How can the regulator show evidence of (no) risk selection in health insurance markets? Conceptual framework and empirical evidence

If consumers have a choice of health plan, risk selection is often a serious problem (e.g., as in Germany, Israel, the Netherlands, the United States of America, and Switzerland). Risk selection may threaten the quality of care for chronically ill people, and may reduce the affordability and efficiency of healthcare. Therefore, an important question is: how can the regulator show evidence of (no) risk selection? Although this seems easy, showing such evidence is not straightforward. The novelty of this paper is two-fold. First, we provide a conceptual framework for showing evidence of risk selection in competitive health insurance markets. It is not easy to disentangle risk selection and the insurers’ efficiency. We suggest two methods to measure risk selection that are not biased by the insurers’ efficiency. Because these measures underestimate the true risk selection, we also provide a list of signals of selection that can be measured and that, in particular in combination, can show evidence of risk selection. It is impossible to show the absence of risk selection. Second, we empirically measure risk selection among the switchers, taking into account the insurers’ efficiency. Based on 2-year administrative data on healthcare expenses and risk characteristics of nearly all individuals with basic health insurance in the Netherlands (N > 16 million) we find significant risk selection for most health insurers. This is the first publication of hard empirical evidence of risk selection in the Dutch health insurance market.