Improving take-up of health insurance program: a social experiment in France

This paper is based on a randomised social experiment conducted in order to understand the low take-up rate of a Complementary health-insurance voucher program for the poorest in France (the Aide Complémentaire Santé: ACS). We explore two of the main hypotheses put forward to explain low enrolment: a lack of information about the program and a voucher amount considered to be too low. A sample of eligible individuals living in an urban area in Northern France were randomly split into three groups: a control group who benefited from the standard level of financial aid; a group benefiting from an increase in the value of the voucher; and a third group benefiting from the same increase along with an invitation to an information meeting regarding ACS. We show that the voucher increase has a small but statistically significant effect on ACS take-up. The invitation to the meeting appears to cancel out the positive effect of the voucher increase. Using an instrumental variable model to control for potential selection bias, we find ambiguous evidence of the meeting attendance on ACS take-up. This study confirms the difficulties that are faced in increasing the healthinsurance coverage of poor populations via subsidy programs.