ROSEN AND QUANDT'S DISEQUILIBRIUM MODEL OF THE LABOUR MARKET: A REVISION.

In a recent paper, Harvey Rosen and Richard Quandt present and estimate a disequilibrium model of the aggregate labor market in the U.S. Their parameter estimates are generally plausible, but one of their results is extremely anomalous: the fitted values of labor supply and demand given by the model imply high negative unemployment rates throughout the 1930's. The figure presented in the article compares the unemployment rates generated by the model with the official unemployment rates. Even though the two series are not attempting to measure precisely the same thing, one should expect them to follow similar general patterns. Rosen and Quandt's estimation does in fact yield a large positive income elasticity of labor supply. The model is thus fitting changes in hours of work through changes in assets, with labor supply as the intermediary. Problems that arise from the inclusion of unearned income in the equation for labor supply also arise in other models of the labor market. Rosen and Quandt note that others have found positive coefficients for nonlabor income in both time-series and cross-section studies.