Moral Hazard and Adverse Selection in Private Health Insurance

Moral hazard and adverse selection create inefficiencies in private health insurance markets.  We use claims data from a large firm to study the independent roles of both moral hazard and adverse selection.  Previous studies have attempted to estimate the price elasticity of medical care by assuming that individuals respond only to the spot price, end-of-year price, expected price, or a related metric.  There is little economic justification for such assumptions and, in fact, economic intuition suggests that the nonlinear budget constraints generated by health insurance plans make these assumptions especially poor.  We study the differential impact of the health insurance plans offered by the firm on the entire distribution of medical expenditures without parameterizing the plans by a specific metric.  We study a firm which transitioned from offering only one plan to offering 3 plans that differed only by the coinsurance rate, deductible, and stoploss.  We predict plan enrollment based on observable characteristics and study the changes in medical expenditures for a given set of characteristics for the period in which each person was constrained to enroll in the same plan to the period in which a choice in plans was available.  We use a new instrumental variable quantile estimation technique introduced in Powell (2013) that provides the quantile treatment effects for each plan, while conditioning on the set of observable characteristics for identification purposes.  This technique allows us to map the resulting distributions to the nonlinear budget set generated by each plan.  Our instrumental variable results isolate the impact of moral hazard and provide estimates of the resulting distribution generated by each plan if selection into each plan were random.  The difference in the observed distribution from the estimated causal distribution identifies adverse selection. This method allows us to separate moral hazard from adverse selection and estimate the relative importance of each for the plans in our data. Our results suggest that more generous plans induce consumption of higher health care spending.  However, we find that attributing this additional consumption to the end-of-year price provides a very different distribution of medical expenditures and we can statistically reject that individuals are responding solely to this price.  We also find strong evidence of adverse selection and we able to quantify its impact.