The incentive effects of sickness abscence compensation: analysis of a "natural experiment" in Eastern Europe

This research project examined the effect of halving of the maximum sick benefit provided by statutory health insurance in Hungary. This policy change in 2011 sharply decreased benefits for high earners, while leaving the incentive to claim sick pay unchanged for those with lower earnings, providing us with a “quasi-experimental” setup to identify the incentives effect of sickness benefits.

We used a difference-in-difference type methodology to evaluate the short-term effect of the reform, relied on high-quality administrative data and analysed a sample comprised of prime-age male employees with high earnings and stable employment. Our results show that the number of days spent on sick leave fell substantially for those experiencing the full halving of benefits, while the incidence of sick leaves decreased only moderately. Estimating the response of the number of sick days with respect to sick pay, we find a significant elasticity of -0.45.

If we look at the health status of our sample based on prior health care spending, we see that the number of days spent on sick leave decreased slightly less in the group that has a worse health condition. The fact that those likely in better health condition responded more elastically to the sickness benefit cut indicates that there might have been a reduction in fraudulent sickness benefit use. At the same time those in worse health also reduced the number of days spent on sickness benefits, so many employees might return to work earlier than they would fully recover.

Finally, it is worth noting that due to the decrease in the maximum sickness benefits, there was a 40 percent reduction in sickness benefit spending on those potentially affected. However, 60 percent of this decrease resulted from the cut in the sickness benefit, and not due to a change in behaviour. 

This research project was financed by the Hungarian Scientific Research Fund, and the paper is forthcoming in Research in Labor Economics Vol. 47.