Supporters for the Romney-Ryan approach to Medicare have a new talking point. They say a new study by “three liberal Harvard economists” proves that the plan’s competition will reduce health care costs without harming beneficiaries. But the study doesn’t say that.
And I should know. I’m one of the economists who wrote it.
Both Mitt Romney and Paul Ryan have said they would like to convert Medicare into a "premium support" (nee voucher) system. Their plans are different, and Ryan himself has proposed several versions. But they share a basic architecture. Starting ten years from now, new retirees would not receive a Medicare card, as they would today. Instead, they would receive a voucher and shop for an insurance policy in a specially regulated market.
The voucher would equal the price of the second-cheapest plan in the market, although its value would be less if insurance prices rose faster than a pre-determined spending cap (of gross domestic product plus half a percentage point)—as they are projected to do. Both Romney and Ryan now say that traditional Medicare, the government-run insurance program, would be among the options in the marketplace. But they would not guarantee that voucher can pay for it. In fact, that’s very much the point of the proposal: To create more competition between Medicare and private plans, even if that means Medicare ends up costing more than the vouchers are worth.
How would this affect seniors? In particular, how many seniors would end up paying more to stay in traditional Medicare?
That’s the question that Zirui Song, Michael Chernew, and I set out to answer in the study, which was published in the Journal of the American Medical Association. To do this, we examined what would have happened if, today, something like the Romney-Ryan plan were in place: In other words, if today’s seniors were getting vouchers, how much would those vouchers be worth?
We found that 24 million seniors, or about two-thirds of the people presently enrolled in the traditional Medicare program, would have to pay more—specifically, an average of $64 per month or $768 per year. Some seniors already enroll in private plans, as part of the “Medicare Advantage” option that has existed, in one form or another, for many years. About 7 million seniors or more than 90 percent of that group would have to pay more.
Supporters of voucher schemes have taken this to be vindication. Among them are James Capretta and Yuval Levin, writing for the Weekly Standard. After all, hadn’t we found that the private plans were cheaper than old-fashioned, government-run Medicare? Doesn’t that prove competition can really lower costs?
Here’s how Capretta and Levin put it:
The Harvard researchers looked at the (limited and constricted) private-plan option already operating in Medicare today—a program called Medicare Advantage … and found that, on average, the Medicare Advantage plans cost far, far less than federally run fee-for-service Medicare.
This is the opposite of what Democrats were saying a year ago. Then, they were touting a Congressional Budget Office study that estimated the private plans offered to Medicare beneficiaries in the system Ryan envisions would cost much more than traditional fee-for-service Medicare, and thus require higher premiums—$6,400 higher in 2022—to be paid by beneficiaries. This new study shows otherwise, and proves the very point that champions of premium support have been making for years.
The $64-per-month estimate is based on the study’s finding that private plans can deliver the full Medicare package of benefits at a significantly lower cost—nearly 10 percent lower, on average—than the government-administered fee-for-service program. That’s precisely the win-win proposition Paul Ryan has been touting: Beneficiaries could get their comprehensive Medicare benefits for no additional premium if they selected the less expensive private plans, and taxpayers would spend 10 percent less on the subsidies for the Medicare program.
But this is a distortion of our findings, for several reasons. First, it confuses costs and payments. Medicare Advantage plans bid less than traditional Medicare, but they are paid more. The plans are officially supposed to use these higher payments to sweeten the pot—add additional benefits, reduce cost sharing, and the like—though some likely go for profit as well. This is why the Affordable Care Act reduced the amount that the government pays to managed care plans, over howls of protest from conservatives. Bidding less does no good for the program if the government then overpays relative to what was bid.
Second, they miss a key part of the reason why the Congressional Budget Office estimated that Ryan’s voucher proposal would cost seniors more. Medicare Advantage plans can only cost what they do because the traditional Medicare program is in place to help them. Specifically, Medicare sets very low payment rates to providers, and Medicare Advantage plans bargain up a bit from those rates. Get rid of the traditional Medicare program, or even reduce its enrollment substantially, and the estimated cost of Medicare Advantage premiums skyrockets.
Third, determining whether the private plans are really more efficient than traditional Medicare requires more than just knowing that they bid less. The question is why private plans come in cheaper. We take a cautious, nuanced view:
Private plans can cost less than traditional Medicare because: (1) they may use medical resources more efficiently; (2) they may enroll healthier patients relative to the risk-adjusted payment; or (3) their negotiated prices may not fully reflect the costs of indirect medical education or payments for disadvantaged hospitals, which traditional Medicare explicitly pays. The magnitudes of efficiency, selection, and avoided add-on payments are unclear... To the extent that the 9% cost advantage reflects efficiency, it suggests there are better ways to provide the traditional Medicare benefit.”
To put it a bit more plainly, it’s possible that the private plans are cheaper because they really do offer the same benefits at a lower cost. It’s also possible that the private plans are cheaper because the insurers are very good at attracting the best risks—that is, the healthiest seniors least likely to run up medical bills—or because they don’t also subsidize other parts of our health care system, such as medical education. In effect, they may be gaming the system. At this point, we really don’t know which answer is correct, although it’s entirely possible all three are true, to an extent.
Making the wrong assumption here could be fatal, particularly to those seniors with the gravest health needs. If managed care plans are able to select healthier enrollees—by skimping on benefits in ways that get around whatever regulations (if any) the Romney-Ryan plan put in place—traditional Medicare will end up with less healthy seniors, driving up its costs. The system will spiral out of control. The costs will proportionately rise and the guarantee of benefits that is the core of the Medicare program would erode. That is why many economists are wary of a pure premiums support model.
At the very least, it makes sense to see how premium-support works in the non-elderly population, since their health needs overall are less severe. The Affordable Care Act does that, by creating “exchanges” for people who don’t have employer-sponsored coverage. Watching and learning from that initiative would help in designing a workable system for the elderly. That is why, on many counts, the biggest lesson is that allowing the Affordable Care Act to work—rather than trying to take it off the books—might be the best way for premium support to succeed.
David Cutler is Otto Eckstein Professor of Applied Economics at Harvard University. In 2008, he was senior health care advisor to the Obama Presidential campaign.