Alternative Ways of Measuring and Interpreting Worker Flows

The present paper provides empirical evidence compatible with a proposed theoretical framework to explain the joint determination of two components of worker flows: worker replacement and job creation. We show that a negative correlation between job creation and replacement across firms emerges from such a framework. An empirical model is specified and its parameters are estimated taking into account two serious problems: measurement error and endogenous regressor. We take advantage of a matched employer-employee longitudinal database with detailed information on job and worker characteristics to tackle both issues. Our estimates confirm the negative correlation predicted by the theory.