Essays on Health Shocks and Social Insurance

Abstract: This thesis consists of four self-contained essays.Essay 1 (with Petter Lundborg and Johan Vikström): This paper provides new evidence on heterogeneity in the impact of health shocks by using register-based data on the entire population of Swedish workers. We formulate a Difference-in-Difference design, where we compare the change in labor earnings across matched high and low-educated workers who experience the same type of health shocks. This design takes away the requirement to define control groups of workers not suffering from health shocks. Our results suggest large heterogeneity in the effects, where low-educated individuals suffer relatively more from a given health shock. These results holds across different types of health shocks and becomes more pronounced by age. Low-educated workers are also more likely to leave the labor force and take up disability insurance, sickness insurance, and unemployment benefits following a health shock. Our results suggest that socioeconomic heterogeneity in the effect of health shocks offers one explanation to how the socioeconomic gradient in health arises.Essay 2: Reductions in SI replacement levels have been a widely used instrument to lower sickness absence rates. The idea is that increasing the direct cost of absence would lower the absence rate. This paper explores a reform of the compulsory Swedish SI scheme. The reform reduced the replacement level from 90 percent of foregone earnings to 65 percent for the first 3 days of a sickness absence spell and to 80 percent for days 4 - 90. From day 91 onwards the rate remained at 90 percent. In this paper, I show that the reform had both a direct and an indirect effect on sickness absence behavior. Using detailed register data on the complete account of all sickness absence spells during the period, I am able to identify and estimate the indirect effect on absence-to-work transitions. I find that the indirect effect significantly reduces the probability to end an absence spell, creating a locking-in effect in sickness absence.Essay 3 (with Per Johansson): Sweden has obligatory sickness and disability insurance which is both financed and administered by the government. In order to receive sickness benefits, insured individuals must have certificates issued by a medical doctor. Since health care is administered at the county level, this means that monitoring is, to some extent, decentralized at a lower jurisdictional level than the funding and governance of the insurance. This paper studies one consequence of such decentralization: the effect on individual sickness absence when such certificates are not approved by the Swedish Social Insurance Agency (SIA) and are instead re-remitted to the doctor for completion and, potential, approval by the SIA. We find that this re-remission increases the length of sickness absence spells by an average of 30 percent. Given that added resources improve the quality of the patients' medical certificates, directed intergovernmental grants from the state to the counties would be cost saving.Essay 4: Injuries due to road accidents represent a large welfare loss to society. Damage costs include a variety of expenses related to, medical treatment, material and immaterial damage, law enforcement, production loss, and loss of time. Using use detailed Swedish longitudinal administrative data, and taking an incidence-based approach this paper provides an estimate of lost productivity not restricted to prevalence and the tenuous assumption that accidents are random. Total lost productivity is estimated to be 900 million SEK (€ 100 M), 408 million SEK (€ 45 M) of which is the burden on public insurance. The study also finds that there is a clear negative selection for road accidents. Accident victims are, on average, less educated, have a lower employment rate, and lower labor earnings than those who where not involved in a road accident. If this selection is not taken into account in the calculations, the cost will be overestimated.