Shrimp/Turtle was the only case since the establishment of the WTO in which an ETM which aimed at protecting the global environment became the issue of the case.186 Unlike other environment-related cases brought to the WTO, the AB invoked the “means and ends” test in its analysis. In this case, the AB invoked the “means and ends” test twice, in the analysis of Article XX (g) and in the chapeau analysis. First, the AB scrutinized whether there is “close and real” relationship between the purpose of sea turtle protection and the ends employed under Section 609.187 The AB found that the scope of Section 609 was not “disproportionately wide” in comparison with the purpose of sea turtle protection. Second, in the chapeau analysis, the AB found that requiring other countries to adopt essentially the same regulatory program as the U.S. without taking into account different conditions in other countries was not an adequate means to achieve the policy goal of sea turtle protection.188 Furthermore, the AB found that the fact that the U.S. did not allow the import of shrimps from non-certified countries even if the particular shrimps were caught using methods identical those employed in the U.S. could not be reconciled with the purpose of sea turtle protection.189
Unlike the LTRA test, the nature of the “means and ends” test is the balancing of values. The “means and ends” test juxtaposes trade harms and environmental benefits and compares them. In the context of ETMs for preserving the domestic environment, the AB&P have avoided “comparing” values. However, they did balance them in analyzing an ETM for protecting the global environment. The adoption of the “means and ends” test could be explained in light of how distributional, philosophical, and moral problems appear in the context of ETMs aimed at protecting the global environment.
First, when countries invoke ETMs in order to protect the global environment, distributional concerns are not as significant as when they impose ETMs in order to protect their domestic environment since countries other than the ETM-invoking countries also obtain benefits from the measure. As a result of global-environment-oriented measures, outsiders are not necessarily worse off. Hence, ETMs aimed at protecting the global environment could not be considered as a kind of “beggar-thy-neighbor” policy as domestic-environment-oriented measures are. In other words, the necessity of minimizing trade harms on outsiders is not so strong. In this regard, it should be noted that the “means and ends” test excludes administrative costs from its consideration. Administrative costs are exclusively borne by measure-imposing countries. On the other hand, trade harms and environmental benefits are shared by ETM-invoking countries and other countries. That is, the “means and ends” test compares the only part of the benefits and costs which are shared by the international community.
Second, with regard to the global commons, no countries can enjoy the sovereign right to decide their appropriate level of protection. Some denounce the invocation of ETMs for protecting the global environment because the imposition of ETMs means that measure-invoking countries force other countries to adopt their preferences concerning appropriate levels of protection for the global environment.190 If we assume this argument is true, the other side of the coin must be also true: prohibiting ETMs means that exporting countries force other countries to accept their preferences for low levels of protection for the global environment. As Lowenfeld observes:
[i]n a situation involving more than one country, if country A wishes to regulate and country B does not, it seems…equally fair to criticize B for attempting to impose its will beyond its borders as it is to criticize A for attempting to exercise its jurisdiction extraterritorially. Similarly, it seems equally fair to restrain states that cause global environmental harm as it is to restrain states that seek to use extra-territorial trade measures ensure that their own markets do not contribute to global environmental harm.191
As a result, in the context of global environment protection, the LTRA test does not have the philosophical, moral, or ethical weight it has in the context of domestic environment protection. As Dunoff maintains, applying the LTRA test to measures aimed at protecting the global environment may imply that trade values are elevated to a pre-eminent status in comparison with environmental values:192 trade harms have to be minimized, however, environmental harms do not have to be minimized. Therefore, balancing tests such as the “means and ends” test are more appropriate in the context of global environmental protection.
Third, by juxtaposing trade harms and environmental benefits, the “means and ends” test raises great concerns about the incommensurability of values and the interstate comparison of utilities. However, in comparison with cost-benefit analysis, these problems are less significant because the “means and ends” test is a balancing test with a moderate level of precision. By roughly juxtaposing values, the “means and ends” test creates room for the consideration of non-monetizable and non-commensurable values and avoids trying to produce a precise comparison.193
Because of these reasons, we can conclude that the AB’s use of the “means and ends” test for ETMs aimed at protecting the global environment is appropriate.