Aetna withdraws from 500 counties in ACA marketplaces. Now what?

My take over at… Some tidbits below.

This Monday, Aetna announced that it will reduce the number of counties where it offers marketplace coverage from 778 down to 242. Aetna’s announcement sent shockwaves through the health policy world. The timing also raised eyebrows among ACA supporters wondering if this was deliberate payback for the Justice Department’s recent decision to oppose Aetna’s proposed $37 billion acquisition of Humana.

This April, Aetna’s CEO Mark Bertolini told investors that the marketplaces were “a good investment.” But then the Huffington Post‘s Jonathan Cohn and Jeffrey Young excavated a July 5letter to the Justice Department’s antitrust division, in which Bertolini wrote:

[If the Department of Justice blocks the proposed merger] it is very likely that we would need to leave the public exchange business entirely… . By contrast, if the deal proceeds without the diverted time and energy associated with litigation, we would explore how to devote a portion of the additional synergies (which are larger than we had planned for when announcing the deal) to supporting even more public exchange coverage over the next few years.

Nice ObamaCare marketplaces you have there, Mr. President.

The timing and tone of Aetna’s announcement seem dubious. But it’s depressing to see liberals and conservatives respond so predictably along the grooves of our own partisanship to a more complicated situation. If marketplaces were profitable, Aetna would not have done this, and we wouldn’t have cared if they did…

Marketplaces face real challenges that require real adjustment. So I offer several ideas about how to improve them here.

Author: Harold Pollack

Harold Pollack is Helen Ross Professor of Social Service Administration at the University of Chicago. He has served on three expert committees of the National Academies of Science. His recent research appears in such journals as Addiction, Journal of the American Medical Association, and American Journal of Public Health. He writes regularly on HIV prevention, crime and drug policy, health reform, and disability policy for American Prospect,, and other news outlets. His essay, "Lessons from an Emergency Room Nightmare" was selected for the collection The Best American Medical Writing, 2009. He recently participated, with zero critical acclaim, in the University of Chicago's annual Latke-Hamentaschen debate.

24 thoughts on “Aetna withdraws from 500 counties in ACA marketplaces. Now what?”

  1. Most of the changes you recommend require a sane, functioning GOP. (And it's not entirely clear that Aetna wouldn't have done this anyway, since some of the exchanges it abandoned did in fact produce significant pretax profit.)

  2. Basically what we're looking at here is the much derided "death spiral" in action. (Aetna pulled out of some markets where they were still making a profit, because they could see that they, too, were soon going to be money pits.)

    The ACA set up a welfare benefit, subsidized health insurance. In order to keep it off budget, health insurance companies were compelled to provide the benefit by regulation. In order to pay for the subsidies, the insurance companies had to charge other customers higher than actuarially justified rates.

    1. I think, for once, we may all be seeing the same, or at least a similar, version of reality. So, hurrah for that!!

      I would add though that the pre-existing huge subsidy program for the health insurers — Medicare, which pays for the most expensive customers, thereby allowing the insurers to appear to be covering large parts of the population when they really just cover employed younger people, which is why politically, they are allowed to exist — really needs to be part of your picture.

      At least though for once we seem to be talking about the same thing. My solution would be, let anyone buy into Medicare, and I guess yours would be, get rid of all programs except for Medicare? Though actually I don't know — maybe you are one of these let's deregulate, catastrophic coverage only (let poor people die) types. (Not trying to be hostile, just… that's what that plan seems like to me …) We sort of agree though, perhaps, that the ACA was in fact a huge bandaid on a cr*ppy system. I just don't feel sorry for insurers – I think they shouldn't exist.

      1. I would not call this a "subsidy", because, absent the ACA, insurers would be permitted to charge actuarially justified rates, so there wouldn't be any class of customer they'd lose money on.

        Following my analogy, it's like saying that food kitchens subsidize grocers and restaurants. Not remotely true, absent some legal imposition of a duty to feed people who can't pay.

        1. absent the ACA, insurers would be permitted to charge actuarially justified rates, so there wouldn't be any class of customer they'd lose money on.

          That's the problem, Brett, not the solution. It's nice to talk about "actuarially justified" rates, etc. Sounds very technical and so on. But it actually doesn't mean much, and to the extent it does, what it means is that some people – a fair number – won't be able to afford health insurance.

          Further, the fact is that most people don't pay "actuarially justified" rates and never did. Employer based insurance typically is priced per employee. It's the same whether a particular employee is an Olympic rower or a fifty-five year old diabetic.

          Besides, you can't talk about "actuarially justified" without talking about the time period over which the calculation is done. Once you start doing that, you quickly realize that the notion is not nearly so helpful as you think.

          1. Your last sentence is completely false. The ACA was not structured the way it was because liberals wanted to hide the cost. It was structured the way it was because centrists wouldn't accept any other system. I didn't want to hide the cost; I accepted what was on offer because there was no other way to get the 60th vote in the Senate. If you don't like the ACA as it is and think we should be paying higher taxes in order to have more of it paid for directly by the government, blame Joe Lieberman and Mark Prior.

            However, I don't think you actually believe that, which is why you are making the bogus arguments that you are.

          2. It makes sense to set a floor we won't let anyone drop beneath,

            OK. But that has to be paid for, and I'm not convinced you are really willing to do that. I am totally convinced that the GOP would like to lower that floor, almost to sea level, in the interest of cutting taxes on the wealthy. They've said so, in policy proposal after policy proposal. What's your take?

          3. JMN has dealt with the "hide the cost" fallacy pretty thoroughly. ACA is what the marginal Senators wanted, in preference to other possibilities.

            Let's go back to "actuarially reasonable." Who is this "identifiable group?" And over what period of time should the costs cover the expected benefits? The difficulty of answering that question is what I mean when I say the phrase is meaningless in and of itself. An "actuarially reasonable" premium for a 20-year, or 50-year, policy is quite different from one fro a one-year policy. And in the former case there is no reason for annual premiums to reflect annual expected cost. The premiums can be scheduled in a way that avoids them becoming unaffordable once the insured gets to be 60 years old or so.

            Please understand that this is a central problem with health insurance. If you let the insurer charge an actuarial rate that is determined when the insured turns 50 or 60 insurance is not going to be affordable for lots and lots of people. Ideally, there is a lifelong guarantee of affordable rates, and one way to do that is "overcharge" (not really, though it looks like it) younger people and undercharge (not really) older ones. It is absolutely stupid to have a system where rates are set annually, as they are for car or homeowners' insurance.

    2. I am not sure what you mean by "off budget". There are the ACA Premium Tax Credits (for households between 100% and 400% the federal poverty level), the Medicaid expansion, and Cost Sharing Reduction subsidies (for households on a silver plan and between 100% and 250% of the federal poverty level). About 85% of people insured through exchanges receive Premium Tax Credits; about 57% received Cost Sharing Reduction subsidies.

      The ACA does require insurers to offer standardized plans, largely regardless of individual risk and cost. But that's nothing new for businesses; Netflix charges the same regardless of whether you are binge-watching blockbusters or a cheap classic once a month. Every all-you-can-eat buffet has learned to handle the underlying actuarial principles. If that were all there is, Aetna could simply raise the rates enough to make a profit. Also, if you insist on calling that a subsidy, it's not actually insurers subsidizing the system; it's the healthy people "subsidizing" those who get ill. But that is normal, and happened to a large extent even before the ACA. It's what risk pooling is all about and calling it a subsidy is a misuse of terminology. The rates themselves are mostly only constrained by the market and a lower limit of 80% on the MLR.

      What we're more likely seeing is that – as in many newly established markets – the exchanges experience a very high degree of competition, with competitors operating at low or even negative profit margins in order to establish a foothold. Aetna is not raising the rates – or feels they can't do that, or at least not by enough – because their competitors aren't. Evidence for this is that rates in general are below what was predicted and plans offered on exchanges are often carry the marks of cost-slashing efforts such as narrow provider networks.

      Group plans for employers are a more attractive market; not only are at least major employers somewhat less price-sensitive (as the cost of health insurance is a relatively small part of total labor costs), but most employers that aren't called Walmart want be able to offer attractive health insurance plans during the hiring process in order to attract good employees. That makes for a much more favorable market for insurers than selling off-the-shelf insurance plans to customers who carefully consider every penny of their expenditure (remember that the vast majority are below 400% of the FPL).

      1. The ACA does require insurers to offer standardized plans, largely regardless of individual risk and cost. But that's nothing new for businesses;…. If that were all there is, Aetna could simply raise the rates enough to make a profit. Also, if you insist on calling that a subsidy, it's not actually insurers subsidizing the system; it's the healthy people "subsidizing" those who get ill. But that is normal, and happened to a large extent even before the ACA.

        Even more to the point, employer-based plans generally use the same system. The company premium is a fixed amount per employee, regardless of age or health status. So younger healthier workers are "subsidizing" older sicker ones in those systems, and have been for many years.

        1. Prior to the ACA, if you hired on at a company, already suffering from an illness, if there had been any interruption in your insurance, that particular illness would not be covered. Get laid off from your job, can't afford COBRA, find you have diabetes, find a new job? Your new insurance won't cover the diabetes, it's a "pre-existing condition". Other conditions not related? You're covered.

          The ACA outlawed that. You get cancer while not ensured, you sign up for insurance, it has to cover the cancer and not factor it into the cost of the insurance. A really great deal, it's like buying a ruined house, and getting homeowners insurance that covers bringing it up to spec for the same price as the guy next door whose home is in good shape.

          Good deal for you, guaranteed loss for the insurance company. At least until the next year, when it's a guaranteed loss for all your co-workers.

          Now, don't get me wrong. This genuinely is a problem for the person with the pre-existing condition. But, at the same time, the insurance companies aren't public charities. They're not in the business of helping people who had something bad happen to them when they weren't their customer. That's not insurance. That's welfare.

          We need a solution for this that doesn't involve legally mandating that private companies implement welfare programs for the government, and then legally mandate that everybody buy their products to keep the private companies from going under.

          I think this has to be a combination of explicit welfare, and reforms to the insurance market. The market for health insurance is terribly warped by the way insurance has been linked to employment. (This being a consequence of WWII era wage controls.) We need to break that link. And, you'll doubtless think I'm insane, Trump actually has some sensible proposals in that area.

          1. 1. That does not generally accord with my experience. No new employee at any company I have been involved in was ever required to take a physical or fill out any sort of form describing pre-existing conditions. I have had employees facing a problem who actually quit and went to work elsewhere precisely because the new employer had more generous health benefits. Insurers usually assume, I believe, that the employees represented a more or less random group, healthy and unhealthy, etc.

            2. What difference would continuous coverage make? If someone is sick, and is going to have claims, then they are going to have claims. I have a friend who has been diabetic for decades. He has switched jobs a few times with no problem.

            3. Even your argument does not address the fact that employees, healthy or not, have varying probabilities of getting sick. Some are old, some young, some fat, some thin, some couch potatoes, some exercise enthusiasts, etc.

            4. If your employer insurance is "experience-rated," and not all are, then yes, the rates will rise somewhat. Of course they will also rise if an already employed worker has a large claim. Very large claims are often reinsured in any case, so that the risks are spread much more widely than over the individual company.

            5. The "pre-existing condition" problem arose mostly with individual policies, where even the slightest health problem led to difficulties getting insurance.

            6. This is precisely why health insurance needs to be a long-term, preferably lifelong, arrangement. Sooner or later everyone gets a pre-existing condition. If that causes you to be unable to get coverage then the system is simply nonfunctional – worthless.

          2. The link shows what was allowed, not what actual practice was. Even there, in most cases,there were limits on the duration of the exclusion and the look-back period.

            I understand the adverse selection issue, but its application in an employment situation, as opposed to buying an individual policy, is pretty tenuous, which is probably why lots of companies ignored it. Another factor was that benefits play a role in deciding to take a job, and you don't want to lose someone you'd like to hire over this.

            I know of no proposal by Trump that addresses the issue. Giving individuals tax breaks does nothing towards that end.

            Large risk pools based on the kind of groups you list used to exist. They suffered from major adverse selection problems and more or less disappeared. Need insurance? Hey, just join the NRA. Unions, of course, had their own arrangements.

          3. Of course it would make it more affordable, Brett.

            That's not the point I was addressing. I was talking about the need for a long-term arrangement that guarantees the availability of affordable health insurance even when you need it. The tax deduction does not help much when the premium equals your annual income. Besides, only about 30% of taxpayers even itemize, and many of them have employer-based insurance. In addition, self-employed taxpayers can already deduct the premiums. So no big deal there, even if the deduction were to get a few more people into itemization territory.

            What is there in Trump's plan that assures people that they will have affordable insurance available at, say, age 50, even if they develop serious health problems before that? Nothing. I just read it. It's a lot of "free-market," babble.

            Affinity insurance? Do you think adverse selection didn't hurt it? Do you think employer-paid insurance wouldn't have hurt it, even without the tax treatment? (Yes, I understand about how it's compensation, etc.)

          4. You are amazing Brett.

            If they gave a gold medal for rationalization you'd win in a walk.

            Not a panacea? It solves almost nothing, creates further problems, and would be far worse than ACA.


  3. Since the natural response to unreasonably high insurance rates is to self-insure, the mandate was added to force people to buy insurance even if it was over-priced. However, it was politically infeasible to set the fine for not buying insurance high enough that a substantial number of people wouldn't chose to pay the fine rather than purchase the insurance.

    So you get the "death spiral": Really sick people, for whom the subsidized insurance is a great deal, sign up. Healthy people who were supposed to be on the losing end of this scheme drop their insurance. The more who drop it, the higher the rates have to go to extract enough excess from those who remain to pay for the subsidies. And the more people decide to pay the fine rather than buy the ever more expensive insurance.

    All this was perfectly predictable from the start, and was predicted from the start.

    1. Except, of course, this is not what's happening. Premiums are still far below what was predicted when the ACA was passed. They would need to go up by about 25% to equal those projections. If the premiums were where the law's drafters assumed they would be, it would be a different story, as they would be profitable. My suspicion is that premiums are going to climb into the range of what was predicted, and you'll see insurers come back into the market once they do.

      1. Did the advocates of the ACA predict any of this? Premiums going up? Exchanges collapsing? Websites not working? That, no, if you liked your insurance you still wouldn't be able to keep it?

        And this is with major components of the ACA having their implementation delayed for years.

        No, I'd have to say the program's critics were more accurate in their predictions.

        1. I'll try again. The advocates of the ACA, at least those involved in making the projections, argued that premiums would be higher now than they actually are. Increases that are, or will be, greater than they projected at the time is entirely an artifact of increases being lower between when the law was enacted and now. They absolutely are not because premiums are higher than they predicted. So, screaming about what premiums are doing now involves totally ignoring the past and pretending that it never happened.

          To the extent that exchanges are collapsing, and that's probably an exaggeration, it is because insurance companies priced their policies too low to make money. In this case, it appears that the policy wonks who designed the law and projected its effects had a better idea as to what insurance companies should charge than the insurance companies did, so this isn't much of an argument in favor of a more market oriented approach.

          My prediction now is that, as premiums rise, you'll see insurers re-enter the exchanges and that, after some short term pain, they'll be fine. Whether the premium increases necessary to reach that point constitute a failure for the ACA depends entirely upon where they settle relative to the projections made by its advocates; unless they end up higher, this does not constitute a failure on their part.

          As to whether or not I ought to be able to keep the insurance I had before, that's a nonsensical question, because I was not able to get a health insurance policy prior to the ACA. Now, I have one that I can afford. So, pardon me if I react to your complaints with contempt.

          As for whether or not we should put the subsidies entirely on budget, I have no idea what, specifically, you mean. I can't tell whether or not it would be a good idea without some specifics. However, I'll reiterate my claim that you are arguing dishonestly, because as soon as we agreed that your plan sounds good, you'd immediately object to paying the taxes needed to implement it. You don't really want the government making sure that everyone can have health coverage at all.

  4. I gather your proposal is to increase the penalty for not buying insurance, and provide subsidies on budget for buying it, to interrupt the spiral.

    Increasing the penalty is still politically infeasible. And if you're going to subsidize insurance on budget, why bother with the ACA, whose only purpose was to hide the subsidy?

    Would it not be simpler to just admit that the lower than actuarially justified rates themselves are subsidies, put them on budget where they should have been in the first place, and restore something vaguely resembling a normal market with rational prices? Rather than doubling down on the original intervention?

    I suspect the idea that the ACA can't be repealed at this point is mostly wishful thinking. All it would take is a President who wouldn't veto the repeal, and it would be gone.

    It's a welfare program. Just admit it, and put it on budget. We don't force grocers to sell food below cost to the poor, and then fine people who don't shop at them. We have food stamps. Why not treat insurance the same way?

  5. Perhaps before you are too critical of the prediction abilities of ACA supporters you could read, by following the links, about Rubio's efforts to kill the corridors.

    Maybe direct subsidies are a better way to go, though I doubt you're going to get much support from the GOP on that, but the problem was not entirely unanticipated. As usual, Republican strategy is to proclaim loudly that the government can't do anything, while working hard to make that come true.

  6. I recall some interesting speculation on NPR awhile back that perhaps the individual ACA marketplaces might come to be dominated by smaller carriers who are better able to exploit the market segment than are the huge firms. This does not strike me as a bad thing, if it is in fact plausible.

  7. An ill wind. The implosion of the Trump campaign has opened up the possibility (no more) of Democratic control of both houses of Congress, for the two years before the probable return of the Senate to GOP control in 2018. That would open up far more options for ACA reform than previously thought. The trouble is the narrowness of the legislative window. There will be no time for a thorough rethink. The only feasible ideas are ones already on the table: Medicare buy-in and the public option. The Swiss solution – making the basic health insurance package nonprofit – is probably too outré.

    Brett: The emerging Democratic rainbow supermajority no longer treats the word socialism as a taboo, and is perfectly happy with more federal intervention and redistribution in the US health care system to make it more like those of other OECD countries. Kamala Harris is going to be closer to Elizabeth Warren than to Ben Nelson. The RBC bloggers are probably now to the right of the centre of gravity of the Democratic party. You should be more worried about the survival of Aetna than of ACA.

  8. Harold, you said: " If marketplaces were profitable, Aetna would not have done this." Not true. Aetna is facing the loss of a billion dollars as a breakup fee if its proposed merger is blocked by the Justice Department. So, they withdraw from the market this year (they were losing about $450 million, but, based on their April statements, believed that they could turn that around). If the Justice Dept. caves, they've saved $1 billion but lost only the possible income for a year (or avoided an operating loss)/ After all, they can always go back into the market next year.

Comments are closed.