The Consequences of (Partial) Privatization of Social Insurance for Individuals with Disabilities: Evidence from Medicaid

Over 60% of Medicaid enrollees are enrolled in a private comprehensive managed care plan. While previous research has studied how privatization has affected mothers, children, and babies, little is known about the consequences of privatization for individuals with disabilities. This is an important omission in the literature in that Medicaid spending for this population amounted to almost $90 billion in 2014. This equates to 40% of total Medicaid spending, despite this population only making up 13.5% of total Medicaid enrollment. This population is also the most policy relevant population when it comes to the public vs. private question, given that many states are currently grappling with the question of whether to move these individuals to private managed care plans and that it is precisely in this sick and vulnerable population where managed care is most likely to have important effects.

In this paper, we study the consequences of Medicaid privatization for adults with disabilities leveraging natural experiments in New York and Texas. Both states rolled out managed care to the disabled population in the mid-2000s. Importantly, they only rolled out managed care in a subset of counties, maintaining the traditional fee-for-service Medicaid program in the rest of the state. We use this experiment to study the consequences of privatization in a difference-in-differences design, comparing changes in outcomes in counties that transitioned to managed care to changes in outcomes in contiguous counties that maintained the public fee-for-service program. We find sharp variation in enrollment in private managed care plans. We use this variation to first estimate the consequences of privatization for total spending by the Medicaid program, finding that managed care increased spending by $1,279.32 per person, per year, or around 12-13% of baseline spending. About one-third of the spending increase was due to an increase in spending on “carved-out” services not included in the managed care plan contract, while the other two-thirds of the increase was due to premium payments to managed care plans that exceeded counterfactual fee-for-service spending for the transitioned individuals. We also find dramatic increases in (carved-out) prescription drug spending on the order of 40-50%. These increases were concentrated in drugs used to manage chronic conditions that are prevalent in this population such as diabetes, heart disease, asthma, mental illness, and pain. In addition, we find strong evidence of an increase in outpatient care and office visits and a decrease in inpatient spending. We conclude by assessing the effects of privatization on avoidable hospitalizations and other measures of enrollee health.