2. Legal Arguments

(a) "directly competitive or substitutable"

27. In order to determine which products belong to this category, the following criteria should be applied³⁰⁷:

1. Physical characteristics;
2. common end-uses;
3. tariff classification;
4. the "market-place".

(i) Physical Characteristics

28. Mexico disagrees with Chile's assertion that the products under consideration have virtually no characteristics in common. The two characteristics mentioned in the Panel Report on Korea � Taxes on Alcoholic Beverages ³⁰⁸ are fully applicable to the comparison between pisco and tequila as well: they are both distilled spirits, bottled and labelled in a similar manner.
29. Mexico notes that the Panel Report on Japan � Taxes on Alcoholic Beverages II states that the physical characteristics are not the decisive criterion for determining whether products are competitive or substitutable. ³⁰⁹ However, Mexico agrees with the European Communities when it asserts that "if two products have sufficiently similar physical characteristics, such similarity may of itself be sufficient to conclude that the products in question are apt to serve for the same end-uses".³¹⁰

(ii) End-uses

30. Mexico notes that the European Communities, in their submission, refers to the findings of the Panel in Japan � Taxes on Alcoholic Beverages II according to which the end-uses are the "decisive criterion" for establishing whether two products are directly competitive or substitutable.
31. Mexico further notes that the 1997 SM Survey ³¹¹ shows that tequila and pisco have the same end-uses in at least the following respects:

1. drinking styles: both spirits are most likely to be consumed with a mixer beverage (such as cola);
2. drinking occasions: Chilean consumers of spirits considered "parties", "with friends", "family meetings" and "weekends" to be the most common occasions for the consumption of pisco or tequila, while the categories "after work", "aperitif", "during week" and "digestive" were considered less common;
3. drinking places: both spirits are mainly consumed "at home" or at a "friend's house";
4. availability in sales channels: the most common sales channels for both beverages are "supermarkets" and "liquor stores", while "gift shops", "duty-free", "others" and "airlines" are uncommon as sales channels in both cases; and
5. types of consumers: in the Chilean market, women tend to drink more pisco and/or tequila than men.

32. Furthermore, the recipe brochures mentioned by the European Communities ³¹² suggest that the producers of pisco themselves perceive pisco and tequila as products with a common final use, i.e. the preparation of "margaritas".

(iii) Tariff Classification

33. Mexico notes that both beverages come under HS subheading 2208.90, i.e. they are at the same six-digit level, which is the Harmonized System's most advanced and specific classification. In Mexico's view, it is also worth noting that the Appellate Body in Japan � Taxes on Alcoholic Beverages II pointed out that if "sufficiently detailed, tariff classification can be a helpful sign of product similarity".³¹³
34. Regarding the mutual substitutability of tequila and pisco in the Chilean market, Mexico refers to the EC's Exhibit 22, Table 4.1.2, which purportedly shows that 17 per cent of consumers of Chilean spirits would buy tequila if they intended to buy pisco and did not find it, while 56 per cent would buy pisco if they could not find tequila.
35. As regards cross-price elasticity, the analysis submitted by the European Communities ³¹⁴ shows a 44.8 per cent change in response to a 27 per cent tax on all spirits instead of the tax currently applied. The table shows a considerable price elasticity.

(iv) Recognition of Government of Chile

36. Mexico points out that in addition to the evidence presented by Mexico, it should be noted that Chile implicitly accepted that pisco and the other spirits at issue in this case are directly competitive or substitutable. More specifically, Chile states that the examples provided by the European Communities concerning the background to the new law "show that legislative representatives of the regions that produce pisco in Chile were seeking to minimize the adverse effects of a New Chilean System on pisco producers and, because adverse effects could not be avoided, also sought other government help for their constituents". ³¹⁵ Mexico asks rhetorically, what adverse effects would the new system have if the products were not "directly competitive or substitutable"?

(b) "not similarly taxed"

37. As can be seen in Table I of Mexico's submission, the tax differentials show a margin of discrimination of 120 per cent and 174.4 per cent between pisco and tequila under the "Old Chilean System" and the "New Chilean System" respectively.
38. Moreover, Mexico agrees with the panel and the Appellate Body in Japan � Taxes on Alcoholic Beverages II when they argue that the amount of differential taxation must be more than de minimis, as determined on a case-by-case basis. ³¹⁶ In the case at issue, the differentials in the rate of taxation suffice to establish that they are in excess of any de minimis criterion.

(c) "so as to afford protection"

39. Mexico notes that the panel in Korea � Taxes on Alcoholic Beverages ³¹⁷ endorses the conclusion of the Appellate Body in Japan � Taxes on Alcoholic Beverages II, focussing on the objective factors underlying the tax measure in question including its design, architecture and the revealing structure.
40. In the case at issue, the following elements show that Chile imposed its tax scheme in order to afford protection to its domestic industry:

1. the amount of the differential taxation (see Table I);
2. the structure of the Chilean tax system;
3. the fact that the great majority of distilled spirits produced on the Chilean market have an alcohol strength of 35� or less;
4. the fact that most imported distilled spirits have an alcohol strength of more than 35� and, in many cases, more than 39�.

41. Mexico notes that as regards design, architecture and the revealing structure of the measure, Chile accepted that its new system was "biased" against relatively higher alcohol products. Chile denied however, that this was a means of affording protection to domestic production ³¹⁸ on the grounds that the producers of other spirits could benefit from the system by diluting their products with water or switching their exports to beverages benefiting from a lower level of taxation. ³¹⁹
42. According to Mexico, Chile's proposals are obviously not only unworkable, but also irrelevant. The product that Mexico is interested in exporting to Chile is tequila. Apart from the fact that under Chilean law, tequila must have a minimum alcohol strength of 40�³²⁰, and that it must comply with the corresponding Mexican Official Standard to be marketed under that name, it is obvious that the consumer of tequila wants tequila and not water. Mexico points out that if its intention had been to export spirits with a strength of 35� or less, it would not have bothered to participate as a third-party in this dispute. In Mexico's view, these arguments by Chile simply confirm that the protection afforded to national production is so high that the only way of competing on an equal footing with Chilean beverages would be to change the quality of the imported products or the products themselves.
43. Mexico further notes that Chile agrees with the European Communities that in its system, different levels of alcohol strength are taxed differently, but argues that Article III:2 of the GATT 1994 has never been interpreted to require a direct proportionality rule. It points out that the Panel in Japan � Taxes on Alcoholic Beverages I, only suggested that taxation must be based on objective criteria. The Chilean system of taxation is not, nor can it be considered to be, based on "objective criteria". Chile gives no satisfactory explanation of how a margin of discrimination of 174.07 per cent ³²¹ applied to the difference of little more than 4� in alcohol strength under the Chilean tax structure (between 35� and 39�) can be seen as being based on "objective criteria". Mexico is not challenging the fact that the Chilean tax varies according to the alcohol strength of the beverages: it is challenging the way in which that tax is applied. Mexico further argues that, the purpose and effect of the Chilean system is to protect national production against imports of products that are directly competitive with or substitutable for Chilean beverages.
44. Mexico argues that in considering the fact that the national production of spirits in Chile is dominated by pisco with an alcohol strength of 35� or less, it is interesting to refer to Table II below. Mexico points out this table was prepared on the basis of Table 8 which forms part of Annex III of Chile's written submission, and shows, in terms of volume, the share of pisco with a strength of 30� and 35� in the Chilean market. Moreover, according to the European Communities, with the exception of 1992, the share of pisco in the Chilean spirits market grew constantly from 44.1 per cent in 1982 to 73.8 per cent in 1996.
45. Mexico notes that Chile, in its own written submission, ³²² reveals the substantial share of imports of spirits with a strength of 40� or more in the Chilean market. It should be observed that tequila is not expressly mentioned³²³, so that the percentage shares are in fact higher than those shown. ³²⁴
46. Mexico further notes that according to the European Communities, imports of whisky and tequila (the two most popular beverages after pisco) account for 93.6 per cent and 100 per cent respectively of the Chilean market for those spirits. ³²⁵ It is curious that under the New Chilean System, these two products are taxed the most heavily; and, the tax applied to tequila will now increase from 30 per cent to 47 per cent. This in itself should be sufficient evidence that the Chilean system is designed to protect domestic production.
47. Mexico concludes that it has proved that the Chilean tax system is contrary to the provisions of Article III.2, second sentence, of the GATT 1994, and therefore requests that the Panel:

1. find that the Chilean system for the taxation of spirits violates the second sentence of Article III.2 of the GATT 1994 in order to help the DSB in making the recommendations or issuing the resolutions provided for under the GATT 1994;
2. find that the measure at issue nullifies or impairs benefits under the GATT 1994 by favouring pisco in a manner contrary to that Agreement.

Table I

Margins of Discrimination

	Pisco	Tequila	Margin of discrimination
Tax applied under the "old system"	25%	30%	120%
Tax applied under the "new system"	27%	47%	174.07%

Table II

Share of Pisco with a Strength of 35� or Less in the Chilean Spirits Market

	1991	1992	1993	1994	1995
Pisco 30�	64.51%	60.50%	52.29%	41.57%	33.53%
Pisco 35�	5.27%	8.45%	17.56%	26.69%	35.57%
Total of the two piscos	69.78%	68.95%	69.85%	68.26%	69.10%

Table III

Market Share of Spirits with a Minimum Strength of 40�

	1991	1992	1993	1994	1995	1996	1997
Volume	82.65%	81.31%	76.37%	66.81%	60.00%	50.83%	54.37%
Value	86.57%	84.69%	79.38%	78.02%	71.09%	59.5%	59.63%

C. Peru

48. Peru briefly stated that it considers that the Chilean system of taxation of alcoholic beverages is discriminatory, contrary to Article III:2 of GATT 1994, and causes harm to Peruvian exports of alcoholic beverages to Chile.
49. Peru also referred to an issue it has raised in two DSB meetings, regarding the propriety of the use of the term pisco by Chile. ³²⁶ Peru stated that it was an exporter of pisco to Chile and since the Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS) would only enter into force in Chile and Peru in the year 2000, Peru wished to reserve its rights to invoke Article 22.1 of the TRIPS Agreement and other provisions related thereto. Peru considered that the denomination geographical indication "Pisco" was Peruvian and as such gave Peru exclusive rights.

D. United States

1. Introduction

50. The United States asserts that Latin America is a key growth market for its distilled spirits producers, who have undertaken significant efforts to promote U.S. products in the region. U.S. exports of distilled spirits to Peru and Venezuela have grown by 52% and 116% respectively over the past two years, following the elimination of discriminatory measures. ³²⁷ However, despite significant economic growth in Chile, U.S. distilled spirits exports have not witnessed similar growth, due to Chile's history of discriminatory taxation.
51. The United States further asserts that in 1996, it exported approximately $1.1 million in distilled spirits to Chile. ³²⁸ Whiskey, which includes Bourbon and Tennessee whiskey, two distinctive types of American whiskies produced from fermented grains, accounted for 29.7% of the total U.S. exports of the distilled spirits at issue in this dispute. ³²⁹ Other traditional spirits at issue are rum, gin and vodka, which accounted for 61.4% of these exports. The remaining 9% of U.S. exports to Chile include liqueurs (3.3%) and pre-mixed cocktails. ³³⁰
52. The facts developed by the European Communities conform to the experience of the U.S. exporting industry. Chile�s measures at issue consist of two elements: (1) the current Transitional System, which applies different tax rates for particular product categories, and (2) the New Chilean System, which will tax on the basis of alcohol content (through Law 19.534, an amendment to its Decree-Law 825/1974).
53. As a legal matter, the United States considers that:

1. The Transitional System on distilled spirits, applicable until 30 November 2000, is inconsistent with the second sentence of Article III:2 of GATT 1994 because it provides for lower internal taxes on the domestic spirit "pisco" than on directly competitive or substitutable imported spirits falling into the tax categories of "whisky" and "other spirits", and is applied so as to afford protection to Chile's domestic production of pisco; and
2. The New Chilean System on distilled spirits, applicable as of 1 December 2000, is inconsistent with the second sentence of Article III:2 of GATT 1994 because it results in the imposition of lower taxes on domestic spirits with an alcohol content of 35 degrees or less than on directly competitive or substitutable imported spirits that have a higher alcohol content, and is applied so as to afford protection to Chile's domestic production.

2. Legal Arguments

(a) General

54. As the European Communities notes in its submission, the Appellate Body has clarified the interpretation of Article III:2, second sentence in Japan - Taxes on Alcoholic Beverages II. The Appellate Body stated that in order to find an internal tax measure inconsistent with the second sentence of Article III:2, three separate elements must be satisfied:

1. the imported products and the domestic products must be "directly competitive or substitutable products" which are in competition with each other;
2. the directly competitive or substitutable imported and domestic products must be "not similarly taxed"; and
3. the dissimilar taxation of the directly competitive or substitutable imported domestic products must be "applied ... so as to afford protection to domestic production", ³³¹ which requires an examination of "the underlying criteria used in a particular tax measure, its structure, and its overall application..." ³³² These underlying criteria include the "design, the architecture, and the revealing structure of a measure", the "magnitude of the dissimilar taxation", and "all the relevant facts and all the relevant circumstances in any given case".³³³

55. The United States notes that, with respect to the first element -- whether two products are "directly competitive or substitutable" -- the European Communities argues convincingly that all types of pisco, regardless of alcohol content, are the same product; that pisco and all other distilled spirits share the same basic physical characteristics; that pisco and all other distilled spirits have the same end-uses; that pisco and all other distilled spirits fall within the same HS heading, HS 22.08; that pisco and all other distilled spirits are sold in the same sales channels; and that there is significant cross-price elasticity between pisco and the all other distilled spirits. In fact, the European Communities establishes that not only is there a close competitive relationship between imported spirits and domestic pisco, but that all distilled spirits at issue are directly competitive or substitutable with each other.
56. According to the United States, there is also ample evidence that both the Transitional System and the New Chilean System involve dissimilar tax treatment as between domestic and imported products, and afford protection to domestic production.

To continue with Old Chilean System: Background

³⁰⁷ These criteria were established by the panel and adopted by the Appellate Body in Japan - Taxes on Alcoholic Beverages II , supra., p. 25.

³⁰⁸ Panel Report on Korea � Taxes on Alcoholic Beverages, para. 10.67.

³⁰⁹ Panel Report on Japan � Taxes on Alcoholic Beverages II, para. 6.22.

³¹⁰ See EC First submission, para. 109.

³¹¹ See EC Exhibit 21.

³¹² See EC First submission, para. 140.

³¹³ Appellate Body Report on Japan � Taxes on Alcoholic Beverages II, supra., p. 21. Various panels shared that opinion, such as EEC � Measures on Animal Feed Proteins, supra., Japan � Taxes on Alcoholic Beverages I, supra., and United States � Standards for Reformulated and Conventional Gasoline, supra.

³¹⁴ See Table 4.2.2, EC Exhibit 22.

³¹⁵ See Chile First submission, para. 71.

³¹⁶ Appellate Body Report on Japan � Taxes on Alcoholic Beverages II, supra., p. 27. This same argument was used by the Panel in Korea � Taxes on Alcoholic Beverages, supra., para. 10.100.

³¹⁷ Panel Report on Korea � Taxes on Alcoholic Beverages, para. 10.101.

³¹⁸ Chile First submission, para. 68.

³¹⁹ Ibid., para. 73.

³²⁰ See Article 12 of the Regulation Implementing Law No. 18455, published in the Official Journal of 23 October 1986 and appearing as Appendix 2.2 of EC Exhibit 12.

³²¹ See Table I.

³²² Ibid., Annex III, Tables 1-7.

³²³ The table provided by Chile only mentions the following spirits as having a strength of 40� or more: whisky, rum and other white spirits, gin and geneva, and, for 1996 and 1997, vodka as well.

³²⁴ See Table III.

³²⁵ Tables 9.A and 9.B of EC First submission, p. 36

³²⁶ Peru has previously expressed its position at two DSB meetings on 18 November 1997, and 25 March 1998 (see WT/DSB/M38 & WT/DSB/M/44).

³²⁷ According to the United States, in June 1993 Peru replaced its discriminatory selective consumption tax (10% for pisco and 50% for other spirits) with a single ad valorem tax of 10%. Similarly, Venezuela presently assesses a rate of Bs.10 per liter for all alcoholic beverages.

³²⁸ See U.S. Exhibit 1, the 1996 data on U.S. distilled spirits exports to Chile (by class). The 1996 data are the most recent and most complete available.

³²⁹ See 27 CFR Sec. 5.21.

³³⁰ The United States notes that the majority of the 5.7% of the "other" category in U.S. Exhibit 1 consists of ethyl alcohol, a product used for industrial purposes and as a primary input for the production of other distilled spirits; ethyl alcohol is not at issue in this dispute. A de minimis portion of the "other" category includes pre-mixed cocktails.

³³¹ Appellate Body Report on Japan - Taxes on Alcoholic Beverages II, supra. The three elements approach was most recently used in the Panel Report on Korea - Taxes on Alcoholic Beverages, supra, paras. 10.34-10.102.

³³² Ibid., p. 29.

³³³ Ibid., p. 29.