Reclassification Risk in the Small Group Health Insurance Market

Abstract

We evaluate reclassification risk in the small group health insurance market from a period before ACA community rating regulations. Using detailed individual-level data from a large insurer, we find a pass through of 16% from health risk to premiums with enrollee fixed effects, and 70% without fixed effects. The fixed effects estimates identify the extent to which the insurer passes through changes in risk to changes in premiums while the higher estimates without fixed effects may be due to more risk rating for new accounts. Our estimates control for selection into insurance take-up with a non-parametric selection model, using individual risk and industrial sector as exclusion restrictions. Our results are also robust to other possibilities, including potential measurement error of risk scores and slow pass through over time. We seek to explain why our fixed effects estimates are much closer to community rating than full experience rating. The limited pass through may be due to implicit “guaranteed renewability” contracts with one-sided pricing commitment on the part of the insurer, as our results are broadly consistent with the equilibrium pass through that would occur under these contracts. Our results cannot be explained by market power, search frictions, or slow pass through over time. We simulate the value that is generated by the insurer’s pricing policy relative to counterfactual pricing policies. The insurer’s policy generates 60% of the welfare gain from community rating relative to full experience rating. Even community rated plans generate substantial reclassification risk due to high out-of-pocket costs.