This article investigates the liberal idea that trade between two states reduces the likelihood of militarized conflict between them. Richard Rosecrance's argument that industrial-technological developments have made peaceful trading strategies more efficient today is examined in connection with the empirical literature on trade and conflict. Development affects the utility calculations of states: since the costs of seizing and holding a territory increase with increased development, and the relative utility of occupying the territory decreases, the chance that the expected utility of occupation will exceed the expected costs decreases with increased development. Likewise, since the utility of trade increases with increased development, then increased development also makes it more likely that the expected costs of breaking the trade bonds will exceed the gains to be expected from occupation. Consequently, the relationship between trade and conflict is contingent on the level of development. Using Cox regression, and introducing a new measure of interdependence based on a gravity model of trade, I demonstrate that there is a clear negative relationship between trade and conflict. However, this relationship is basically restricted to dyads consisting of two developed dyads. Development itself is strongly associated with peaceful behavior. The results also suggest that the democratic peace requires a minimum level of development to be efficient.