Imperfect Information, Bayesian Learning and Capital Accumulation

This paper examines the consequences of introducing a simple informational imperfection into a growth model with overlapping generations, in which agents learn the technological parameters in a Bayesian fashion. In this setup, we study the properties of the equilibrium dynamics of beliefs and the capital stock. Under mild sufficient conditions, beliefs converge to the true value of the technological parameters. Even short-lived informational imperfections could have lasting effects, as they alter the long-run equilibrium levels of the capital stock. Therefore, learning dynamics may explain the observed differences in the performance of countries with otherwise similar economic characteristics.