Determinants of lifetime unemployment: a micro data analysis with censored quantile regressions

Building on a large administrative micro data set for the time span 19752004 we look at lifetime unemployment for West German birth cohorts 1950 to 1954. Descriptive evidence shows a highly uneven distribution of unemployment in West Germany more than 60% of the individuals in our sample were not unemployed for a single day over the better part of their professional career while almost half of the total amount of unemployment fell upon 5% of the individuals covered. We employ censored quantile regressions to explain the amount of individual lifetime unemployment. Explanatory variables are either characteristics of the individual (like education), or of the job (like the wage) or the employer (like the size of the firm) early in the professional career. A particular emphasis is placed on the importance of the occupation: we find that for men working in a disadvantageous occupation at age 25 ceteris paribus leads to a significantly higher amount of lifetime unemployment. Educational attainment or the wage earned at age 25 are also related to the amount of men's lifetime unemployment, amongst others. Some of these variables show very interesting patterns when looking at different quantiles. For women results are in general less clear-cut.