Sickness Insurance: Design and Behavior

Sammanfattning: The first essay presents the results from a questionnaire study of sickness insurance and sickness absence behavior in Sweden. In the first part, people’s preferences for the design of a sickness insurance are discussed. Three combinations of qualifying days and compensation rates, with approximately the same costs within the insurance system, are compared. From the questionnaire it is concluded that a system with one qualifying day is the most preferred, compared to no qualifying day at all (and lower compensation) or three qualifying days (and higher compensation). The second part of this paper focuses on factors that influence people’s decision whether or not to report sick. In the questionnaire individuals were asked whether they would go to work or not, presupposing that they actually feel ill. Respondents were asked the same questions under different hypothetical compensations. The results indicate strong effects of factors related to the financial loss of being absent on the propensity to report sick. The second essay analyzes to what extent Swedish employees choose to take a holiday instead of reporting sick, in an attempt to avoid the costs related to sickness absence. The data used comes both from a questionnaire survey and from the Labour Force Survey by Statistics Sweden. The data from the questionnaire shows that individuals sometimes actually choose a holiday instead of reporting sick, with the intention of avoiding costs. The Labour Force Survey study does not confirm that this type of behavior has a considerable effect on the overall sickness absence. The third essay focuses on the main differences between insurance with uniform premiums and insurance with experience-rated premiums. The purpose is to explore the complex interaction between, on the one hand, probability of failure, risk-aversion, discount rate and probability of changing risk group, and, on the other hand, people’s preferences for uniform premiums and experience-rated premiums. In the paper, both a theoretical model and a simulation model illustrate the differences. One important conclusion is that the complex nature of experience-rated insurance precludes simple and general conclusions about the preferences for this insurance. It is shown that an insurance with quite considerable redistribution between risk-groups might be preferred even by good risk individuals. This can be seen as support for social insurance. The fourth essay is a theoretical study of temporary sickness absence and insurance. The purpose is twofold; first to refine the theoretical model for temporary sickness absence and insurance, and second to analyze the optimal level of compensation in an employer-provided sickness insurance and a public sickness insurance, and to compare these two. It is found that full compensation, under some circumstances, could possibly be optimal in both employer-provided and public sickness insurance.