The Agenda

The GOP Budget

I’m writing my Friday column on Rep. Paul Ryan’s budget proposal, so I don’t want to get ahead of myself. Naftali Bendavid has written a useful scene-setting piece for the Wall Street Journal.

I strongly endorse the frameworks advanced by Tyler Cowen and Megan McArdle for how to think about the unfolding debate about government health expenditures.

To get a sense of what opponents of Rep. Ryan’s approach will say in the coming weeks, I strongly recommend reading Ezra Klein’s posts on the subject:


In Medicare’s case, the reform is privatization. The current Medicare program would be dissolved and the next generation of seniors would choose from Medicare-certified private plans on an exchange. But that wouldn’t save money. In fact, it would cost money. As the Congressional Budget Office has said (pdf), since Medicare is cheaper than private insurance, beneficiaries will see “higher premiums in the private market for a package of benefits similar to that currently provided by Medicare.”

It is really important to clarify what we mean when we say this proposal will or won’t save money. There are total expenditures on medical care and there are taxpayer-funded expenditures. The Medicare program is currently designed to meet a very large share of expenditures on medical care for all Americans over the age of 65. Medigap plans and modest co-pays are designed to help fill the gap. 

One of the ideas behind moving to a defined contribution model is to change the system so that cost-sharing plays a bigger role. The idea is that this cost-sharing will encourage Medicare beneficiaries to choose plans that offer more bang-for-the-buck, and that this in turn will encourage insurers and providers to deliver more bang for the buck. 

The vouchers are designed to grow at a fixed pace, broadly in line with the increase in the size of the overall economy. A given package of benefits may well prove more expensive than the defined contribution, and beneficiaries will be expected to pay the difference between the defined contribution and the ultimate cost out of pocket. Or they beneficiaries could choose a somewhat less generous plan or a plan that embraces a more efficient delivery model. 

If the defined contribution model does not encourage more cost-consciousness and does not encourage business-model innovation, this approach won’t work very well. Much depends on how you think markets in this space will evolve in the medium term. Some people, including Ezra, believe that we can evaluate this effort to revamp the entire Medicare system by looking to the experience of Medicare Advantage plans. Others believe that the persistence of Medicare FFS has an impact (a distorting impact, one could say) on the broader healthcare marketplace that can’t be discounted.

Ezra continues:

In Medicaid’s case, the reform is block-granting. Right now, the federal government shares Medicaid costs with the states. That means their payments increase or decrease with Medicaid’s actual rate of spending. Under a block grant system, that’d stop. They’d simply give states a lump sum at the beginning of the year and that’d have to suffice. And if a recession hits and more people need Medicaid or a nasty flu descends and lots of disabled beneficiaries end up in the hospital with pneumonia? Too bad.

It’s a bit more complicated. Medicaid’s actual rate of spending depends on a lot of different things, including the eligibility rules established at the state levels. As you can imagine, some states are a lot more generous than other states. Right now, the federal government will give more money on a per person basis to the most generous states than to the least-generous states. This creates a funny problem: in a lot of cases, rich states get much more money than poor states. Now, you’d think that rich states would be better able to sustain the burden of caring for poor and near-poor citizens by drawing on their own tax base. And you’d be right. But Medicaid doesn’t work that way right now. By giving states a predictable lump sum, you can, in theory at least, give states much more flexibility in deciding how best to deploy those dollars. Some states might design to build a better network of publicly-funded clinics to meet basic health needs while others might want to experiment with HSAs and cost-sharing. That is the biggest advantage of a block grant.

Ezra does raise the important point about countercyclical policy. What happens if a recession or a nasty flu hits? Fortunately, there are a lot of options. A block grant could be designed in many different ways, e.g., it could guarantee states $11,000 for everyone below a certain level of income. A recession would presumably increase that number, and so the block grant would increase the next year. Ah, but what about this year? This points us to a much bigger problem about countercyclical policy that Christopher Edley Jr. proposed solving last year: we could allow state governments to take advances against future transfers.

As for a nasty flu, I think much depends on the nature of the flu. If it’s a rough flu season that raises costs at the margin, state governments will have to find a way to work around it, perhaps by trimming other expenditures. If it’s a serious outbreak that impacts much of the country, I assume that a democratically elected legislature would intervene. A block grant would not prevent the federal government from taking emergency measures to fight against, say, avian flu. 

The core idea behind block grants is to fix the misalignment of incentives.

Ezra continues:

Remember how the Affordable Care Act was really, really, really long? There was a reason for that. It was full of delivery-system reforms meant to make the health-care system cheaper and more efficient — things like bundling payments for illnesses and reducing reimbursements to hospitals with high rates of infection and creating a center tasked with seeding cost-control experiments throughout Medicare and encouraging the formation of Accountable Care Organizations. The hope is that those reforms will cut costs, which will make the rest of the bill’s cuts possible (more on thathere). Republicans, notably, have been skeptical that these reforms will work, and have argued that the cuts won’t stick because beneficiaries will revolt.

The difference here is about how to make delivery-system reforms work. Do we relax restrictions on business-model innovation — PPACA actually creating new protections for general hospitals and put in place barriers that could stymie business-model innovation — or do we take a more centralized approach in which we rely on the best evidence available to experts working alongside the federal government? Many of us suspect that a decentralized, trial-and-error process that uses price signals rather than administrative guidance is the best way to go. 

To my knowledge, Ryan’s budget doesn’t attempt to reform the medical-care sector. It just has cuts. The hope is that those cuts will force consumers to be smarter shoppers and doctors to be more economical and states to be more innovative. But all that’s been tried, and it hasn’t been enough. 

It isn’t obvious to me that the defined contribution model has been tried. If anything, we’ve moved away from its core tenets in recent years, e.g., the way the tax treatment of FEHBP was “reformed” moved us in a very different direction, the way in which Medicare FFS was never placed on an equal footing with Medicare Advantage plans as Coulam, Feldman, and Dowd proposed, etc. But there’s room for interpretation here. 

This promises to be an interesting conversation. 


The Latest

Is Trump buckable? &c.

Is Trump buckable? &c.

On internal GOP politics; a Chinese artist and the Olympics; Ukraine and its right to exist; Joan Didion and other Buckley-hired talents; and more.