Continuity in work life and the firm

The influence of the firm and the former work life on the probability to in terrupt ones current employment between 1993 and 1995 is analyzed. For reco nstructing individual work lifes, we use data from the "German Employment S tatistic Register Sample 1975-1995" of the Institute of Employment Research (IAB). Characteristics of the firm were taken from the "firm panel" (Betri ebspanel) of the same institution between 1993 and 1995. Both samples were merged to create a "linked employer-employee" data-set. Therefore, firm rel ated variables like number of employees, economic situation and industry ar e independent of individual characteristics. In our empirical model we incl uded the change of the employer, the interruption of work and unemployment as competing risks. Our analysis shows that the former work life has an ind ependent influence on the risk of interrupting work or to become unemployed between 1993 and 1995. Furthermore, qualified persons who experienced job shifts within their last firm and who are employed by larger employers whic h invest in their employees have a greater probability to move directly to another employer. In sum, our results indicate a "reproduction" of disconti nuous employment patterns over the work life.