Abstract:
This paper documents earnings dynamics over the life-cycle and income level using a large administrative database from German tax records. I find that labor earnings display important deviations from the typical assumptions of linearity and normality. For the bottom earners, large income changes are driven equally by hours and wages which is consistent with transitions between labor status or jobs, whereas for those at the top, earnings changes are mainly induced by wage rate growth. There are also asymmetries in mean reversion of earnings growth mainly driven by the asymmetric hours dynamics. Finally, there is no evidence of an added-worker effect but government insurance and income pooling can mitigate the pass-through of individual earnings changes to the household level and attenuate the deviations from normality of the male earnings growth distribution.