"This study investigates how the effects of low-wage employment and non-employment on wage prospects vary depending on qualification. We apply dynamic multinomial logit models with random effects and include interactions of the lagged labor market state with qualification to estimate heterogeneity in state dependence. We find that low-wage jobs are stepping stones to high-paid jobs for low qualified workers. In contrast, the chances of workers with a university degree to obtain a high-paid job are the same when being low-paid or non-employed (whereas their risk of non-employment is lower when having a low-paid job). Furthermore, our results suggest that for workers with university degree low-wage jobs are associated with negative signals."

Support: Digital