View all Articles
Commentary By Yevgeniy Feyman

Federalism On Steroids: How To Fix Medicaid

Health, Health Healthcare, Affordable Care Act

Among many other changes to the health care system, the ACA created an expansion of Medicaid – made optional by the Supreme Court in 2012 – funded largely by federal dollars. Thus far, 30 states and the District of Columbia have accepted the Medicaid expansion. And as should be expected, states that expanded the program have seen spending grow much faster than those that didn’t. In a recent report, the Kaiser Family Foundation found that total Medicaid spending grew nearly 18 percent in expansion states, though the state share of growth was relatively low (less than 4 percent). And while health care has remained relatively quiet as a campaign issue, Governors Kasich and Christie – both Republican presidential hopefuls – expanded Medicaid (and both have defended this expansion) in their respective states.

Whatever the merits of Kasich’s and Christie’s decisions, a simple expansion of the Medicaid program is a bad idea. Poor access (to specialists, in particular), relatively low value to beneficiaries, and the program’s growing share of state budgets makes the it ripe for reform.

Of course, not all Medicaid programs are made equal. Some states – like Iowa and Arkansas – are taking the opportunity to privatize part of their Medicaid program using so-called “1115 waivers,” placing expansion beneficiaries into plans purchased on the exchanges.

Nevertheless, to the extent that we can generalize, studies have found that benefits from the program range from non-existent to beneficial for child health outcomes. (My colleague, Avik Roy, has summarized much of the evidence against Medicaid in his book.)

In one recent analysis, perhaps most damning, some of the nation’s leading health economists found that Medicaid’s value to actual beneficiaries is only about 20 to 40 cents on the dollar. Even when adjusting for a higher “value of a statistical life,” – a common tool that economists use to determine value – the upper limit on Medicaid’s value grows to 60 cents.

These latest findings are also the most valuable for determining Medicaid’s benefits. Its data was collected from an Oregon health insurance study which created a lottery for distributing Medicaid coverage in the state. The randomization created an opportunity to measure causal effects rather than simply correlational.

So what’s driving Medicaid’s poor performance? Primarily, it stems from the fact that the uninsured, according to the authors, only pay for about 20 cents of every dollar of health care coverage they obtain.

The implication, then, is that Medicaid spending could be replaced with equivalent spending elsewhere, making beneficiaries significantly better off without affecting overall welfare spending.

But this doesn’t mean states should simply abandon the population otherwise left uncovered by the ACA. Creativity will be key here, and as Paul Howard and I have argued , states can take the lead with so-called “1332 waivers” or “state-innovation waivers.”

These waivers – which allow states to get around many of the ACA’s regulations – work best when paired with waivers for Medicaid (the 1115 waivers that Arkansas and Iowa are using).

The basic idea is simple: 1332 waivers can help save money on the health insurance exchanges—where overregulated and overly-expansive insurance is unnecessarily expensive–and Medicaid waivers can allow changes to existing Medicaid programs. Importantly, this doesn’t necessarily mean moving beneficiaries into private plans, which can sometimes be more costly.

Instead, existing Medicaid programs could simply be modified with a deductible and a health savings account—this is what the Healthy Indiana  program once offered—along with state contributions to the HSA.

The idea behind this is simple. If cost-sharing could be structured in a way so as to reduce unnecessary use of services (without discouraging necessary use), the effect on beneficiaries is minimal. But done properly, this leaves a pot of money for the state to play around with, in order to address the issue of missing value. The best way to do this? Give beneficiaries money!

This is also where it gets a little tricky. There is no simple process by which states could offer direct cash grants to beneficiaries with Medicaid dollars. While a state may be able to request this authority under the existing Medicaid waivers (cash grants aren’t too different from HSA contributions), it isn’t clear whether the administration would sign off on it. But if states were to split savings from Medicaid and the exchanges with the feds, let’s say 50/50, that would still leave a significant chunk of money to play with.

(In an ideal world, there would exist – as Jim Manzi recommends in his book Uncontrolled – a more universal waiver, under which states could waive most, if not all, program requirements where federal funding exists. The goal? To experiment with different approaches of providing services and support for the poor. Paul Ryan’s Opportunity Grants come closest to approximating what this would look like.)

You could do much with the savings – invest in supportive housing, expanded mental health support, etc. Alternatively, states could even implement wage subsidies (a policy my colleague Oren Cass has advocated as being superior to EITC or minimum wage in reducing poverty). The point is that there’s a growing sense that we’re leaving a lot of value on the table with our Medicaid program. Smart reforms can deliver savings to taxpayers, and better lives to beneficiaries.

This piece originally appeared in