Strategic trading and Ricardian comparative advantage

This paper analyses the failure of the traditional Ricardo–Haberlerian (1817; 1936) theory of comparative advantage (RTCA) in a strategic market game à la Shapley–Shubik (1977). In this model, trade is driven, not by comparative advantages, but by strategic considerations. We prove, in a Ricardian framework, the simultaneous existence of two types of equilibria, at both of which active international trade takes place. In the first type of equilibrium, both countries specialise based on comparative advantages. In the other, eachcountry produces only its comparative-disadvantage good. The welfare properties, and policy implications of this result (using the examples of the China–US trade war and Venezuela), are discussed at length in two dedicated sections. We show that the predictions of the RTCA depend, not on the number of agents in the economy, but on the nature of agents: the RTCA fails to obtain even with an infinite number of large players in each country. We prove that the RTCA prevails when agents are price-takers, and establish the conditions under which equilibria of our market game coincide with Walrasian ones.