Liquor Tax Law does not distort conditions of competition between whisky and shochu. Thus, it follows, in Canada’s view that tax/price ratios that are not “roughly constant” show that the Liquor Tax Law distorts conditions of competition between whisky and shochu in favour of domestic shochu production. On the basis of Japan’s own evidence that ostensibly summarizes tax/price ratios using suggested retail prices, the Liquor Tax Law does not yield “roughly” equivalent tax/price ratios between distilled liquors. This is made even more clear using tax/price ratios based on on-shelf retail prices. Put simply, in the consumer marketplace where whisky and shochu compete, the tax/price ratios show that the tax differentials between whisky and shochu are clearly skewed in favour of domestic shochu production. For Canada, in stating that “[a] consumer of any category must pay roughly 20 per cent of the expense for the tax” , Japan acknowledges that on-shelf retail prices — the prices that consumers actually pay -- are the appropriate price variable in assessing tax/price equivalency between categories of distilled liquors. In Canada’s view, Japan relies on the prices offered at the discount retail outlet “Sake Ichiba Yamada”, to show that “at discount prices [the tax/price ratio] is 35 per cent and 40 per cent, respectively, for whisky/brandy and shochu” and thus that “the relation between tax/price ratio of different categories tends to remain the same”. However, Canada argued, the prices submitted by Japan reflect only a single discount retail outlet; significantly, in comparison to the whisky and brandy sold at this store in volumes of either 700 or 750 mil sizes, eight of the nine shochu products listed by Japan are sold in very large sizes ranging from 2.7 litres to 5.0 litres that maximize volume discounts in pricing, thereby “driving down the per unit cost of shochu and consequently “grossing-up” the tax/price ratios of shochu”. The one shochu product selected having a volume of 720 mi has a tax/price ratio of 25.4 per cent that compares to a tax/price ratio for whisky of similar volume that reaches a high of 52.7 per cent. Indeed, tax/price tatios based on on-shelf retail prices in Fapan show that whisky and shochu are subject to a substantial tax differential in favour of shochu: A survey of on-shelf retail prices of shochu Group A, formerly Special Grade whisky and formerly Second Grade whisky (standard and premium labels) shows that the current tax/price ratio. for premium imported whisky is 26.4 per cent, for standard imported whisky is 41.7 per cent, for shochu A is 16.6 per cent and for Shochu B is 9.5 per cent. Based on on-shelf retail prices, the tax/price ratio of Canadian whisky shows a similar pattern to the formerly Special Grade whisky and formerly Second Grade whisky surveyed. WT/DS3/R WT/DS10/R WT/DS1U/R Page 87 - A more recent survey of on-shelf retail prices of shochu in discount, supermarket and smaller shops in four cities, Tokyo, Osaka, Nagoya and Fukuoka demonstrates tax/price ratios ranging from 10.9 per cent to 23 per cent with most ratios falling between 13 per cent and 18 per cent. Canada submitted that the tax/price ratio of whisky and shochu is consistently skewed in favour of shochu. Accordingly, whether considered in absolute terms or in terms of tax/price equivalency, the Liquor Tax Law imposes a substantial tax differential between distilled spirits. Clearly, whisky and shochu, two directly competitive or substitutable products, are not similarly taxed (see paragraphs 4.90 to 4.92 above). 4.160 The Community also argued that, according to the data submitted by Japan the percentage of taxes on retail prices may vary from only 5 per cent to as much as 22 per cent . Whether it is shochu or the other categories of distilled spirits that bear a higher tax/price ratio lacks any relevance. The evidence shows that the Japanese tax system is far from being “neutral” even in the terms defined by Japan and on the basis of Japan’s own price data. Moreover, the method followed by Japan to calculate the tax/retail price ratios grossly and systematically underestimates the ratios for liquors other than shochu while overestimating the ratio for shochu: - It does not take into account sales of domestically produced “whisky/brandy”, “spirits” and “authentic liqueurs”. The retail prices for domestic brands of these categories tend to be lower than the prices for imported brands. As a result, the tax/price ratios of domestic brands are, as a general rule, higher than ratios of imported brands. This difference has been recognised by Japan in a previous estimate provided to the Community during the consultations which shows that, for example, the ratio for imported brands of whisky is within the range of 16.7 per cent to 29.3 per cent, while the ratio for domestic brands may vary from 30.5 per cent to 36.3 per cent. By excluding domestic brands from the calculation, Japan artificially reduces the average ratio for these categories. In the case of “whisky/brandy” and “spirits” this effect is particularly important since domestic brands account for a majority of the total sales. - Moreover, the effects of the tax/price ratios have been calculated on the basis of weighted average prices. This basis could be considered as representative if the prices for the individual brands stood within a relatively close range. However, the prices for individual brands of “whisky/brandy” and of “spirits” vary considerably. For example, on the basis of Japan’s evidence, it may be estimated that the suggested retail prices (“SRP”) for whisky range from ¥2,000to ¥7,000. When calculating the weighted average price for whisky a bottle of ¥7,000 would weigh the same as 3.5 bottles of ¥2,G00. In view of this, it is very likely that a majority of the individual sales of whisky covered by Japan’s cafculation was in fact made below the weighted average price and, therefore, with a tax/price ratio in excess of the one submitted by Japan. Indirect confirmation of this is provided by the fact 14 out of the 20 best selling brands of imported whisky have tax/price ratios above the weighted average ratio. ~ SRPs for shochu A and shochu B vary much less. On the basis of Japan’s evidence, it may be estimated that 19 out of the 20 best selling brands of shochu A retail at SRPs between ¥700 and ¥1,100. The range of SRPs for shochu B is even shorter. This means that, unlike in the case of “whisky” and “spirits”, the actual tax/price ratios for the majority of individual sales of shochu should be close to the weighted average ratios. - The tax/price ratios have been calculated on the basis of SRPs and not of actual retail prices. Discounts on SRPs are widespread and substantial (in some cases they may represent as much as 70 per cent of the SRP) which renders any comparison on the basis of SRPs a purely WT/DS8/R WT/DS10/R WT/DSIUR Page 88 theoretical exercise. Discount margins vary considerably from one category to the other. As a general rule, discounts are lower for shochu brands than for brands of other categories. For example, on the basis of the evidence submitted by Japan, the average discount margins for brandy, whisky and shochu offered at the outlet “Sake Ichiba Yamada” are 57.2 per cent, 42.8 per cent and 37.6 per cent respectively. As a result, and contrary to Japan’s allegations, the relations between the tax/price ratios of different categories do vary substantially depending on whether the calculation is made on the basis of SRPs or of actual discounted prices. For example, at Sake Ichiba Yamada, the tax/retail prices ratios for whisky and shochu are, on the basis of SRPs, 26.6 per cent and 25.5 per cent, respectively. On the basis of discounted prices, the ratios are 42.8 per cent for whisky and 37.6 per cent for shochu, i.e., the difference, which is of one percentage point on the basis of SPRs, increases to five percentage points on the basis of actual prices. The charts submitted by Japan do not take into account the differences in relative prices between bottles of different sizes. Large size bottles tend to be less expensive in relative terms than small bottles and consequently have higher tax/price ratios. Shochu is more frequently sold in large size bottles than the other categories. By lumping together bottles of all sizes when calculating the average tax/prize ratios Japan conceals the fact that the ratio for shochu is much lower than the ratio for other categories when only bottles of the same or similar size are compared. For example, the tax/discount price ratio for the shochu brand Triangle (the only brand sold in botties of 0,72 litre which is included in the sample) is only 28.9 per cent. In contrast the average tax/discount price ratio for the sampled whisky brands (all of which are sold in 0.7 or 0.75 bottles) is 42.8 per cent, i.e., almost 14 percentage points higher. 4.161 The United States submitted that the tax differential today amounts to a 9.6 to 1 differential between whisky and shochu B, 2 6.3 to 1 differential between whisky and shochu A, and a 2.4 to 1 differential between spirits and shochu A. At the respective reference values for each, the tax rate per degree alcohol is ¥24.88 for whisky, ¥6.2 for shochu A, and ¥4.08 for shochu B, amounting to a four-to-one differential between whisky and shochu A and a six-to-one differential between whisky and shochu B. From the US point of view, Japan has not offered any convincing policy rationale for this differential. Instead, Japan has asserted that this discrimination simply does not exist. Japan has claimed repeatedly that “the Japanese tax is consistent with Article III because the tax/price ratio is roughly constant between categories, and would not distort consumer choice”. Japan had asserted that the tax/price ratio is approximately 20 per cent of the pre-tax retail price but Japan’s use of tax/price ratios to gauge tax incidence was misguided. For instance, the table used by Japan to support its argument on consumer neutrality was based on manufacturers’ suggested retail prices for the various types of distilled spirits. There is a wide gap between those prices and actual retail prices. If the objective were to maintain neutrality with respect to the consumer, it is appropriate to use actual prices, not the mamufacturer’s suggested retail price, and to weight the prices per degree of alcohol, in order to account for the fact that most consumers consume distilled spirits in diluted form, diluted to approximately the same strength. The United States referred the Panel to a survey conducted by the European Business Community in Tokyo showing that the tax/price ratio for shochu is 13 to 16 per cent of actual retail prices for shochu A and a mere 9.5 per cent for shochu B. Furthermore, even though actual retail prices for imports had been declining since 1992, the Japanese Government had not reduced the tax rates on whisky, and the liquor tax rates on whisky now represent as much as 35-40 per cent of retail prices. The only plausible explanation is that the Japanese Government has been maintaining a margin of protection for politically-favoured shochu producers. The evidence submitted by Japan shows that the prices for imported whisky and brandy are much more variable than the price of shochu A or B, especially if corrected for differences in alcoholic content. Thus, there appears to be less active price and quality competition in the case of shochu. The Liquor Tax Law’s dampening WT/DS8/R WT/DS1OR WT/DSIUR Page 91 submitted a figure which according to him demonstrates for instance that a bottle of shochu with a net price of ¥500 has a tax of ¥76, which implies that the tax increases the price by 15 per cent. By the same token whisky with a net price of ¥500 is burdened with an additional tax burden of 158 per cent. According to the expert of the Community, there exist a tax discrimination in every single market segment. This is true in particular for the low and medium quality brands of whisky and brandy, while for the premium brands the effect is less important. According to the expert of the Community, an important possible counter argument against the evidence submitted by the Japanese could be that the differences in taxes are justified by the different alcohol content of shochu (25 per cent), imported brandy (43 per cent) and whisky (40 per cent). This argument may not be unreasonable and could be taken to justify a tax on whisky that is 1.6 times higher than that on shochu to take this correction factor into account. This correction does not diminish the highly discriminatory nature of the tax system. For the complainants, the average tax rate for brandy and whisky is low not because the Japanese government set it low, but because expensive brands are what is left on the shelf’, while low and medium. quality brandies are driven out of the market through a tax system that very effectively discriminates against this market segment. 4.165 Canada argued that the portions of the survey relating to imported whisky are particularly tevealing of the fundamental failures in the survey. According to Canada, the whisky samples represented in the survey cover an enormous price range, running upwards from ¥999 to over ¥15,000. Canada noted that since whisky is commonly sold in containers of similar volumes and within a similar range of alcoholic strengths, the tax applicable to each item will be very similar in most cases. Canada stated that with the applicable tax being roughly constant and the prices ranging over a magnitude of 15, the result is necessarily an equally vast range of tax/price ratios. Canada noted by way of example that a 750 ml bottle of 40 degree whisky priced at ¥1500 will have a tax/price ratio of 49 per cent whereas an equivalent premium whisky priced at ¥15000 will have a tax/price ratio of 4.9 per cent. Canada argued that the survey masks this by the artificial imposition of a single tax/price ratio of 17 per cent that purportedly applies to each and every one of the whisky samples in the survey. According to Canada, when examined on an item-by-item basis, the degree to which tax/price ratios diverge in actual situations from Japan’s claimed “rough constant” of 20 per cent becomes obvious. Canada concluded, therefore, that Japan’s attempt to use the survey to support its use of suggested retail price falls apart. 4.166 Japan also argued that a flat tax rate runs counter to the notion of tax equity: a consumer of gin must pay most of his expenditure as liquor tax, while a consumer of whisky is taxed only a small portion of his expenditure. Similar inequity would result if Japan were to apply a flat-rate tax. Japan recalled that under the Liquor Tax Law consumers of shochu and imported whisky are both paying about one fifth of their expenditure as liquor tax. If a flat rate tax were to be applied to these products at the present whisky tax level, however, shochu consumers would be forced to pay half their expenditure as the liquor tax, while whisky consumers’ burden would remain at one fifth of their expenditure. The Tax Commission’s following recommendation to the Prime Minister stresses the importance of the tax/price ratio as a yardstick in achieving an appropriate level of the tax burden: “The liquor tax has specific tax rates. When the prices rise, the level of the burden becomes lower. Therefore, it is necessary to review the level from time to time, and to ensure an appropriate level of the burden”. WT/DS8/R WT/DS10/—R WT/DS1U/R Page 92 b) Cross-Price Elasticity 4.167 Japan submitted the results of the survey mentioned in paragraphs 4.83 to 4.89 above, namely a first survey conducted in March 1995 by “Shakai-Chosa Kenkyujo” (Institute for Social Studies) as well as a Summary of Findings of Statistical Analysis based on data taken from the national household survey (a government survey conducted by the Census Bureau of the General Affairs Agency since 1962 on the Japanese household revenues and expenditures) for the 20-year period of 1975 through 1994, which, in Japan’s view, confirmed that there is no cross-price elasticity between shochu and other distilled imported liquors and if shochu has any cross-price elasticity with another liquor, one may argue that there is cross-price elasticity between shochu and beer, 4,168 For the Community, the Japanese statistical analysis suffered from such problems relating to the data underlying it, the common trends to which they are subject and the limited number of data points (20), that statistical correction techniques reduce the degrees of freedom in the data to such an extent that it becomes highly unlikely that one could ever statistically prove that there is cross-price elasticity of demand between whisky and shochu, even if it existed in reality. The Community also argued that in these circumstances Japan should be required to disprove that there is such cross-elasticity of demand. For the Community, the most important error of the study is that it uses nominal prices instead of deflated ones. This means that statistically the analysis does not distinguish between the price of shochu in 1984 and the price in 1993, which are practically identical in the table. In real terms, however, the price in 1993 was actually substantially lower than in 1984. Taking the average Japanese inflation rate in the 1980s (=1.9 per cent p.a.), there was a real decrease in the price of shochu by Toughly 20 per cent. Evidently, the failure to discern a real price change from a nominal price change can lead to spurious results. A proper analysis should therefore use real (i.e., deflated) values to correct for this problem. A major problem of time series analysis in general is that time series are dominated by trends. In the Community’s view, if one looks at the raw data, one can see for instance that the consumption of sake steadily decreased while that of beer steadily increased. There could be multiple reasons for this, e.g., changes in taste or successful advertising. In the present case, price plays a relatively small role compared to those factors that are not included in the analysis, except for what is labelled under “trend”. The Community continued by arguing that most time series have such an underlying trend. If one took a random time series, e.g., of cucumber production in France, the population of India and shochu consumption in Japan, a naive statistical analysis could “prove” that for the last 20 years shochu consumption in Japan was positively related to cucumber production in France. Evidently, there is no causal relationship just because all three variables follow a common trend. This type of logic is, however, to be found in the Japanese study: Because the consumption of shochu rose at the same time as its price rose (due to inflation), a naive regression will tell you that the higher the price of shochu, the higher will be shochu consumption. This is evidently economic nonsense, but is predictably the result of the regression (expressed as positive own price elasticities in the table). 4,169 In this context, the Community submitted a study by Ames and Reiter.”> They found that an R? (the figure that roughly indicates how well the regression explains the dependent variable; a perfect match would mean an R? of 1) in excess of 0.5 could be obtained by selecting an economic time series and regressing it against two to six randomly selected time series. If one looks at the data of most regressions undertaken in the Japanese study, the R? is lower than that. Thus, most of these regressions do not produce better results than if shochu consumption had been regressed on production Ames, E. and S. Reiter, “Distributions of Correlation Coefficients in Economic Times Series” , Journal of the American Statistical Association, 56, 1961, pp. 637-56. WT/DS38/R WT/DS10/R WT/DSILUR Page 93 of cucumbers. Therefore, the Community argued, in naive time series analysis one can basically show anything. One can econometrically filter out a lot of these spurious relationships, but for every type of correction one looses a degree of freedom in the data. The more complex the analysis is, the more data points you need; 20 data points are hardly enough to correct for the following problems that appear simultaneously: - There normally is a lagged reaction of consumption to prices. This is because it usually takes a while for consumers to find out or get used to the fact that a certain product now has a different price, and shopping goes often by habit. - There is a problem of autocorrelation. This means basically that, if a variable is higher than average in one year, it is likely to be so in the following one. For instance, if you have low growth in one year most tikely growth will be low in another one. This effect makes your regression less reliable. The Cochrane-Orcutt method is one way of correcting for this problem, if applied correctly. - There is a problem of multicollinearity. This means that variables tend to move in a common direction. If, for instance, at the same time that one independent variable goes up and the other always goes down, it is statistically difficult to separate the effect of one from the other. Because all price as well as the consumption variables are determined by a similar trend, exactly this problern occurs. It is unlikely that under those circumstances the regression parameters are statistically significant. - Apart from inflation-caused price increases, variations in real prices appear relatively small. This is important because small variations decrease the likelihood that a parameter can be shown to be statistically significant, i.c., produce reliable estimates. Because there are only 20 data points, in combination with the type of problem listed, it should be clear that it is unlikely that the regression analysis produces significant and/or robust values. 4.170 In response to the Community’s attack on Japan’s econometric analysis, Japan recalled that hypothesis A: “the prices of beer and shochu affect the consumption of shochu” explained the actual data. However, hypothesis B: “the prices of whisky and shochu affect the consumption of shochu” led to a result inconsistent with accepted theory of economics. This is why Japan found hypothesis A reliable and hypothesis B not reliable. Japan recalled that the United States submitted the result of its own analysis of the data it used and conchided that “the annual price indices of whisky and shochu and annual household expenditures do not account for movements in the quantity consumed of these products”. In Japan’s view, this supports its rejection of hypothesis B. As to the Community’s claim that “[t]he most important error in the study is that it uses nominal prices instead of deflated ones”, Japan submitted that for further confirmation, Japan ran an additional regression analysis using the deflated prices and deflated household expenditures as variables. The results were similar to the nominal price analysis. Hypothesis A, based on the prices of beer and shochu, could explain the data in a significant manner. However, hypothesis B which attempts to explain shochu consumption by the prices of whisky and shochu once again led to a result not compatible with the accepted economic theory. In sum, for Japan, the most important alleged error does not alter the conclusion. Even though the Community insists that “in naive time series anatysis one can basically show anything”, neither Japan’s initial method, the second method based on the suggestion by the Community itself, nor the method employed by the United States, succeeded in demonstrating that “prices of whisky and shochu affect