Testing the intertemporal labor supply model: are jobs important?

The neoclassical life-cycle labor supply model assumes that hours of work are determined by labor supply choices alone and does not include a role for employers or job distinctions. An alternative model in which employers have an interest in employee hours and changing jobs is costly may reconcile weak and conflicting evidence regarding the neoclassical model. An important implication of such an alternative is that individuals may face hours constraints on a job and thus can choose hours freely only between jobs. This paper tests several implications of the neoclassical model against this alternative model of hours determination. Using a unique panel of individuals and jobs constructed using data on women from the 1988-1992 Panel Study of Income Dynamics, I compare hours variances and labor supply elasticities measured within and between jobs. Results provide some evidence against the neoclassical model and suggest that ignoring the role of employers in determining hours of work can lead to downward bias in estimates of labor supply elasticities.