A Quantitative Study of the Extent, Efficiency, and Cyclical Behavior of Job-to-job Transitions

Northwestern University, Evanston IL

Investigators

Abstract

Project summary The rate of job-to-job transitions, moves of workers between employers without an intervening unemployment spell, are twice as large as the rate at which workers move from employment to unemployment. Moreover, the relative importance of these job-to-job transitions has increased dramatically in recent decades. This means that job-to-job transitions are now at least as important a means of reallocating labor towards its more productive uses as are the transitions of workers through unemployment. Despite their quantitative significance, job-to-job transitions are rarely given an important role, or even included, in equilibrium search models, which have been for the past two decades the primary and most successful tool used to analyze worker reallocation and the phenomenon of unemployment. Moreover, the search models that do incorporate job-to-job transitions adopt the so-called basic job-ladder model of these transitions. My recent work has demonstrated some serious and previously unrecognized quantitative shortcomings of the basic job-ladder model with respect to recently documented stylized facts. In fact, the proposal shows that there exists no theoretical model of job-to-job transitions that can quantitatively match their salient features. The intellectual merit of this study will be to remedy this significant gap in the literature by developing a new theoretical model of job-to-job transitions that is able to match observed salient features of these transitions, both in terms of their magnitude and of their variation across groups of workers and across stages of the business cycle. The broader impact of the study will involve the efficiency properties of the proposed model and the resulting development of policy implications of theories of frictional labor markets that incorporate an empirically grounded model of job-to-job transitions. Any successful labor-market policy should aim to achieve the reallocation of labor to its most productive uses with the least social cost. Equilibrium search models portray unemployment as a social cost that is necessary to achieve this reallocation. In fact, they predict that, under certain conditions, the unemployment observed in decentralized markets is equal to its socially optimal level. The introduction of job-to-job transitions, in which the reallocation of labor towards its more productive uses takes place without workers having to bear the burden of unemployment, can be expected to challenge these conclusions. It will provide novel insights about the optimality of the observed level of unemployment and of patterns of worker turnover in general. These new insights can then guide policy-makers to adopt new approaches to the long-standing policy concern of unemployment.