In social insurance systems that grant workers paid sick leave, physicians act as gatekeepers, supposedly granting sickness certificates to the sick and not to shirkers. Previous research has emphasized the physician's superior ability to judge patients' need of treatment and potential collusion with the patient vis-á-vis an insurer. What is less well understood is the role of patients' private information. We explore the case where patients have private information about the presence of nonverifiable symptoms. Anyone can then claim to experience such symptoms, reducing physicians' ability to distinguish between sick patients and shirkers. Doubting a patients' reported symptoms may prevent good medical treatment of the truly sick. We show that for all parameter values, the Bayesian Nash equilibrium is that some physicians trust all claims of nonverifiable symptoms, sicklisting shirkers as well as sick; for many values, every physician is trusting. In particular, if physician strategies are observable by patients, extremely strong gatekeeping preferences are required to make physicians mistrust. To limit unwarranted sicklisting, policies reducing the benefits of shirking for healthy workers may be better suited than attempts to convince physicians to be strict.
Keywords: moral hazard; physician gatekeeper; sick leave.
© 2020 The Authors. Health Economics published by John Wiley & Sons Ltd.