6. "So As to Afford Protection"

1.308. Korea notes the EC statement that Korea has dealt with the 'so as to afford protection' matter in a 'perfunctory' manner.²⁷⁹ According to Korea, while it is true that Korea does not spend a great deal of time on the 'so as to afford protection' facet of this case, this is mainly because Korea is so convinced that there are no 'like' or directly competitive or substitutable relationships in this case, such that the 'so as' part of this case will never be reached.

1.309. However, Korea has also argues that taxes in this case are not 'so as to afford protection' -- that the price differences in this case are too great for the tax to have afforded any protection to domestic production, in particular the production of standard soju. Korea notes that according to the European Communities, it is illogical to mention these price differences here: if the products concerned are found to be directly competitive or substitutable 'despite the pre-tax price differentials',²⁸⁰ then these price differentials are not really meaningful. In Korea's view this is not a very compelling point, coming from party which has been arguing that price differences are not relevant to a directly competitive or substitutable finding, and that it is sufficient for the Panel to look at physical characteristics only. Thus, according to Korea, if this Panel, despite the force of precedent, followed the EC view on directly competitive or substitutable, the absolute differences in pre-tax prices would be an entirely separate issue to be addressed in connection with the 'so as to afford protection' requirement.

1.310. Because the differences in price are so large, Korea contends that the additional price difference that can be attributed to the tax would not have any effect on consumer behaviour, and therefore could not afford protection to a domestic industry. Korea states that apart from its dubious appeal to logic, the European Communities does not show how the liquor taxes can afford protection, to standard soju in particular, given the considerable differences in pre-tax prices with western-style liquors.

1.311. Korea notes that, in contrast, the United States has recognised the pertinence of the price differences to this third leg of Article III:2. While it also contests the relevance of Korea's price-based arguments to the determination of 'like' or directly competitive or substitutable relationships, it argues that 'differences in prices . . .[are] more appropriately considered in the third element of analysis under Article III:2: whether the taxes are applied so as to afford protection to domestic production.'²⁸¹ However, Korea notes that the US has not addressed the price differences cited by Korea in connection with the 'so as to afford protection' requirement.

1.312. Accordingly to Korea, it is clear that the complainants have not carried their burden of proof regarding this third requirement of Article III:2 either. They would have had to show that, despite the much higher pre-tax prices of western-style liquors, the tax differential had protected domestic production of standard soju.

1.313. In respect of distilled soju Korea's defence has been different, as the pre-tax prices of distilled soju are even higher than or in about the same range as western-style liquors. However, according to Korea, the sales of this traditional, artisanal product are minimal. The complainants have not shown how the continued presence of this product, taxed at the current rate, would hurt them.

7. Comments of Korea to the EC Trendscope Survey

1.314. Korea notes that at the second substantive meeting of the Panel, the European Communities submitted a new consumer survey which Korea had not seen before.

1.315. Korea submits that interpretation of this survey is not easy, and that the European Communities, by presenting the survey so late in the proceedings, has deprived Korea of the possibility of exploring these problems by questioning specific aspects of it before the Panel.

1.316. Korea claims that some of the "Key Findings of the Consumer survey"²⁸² strongly support points made by Korea by showing, for example, the considerable degree of specialization of Korean outlets for these products. According to Korea, the Trendscope Survey shows that soju is almost never consumed in hotel bars, western restaurants, cafés, room salons and night clubs, but is the drink of choice in Korean restaurants. The survey also allegedly backs up Korea's assertions that whisky is almost never consumed in Korean, Japanese or Chinese restaurants.

1.317. Korea states that the survey also contains results that are inconsistent with Korea's understanding of Korean drinking habits, and is puzzled that 66% of Trendscope respondents claim to drink soju "without a meal" (though only 3% "without food"); and that 86% claim to drink whisky "with food" (though only 7% "with a meal").

1.318. Korea also notes that only 34% of those surveyed by Trendscope responded that they drank soju with their meal. According to Korea, the respondents might have been thinking about meals consumed at home, where according to the Nielsen study, 29.3% of the respondents consumed diluted soju with their meal.

1.319. Korea argues that the Trendscope result is inconsistent with Korea's understanding of the market as far as "on-premise" consumption is concerned. Korea refers to the Nielsen study, wherein it was found that 73% of consumers drank diluted soju with their meals in Korean restaurants.

1.320. Korea further argues that, to interpret the Trendscope results properly, it is necessary to know how respondents distinguished between "meal" and "food". According to Korea, the distinction between "food" and "meal" is vague and dependent on context in Korean and English. Korea recalls that in referring to the main end use of soju as meals, it has excluded snacks.

1.321. Korea maintains that western-style liquors like whisky are not normally drunk with meals (i.e. lunch or dinner) whereas most soju is consumed with meals.

1.322. Korea also argues that another ambiguity arises from the Trendscope Survey's use of the term "umsik" for "food". According to Korea, the usual meaning of "umsik" in Korean is "food and drink", and as such "umsik" can be interpreted broadly to include drinks as well as snacks. In Korea's view, this ambiguity would mean that respondents who, for example, eat peanuts with their whisky may well have answered that they drink whisky "with food".

1.323. Korea asserts that there is also an ambiguity in the definition of "with meals". Korea poses the question whether "with meals" was necessarily interpreted by their respondents as eating contemporaneously with drinking, or whether it was interpreted more liberally to include "before dinner" and "after dinner" meals.

1.324. Korea points out that there is also an ambiguity in the sense that it is unclear whether the respondents, being faced with questions about food and meals at the same time, would have considered the questions as mutually exclusive. In Korea's view, some respondents might have thought that the questions about "food" were intended not to include "meals", while others might have assumed that because meals are food, then soju which is usually consumed with meals is as of necessity consumed with food too.

1.325. Korea submits that since the complainants have submitted this survey too late, it did not have the opportunity to pursue the ambiguities with questions through the Panel. Korea adds that this late submission of this survey also means that there is insufficient time to evaluate its evidentiary value, and therefore, the panel should disregard it.

D. EC and US Comments on the Nielsen Study

1.326. The complainants responded to this survey, which was commissioned at the instance of Korea, by pointing out that there were several categories of overlapping end-uses. They state for example, that all Japanese restaurants served soju and 40% of them served whisky; a further 6.7% served brandy or cognac. According to the complainants, of the responding Western-style restaurants and cafés, 90% served whisky and a lesser number served other types of western-style beverages, and 21.7% served soju.

1.327. The complainants also note that while only 1.7% of the individual respondents drank whisky at home with meals, only 29.3% of all respondents consumed any alcoholic beverages at home with meals. In the complainants view, the proper comparison was that between the 1.7% and the 29.3%, thus leaving 5.8% of all respondents who consumed alcoholic beverages at home as drinkers of whisky with their meals.

1.328. The complainants have also questioned some of the findings of the Nielsen study, arguing that these results are actually indications of overlapping end-uses. The complainants note that there were almost no western-style beverages in Korea until the last five years following changes in the duty rates on imported distilled beverages.

1.329. The complainants argue that alcoholic beverages, like many foods and beverages, are habit based products. In their view, people tend to purchase what they are used to and change their tastes only over a period of time. They must become familiar with the taste of new products and will only make minor substitutions for the familiar product at first and more significant changes will tend to occur over a period of time until a fairly stable rate is achieved.

1.330. According to the complainants, the trends shown in the Nielsen study, as well as the substitutability shown in the EC market survey (the Dodwell study), show unmistakable evidence of the beginnings of substitutability and common end-use by imports.

1.331. The United States claims that the Nielsen study also contradicts Korea's emphasis that soju is only drunk straight in alleged contrast to Western spirits. According to the United States, survey respondents reported that in addition to being drunk straight, standard soju is consumed mixed with cola, cider, juice or some other manner. The results confirm that even if standard soju were predominantly consumed straight by survey respondents, it was not consumed only straight. Moreover, survey respondents were asked how they "normally" like to drink standard soju and whisky, and they were limited to a single response, suggesting that the reporting of multiple styles of consumption were avoided by the design of the question.

1.332. The United States further notes that the results of the Nielsen study also run counter to the supposedly rigid styles of whiskey consumption Korea paints in its submissions and first oral statement. For example, one-third of the survey respondents reported preferring consuming whisky straight and a smaller percentage indicated, similar to standard soju, a preference for whisky mixed with cola, cider, and juice. According to the United States, with a third of respondents preferring their whisky straight, just as Korea claims standard soju is largely consumed, it is clear that whisky and soju consumption styles overlap significantly.

1.333. The United States further claims that the Nielsen study also directly undermines Korea's assertion that soju is "never drunk mixed." Korea says this assertion was based on "common knowledge,"²⁸³ but apparently such common knowledge may not be a reliable source of information. Korea has insisted that the burden is for the complaining parties to show its factual assertions to be untrue. In our view, the Nielsen study's direct contradiction of Korea's so-called common knowledge places doubt on all similar unsubstantiated assertions by Korea concerning consumption in its market.

1.334. The United States asserts that, in response to the Korean Nielsen study, the U.S. Embassy in Korea identified nine large traditional Korean restaurants in Seoul that serve traditional Korean food based on the common knowledge among Embassy staff. U.S. embassy employees then asked whether the establishments served whiskey and whether they served soju. It turned out that every single one of these nine restaurants serves both soju and whisky.²⁸⁴ The United States does not claim to have taken a representative market survey, but such a survey is not necessary for the point at hand. We are simply saying that Korea's depiction of stratified end uses in different restaurants in Korea is contradicted by a simple tour of its capital city. These observations based on simple anecdotal evidence stand in stark contrast to the Nielsen finding that no surveyed Korean style restaurants served whisky and soju, whether in Seoul or elsewhere. This tends to confirm that the Nielsen study selectively sampled the restaurants known to limit the variety of spirits, representing an additional bias in the sampling. Moreover, the Trendscope survey, which the EC has described in detail earlier, further contradicts the Nielsen results by also showing that whisky is consumed at venues such as Korean restaurants where soju is the main item. Additionally, the Trendscope study shows that soju, similar to whisky, is most often enjoyed without meals.

1.335. According to the United States, even if the Panel were to accept these contradictions and biases, the Nielsen survey, highlighted by the EC, survey in fact lends credible support to the U.S. position in this case by indicating significant overlaps in usage between soju and Western spirits. Again, given the recent removal of barriers to entry, it is not expected that Western spirits will be consumed in every venue and in equal proportions to the traditional Korean spirit of soju. Nevertheless, the survey indicates whisky is consumed with meals, whisky is served in three out of six types of restaurants surveyed (four out of seven if you include hotel bars), and whisky and soju are both consumed straight, or mixed with cola, cider or juice.

To continue with Answers to questions

²⁷⁹ EC written rebuttal, at para. 1.

²⁸⁰ EC written rebuttal, at para. 166.

²⁸¹ US written rebuttal, at para. 70.

²⁸² Trendscope survey, Charts entitled "Key Findings of Consumer Survey".

²⁸³ Korean response to a U.S. Question.

²⁸⁴ US Exhibit S.