5. Soju and the other Distilled Spirits are Directly Competitive and Substitutable Products

(a) Standard for the Interpretation of "Directly Competitive and Substitutable"

1.49. The European Communities notes that at several points of its submission, Korea argues that the notion of "directly competitive and substitutable products" must be applied "strictly".

1.50. According to the European Communities, the restrictive interpretation of the terms "directly competitive or substitutable" advocated by Korea finds no support in the text of Article III, and that it is refuted by the drafting history of GATT 1947 as well as by prior panel reports.

1.51. According to the European Communities, during the discussions within the Geneva Preparatory Committee and subsequently at the Havana Conference, the delegates discussed a number of examples of "directly competitive or substitutable products" which indicate clearly that the drafters had in mind a rather broad interpretation of those terms. Those examples included apples and oranges;¹⁷⁹ linseed oil and tung oil;¹⁸⁰ synthetic rubber and natural rubber;¹⁸¹ coal and fuel oil;¹⁸² and tramways and buses.¹⁸³

1.52. The European Communities argues that past panels which have interpreted the notion of "directly competitive or substitutable products" have also refrained from taking the narrow approach advocated by Korea. The European Communities also argues that the two Panel reports on Japan - Taxes on Alcoholic Beverages reached the conclusion that all distilled spirits were directly competitive or substitutable products. Another example, according to the European Communities, is provided by the Panel report on EEC - Measures on Animal Feed Proteins, which concluded that vegetable proteins and skimmed milk powder were "directly competitive or substitutable" products for the purposes of applying the second sentence of Article III:5.¹⁸⁴

1.53. The European Communities also notes that Korea, in order to justify its restrictive interpretation of "directly competitive or substitutable", argues that the purpose of Article III:2 is not to harmonise tax policies but to avoid protectionism.

1.54. The EC view is that it would agree that the purpose of Article III:2 is to avoid protectionism, but nevertheless takes issue with Korea's contention that this purpose commands a "strict" reading of the notion of "directly competitive and substitutable" product.

1.55. The European Communities notes that in Japan - Taxes on Alcoholic Beverages II, first the Panel and then the Appellate Body concluded that the notion of "like product" must be construed "narrowly" in the first sentence of Article III:2. According to the European Communities, this interpretation was deemed necessary in view of the fact that, as put by one of the complainants in that dispute, Article III:2, first sentence, is like a "guillotine": once it has been established that two products are like, any tax differential between them is deemed prohibited, without it being necessary to ascertain whether the tax differential is applied "so as to afford protection".¹⁸⁵

1.56. The European Communities notes further that in contrast, there is no indication in Japan - Taxes on Alcoholic Beverages II that the notion of "directly competitive or substitutable" product must also be construed "narrowly" or "strictly". According to the European Communities, this reflects the different wording and structure of the second sentence of Article III:2. Unlike the first sentence of Article III:2, the second sentence makes express reference to the first paragraph of Article III. This means, in the EC view, that in order to establish a violation of Article III:2, second sentence, it must be determined first, as one of three separate requirements, that the tax differential is "applied .... so as to afford protection to domestic production". According to the European Communities, therefore, a "strict" or "narrow" reading of the terms "directly competitive or substitutable" is not warranted in order to ensure that only protectionist measures are condemned.

1.57. The European Communities notes that in the same vein, Korea also argues that Article III:2 second sentence applies only where there is "actual" competition, as opposed to "potential" competition. According to Korea, this interpretation is "strongly suggested" by the Interpretative Note to Article III:2 and, in particular, by the terms ".... where competition was involved". According to the European Communities, those terms, however, refer to "competition" only, without requiring that it must be "actual" competition. "Potential" competition is already "competition" within the meaning of the Note.

1.58. According to the European Communities, the use in the Interpretative Note of the terms "competitive" (and not "competing") and substitutable (instead of "substitute") is a further indication that the GATT drafters envisaged the application of Article III:2 not just in instances of "actual" competition but also where there is "potential" competition. This is allegedly even clearer in the equally authentic French and Spanish versions which refer to ".... un produit directement concurrent ou un produit qui peut lui être directement substitué.. ." and "... un producto directamente competidor o que puede substituirlo directamente...", respectively.

1.59. The European Communities argues that Korea's interpretation is also refuted by prior Appellate Body and Panel reports which have recognised the relevance of "potential" competition for the purposes of Article III:2.

1.60. The European Communities recalls that Japan - Taxes on Alcoholic Beverages I stressed that internal tax measures should not be used to "crystallise" consumer preferences for domestic products.¹⁸⁶

1.61. The same view was reiterated in Japan - Taxes on Alcoholic Beverages II:

1.62. The European Communities also notes that the relevance of potential competition has also been recognised by the Appellate Body in Canada - Certain Measures Affecting Periodicals:

1.63. According to the European Communities, the relevance of potential competition for the application of Article III:2, second sentence, flows from the well established principle that Article III does not protect export volumes but expectations on the competitive relationship between imported and domestic products.¹⁸⁹ Those expectations may exist even if there is no "actual" competition yet between imported and domestic products due to protective tax measures.

1.64. The European Communities further argues that, Korea's position would have the absurd result of actually rewarding those Members who apply the most protectionist tax measures. If a Member applies a tax in such a way as to completely exclude imports of a competitive product, it would never be possible for other Members to show that there is "actual" competition between that product and the protected domestic product and, therefore, that such measures violate Article III:2. Meanwhile, a Member applying a less protectionist tax measure which restricts but does not pre-empt "actual" competition between domestic and imported goods would be found to violate Article III:2

1.65. The European Communities notes that Korea, in support of its peculiar interpretation of Article III:2, emphasises that this provision is perhaps the GATT provision "that treads most heavily upon national sovereignty." Korea implies that since taxation is at the core of the Members' sovereignty, Panels should adopt a deferential standard whenever taxes are concerned.

1.66. According to the European Communities, this argument is totally misguided. In the EC view, it is instructive to compare the wording of Articles III:2 and III:4 of GATT. Article III:4, the general national treatment provision, is concerned only with discrimination between "like products". In contrast, Article III:2 is concerned with discriminatory taxation not only between "like products" but also between the larger category of "directly competitive and substitutable" products. In the EC view, this shows that the GATT drafters were well aware that discriminatory taxation may be one of the most pernicious forms of protectionism and, for that reason, aimed to provide stricter rules with respect to internal tax measures than with respect to other internal regulations, rather than the opposite.

(b) Physical Characteristics

1.67. The European Communities notes that Korea's submission lingers upon the differences in physical characteristics between soju and other spirits. Korea implies that those differences are sufficient to conclude that soju and other spirits are not "directly competitive or substitutable" with each other.

1.68. The European Communities states that it is necessary to recall that two products do not have to be similar in terms of physical characteristics in order to be "directly substitutable and competitive". As noted in Japan - Taxes on Alcoholic Beverages II,

1.69. The European Communities argues that it is obvious that if two products have similar physical characteristics, this constitutes a strong indication that they are "directly competitive or substitutable". According to the European Communities, in the present case, the similarity between soju and the other distilled spirits is such that it is a sufficient reason for the Panel to conclude that all of them are "directly competitive or substitutable".

1.70. The European Communities does not deny the existence of differences in physical characteristics between soju and other spirits. According to the EC view, if there were no such differences, it would have claimed that soju and the other distilled spirits are "like" and not simply "competitive or substitutable".

1.71. According to the European Communities, the differences invoked by Korea are, in essence, the same shown in EC Annex 9. In the EC view therefore, there seems to be no disagreement with Korea as to the nature of the differences, only as to their significance.

1.72. In the EC view, those differences are relatively minor and do not prevent soju from being "directly competitive or substitutable" with other distilled spirits. Indeed, according to the European Communities, many of the differences invoked by Korea, such as differences in alcohol content, colour or flavour, would not be sufficient even to exclude a finding of "likeness".¹⁹¹

1.73. The EC claims its position is supported by the two Panels on Japan - Taxes on Alcoholic Beverages, which allegedly concluded that all distilled spirits were "directly competitive or substitutable" on the Japanese market, notwithstanding their different physical characteristics. Even if, as claimed by Korea, Japanese shochu was not "like" Korean soju, it remains that the differences between Korean soju and the other types of distilled spirits alleged by Korea are the same as the differences between Japanese shochu and those spirits.

(c) End Uses

1.74. The European Communities notes Korea's argument that soju and the other spirits are not used in the same way by Korean consumers and, for that reason, are not "directly competitive or substitutable." Korea also points to differences in drinking style, drinking occasion and place of consumption.

1.75. According to the European Communities, firstly, Korea seems to rely on the mistaken premise that in order to be "directly competitive or substitutable" two products must compete or substitute each other in respect of all possible economic uses. According to the European Communities, that narrow view has been rejected by the Appellate Body in Canada - Certain Measures Concerning Periodicals.¹⁹² In that case, Canada argued that US magazines were not "directly competitive or substitutable" with Canadian magazines because, while they provided a reasonable substitute as an advertising medium, they were poor substitutes as an entertainment and communication medium. Thus, according to Canada, US and Canadian magazines were only "imperfect substitutes". The Appellate Body dismissed this argument by pointing that a case of "perfect substitutability" would fall within Article III:2, first sentence and ruled that all the magazines concerned were directly competitive or substitutable.

1.76. The European Communities further argues that similarly, in Japan - Taxes on Alcoholic Beverages I, the Panel based its conclusion that Japan's Liquor Tax Law violated Article III:2, second sentence, on the finding that there existed direct competition or substitutability among the liquors concerned, "even if not necessarily in respect of all the economic uses to which the product may be put"¹⁹³

1.77. According to the European Communities, these two reports make it very clear that in order to establish that two products are "directly competitive or substitutable", it may be sufficient to show that they are "directly competitive or substitutable" in respect of certain uses.

1.78. Secondly, according to the European Communities, the consumption patterns of soju and the other spirits concerned in this dispute are much more diverse and flexible than Korea's simplistic presentation would suggest. According to the European Communities, Korea would have the Panel believe that soju is always drunk straight and with meals and only at certain types of outlets, whereas western spirits are always drunk in styles other than straight, before or after meals and at different types of outlets. In the EC view, the reality is much more complex than this black-and-white picture. According to the European Communities, although the consumption patterns of soju and western spirits are not identical, there is a substantial degree of overlapping and, therefore, competition between the two categories.

1.79. The European Communities further argues that Korea's submission focuses exclusively on the most traditional consumption patterns and disregards the rapid emergence of a new drinking culture which is increasingly comparable to that of other developed countries, especially among young consumers.¹⁹⁴

1.80. Thirdly, in the EC view, consumption patterns are affected by prices, which in turn are affected by tax differentials. It is certainly not a coincidence if in the present case the less taxed product is consumed more often with meals¹⁹⁵ or if the more taxed products are more often found at expensive outlets than at less expensive ones. According to the European Communities, it is necessary to discern those differences which may reflect the genuine "tastes and habits" of the Korean consumers from those which have been created or, at least, "fossilised" by discriminatory taxation.

1.81. Finally, the EC view is that Korea's sweeping and categorical statements regarding the end uses of the different types of spirits are not supported by any evidence whatsoever, even though Korea is in possession of the necessary market surveys. According to the European Communities, the Panel should draw appropriate inferences from Korea's refusal to disclose those surveys.

(d) Drinking Styles

1.82. The first difference alleged by Korea is that soju is always drunk straight. The EC view is that this is simply not true. The most frequent style for drinking soju is straight. But soju is also consumed in other styles, including mixed with other non-alcoholic beverages, especially by young consumers. This is attested by the growing sales in recent years of soju-based pre-mixes, of which some samples have been provided to the Panel. For tax purposes, soju-based pre-mixes are not considered as soju but as liqueurs. Nonetheless, the success of those pre-mixes proves that Korean consumers enjoy drinking soju mixed with other beverages.¹⁹⁶

1.83. Furthermore, western-style spirits also are consumed straight. In particular, straight is the most frequent drinking style of brandy. Similarly, whisky is most often consumed alone, diluted with water or on the rocks, all of which are effectively "straight".

(e) Drinking Occasion

1.84. The European Communities notes that, according to Korea, soju is consumed always with meals. In the EC view, this is an over-simplification. Soju is often, but not always, consumed with meals. For instance, according to the European Communities, sojubangs are one of the most typical places for drinking soju. Yet sojubangs are bars and not restaurants. In addition, it is the EC view that soju-based cocktails are rarely consumed with meals. On the other hand, other spirits can be and sometimes are drunk with meals, even if admittedly less frequently than soju.

1.85. The European Communities further notes that according to Korea, the reason why Korean consumers drink soju and no other spirit with their meals is because soju has a unique "harsh" (or "rough") flavour that goes well with Korean hot and spicy food.

1.86. According to the European Communities, this "culinary exception" argument, however, is far from convincing. In the EC view, the traditional tastes and habits of Korean consumers alone are not sufficient to explain why soju is consumed more often with meals. Koreans do not eat hot and spicy food all the time. Nor does all soju respond to the "harsh/rough" description, as allegedly evidenced by the advertisement for the soju brand Jinro Bisun which praises itself on being "mild".¹⁹⁷ In the EC view, there are other spirits which may go just as well with hot and spicy food, such as vodka or, as pointed out by Mexico in its oral intervention before the Panel, tequila. Finally, the European Communities states that the peculiarities of Korean cuisine would not explain why, according to Korea, soju predominates in "inexpensive Chinese restaurants" but not in the rather more expensive Japanese restaurants.

1.87. According to the European Communities, one of the main reasons why soju is more often consumed with every-day meals than western-style spirits is simply because western-style spirits are more expensive, to a large extent as a result of protective taxation. In the EC view, if western-style spirits were taxed as soju, they would be less expensive and Korean consumers could afford to drink them with meals more often.

1.88. The European Communities also notes that Korea does not claim that distilled soju is consumed with meals, yet until the 1960s most soju was distilled soju. The European Communities asks whether this means that Koreans' allegedly traditional habit of drinking soju with meals did not start until the 1960s? or rather, as it seems more plausible, that most Koreans cannot afford to drink their every-day meals with the now rather expensive distilled soju and instead tend to keep it for special occasions and drink it in smaller doses than required to accompany a meal, just like they do with western-style spirits?

To continue with Place of Consumption

¹⁷⁹ EPCT/A/PV.9, at 7.

¹⁸⁰ E/CONF.2/C.3/SR.11 p. 1 and Corr.2.

¹⁸¹ Ibid., p.3

¹⁸² E/CONF.2/C.3/SR40, at 2.

¹⁸³ Ibid.

¹⁸⁴ Panel Report on EEC - Measures on Animal Feed Proteins, supra., para. 4.3.

¹⁸⁵ The European Communities states that giving a narrow meaning to "like products" is also justified by the inescapability of violation in case of taxation of foreign products in excess of domestic products.

¹⁸⁶ Panel Report on Japan - Taxes on Alcoholic Beverages I, para. 5.7. At the same paragraph, the Panel added that:

¹⁸⁷ Panel Report on Japan - Taxes on Alcoholic Beverages, para. 6.28

¹⁸⁸ Appellate Body Report on Canada - Certain Measures Concerning Periodicals, supra., at 28.

¹⁸⁹ See e.g. Working Party on "Brazilian Internal Taxes", adopted 30 June 1949, II/181, 185, at para. 16; Panel Reports on United States - Taxes on Petroleum and Certain Imported Substances, supra., para. 5.1.9, United States - Measures Affecting Alcoholic and Malt Beverages, supra., 271; and United States - Measures affecting the Importation, Sale and use of Tobacco, adopted on 4 October 1994, DS44/R, paras. 99-100.

¹⁹⁰ Panel report on Japan - Taxes on Alcoholic Beverages II, supra., para. 6.22.

¹⁹¹ Ibid., para. 5.9.

¹⁹² Appellate Body Report on Canada - Certain Measures Concerning Periodicals, supra, p.28.

¹⁹³ Panel Report on Japan - Taxes on Alcoholic Beverages I, supra, para. 5.7.

¹⁹⁴ See the Sofres Study, pp 3-7. According to the Sofres Report (at p. 4), the main changes in the Korean drinking culture are:

¹⁹⁵ The EC claims that Korea defies economic logic when it states that "In Korea, standard soju is the drink one finds on the dinner table, the drink that is consumed with meals. As such it is an inexpensive beverage" (Korea's First Submission, para. 130)

According to the European Communities, it seems more logical that soju has acquired the status of "every-day meal drink" because it is inexpensive, rather than the opposite. If soju had been subjected to import quotas and then to dissuasive import duties and very high discriminatory internal taxes, it could have never achieved that status.

¹⁹⁶ The Sofres Report describes this development in the following terms, at p.24:

¹⁹⁷ Attachment 6 to Korea's First Submission.