Moving or Staying? Job Mobility as a Sorting Process

This thesis constists of three papers dealing with labour mobility. In paper [I], I study the effects of job change for wage growth. Job change is assumed to influence wage growth and productivity for both job movers and job stayers. I find that returns on work experience are higher for job movers, both before and after job change. This contradicts the hypothesis that job stayers have higher returns to work experience as an effect of investing time in aqcuiring job-specific skills. In addition, I find that the returns on work experience do not improve for job stayers, when job changers quit. This means that productivity among job stayers is not affected by selective termination of bad matches. I find no evidence to support that job mobility increases productivity for job movers, which indicates that job mobility acts mainly as a sorting device, sorting workers into jobs they are most suited for. In paper [II], I study the effect of job change on wages and income, as well as on wage and income growth, allowing for multiple job changes. I assume that match quality is partly due to outcome of the individual's career decisions up until the time of observation. The individual's career history can signal to the employer the quality of current and previous matches, which may affect the wage path. I find that each job change increases wage and income growth between 1981 and 1991, but that there is no evidence to suggest that job changers have higher wage and income levels in 1991. This result is constistent with a scenario where job changers, through mobility, catch up the higher wages earned by job stayers. I also find some evidence to suggest that job movers are adversely selected, i.e. that job changers are drawn from a pool of workers who were predominantly 'bad matches' in 1981. In paper [III] I study the determinants of long and short distance migration. I estimate a multinomial logit model, to ascertain whether the determinants of migration differ between the choices of migrating long or short distances. I find that increases in pre-migration income decreases probability of migrating a long distance, while pre-migration income has no effect on the decision to move a short distance. The higher the education level, the higher the probability of long distance migration, while education level has no effect on the probability of short distance migration. Dependence on unemployment benefits increases the probability of both short and long distance migration. I find no effect of local fiscal characteristics or local labour market characteristics on the probability of either long or short distance migration.