Efficiency in Plan Choice with Risk Adjustment and Premium Discrimination in Health Insurance Exchanges

Adverse selection, a major concern in health insurance markets, induces two forms of inefficiency. First, insurers may manipulate plan policies to attract profitable enrollees. Second, when premiums do not reflect the marginal cost of consumer choice, consumers make inefficient choice among plans with different levels of generosity. In theory, two payment system policies address selection: risk adjustment contends with insurer actions to affect selection, and risk-based premiums can improve the efficiency of consumer plan choice.

While risk adjustment has long been implemented in practice, such as in the U.S. Medicare program, and the national health care systems of the Netherlands and Germany, premiums are not commonly adjusted for risk.  One recent change is that premiums may be based on enrollees' age, which is highly related to health status, along with a few other individual characteristics in the Health Insurance Exchanges as part of the Affordable Care Act (ACA). Thus, risk adjustment and premium discrimination, for the first time, will work together through the new Exchange payment system.

In this paper, I analyze how the efficiency of plan choice is affected by risk adjustment and premium discrimination in a model of the Exchanges. I begin by constructing a model describing plan sorting in equilibrium: consumers with expected health care expenditures greater than or equal to a threshold enroll in the more generous insurance plans, and with expenditures smaller than the threshold enroll in the less generous plans. In order to understand how risk adjustment and premium discrimination work separately and jointly, I study four cases: 1) No risk adjustment or premium discrimination is implemented; 2) Only risk adjustment is implemented; 3) Only premium discrimination is implemented; 4) Both are implemented. Policy impacts on plan choice are characterized by comparing plan enrollments and premium levels based on the thresholds in the four cases. The model shows that premium discrimination could improve or impair efficiency in general, while risk adjustment always improves efficiency.

I then apply the model to a potential Exchange population drawn from five waves of the Medical Expenditure Panel Survey (MEPS), and simulate the equilibrium in each case. The results show that both policies improve efficiency of plan selection for the overall population in the context of Exchanges.

I also construct two measures to assess welfare impact for consumers. One is the fraction of the population enrolling in the more generous plans, and the other is the average “risk premium” borne by the population. I find that risk adjustment and premium discrimination increase consumer welfare separately and jointly, and welfare loss of consumers is minimized when both policies are implemented.