« Back to Results

Market Power, Consumer Behavior, and Welfare in Health Care Markets

Paper Session

Sunday, Jan. 3, 2021 3:45 PM - 5:45 PM (EST)

Hosted By: American Economic Association
  • Chair: Robin Lee, Harvard University

Optimal Health Insurance in the Presence of Market Power

Kate Ho
,
Princeton University, NBER and CEPR
Robin Lee
,
Harvard University and NBER

Abstract

Much of the literature concerning optimal coverage design in insurance markets focuses on the trade-off between risk exposure and incentives. Early papers (Arrow 1963, Pauly 1968) argue that risk averse consumers benefit from insurance, but if they are responsive to price at the point of care (ex post moral hazard), an increase in coverage generates an increase in spending that offsets the benefits. Absent externalities from care, full insurance is socially optimal if consumers are risk averse but do not exhibit moral hazard. With moral hazard but no risk aversion, again abstracting from possible externalities, any medical utilization in response to insurance coverage is socially inefficient and therefore any insurance coverage is suboptimal.  These essential conceptual points, present in many recent models, imply that partial insurance with some degree of cost-sharing is optimal in settings with both risk aversion and moral hazard. We add insights from the industrial organization literature regarding health care provider oligopoly and markups. If medical providers charge insurers prices that are above true marginal (social) costs, some portion of consumers' incremental utilization due to insurance coverage may generate marginal benefits that are greater than these costs (and hence desirable), but still less than providers' prices. This provides an additional motivation for a social planner to provide more generous insurance coverage than would be prescribed in the absence of provider markups.  We use a model of insurance plan choice and health care utilization from Ho and Lee (2020), estimated using data on employees at a large firm, to quantify the increase in the socially optimal coverage level that accounts for provider markups. We show that adjustments are economically significant.

Product Proliferation under Rational Inattention: Application to Health Insurance

Zach Y. Brown
,
University of Michigan
Jihye Jeon
,
Boston University

Abstract

In markets with complicated products such as insurance, firms often offer many similar products. We develop a simple model showing that firms can increase profits by offering very similar or duplicative products when consumers are rationally inattentive. Focusing on prescription drug insurance, we document that there are many similar plans offered by the same insurer and examine how these similar plans affect choice. Using a demand model with rationally inattentive consumers, we use counterfactual simulations to examine how limiting similar plans affects consumer surplus.

Can Automatic Retention Improve Health Insurance Market Outcomes?

Mark Shepard
,
Harvard University and NBER
Adrianna McIntyre
,
Harvard University

Abstract

There is growing interest in market design that uses default rules and other choice architecture principles to steer consumers toward desirable outcomes. Using data from Massachusetts’ health insurance exchange, we study an "automatic retention" policy intended to prevent coverage interruptions among low-income enrollees. Rather than disenroll people who lapse in paying premiums, the policy automatically switches them to an available free plan until they actively cancel or lose eligibility. We find that automatic retention has a major impact, switching 26% of consumers annually and differentially retaining healthy, low-cost individuals. We discuss policy tradeoffs and potential impacts on insurer competitive incentives.
Discussant(s)
Martin Gaynor
,
Carnegie Mellon University and NBER
Joonhwi Joo
,
University of Texas-Dallas
Fiona Scott Morton
,
Yale University and NBER
JEL Classifications
  • D2 - Production and Organizations
  • D4 - Market Structure, Pricing, and Design