We show that lagged duration dependence is non-parametrically identified in mixed proportional hazard models for duration data, in the presence of competing risks and consecutive spells.

We study the identification of a mixed proportional hazard model with lagged duration dependence when data provide multiple outcomes per stratum. Within strata variation is exploited to non-parametrically identify lagged duration dependence in more general models than in the literature.

In 1996, by law the maximum standard workweek in Portugal was reduced from 44 h to 40 h. We find that for workers involved this change reduced the job separation rate and increased hourly wages, keeping monthly earnings approximately constant. The working hours reduction also affected...

In a field experiment in two Danish counties about half of the unemployed workers were randomly assigned to a mandatory activation program. Analyzing data from this experiment we show that activation can be very effective in reducing unemployment durations.

We analyze the determinants of reading literacy, mathematical skills and science skills of young immigrant children in The Netherlands. We find that these are affected by age at immigration and whether or not one of the parents is native Dutch.

This paper studies the ex-ante effect of worker separations on wage negotiations using matched worker-firm data from The Netherlands. We find that wage negotiations aim to prevent separations; workers with a high propensity to quit are offered higher wages, while workers with a high layoff...

We study the effect of an unexpected reduction in maximum unemployment benefit duration on the inflow into unemployment of elderly workers in the Netherlands. We analyze the reduction as if it was a natural experiment finding substantial effects.