Conventional wisdom tells us that in Korea–Beef, the Appellate Body interpreted the word ‘necessary’ in GATT Article XX to require a cost–benefit balancing test. The Appellate Body is supposed to have applied this test also in EC–Asbestos, US–Gambling (involving GATS Article XIV), and Dominican Republic–Cigarettes. In this article I demonstrate, by detailed analysis of the opinions, that the Appellate Body has never engaged in such balancing. They have stated the balancing test, but in every case they have also stated the principle that Members get to choose their own level of protection, which is logically inconsistent with judicial review by cost–benefit balancing. And they have decided every case by reference to the ‘own level of protection’ principle. The Appellate Body is right not to balance. Balancing is not authorized by the treaty texts, and it is not needed to prevent inefficient harm to foreign interests.

Economists have increasingly become involved in trade remedy and litigation matters that call for economic interpretation or quantification. The literature on the use of econometric methods in response to legal requirements of trade policy is rather limited. This article contributes to filling this gap by demonstrating the efficacy of using a simple ‘probabilistic’ model in analyzing the ‘likelihood’ of injury to the local industry concerned, following a finding of continuation or recurrence of dumping (or countervailable subsidies). The legal concept of ‘likelihood’ is not only particularly well-suited to illustrate the systemic need for trade lawyers and economists to cooperate. It is also of imminent practical relevance with a groundswell of ‘sunset’ reviews looming on the horizon. We discuss the significance of economic analysis for trade remedy investigations by reviewing the literature, the applicable WTO rules, and, in particular, the pertinent case law. The potential value of probabilistic simulations for ‘likelihood’ determinations is exemplified using a real-world application. Using data from past United States International Trade Commission investigations, we find that a probabilistic model that takes account of the uncertainty surrounding economic parameters reduces the risk of misjudging the effect on the domestic industry of a termination of trade remedies.

It is commonly perceived that much of the initiative for removal of distortions in trade in agriculture has to come from the developed world. This generalization is not valid across all agricultural products, in particular for tropical products that are predominantly produced in developing countries and constitute a large share of their exports. It emerges that export of most of these products suffers not because of barriers in advanced countries, but from those in the developing world. This implies reciprocal benefits from growth in counter-trade if developing countries agree on more stringent trade liberalization measures for them. Cotton, sugar and bananas are important exceptions; they face trade barriers in some developed economies, so they deserve specific attention of trade negotiators.