Timothy Layton (Harvard)

May 3, 2017 @ 1:30 pm – 3:00 pm
Roosevelt House Public Policy Institute
47 E 65th St
New York, NY 10065
Karna Basu

Timothy Layton is an Assistant Professor in the Department of Healthcare Policy at Harvard Medical School.

Seminar topic: “Supply Side Incentives and Health Spending: Evidence from Random Plan Assignment in Medicaid”

Abstract: Contracting out the provision of government services to competing private firms has become increasingly popular in the U.S. over the last 25 years. This is especially true in the Medicaid program, the largest insurer in the United States, where over 60% of beneficiaries receive their Medicaid benefits through a private Medicaid managed care plan. While some research exists comparing private vs. public provision of Medicaid benefits (i.e. fee-for-service vs. managed care) little is known about the underlying economics of Medicaid managed care markets. In this paper, we study the robust Medicaid managed care market operating in New York City. We leverage random assignment of Medicaid beneficiaries to managed care plans to reveal significant variation in spending across plans despite regulations virtually eliminating cost sharing: Spending on identical beneficiaries varies by as much as 30% between the highest- and lowest-spending plan. We show that differences in negotiated upstream prices explain only a small fraction of this difference, with almost the entire difference in spending attributable to differences in the quantity of services provided. We also show that beneficiary choices follow spending, with the vast majority of beneficiaries choosing to switch out of the low-spending plans switching to high-spending plans. Finally, we develop a stylized model of insurer competition in a market with no premiums, constant per-person payments, and a subset of beneficiaries who are either randomly assigned to plans (as in New York) or who choose poorly. We show that the model produces an equilibrium like the one we observe where there exist both high-margin/low-enrollment plans and low-margin/high-enrollment plans. We then conclude by using the model to show that alternative auto-assignment policies can improve overall consumer welfare by improving the quality of all plans in the market while holding the overall cost to the state constant.