The effect of economic incentives on worker absenteeism is analyzed using panel data on work absence behavior for each day during 1990 and 1991 (i.e. 730 observations in the time dimension) for a representative sample of 1,396 Swedish blue collar workers. During the observed time period, a major reform of the sickness insurance as well as a tax reform were implemented, both of which affected the worker's cost of being absent from work. We differentiate between the dynamic dependence conditional on whether the worker is in the work absence state versus in the work presence state. We also control for unobserved heterogeneity. The results show that the cost of being absent has a significant effect on work absence behavior and the importance of considering unobserved heterogeneity when modeling individual work absence behavior.