Switch-On and Switch-Off Effects of Sick Pay Reform on Absence from Work and on Health-Related Outcomes

Abstract

We evaluate the switch-on and switch-off effects of a natural experiment that reduced sick pay in Germany from 100 to 80% of the wage rate but that effectively only applied to workers without a collective bargaining agreement. Two years following implementation of the reform, a newly elected federal government repealed it. We estimate the reform’s impact on annual days of absence by applying a difference-in-differences strategy to person-level data from the German Socio-Economic Panel. We find a 2-day reduction in the number of days of absence, almost a quarter of the pre-reform mean. The reform also reduced the average number of days spent in hospital by almost half a day, but at the same time did not have an effect on subjective health outcomes. Quantile regression reveals higher point estimates (both in absolute and relative terms) at higher quantiles, meaning that the reform predominantly reduced long durations of absence. A further analysis of the switch-on and switch-off effects shows higher point estimates for switch-off than switch-on effects. This finding is consistent with experimental evidence on the effect of incentive contracts on voluntary cooperation.