• It’s not the age bands, ctd.

    “It’s not the age bands,” writes Harold. He’s right. It isn’t.

    I have no disagreement with what Harold wrote. However, I may be more open to additional modifications to community rating in the individual market if the Affordable Care Act’s current structure fails. I am not predicting failure. I am not supporting failure. I am hoping things turn out just fine.

    But, what if they don’t? Then, I welcome all ideas on the table, including those scholars offered recently at AEI, and about which Harold and I have already gone back and forth (here and here). That we evaluate some ideas in advance of the possibility (and by no means certainty) of trouble doesn’t bother me in the least. I think that’s entirely appropriate. With this, Harold agrees.*

    The value of the plan is three-fold. First, it affirms many of the health reform principles held by most progressives and many (most?) conservatives. Second, it demonstrates that scholars of varying political persuasions can commit themselves to addressing an important set of problems with the U.S. health system. In large part, a lack of such a commitment as the first principle is what gives rise to the acrimony over the ACA and its implementation. Third, with respect to risk (or “experience”) vs. community rating, it encourages us to expand our view of the possible, as well as, perhaps, contract it, all on the merits.

    For example, it encouraged me to seek and be receptive to research relevant to risk rating, as cited in my post. What I found and reported was that it empirically isn’t about the age bands. Even loosening them up all the way to year-specific risk rating would not go very far toward encouraging young people to enroll. Consequently, should selection into marketplace plans prove to be an issue, one must consider alternatives, or accept whatever market failure may result. One idea is some additional risk rating in another dimension. Another is a more stringent mandate. Indeed, Handel, Hendel, and Whinston do demonstrate that cranking up the mandate penalty would encourage enrollment of young people, as one would expect.

    My point is that I want to see more evidence and analysis like this. Harold agrees. So, let’s have at it!

    Still, as I wrote, I do not believe the full AEI plan would ever be implemented, and Harold suggests some reasons why. It may be some Americans’ way without being the American way. If I recall correctly, even at the unveiling event itself, it was acknowledged that consideration of partial movement (whether by age or otherwise) toward full risk rating might be useful, and it might be more broadly acceptable. Indeed, as we all know, the ACA is not a full community rating scheme as it is. Is smoking status the only non-age dimension of risk on which we might agree to rate (conditional on selection being a  problem in need of a solution)?

    Where I think Harold and I may disagree is that he’s more confident than I am right now that solutions other than additional risk rating would be better, apart from whether they’re politically or culturally acceptable. Where we probably agree is that other solutions, like a larger mandate penalty or some alternative inducement, than additional risk rating are far more likely to be implemented, whether better or not. I should point out, even if that is the case, such solutions may be an improvement over doing nothing. On those grounds, if not others, we might both support them.

    This is an often overlooked nuance in health policy debate. In our confidence on what would be ideal or a blunder, we can agree to disagree. But at the end of the day, we can agree to move forward with something that’s nobody’s ideal but better than the status quo. We can share a principle of ambition that motivates compromise on the details. Harold and I already share that, as do, I think the authors of the new proposal we’ve been discussing. How refreshing it would be if this were a broadly shared cultural value, especially inside the Beltway where it is, I’m afraid, dead on arrival.

    * Indeed, the frame that this is an argument is incorrect. Harold and I are having a very valuable and enjoyable discussion. You get to watch.


    • I’m jumping in this a bit late, but I think that one point that needs to be made is that even though there may be deadweight losses associated with adverse selection, some amount of adverse selection can still be considered socially optimal in the Mirrleesian sense. That is, the main objective here should be to maximize social welfare, not coverage per se, which means we want to redistribute in a way that balances healthcosts across risk types, subjec to participants’ incentive compatibility constraints

      A pure community rating is unlikely to be the precise optimal. But a modified community rating like the ACA has, plus the uninsurance tax that the ACA has, could be a close approximation to the optimal policy function. For example, consider three types of individuals of decreasing health risk types. We impose a community rating that causes the low risk type to drop out of the market. The community rating causes redistribution from the medium risk to the high risk types within the insurance market, and then redistributes from the low risk types to the medium and high risk types by way of the tax code.

      Under the optimal policy, low risk types would not drop out of the market. But between the uninsurance tax and guaranteed issue, it’s kind of like they are still in the market.

    • Austin (and Harold),

      What i find interesting:

      If we lived in a parallel universe where experience rating was the norm, a proposal for community rating with attendant x-txrs from insurance company A to B (risk corridors, RA, reinsurance) would be viewed equally as disruptive and illogical.

      I would guess, community rating became world standard long ago–not because of best in class–but because we did not have the technology, understanding, or broad palette of diseases to utilize to underwrite. On the latter, how many “ICDs” existed in 1935; how many now?


    • Writing that “It’s not the age bands” risks missing what may be the core question on which this debate turns. The young and healthy can either purchase insurance or not. They must choose between having insurance and paying some premium or not having insurance and paying some penalty. Many policies push on one side of this ledger or the other. The fact that full age rating may not, by itself, bring the young and healthy into the market isn’t really the point. The point is that it will reduce their premiums non-trivially, bringing them closer to the margin of purchasing insurance. Full age rating thus reduces the size of the penalty required to hold the market together.

      To put the point another way, It is also likely true that, absent any age rating (i.e., under pure community rating), increasing the penalties by two or three thousand dollars would be insufficient. In this case as well it would present an incomplete picture to write “it isn’t the penalty.”

      Also relevant to this discussion are policies designed to remove particularly high risks from the exchanges. High risk pools, covering the disabled through Medicare or Medicaid, and other such policies will also make it more likely that the market holds together with modest penalties and incomplete age rating. The following is a hypothetical worth thinking about in this context: What would happen to the optimal penalty and age-rating policy if we increased Medicare’s eligibility age and expected individuals aged 65-70 to buy insurance in the exchanges?

      • We currently do community rating through employer-provided insurance. Without it, it seems like no one wants comfortable middle-class people to be able to protect their homes and lifestyles from a catastrophic medical disease.

        “Covering the disabled through Medicare and Medicaid” presumes that someone who’s worked hard and has a comfortable income has to sell their house and become poor before they can get health insurance after they’ve acquired a horrendous risk factor, like, say, a child with a heart valve defect.

        Long term insurance contracts means what? The kid I knew was 13, health and active, but had had 2 or 3 open heart surgeries in his past, a few more in his future and 13 years before he aged off his parents’ policy (they worked for the government because of its great health insurance).

    • On your meta-issue, let me second what you are doing. It has been disappointing that so many on the right have been unable to deal with the ACA as passed legislation. Rather than participating to try to improve it in ways that they would find acceptable, they have just become oppositional. it is inevitable that those opposed to the ACA will eventually regain control of the political process. At that time, ACA supporters (in the broad sense) will have to deal with proposals from the right. Rather than also becoming oppositional, I strongly support an effort to look at what those plans might be, the AEI plan being one possibility, and try to find what could be supported in the plan and perhaps modified. I truly hope that prominent policy advocates who have supported the ACA will not take the course taken by those who opposed it.