This paper contributes to the recent empirical debate about the effectiveness of the GATT and the WTO in promoting trade. We use gravity models to explore the impact of the GATT/WTO on bilateral trade in a sample of 46 countries over the period 1965-1997. Our data enable us to disaggregate trade by broad commodity aggregates. The results for total trade are similar to those reported by Rose (2004). However, the disaggregated estimates reveal that the GATT/WTO has had a positive and statistically significant impact on trade in capital-intensive commodities, but that it has had no statistically significant impact on trade in other commodities. The paper demonstrates that simple modifications of Rose's approach lead to results that are much more 'common sense' than his.