Discussion Paper Series

Farber, Henry S., "Evaluating Competing Theories of Worker Mobility" NLS Discussion Paper, Report: NLS 92-5

Objectives
This study uses a sample of over fourteen thousand full-time jobs held by workers in the National Longitudinal Survey of Youth (NLSY) to examine mobility patterns and to evaluate theories of labor mobility (defined as change of employer). In particular I investigate the following questions:

• How important is heterogeneity in determining mobility rates for young workers?

• Can heterogeneity in mobility rates be characterized as fixed differences across workers or as variable with workers changing types over time (either systematically or otherwise)?

• How important is state dependence in mobility rates? In other words, does mobility vary importantly with how long a worker has held his or her job?

• Does mobility decline systematically with how long a worker has his or her job, or are there periods where likelihood of mobility increases?

• What do the facts discovered about the nature of the relationships between mobility and both heterogeneity and state dependence tell us about what actually causes mobility? Specifically, how important is the accumulation of specific capital, how important is the quality of particular matches between workers and firms, and how important is the underlying variation in the stability of workers?

Findings
There are five main findings. The first three relate to heterogeneity across workers in mobility rates. The last two are about state-dependence or how mobility rates vary with how long a worker has held their job.

1.) Mobility in a new job is strongly positively related to the frequency of job change prior to the start of the job. There is substantial heterogeneity across individuals in mobility rates, and this persists throughout subsequent jobs.

2.) Job change in the most recent year prior to the start of a new job is more strongly related than earlier job change to mobility on the new job. Thus, heterogeneity in mobility rates is not fixed over time, and workers seem to change, becoming fundamentally more or less mobile over time.

3.) Females hold significantly fewer jobs than do males in a fixed period of time early in their careers. Thus, females who have committed to the labor force exhibit less mobility than otherwise equivalent males. This results seems to be drive by a lower exit rate for females from the first job after entry. Potential links between the fact that the sex differential in wages rises with experience and the lower mobility of young females needs to be investigated further.

4.) Mobility rates are very high early in jobs. One-third of all jobs end within the first six months, and one-half are over within the first year. Mobility rates are much lower late in jobs. These findings demonstrate the importance of detailed analysis of mobility early in jobs, and they suggest the importance of accumulation of specific capital (in the form of both job-specific skills and job-specific information) on the job.

5.) The monthly hazard of job ending is not monotonically decreasing in tenure as most earlier worked using annual data has found, but it increases to a maximum at three months and declines thereafter. This finding is robust to controlling for worker heterogeneity, and it appears in jobs starting at any point after entry. This finding is consistent with a situation where workers and employers learn about the quality of the worker-firm match over the first several months on the job.

Copies of this and other papers in this series are available from BLS by contacting Rita Jain at Jain.Rita@bls.gov or at (202) 691 - 7405.

Last Modified Date: July 09, 2003