What's New?
 - Sitemap - Calendar
Trade Agreements - FTAA Process - Trade Issues 

espa�ol - fran�ais - portugu�s
Search

Korea - Taxes on Alcoholic Beverages

Report of the Panel

(Continued)


    VII. Answers to questions

    A. European Communities

  1. In response to a question concerning levels of cross-price elasticity, the European Communities states that:
  2. (a) As noted by the Appellate Body in Japan - Taxes on Alcoholic Beverages II, how much broader the category of "directly competitive or substitutable products" may be in a given case is a matter for the panel to determine based on all the relevant factors in that case. 285 Accordingly, it would be inappropriate to try and define a standard of general application, and in particular a quantitative one.

    (b) There is no support for Korea�s contention that the notion of "directly competitive or substitutable products" must be interpreted "strictly". On the contrary, as demonstrated by the complainants, an examination of the drafting history of GATT and of previous Panel and Appellate Body reports (including the two cases on Japan - Taxes on Alcoholic Beverages II) shows that in practice the "directly" has been given a rather broad interpretation.

    (c) In Japan - Taxes on Alcoholic Beverages II, the Appellate Body made it clear that cross-price elasticity is not "the decisive criterion" 286 for establishing whether two products are directly competitive or substitutable. According to the Appellate Body, cross-price elasticity is but one of the means of examining a relevant market. In turn looking at competition in the relevant markets is just "one among a number of means" 287 of identifying the products that are directly competitive or substitutable in a particular case. The other means mentioned by the Appellate Body are the physical characteristics, the end uses and the customs classification of the products.

    (d) Furthermore, the relevance of a particular level of cross-price elasticity may vary according to the circumstances of each case. For instance, if two products have been sold for a long time and under similar conditions on the same geographical market, a "very low rate of cross-price elasticity" could be an indication that they re not "directly competitive or substitutable".

    (e) On the other hand, in a situation where one of the products concerned has dominated a geographical market for a long time and the other product is a new entrant in that market (e.g. because until then it has been excluded therefrom by import and/or tax barriers), it would be unwarranted to conclude from the mere fact that the initial cross-price elasticity is relatively low that the two products are not "directly competitive or substitutable". The more so in the case of products such as spirits, where market penetration is slow and short-term reactions to price changes tend to be relatively low.

    (f) Korea�s position in this case appears to be that Article III:2, second sentence, would apply only if and when imported products succeed in establishing themselves in a market. Foreign products would have to achieve first a level of market penetration such that it is possible to prove statistically a high rate of cross-price elasticity. This approach, however, disregards the obvious fact that protective taxes may be a factor that delays or prevents imported products from ever reaching that level of market penetration. Clearly, such an approach is at odds with the well-established principle that Article III protects "competitive opportunities". There is no reason to limit such "opportunities" to those that are available to a product in the very short term. Article III protects any competitive opportunities that a given product may have by reason of its inherent characteristics.

    (g) In connection with this question, the European Communities further argued that the Dodwell study does not purport to provide a precise measurement of cross-price elasticity. In order to do that, it would have been necessary to carry out an econometric analysis based on historic sales and price data. In the present case, however, that type of analysis was precluded by the fact that western spirits have been virtually excluded from the Korean market until only a few years ago. This means that the available sales and price data are too few to allow a statistically valid analysis.

    (h) The Dodwell study has a more modest purpose. It aims at testing by means of a consumer survey the hypothesis that a reduction in the prices of western spirits and/or increase in the prices of soju resulting from the elimination of the existing tax differentials will lead to an increase in the consumption of western-style spirits at the expense of soju. The results of the Dodwell study clearly validate that hypothesis.

    (i) Additionally, the study indicates that the extent of potential substitution could be significant. For example, in one of the possible after-tax harmonization price scenarios the percentage of respondents who would choose whisky instead of standard soju would increase from 14.2% to 23.8%. While, for the reasons explained elsewhere, it was appropriate to distinguish in the survey between premium and standard soju, this has the consequence that the survey only measures the shift of respondents from standard soju to western spirits, and not the additional shift from premium soju to western spirits. Furthermore, since the prices of premium soju are not increased in parallel with the prices of standard soju (so as to reflect the fact that taxes would be increased on all diluted soju), but rather decreased, the survey overestimates the shift from standard to premium soju at the expense of the shift from standard soju to western spirits. In comparison, in the same scenario, the percentage of respondents choosing premium soju would increase from 12.6% to 19.8%. Thus, the Dodwell study suggests that the elasticity of substitution between standard soju and whisky could be higher than the elasticity of substitution between standard soju and premium soju, two products which have been described by Korea as being close "substitutes".

    (j) Moreover, it should be borne in mind that a consumer survey like the Dodwell study necessarily underestimates the degree of potential competition between soju and western spirits.

    (k) In the first place, the Dodwell survey can show only the immediate reaction of consumers to price changes. Yet, spirits consumption is to a large extent based on habits, which only change gradually. This means that, over a certain period of time, the price changes resulting from the elimination of tax differentials will lead to a more substantial shift from soju to western spirits than the one shown in the Dodwell study.

    (l) Secondly, it must be recalled that western spirits are new entrants in the Korean market and still hold only a small share of that market. This has two implications. The first one is that the respondents are generally less familiar with western spirits than with soju. This leads to a lower response in the survey than in the case of two products which were both well known to the respondents. The second implication is that western spirits still have considerable potential to increase their share of the market through marketing efforts (in advertising, distribution, etc.). The impact of those efforts would be considerably boosted by the price changes envisaged in the Dodwell study. On the other hand, the continued application of protective taxation would discourage such efforts. The interaction of these two factors, however, is not and cannot be addressed by a survey like the Dodwell study.

    (m) Finally, the Dodwell study considers only the price changes that could result directly from the elimination of the existing tax differentials. It does not take into account that the elimination of tax differentials could lead as well to a decrease of the pre-tax prices of western spirits and, consequently, to further substitution.

  3. In response to a question as to whether there is a de minimis standard in assessing the question of "so as to afford protection", the European Communities states that:
  4. (a) In Japan - Taxes on Alcoholic Beverages II, the complainants expressed different views with respect to the interpretation of the third element of Article III:2, second sentence. The European Communities argued that the measures at issue afforded protection to domestic production because a majority of the sales of the less taxed product (shochu) were domestically produced in Japan. In turn, the United States argued that the measures afforded protection to domestic production because their structure and design evidenced that they were not aimed at achieving any legitimate policy objective but only at providing an advantage to Japanese shochu. The approach established by the Appellate Body in Japan - Taxes on Alcoholic Beverages II appears to combine both positions. While putting the emphasis on the objective aim of the measures, as revealed by their structure and design, 288 the Appellate Body also noted the fact that Japanese shochu was "isolated" from imports of shochu289. Thus, the Appellate Body seems to have considered that the demonstrated actual protective effects of a measure may be taken into account as an indication that a measure is aimed at protecting domestic production. 290

    (b) On the other hand, there is no suggestion in Japan - Taxes on Alcoholic Beverages II that in order to meet the third element of Article III:2, the tax measures must afford a minimum "degree" of protection to the less taxed product. The Appellate Body agreed with the Panel that the reasoning required under Article III:2, second sentence, is the following:

    If directly competitive or substitutable products are not "similarly taxed", and if it were found that the tax favours domestic products, then protection would be afforded to such products, and Article III:2 second sentence is violated. 291

    (c) Thus, the third element of Article III:2, second sentence, is concerned only and exclusively with the question whether, by taxing one product less than another directly competitive or substitutable product, the measures "favour" domestic production over imports, not with the extent of the "protection" afforded to the less taxed 292 product. In other words, the third element of Article III:2, second sentence, is not about how much protection is afforded, but rather about who is protected.

    (d) The view that in order to establish a violation of Article III:2 it is necessary for the complainants to show that the measures actually afford a certain "degree" of protection by effectively reducing sales of imports above a de minimis level would be in contradiction with the well established principle that GATT Article III is concerned with the protection of competitive opportunities and not of actual trade flows. More specifically, according to the Appellate Body:

    [i]t is irrelevant that the "trade effects" of the tax differentials between imported and domestic products, as reflected in the volumes of imports, are insignificant or even non-existent: Article III protects expectations not of any particular trade volume but rather of the equal competitive relationship between imported and domestic products. 293

    (e) In sum, the European Communities is of the view that the third element of Article III:2, second sentence, does not introduce another de minimis threshold, in addition to those which result from the application of the first and the second element. If two products are "directly" competitive or substitutable, any tax differential above the de minimis level must be deemed to affect that competitive relationship and, as result, to "protect" the less taxes product. The only remaining issue is then whether the "protection" given to the less taxed product favours a "domestic production".

  5. In response to a question concerning the price mix of exports to Korea compared to other markets, the European Communities states that:
  6. (a) The pre-tax prices of western spirits are not higher than the pre-tax prices of soju only because of Korea�s Liquor Tax system. The differences in prices reflect also differences in production and transportation costs as well as the impact of tariffs. Nevertheless, the pre-tax prices of western spirits are higher than they would be under a neutral tax system.

    (b) As already explained in the EC submissions, one of the effects of Korea�s tax regime is that higher priced premium brands account for a disproportionate share of the sales of western spirits. For example, as shown by Annex 1, premium brands account for as much as 70 per cent of all sales of Scotch whisky in Korea. The same Annex shows that in a number of representative export markets with a neutral system of taxation the proportion of premium brands is much lower: 3 per cent in Australia; 6 per cent in New Zealand and 14 per cent in Venezuela. The preponderance of premium brands in the Korean market as compared to other export markets is further confirmed by the fact that whereas in 1997 the average unit price of all exports of Scotch whisky (for 70 cubic litre bottles at 40% volume) was �2.79, the average unit price of the exports of Scotch whisky to Korea was �4.42.

  7. In response to a question concerning comparison of legal standards under competition law with standards under Article III, the European Communities states that:
  8. (a) The basic criteria applied in order to define the relevant product market for the purposes of EC Competition law are the same as those applied in order to establish whether products are directly competitive or substitutable for the purposes of GATT Article III:2, second sentence".294

    (b) There is, nevertheless, an essential difference. When applying GATT Article III:2, first sentence, Panels must take into account the "potential" competition which would materialise between the products concerned in the absence of the tax differential in dispute and not the "actual" competition existing under current taxation conditions. In contrast, competition authorities tend to consider tax differentials as a permanent barrier to competition and disregard any additional competition which may arise from removing that barrier. 295 As a result, the scope of the "relevant product" markets defined for competition purposes will generally be narrower than the scope of "directly competitive products" defined for the purposes of Article III:2, second sentence.

    (c) It must also be borne in mind that the notions of "competition" and of "substitutability" are relative ones. From an economic perspective, two products are not either "competitive" or "non competitive". Rather, products are "more or less" competitive. For that reason, as important as the criteria for defining a relevant market or for defining the notion of "directly competitive or substitutable products" is the degree of competition which is deemed relevant in each case. That degree will determine the standard by which the criteria are to be interpreted. That standard may vary depending on the purpose of the legal provision to be applied. It may vary also from one jurisdiction to another.

    (d) This link is expressly recognised in an EC Commission Notice on the definition of relevant markets for Competition law purposes, which states that "the concept of relevant market is closely related to the objectives pursued under Community competition policy". 296 Further, that Notice acknowledges that the definition of the relevant market may vary depending "on the nature of the competition issue being examined." 297

    (e) In this regard, it is clear that the objective of competition law is very different from the objective pursued by GATT Article III:2, and more generally by the WTO Agreement. The general objective of competition law is to preserve a certain degree of competition against action by the market participants. If the competition authorities of a certain country aim at maintaining a high degree of effective competition, they will apply the relevant criteria strictly, thereby arriving at a narrow definition of the relevant market.

    (f) On the other hand, the purpose of GATT Article III:2, second sentence is to prevent Members from applying internal taxation so as to afford protection to domestic production. Unlike the objective of competition law, the objective of Article III:2, second sentence, is furthered by a broad interpretation of the relevant criteria, rather than a strict one.

    (g) For the above reasons, the EC is of the view that the decisions taken by its competition authorities with regard to the definition of relevant product markets are devoid of relevance for the purposes of applying the notion of "directly competitive or substitutable products" in this dispute.

    (h) In this regard, a parallelism can be drawn to the notion of "like product". The criteria for applying the notion of "like products" are the same in all the GATT provisions where that notion is found. Yet, in Japan - Taxes on Alcoholic Beverages II, the Appellate Body confirmed that the notion of "like products" is a relative one which may have a different scope in each GATT provision concerned. 298 In Article III:2, first sentence, it must be construed narrowly. In other GATT provisions, it may be construed more broadly. A fortiori, the notion of "directly competitive or substitutable" may also have a different scope in Article III:2 of GATT and in the competition laws of Members, which pursue an altogether different objective.

    (i) More relevant for the interpretation of GATT Article III:2 is the case law of the European Court of Justice (ECJ) regarding the application of Article 95 of the EC Treaty, 299 whose wording is almost identical to that of GATT Article III:2 and therefore, unlike EC Competition Law, shares a similar purpose. In a long line of cases, the ECJ has concluded that all distilled spirits are either "similar" (the equivalent concept of "like") or "directly competitive or substitutable".300

  9. In response to a question concerning the relevance of production processes to assessing whether products are like or directly competitive or substitutable, the European Communities states that:
  10. (a) Similarities or differences in production processes may be relevant only to the extent that they affect the characteristics of the products. This principle flows clearly from the Panel Report on US - Standards for Reformulated and Conventional Gasoline. 301 Although that Panel report is concerned with GATT Article III:4, the European Communities is of the view that the same principle applies also with respect to Article III:2.

    (b) In the present case, it is relevant that all distilled spirits are obtained by the same manufacturing process (distillation) because it has the consequence that all of them share the same basic physical characteristics. On the other hand, differences regarding the method of distillation (continuous or pot-still), the method of filtration (through white birch or other methods) or the production volume (artisanal v. industrial) are irrelevant because they have either no impact at all or only a minor impact on the physical characteristics and end uses of the products.

To continue with United States


285 Appellate Body Report, supra., p. 25.

286 Ibid.

287 Ibid.

288 Ibid., p. 29.

289 Ibid., p. 31.

290 See also the Appellate Body Report on Canada - Certain Measures Concerning Periodicals, supra., pp. 31-32.

291 Ibid., p. 29.

292 In contrast, under Article III:2, first sentence, it is not necessary to demonstrate that a difference in taxation between two types of "like" products leads to discrimination between imports and "domestic production". Like products must always be taxes equally.

293 Ibid., p. 16.

294 The applicable EC competition regulations define the notion of "relevant product market" for the purposes of EC Competition law as follows:

A relevant product market comprises all those products and/or services which are regarded as interchangeable or substitutable by the consumer, by reason of the products' characteristics, their prices and their intended use.

See the Commission Notice on the definition of relevant market for the purposes of Community competition law (published in OJ of 9.1.97, C 372/5, hereafter "the Notice"), para 7.

295 Ibid. at para 42. The decision in the Case No IV/M 938 - Guinness/Grand Metropolitan mentioned by Korea in one of its questions to the EC provides an excellent illustration of this difference. The parties to the merger had provided to the Commission consumer surveys which suggested that all spirits were within the same relevant market. The Commission, however, disregarded those surveys because:

where those surveys (most of which were originally aimed at addressing taxation issues) employed price-change data, the overall levels of change (which mainly reflected changes in taxation) were much higher than those normally used by competition authorities as an aid to market definition (para 10).

296 Notice, para 10.

297 Notice, para 12.

298 Appellate Body Report on Japan - Taxes on Alcoholic Beverages II, supra., pp. 21-22

299 Article 95 of the EC Treaty reads as follows in the pertinent part:

No Member State shall impose, directly or indirectly, on the products of other member States any internal taxation of any kind in excess of that imposed directly or indirectly on similar domestic products.

Furthermore, no member State shall impose on the products of other Member States any internal taxation of such nature as to afford indirect protection to other products.

300 The standard reasoning followed by the ECJ in those cases is the following:

"There is, in the case of spirits considered as a whole, an indeterminate number of beverages which must be classified as "similar products" within the meaning of the first paragraph of Article 95, although it may be difficult to decide this in specific cases, in view of the nature of the factors implied by distinguishing criteria such as flavour and consumer habits. Secondly, even in cases in which it is possible to recognise a sufficient degree of similarity between the products concerned, there are nevertheless, in the case of all spirits, common characteristics which are sufficiently pronounced to accept that in all cases there is at least partial or potential competition. It follows that the application of the second paragraph of Article 95 may come into consideration in cases in which the relationship of similarity between the specific varieties of spirits remains doubtful or contested" (Judgement of the ECJ of 27 February 1980, Commission of the European Communities v Kingdom of Denmark, Case 171/78, ECR 1980, 447, at par. 12)

301 Panel Report on US - Standards for Reformulated and Conventional Gasoline, adopted on 20 May 1996, WT/DS2/R, para 6.11-6.12