• “Individualized premiums.” We’re already part way there

    I have been disappointed but not surprised by the response to the health reform proposal offered by health policy experts at the American Enterprise Institute (AEI) last week. Some conservatives, I am told, were irate that the plan bears some structural similarity to the Affordable Care Act (ACA). Some progressives, I have observed, seem irate on the grounds that offering something different from the ACA is itself an abandonment of the principles that motivate it.

    There are other objections too, as expressed by Henry Aaron and echoed by Harold. Though Harold was “gentler,” he called the idea of experience rating a “gigantic blunder.” In many ways I agree with Henry and Harold (more on that below). But in one way, I think we’re all in danger of making a gigantic blunder.

    I doubt many health policy experts would disagree with me that the politics of health care policy these days are dreadful or worse. There are few health policy ideas that can be openly discussed without attracting an angry swarm of charges and counter-charges of playing politics. Even when scholars aren’t directly making such accusations, I can’t help but feel that the high political tension has infected what might otherwise be more deliberative entertainment of ideas.

    In a sense, this is entirely understandable. The stakes are very high. ACA implementation is important. It’s important to me, as I know it is to Henry and Harold, as well as for at least some of the authors of the new proposal. (I don’t know them all.) For reasons I do not fully understand, it is extremely important to others that the ACA not be implemented. In this context, any proposal that departs significantly from the ACA can be seen as a threat to the law’s supporters. Any proposal that doesn’t fully repeal it can be seen as a threat to the law’s opponents.

    Enter the proposal offered by scholars at the AEI. Despite how it was received, there is a lot in it that is not controversial among health policy experts. For instance, most experts are open to eliminating the preferred tax treatment of employer-sponsored insurance. Many would support removing the two-tiered nature of health care that Medicaid promotes. Nobody proposes a health insurance plan that doesn’t include subsidies that vary by income and provisions that make care more affordable to those with higher health care expenses. Though the goal of universal coverage is not uniformly shared, few are willing to publicly and explicitly reject it.

    So, what’s wrong with the new proposal? I was not alone in characterizing it as politically dead on arrival. Harold put his finger on the most challenging, substantive aspect of it: the call for an entirely experience-rated system. This is radical. Does that make it a “gigantic blunder”? Considering the plan also includes income- and health-based subsidies and multi-year contracts to guard against reclassification risk, I’m not so sure. I think it warrants consideration, discussion, and analysis.

    I would like to see the scholarly community, myself included, calm down and undertake that work. Let us evaluate it on the merits. This cannot be done in a few days. Any full-throated rejection of the idea right now seems to me to be premature.

    Indeed, it doesn’t take lengthy consideration for one to opine that the subsidies in the proposal are less generous than one may like, as Harold pointed out. But as the authors articulated at the AEI event, they are not wedded to them. They’re just a placeholder for whatever society finds more palatable. We shouldn’t attack the subsidy level. We should discuss what it should be.

    But should we attack the plan’s structure? Is experience rating, in this context, so bad? I think Harold was right-on in writing that the “ACA’s most immediate challenge in this area concerns allowable age-gradients in premiums.” This alone admits at least the possibility that the ACA’s modified community rating structure may need to be reconsidered. It’s already impure community rating, allowing premiums to vary by age (up to a 3:1 ratio) and smoking status. Should there be a problem with the selection of risk into marketplace plans and to the extent that is due to too few young people enrolling, a natural solution would seem to be to reconsider the degree of age rating, along with subsidy levels. Perhaps instead of a 3:1 ratio, permitting premiums to vary by, say, 5:1 would be a helpful step.

    Recognize that what is being considered here is a different degree of experience or community rating. Yet, Blumberg and Buettgens find that loosening the age bands would have very little effect on out-of-pocket premium costs for younger consumers. So, this may not, in fact, be a very productive direction. (H/t Adrianna for drawing my attention to this analysis.) Perhaps for this reason, Handel, Hendel, and Whinston find that even under full age-based pricing, about one-quarter of the population would not participate (ignoring the mandate), skewed heavily toward lower ages and leading to selection-based market unraveling. (H/t Aaron Shwartz for drawing my attention to this one. Also, caveat: It’s a working paper.) Not that we should take these two studies as definitive, but see what a little analysis can do?

    Of course there are many other solutions to the problem of adverse selection into marketplace plans resulting from low take-up. We could make the mandate penalty more severe. We could impose delayed enrollment penalties. One could propose auto-enrollment with an opt-out. Etc. Not everyone would be happy with all elements of any additional reform. But the role of the scholar is not to be happy. It’s to analyze ideas dispassionately and to assess the extent to which they might address specific problems we may face.

    What the new proposal encourages us to do is to think further beyond the current (modified) community rating structure of the ACA. A greater degree of experience rating, with subsidies and tied to multi-year contracts (as the proposal offers) are solutions to certain problems. If those problems arise, these solutions may be useful, though not unique. Today may not be the day to fully embrace such ideas. By the same token, it may not be the day to fully reject them either.


    Comments closed
    • This is the most thoughtful commentary on this issue I have seen and exactly why I come to TIE daily.

      I’m pretty sure that the consumer-oriented folks and the progressives (if the two sides can be categorized so clumsily) talk past each other. Both seem to have very similar objectives. However, the progressives are looking mostly for first-order effects to achieve their objectives, while the consumer-oriented folks believe the power is in the second- and third-order effects. I rarely see discussion involving (or acknowledging) both.

    • The authors at the AEI symposium did not describe their method of indemnifying against re-classification risk as straightforwardly as Mark Pauly or John Cochrane have.

      I was at the event and asked what they would do about someone who did not have insurance, contracted HIV/AIDS and then shopped for insurance in the proposed national exchange. I asked:

      ” Would I go to the exchange, tell them I have HIV/AIDS, be quoted an annual premium of $150,000, and then be directed to the IRS to apply for a tax credit of $135,000, for example?”

      The answer was that my income would be immediately adjudicated by the national exchange crunching data with the IRS. But that does not do the trick because it confuses income-based subsidy with risk adjustment.

      In Medicare Advantage or the forthcoming Obamacare exchanges, there is re-insurance, but this is invisible to the beneficiaries. Income-based premium differences (in Medicare Advantage) or tax credits (in exchanges) are visible to the beneficiaries.

      Leaving aside the question of prospective versus retrospective risk adjustment, I struggle to see the economic difference between executing the reinsurance at the “retail” level (as proposed by the AEI symposium) and executing it “wholesale”, as in Medicare Advantage and Obamacare exchanges.

      Politically, I cannot imagine that “retail” re-insurance would be preferable to “wholesale” re-insurance.

    • Austin, as usual, this is a thoughtful commentary. You’re of course correct that the AEI proposal should spur discussion and academic analysis.

      But beyond the intellectual exercise, or perhaps informing a reform that may happen years from now, what’s the point? AEI’s proposal would have been useful four or five years ago. But right now, like or not, for better or worse, the ACA is the reform that’s being implemented. It would be far more valuable if AEI and other right-leaning groups accepted reality and discussed ways to make the ACA better and more effective. Instead, we get an intellectual exercise that has no relationship to current political realities.

      Except, perhaps, for the political reality that makes it verboten for any right-leaning groups to acknowledge that the ACA is going to happen and suggest ways to improve it. It’s still all or nothing . . .

      • The “current political reality” is that PPACA is falling apart before our eyes. It is less likely that Republicans will accommodate it today than it ever was.

        What the environment will look like 12 months from now, I hesitate to predict.

    • Very commendable post.

      As much as I hope that it will catalyze the sort of discussion and analysis that you desire, if I were placing bets I’d bet against such an outcome.

      My conversations with people who view the world in terms of the oppressor-oppressed axis (to borrow from Kling) have persuaded me that for them, the mechanism by which a redistribution from oppressor to oppressed is achieved is as important, if not more important than the magnitude of the redistribution, and economic efficiency takes a distant back seat to perceived equity. For exhibit A, just look at the dialogue regarding the merits of the negative income tax versus the minimum wage as a redistributive mechanism. Even those who acknowledge that a negative income tax is vastly more efficient have a great deal of difficulty endorsing such a mechanism for bringing about a redistribution because it violates many of the notions of fairness and equity that they value above any consideration of efficiency.

      My prediction is that much the same fate will befall the central tenets of the AEI plan, for precisely the same reasons.

      • It feels like you’re mixing up several ideas. One is analysis by academics and experts capable of conducting it. I do believe that will occur. Another is dissemination of that work in a manner that makes it comprehensible to a wider audience. Some of us do that work, but far too few. Finally, there is the wider, honest and open discussion we’d like to see among the broad populace. That is very unlikely to occur. I was mostly talking about the first of these three because it generates the inputs I need to do what I do.

        • Does the ideological chasm between, say, the freshwater and saltwater schools of econ endow you with any particular confidence that people who should be capable of engaging in dispassionate analysis will necessarily do so? Macro-land is a platonic ideal compared to health policy. Is there anyone who can plausibly claim that ideological priors don’t ultimately determine everything from the models that the main players in the field favor to the the policy prescriptions that they champion in public?

          Is there anything in particular that justifies your confidence in the health econ commentariat’s capacity for impartial analysis?

    • IMHO the ACA to be politically advantageous to the politicians who passed it it had to be structured to fool the voters and so knowledgeable people even if they support ACA as better than nothing should be shouting how badly structured it is and how the politicians are scamming the voters. ACA cost much more than it should, solves much less of the problems that it should. It creates very bad incentives for employers! It creates bad incentives for patients and it will accomplish very little in improvement in heath. IMO Democrats are too willing to support bad very high cost legislation to get a tiny theoretical gains in some areas. Consider CAFE standards, MIT says it costs 6-13 time more per gallon of gas saved (co2 put in the air) than a visible tax yet even knowledgeable Democrats talk it up. You should always be pointing out how corrupt our politicians (and to a lesser extent voters) are. Trying to lower the status of politicians.