Here's more from Todd Tucker on the Clove Cigarettes ruling:
While menthol and clove are both "flavored cigarettes," so are cola- and candy-flavored cigarettes. The U.S. argued, plausibly in my mind and to paraphrase Big Bird, that "one of these things is not like the other." Specifically, menthol. Why? Significant numbers of adults smoke them, particularly in the African American adult community. For that reason, it poses significant adverse effect risks that the others did not.
Cloves and candy flavored cigarettes, however, are not only flavored, but they are trainer cigarettes that appeal to teenagers in significant numbers, but not to adults in significant numbers....
In short, the WTO should have compared the treatment of domestic candy- and cola-flavored cigarettes to imported clove cigarettes - both banned under the FSPTCA. It was only by choosing to compare menthols to cloves that the panel could come up with a finding of differential, let alone discriminatory, treatment.
Todd's post provides a good basis for jumping into the issue of "likeness" in the case. As I see it, there are two big issues: (1) Selecting which products are potentially to be included in the group of "like products"; and (2) the proper standard for determining "likeness." The issues are related, but have some different aspects, so I deal with them separately.
Selecting the group of "like products" in the Clove Cigarettes case
With regard to the appropriate group of like products, Todd says: "While menthol and clove are both 'flavored cigarettes,' so are cola- and candy-flavored cigarettes." The issue of which cigarette products were "like" was a complicated one in this case. Is it just cloves and menthols? Is it all flavored cigarettes? Is it all cigarettes?
The Panel's consideration of the issue was probably shaped by the parties' arguments. Here's what Indonesia argued:
7.42. ... Indonesia argues that clove cigarettes are "like" "all other domestically produced cigarettes, generally, and menthol cigarettes, in particular", because they share the same physical properties, end-uses, consumer preferences and tariff classification. In response to a question from the Panel, Indonesia clarified that it requests the Panel first to conduct a like product analysis of clove cigarettes vis-à-vis menthol and tobacco flavoured cigarettes produced in the United States. In the event that the Panel decides to limit the likeness analysis to cigarettes with a particular "characterizing flavour", Indonesia argues, clove cigarettes are "like" domestic menthol cigarettes.
...
7.45 Furthermore, Indonesia clarifies that the so-called "candy flavour" cigarettes (e.g., chocolate, strawberry) may not be "like" regular, menthol or clove cigarettes, because they may pose greater health risks as they encourage youth to start smoking. Indonesia has indicated that it "is not asking the Panel to include candy-flavoured cigarettes in its like product analysis".
In essence, Indonesia wants its clove cigarettes compared to menthol and regular cigarettes (the cigarettes that may still be sold), and not compared to candy flavored cigarettes (which, like cloves, are banned). That seems like a reasonable position for Indonesia to take, as it makes its claims more compelling.
But stepping back from the parties' views, which of the various products here are best described as "like"? The Panel seemed to think that cigarettes with a "characterizing flavour" was a relevant grouping:
7.119 ... We rather think that the weighing of the evidence relating to the likeness criteria should be influenced by the fact that Section 907(a)(1)(A) is a technical regulation having the immediate purpose of regulating cigarettes with a characterizing flavour for public health reasons.
However, despite the general reference to cigarettes with a "characterizing flavour", the Panel concluded that the two products at issue were only cloves and menthols:
7.147 ... we feel compelled to conclude that we are bound by Indonesia's summary of the legal basis of its national treatment complaint, which identifies the products at issue as imported clove cigarettes versus domestic menthol cigarettes. In our view, we would be exceeding our terms of reference if we were to expand the scope of Indonesia's national treatment claim by including domestic regular cigarettes in our examination.
Basically, because of the way Indonesia framed the issue in its panel request, regular (tobacco) flavored cigarettes were not included in the analysis (and nor were the other flavored cigarettes included in the analysis).
In spite of how the claim was set out, would it have been possible for the Panel to look generally at the products at issue, so as to establish the entire group of like products? Perhaps the answer would have been that all "flavored cigarettes" should have been considered. As the Panel noted:
7.232 It is our view that the various studies mentioned above and their conclusions indicate that in the mind of youth, flavoured cigarettes, including those flavoured with clove or menthol, are similar. ...
Or maybe even all cigarettes generally should have been considered "like." In para. 7.119, the Panel emphasizes that the purpose of the measure is "regulating cigarettes with a characterizing flavour for public health reason." But this may not be the best way to describe the measure. It could also be said that the measure regulates all cigarettes, by prohibiting those with certain characterizing flavors and thus dictating what can be contained in a cigarette (i.e., not those flavors).
With all that in mind, let me get back to Todd's point: "the WTO should have compared the treatment of domestic candy- and cola-flavored cigarettes to imported clove cigarettes - both banned under the FSPTCA. It was only by choosing to compare menthols to cloves that the panel could come up with a finding of differential, let alone discriminatory, treatment." He may be on to something here, in that a focus on cloves and menthols leaves out some comparable products. I'm not sure his cloves versus candy/cola comparison is much better, though.
Regardless, assuming the broader category of flavored cigarettes is considered "like" (or if all cigarettes are considered), would that have changed the outcome? Based on the facts as I understand them, I'm not sure that this would have changed the "less favorable treatment" analysis. Looking at the distribution of imported and domestic products in the various categories, you still get the result that Indonesian products fare very badly in comparison to domestic products.
Of course, the products deemed to be "like" depends on the standard for likeness. Let's look at that now.
The General Standard for Determining "Likeness"
As to the standard for determining likeness, Indonesia argued:
7.43 ... in the context of Article 2.1 of the TBT Agreement, a determination of likeness is fundamentally a determination about the nature and extent of a competitive relationship among products, as established by jurisprudence on Article III:4 of the GATT 1994.
(See also para. 7.86.)
In response, the United States contended:
7.58 The United States argues that Indonesian clove cigarettes are not like U.S.-manufactured tobacco or menthol cigarettes as clove cigarettes are not in a competitive relationship with tobacco or menthol cigarettes and are not substitutable or interchangeable among retailers or consumers. ...
7.59 The United States notes that Section 907(a)(1)(A) makes distinctions among a group of broadly similar products – cigarettes – based on factors relevant to the legitimate objective of protecting public health. Accordingly, a "likeness" determination, in addition to focusing on the competitive relationship of the products, will need carefully to parse the significance of traits that are generally shared among all cigarettes and traits that are significant with respect to the public health provision at issue. ...
7.89 ... the United States also submits that in the context of analysing whether the products at issue in this case are "like products", a relevant factor in this case "is the degree to which differences among the regulated products directly relate to the public health objectives of Section 907(a)(1)(A)." ...
7.90 ... the "likeness" analysis under both Agreements should be informed by the specific context of the TBT Agreement by considering "not only the nature of the competitive relationship among and between products but also the nature of the public health basis upon which the technical regulation at issue is based".
On this issue, the Panel noted several options:
7.91 ... First, we could interpret Article 2.1 of the TBT Agreement following Article III:4 of the GATT 1994 jurisprudence. Under this approach, the jurisprudence under Article III:4 of the GATT 1994, which is mainly focussed on the nature and extent of a competitive relationship between the domestic and imported products at issue, would be directly transposable in its entirety to Article 2.1 of the TBT Agreement, based on the similarity of their respective language.
7.92 Second, we could interpret Article 2.1 of the TBT Agreement in the context of the provision itself and that of the TBT Agreement, without transposing any of the jurisprudence on Article III:4 of the GATT 1994, as this is but one of the concepts of "like products" found in the WTO Agreement. Under this approach, one would not focus on the competition-based approach that has been developed in the jurisprudence on Article III:4 of the GATT 1994.
7.93 Third, we could follow the approach suggested by the parties, which consists of interpreting Article 2.1 of the TBT Agreement taking into account both the jurisprudence under Article III:4 of the GATT 1994 and the context of the TBT Agreement. We could also follow the United States' suggestion to take into account the "public health objectives" of Section 907(a)(1)(A) when interpreting likeness under Article 2.1 of the TBT Agreement.
The basic issue was the balancing of (1) the role of competition between the products and (2) taking into account other factors, such as public health risks of the products. Is likeness solely about competitiveness, or can you take into account the public policy objectives behind the measure?
The Panel decided to give public health risks a fairly large role:
7.109 The fact that Section 907(a)(1)(A) is a technical regulation and has as its immediate purpose to regulate product characteristics (characterizing flavours) for certain types of products (cigarettes) should have some weight, and potentially great weight, in the determination of whether the products at issue are like. Indeed, cigarettes with characterizing flavours are regulated by Section 907(a)(1)(A) as a single group of products.
7.110 ... for the purpose of being regulated and thus prohibited under Section 907(a)(1)(A), a cigarette must have an additive that produces a distinguishing flavour, taste or aroma. Under this interpretative approach, the fact that cigarettes contain such additives should figure in the likeness determination more so than competition-related or other criteria, as it is that precise feature that determines whether or not a cigarette falls within the scope of application of the technical regulation at issue, i.e., Section 907(a)(1)(A).
...
7.119 From our considerations above, we do not believe that the interpretation of Article 2.1 of the TBT Agreement, in the circumstances of this case where we are dealing with a technical regulation which has a legitimate public health objective, should be approached primarily from a competition perspective. We rather think that the weighing of the evidence relating to the likeness criteria should be influenced by the fact that Section 907(a)(1)(A) is a technical regulation having the immediate purpose of regulating cigarettes with a characterizing flavour for public health reasons.
This reasoning seems like a big win for Todd. Instead of focusing just on whether the products compete, the Panel will look at the health risks of the products, thus offering up the possibility of finding that the products are not "like" even though they compete. So if that part seems like a win, how did he lose? Well, based on an examination of the products/measure at issue here, the Panel concluded that cloveand menthol cigarettes are like. Here's a key part of the reasoning:
7.247 The measure at issue in this case plainly regulates cigarettes on the basis of a characteristic that clove cigarettes and menthol cigarettes have in common, which in the words of Section 907(a)(1)(A), is the shared characteristic that they "contain, as a constituent … or additive, an artificial or natural flavor … or an herb or spice … that is a characterizing flavor". In the context of this particular measure, which regulates tobacco products on the basis of this particular characteristic – which may be regarded as perhaps the defining feature of each type of product – we find it very difficult to see how clove cigarettes and menthol cigarettes would not be considered to be "like". As discussed in our findings, we are aware that there are certain differences between clove cigarettes and menthol cigarettes. These differences may well lead to the conclusion that these two products are not "like" in the context of different measures. However, in the context of the measure at issue in this dispute, these differences are less significant, and less relevant. In other words, contrary to what the United States argues, those differences do not relate to the public health objective of the measure at issue and therefore, are not relevant to the like product analysis in this case. In our view, the similarities related to the public health objective of Section 907(a)(1)(A) are highly relevant to the like product analysis in the circumstances of this case.
Basically, for the purposes of the public health objective of reducing youth smoking, the Panel found that clove and menthol cigarettes are like, as the differences between the products "do not relate to the public health objective of the measure at issue."
Of course, Todd wants to redefine the like products to be cloves and candy cigarettes, not cloves and menthols. He says:
Cloves and candy flavored cigarettes, however, are not only flavored, but they are trainer cigarettes that appeal to teenagers in significant numbers, but not to adults in significant numbers.
That's an issue on which I would have liked to see more evidence. I felt that point was not adequately resolved. He also gives the following as a reason not to ban menthols: "Significant numbers of adults smoke [menthols], particularly in the African American adult community". However, this point could just as easily be used as an argument that banning menthols should have been the priority.
All of this gets us into what "likeness" is and should be all about, which gives me an opportunity to recall the 1870s dispute about the treatment of Manila and Russia hemp that I mentioned recently. Recall the opposing sides to the argument there:
Like means the same:
"The words do not mean almost the like articles, nor nearly the like articles, but like articles, or absolutely the same articles. No matter if the word "like," in some exceptional cases, is used in qualified sense, as nearly equal. Here it is used as absolutely "like," or the same, and nothing less nor more."
In other words, any difference between the products -- presumably including differences in public health risks -- can be used to show they are not "like."
Like means in competition with each other:
"[We are satisfied with a definition of "like articles" that means] articles which, being known in trade and commerce by the same name, are so far alike that one can come into injurious competition with another, for any of the purposes of trade or manutacturing, so as to work an unfavorable discrimination against the country with which the treaty subsists."
In this view, by contrast, the issue of likeness is all about competition in the marketplace.
What did the Clove Cigarettes panel think of this issue? Well, they certainly looked at competitiveness, but they also went beyond it, focusing heavily on the public health objectives of the measure, on the basis of which certain attributes of the products were regulated. Arguably, such an approach could bring a lot of uncertainty into the likeness analysis. It is not clear how boundaries could be established for when public policy goals may be used to determine likeness. What, if any, are the limits to this type of argument? It could also lead to overlap and conflict with the "less favorable treatment" element, and, in GATT Article III cases, with the GATT Article XX analysis. But to close on a positive note, this Panel ruling could give the Appellate Body a great opportunity to clarify its views on these issues.
