A Theory of Wage Distribution and Dynamics with Assignment and Pareto Learning

This paper develops a simple model of assignment and Pareto learning. The model gives a unified explanation for the following basic facts of wage dynamics. First, the expected wage change increases with experience but the percentage of wage increase declines with it. Second, more educated workers have higher rates of wage growth. Third, the percentage of workers with a wage gain in any cohort decreases with the cohort experience but this percentage is independent of schooling level. The model also sheds light on the puzzle of serial correlation of changes in wage residuals. This model predicts that serial correlations are positive but small. On the other hand, the serial correlations increase with the importance of job matching and decrease with experience. This offers a potential explanation for why positive serial correlations have only been found in a subpopulation of workforce such as young, professional, and managerial workers and not in large, representative data sets.