Comparative advantage and strategic specialization

This paper shows that a strong comparative advantage is necessary for free trade and specialization in a 2 × 2 symmetric Ricardian model to be achieved in a Nash equilibrium. Governments strategically control labor distribution across industries, and representative agents maximize Cobb–Douglas utilities. A Nash equilibrium with complete specialization is achieved if and only if relative productivity exceeds a key value of 3, which is considered a very large number based on previous empirical studies. This paper also introduces a two-stage game where each government chooses labor distribution first and then tariffs. In this two-stage game, complete specialization is never achieved for any relative productivity level. Finally, by generalizing the Cobb–Douglas model into constant elasticity of substitution (CES) preferences, I show that if immiserizing growth effects exist, complete specialization could not be achieved for any level of relative productivity.