Effect of restricting asset trade in dynamic equilibrium models

This paper shows that differences between the predictions of an international real business cycle model with complete markets and the predictions of a model where agents can trade only risk-free bonds depend heavily on three parameters: discount factor, and degrees of persistence and spillovers in productivity shocks. This sensitivity explains apparently paradoxical results previously obtained in the literature. Also, since empirical work finds that two of those parameters are not estimated precisely, the outcomes of quantitative studies comparing complete-markets and bond economies using only the point estimates of those parameters inherit the substantial uncertainty in the parameter estimates. Copyright 2004 Blackwell Publishing Ltd