Wealth Inequality and Externalities from Ex Ante Skill Heterogeneity

This paper develops an incomplete markets model with state dependent (Markovian) stochastic earnings processes and ex ante skill heterogeneity corresponding to being university educated or not. Using the Wealth and Assets Survey for Great Britain, we find that the university educated group has higher average wealth, higher earnings risk but lower within group wealth inequality. Using estimates of the earnings processes for each group to calibrate the model, we find wealth inequality within and between the groups that is consistent with the data. Moreover, the predictions for overall wealth inequality are closer to the data, compared to the benchmark model with ex ante identical households. In this framework, ex ante skill heterogeneity generates a between-group pecuniary externality which in turn leads to the predicted differences in wealth inequality between the groups and works as an amplification mechanism to increase overall wealth inequality.