Evidence of Selection in a Mandatory Health Insurance Market with Risk Adjustment

This paper aims to identify selection separately from moral hazard in a mandatory health insurance market where enrollees can freely choose their deductible scheme. The empirical analysis uses a unique data set for the period 2010-2013 covering the whole population of the Netherlands at enrollee level, allowing us to use prior health expenses of the enrollees to demonstrate the selection e ect separately from the potential moral hazard e ect. Our estimates show that the enrollees who opt for deductibles are both healthier and have a higher risk-adjusted result (i.e. the di erence between the compensation from the risk-adjustment fund and the actual health care cost) under the prevailing risk-adjustment system. Compared to enrollees who have chosen the lowest available deductible level, enrollees who have chosen the highest deductible level have an average risk-adjusted result that is approximately AC450 higher per enrollee. An option that the Dutch government could consider to fully eliminate the risk-adjustment gain of the deductibles is to include the choice of a voluntary deductible in the risk-adjustment system as one of the characteristics of the consumer. Our detection of substantial selection e ect of deductibles suggests the need of further research to understand in greater detail the relationship between premium discounts and the expected gains on the risk-adjustment for enrollees with a voluntary deductible.