After establishment closure
Individual characteristics that determine re-employment probabilities of displaced workers in Sweden
This paper studies the relationship between individual characteristics of displaced workers and the probability of re-employment. A competing risks hazard model is used, distinguishing between exits from joblessness to self-employment and to paid-employment. All individuals between 25 and 55 years of age, at the time of displacement, that suffered from at least one year of joblessness after being displaced between 1990 and 1998 due to establishment closures that occurred between 1990 and 2001 are included. This allows for the closure procedure to be between one and three years long. Each individual is followed, from the year of displacement until the year of re-employment or at the latest, ten years after displacement. Semi-parametric estimation techniques for discrete time data are used, and in consistency with previous research the results show that subgroups of the jobless individuals experience different re-employment probabilities. The results suggest that a non-immigrant, high income-earner in the mid-thirties, with short tenure at the closing establishment, and who was not displaced early in the closure process and who has lived in the same city for a long time, faces the greatest probability of becoming re-employed. Furthermore, men and individuals with self-employment experience face lower probabilities of re-employment in paid-employments compared to their counterparts. This relationship is however reversed when studying the probability of leaving joblessness for self-employment. Men, immigrants, high income-earners, displaced from smaller establishments, previously self-employed and those with shorter tenure in previous employment are found to be more likely to enter self-employment than their counterparts. A positive duration dependency is prevalent in re-employment probability, suggesting that search activity is increased over time. The probability of self-employment entry is however decreasing the first years of joblessness following displacement, displaying an initial negative duration dependency.