For years, Republican politicians assailed the Affordable Care Act for causing health-insurance premiums to rise, out-of-pocket costs to soar, and insurers to drop out of the marketplace altogether. Yet the congressional GOP has failed to enact legislation to supersede it, despite enjoying majorities in both houses and a Republican president. Nonetheless, incremental reforms have been made. President Trump issued an executive order to make available cheaper insurance plans outside the framework of the ACA’s regulations, and Congress has repealed the individual-mandate penalty, to which individuals purchasing such non-ACA-compliant plans would otherwise have been subject.
Congressional Democrats suggested these reforms would “undermine consumer protections, disrupt the individual health insurance Marketplaces, and expose consumers to great financial risk,” while the Washington Post attacked the new options as “cheap, substandard plans” that portended a “disastrous escalation of costs.”
Mark V. Pauly is a professor of health-care management at the Wharton School of the University of Pennsylvania and one of America’s leading experts on health-insurance markets. Over the past five decades, the publications he has written and the students he has trained have done much to define the field. In a 2010 monograph, “Health Reform without Side Effects,” Pauly discussed the Affordable Care Act’s insurance-market reforms — anticipating many of the difficulties it has faced, and suggesting more appropriate means of fixing problems with insurance markets.
I called Pauly to learn what he thought of the developments of recent years, and to get his assessment of the Trump administration’s recent initiatives. The conversation has been lightly edited for clarity.
Pope: The Affordable Care Act was 974 pages long, and made changes to every element of the health-care system. Which elements of it do you feel have been most successful, most harmful, and most overrated either way?
Pauly: Most successful, I think, has been covering about half of the formerly uninsured. That, presumably, was a primary reason it got passed in the first place. The Achilles’ heel, I think, has been its treatment of risk variation in the individual insurance market. I think that the end state of the exchange will probably be a pool of mostly low-income, high-risk people.
As I have been saying for years, the inclusion of community rating [meaning insurers cannot take individuals’ health status into account when setting premiums] has caused all sorts of unnecessary mischief, which has resulted in both political blowback and difficulty of market functioning in the exchange.
It pains me to say this somewhat, but I think the individual mandate wasn’t that big of a deal one way or another. I still think that it’s wiser to have one than not, but not that individual mandate. An individual mandate would be good, but not the one that was in the law. A lot more has been made about that than is probably necessary, given that the heaviest burden of encouraging people to buy insurance was borne by the heavy subsidies to people regardless of risk level — and to some extent that probably offset the need for the individual mandate for low-risk, low-income people.
Pope: Why did individuals with pre-existing conditions have trouble getting health insurance before the enactment of the Affordable Care Act, and to what extent was this a serious problem?
Pauly: It was a problem. There is no doubt that some people did have this trouble. This sounds a little hard-hearted, but I’d repeat what my father always told me whenever I came to him with a problem: “Well, you should have thought of that beforehand!” In most cases, a person had a problem if they didn’t have health insurance, and then tried to get insurance when they had become a high risk.
The actual number of high-risks who are also in the individual market, who didn’t have group insurance already, and so would have to seek coverage in the individual market was a quite small fraction of the total U.S. population.
The most typical way that people (then and now) got insurance was through their job — and as long as you can hang on to a job, whether you become high-risk or low-risk is not going to affect you. It wasn’t really that much of a threat to people who could remain in the employment-based system. The main problem was for people who (for whatever reason) had been in the employment-based system and had dropped out of it. That is, people who, more or less simultaneously with giving up their job-based coverage, had something happen that made them a high-risk — they were diagnosed with cancer, or had a heart attack, or something like that.
There are 330 million people in this country, so there were probably millions who were in that situation of having given up that security blanket of job-based coverage in the private sector, and then being hit by the event of being a high risk — but it’s a tiny fraction. There were always those who were born with high-risk conditions (birth defects and such), but for the bulk of them, they would have been picked up by disability provisions of existing public programs.
There’s always a big debate over “how many high-risk people are there?” in the sense of people who would have to pay a much higher than average premium. I concluded that it was certainly less than 10 percent of the population, and probably down to 1 or 2 percent of the population. But those are certainly miserable cases — certainly if you happen to be one of them or know people in that situation.
The main thing to note is that this isn’t some kind of situation where everybody in the individual market is above average in terms of risk. The actual number of high-risks who are also in the individual market, who didn’t have group insurance already, and so would have to seek coverage in the individual market was a quite small fraction of the total U.S. population. This was why it didn’t seem sensible to have to torque around the whole health-care system — it’s a serious problem, but one that affects a relatively small fraction of the population.
Pope: In “Health Reform without Side Effects,” you suggested that community rating was “the worst possible way to do a good thing.” Why was that?
Pauly: Well, the good thing was that most people care about their fellow human beings not only having health insurance, but also not receiving too large a reduction in the rest of their disposable income to pay for it. That means we’d like to help out financially the high-risks, especially those with moderate incomes — if Bill Gates gets diabetes I’m not going to worry about it.
The basic economics here has been known for years. The best way to do that is to provide a subsidy to those high-risk people to offset their higher-than-average medical expenses, so that the net premium they would pay would be regarded as socially acceptable or affordable. That’s the straightforward way, and then you fund that subsidy — presumably with the fairest and most efficient tax base you can find.
But, instead, some had the idea of making insurers charge low premiums to high-risk enrollees. Some people think that means the insurers or stockholders will pay for it, but I think most people actually come to the view that it will be the low-risks who have to pay for it through higher premiums.
That has two downsides: One is that it doesn’t seem fair that the cost of paying for this would be put on the low-risks. The other is that when you do put the costs of paying for it on the low-risks, you encourage them to go without coverage and need some other subsidy for them — which creates an inefficiency. You’re effectively collecting money for the subsidies with what is effectively an excise tax on insurance bought by low-risks, and we generally think that excise taxes are not a good idea because they cause distortions in behavior.
There’s a much more straightforward way to do it, which is to use something like the income tax to adequately fund a high-risk pool. Of course, you still have the political decisions of who gets what kind of subsidy in that pool and what constitutes a high enough risk. Although events these days conspire to make me believe otherwise, I still think you still might have a reasonable political decision about those issues.
Pope: Could community rating be made to work with a stronger mandate, better risk adjustment, and more thorough regulation of essential benefits?
Pauly: I suppose it could if you had the political will and administrative expertise to plug all the holes, but it’s one of these things that where you plug one hole the leak comes somewhere else. By effectively making all insurance premiums compulsory (which some countries do), or with an absolutely binding individual mandate, it could work. But effectively raising a subsidy from a head tax on at least some portion of the population — the low-risk people who aren’t poor — usually we don’t think that’s the best kind of tax. There should be some relationship of the contribution to income.
Pope: What value does competition in insurance markets bring if the government is ultimately going to bear much of the risk associated with those who have pre-existing conditions?
Pauly: Probably not very much in terms of pure insurance function. The pure insurance function isn’t that hard or complicated to perform anyway. Competition can probably still be helpful in terms of innovations of product design or compensating providers or organizing networks.
These tends to be stifled, because whenever you permit an innovation in product design, there’s always a really good chance you open a loophole for risk-selection. This means that you have another loophole you have to plug, to make sure that the new product design (let’s say high-deductible health plans or reference pricing) doesn’t appeal more to low-risks rather than high-risks. With perfect risk adjustment [meaning a system of taxes and subsidies to transfer funds between insurers to perfectly offset the variation of expected health-care costs associated with covering enrollees with differing medical risks] you could solve that problem, but that perfection is not to be expected on this earth.
To put it the other way around, I don’t think there’s that much possibility of reaping enormous monopoly profits off insurance in the private sector anyway, unless some firm was given a de facto monopoly. Competition could stimulate innovation — which, heaven knows, we need. But those attempts get stifled by regulation, as you could spin out a story about how almost any innovation will advantage some risks more than others.
Pope: Short-term health-insurance plans are exempt from the ACA’s community-rating requirements. The Trump administration recently sought feedback on a proposed rule that considered repealing a ban on insurers’ guaranteeing their renewability for individuals who subsequently developed serious illnesses. What is the value of guaranteed renewability in health-insurance markets?
If your primary goal is to make sure that people have at least some insurance, no matter what, then it seems to me that the ideal arrangement would be to let them choose what policies they like.
Pauly: Well, it actually solves the problem of dealing with people who become high risks. Effectively, if you buy a plan with guaranteed renewability this year, you’re effectively buying two insurance policies. One says, “If something terrible happens to you this year and you have high medical expenses, in return for a fixed-dollar premium we’ll cover that bad luck.” And then, if you should contract a condition that means that you will have high premiums for the rest of your life, or some number of years in the future, you have bought another insurance policy that says, “We’re going to make payments to offset those high premiums.” It’s called reclassification risk, and is another kind of risk. It might not be the most important thing for most people, but it’s probably something you’d want to be protected against at the extreme. This is a way people can do it without there having to be regulation.
There can be more or less creative ways to protect people from reclassification risk if the industry was allowed to explore them. Maybe the insurer that’s offering you guaranteed-renewable coverage can come up with some kind of wellness program that could reduce future risk. That would be good for them, but right now the incentive to do that is attenuated if any individual insurer thinks that people who they might save from chronic conditions might move on to another insurer.
It does have the downside, in the simplest version, that you have to stay with your insurer. The Germans say that “you’re married to your insurer” — though I think there would be ways to allow an amicable divorce, if people really thought that was an important feature.
Pope: A former Obama-administration official has called short-term plans “junk insurance” and a “license to steal,” which would cause quality insurance to become gradually less available. Should insurers be allowed to sell plans that are exempt from the ACA’s benefit and cost-sharing standards?
Pauly: If your primary goal is to make sure that people have at least some insurance, no matter what, then it seems to me that the ideal arrangement would be to let them choose what policies they like. It’s a license to steal, of course, if people can be duped into thinking that insurance which actually has very skimpy coverage provides more coverage. But that’s a product-quality issue, for which government regulation could help to set standards.
The simplest way to describe an insurance policy is to describe it in terms of the “loading” or “medical-expense ratio” — how much of the money that you put in do you get back. As long as you’re getting back a good share of the money that you put in, there’s no stealing going on.
It is true that if you allow policies with more limited coverage they’ll tend to be more attractive to low-risks and will cause a community-rated pool to unravel. That wouldn’t happen if premiums were risk-rated, of course. As I just said, any innovations of coverage in the presence of community rating could cause problems. But I don’t see them as a license to steal. There are some that I wouldn’t advise people to buy — insurance whose maximum benefits tops out at $1,000.
My political strategy would be to be very light-handed in terms of regulating what coverage is offered, and then if it turns out that a lot of people buy coverage that they end up finding out wasn’t very good for them you can fix that later — but the first thing is to get them to take some coverage. Fix it up later, rather than trying to design the perfect coverage. That doesn’t exist for even a single individual — and certainly isn’t uniform for all individuals, especially given the varying circumstances Americans find themselves in.
Pope: To avoid this problem of a deteriorating risk pool causing premiums on the exchange to spiral upwards, should a mandate be used to push people out of the short-term plans and back onto the exchange?
Pauly: If you stick with community rating, you end up putting the kibosh on all sorts of things that otherwise might be reasonable innovations. On the other hand, if the high-risks were covered in a high-risk pool, then a plan with less generous coverage probably would be more attractive to healthier people who expect to be healthier and have less expense. That would be fine — they would be paying what their coverage costs.
It’s only if you want to force them to pay more than their coverage costs in order to lower the premiums for the people who are paying less than their coverage costs that you would want to prevent people from buying that kind of coverage.
Pope: The individual market is tiny (17 million enrollees in 2017) compared with employer-sponsored insurance (156 million). How should we think about how these two markets interact?
Pauly: The individual market seems to me to almost be bifurcated.
There is some fraction of the population, such as the self-employed people, who aren’t going to be employees and members of large groups, and they deserve to have an opportunity to obtain health insurance. Inevitably, the administrative costs to service people one-on-one are going to be higher than if it was sold to a group — but that’s the quid pro quo for being your own boss.
So far my view is that we know an awful lot about the things that don’t work.
The other part is people who move temporarily out of group coverage into other life situations and then back into group coverage. That, in some ways, is what the short-term policies are already set up for. I’m not personally that concerned about that population. If you want to focus on the individual market, the desire would be to make it work well for people who are going to stay in it for a long period of time because that’s their employment situation.
Pope: The best-possible insurance market is unlikely to make up for a poorly structured delivery system. Do any of the initiatives the Obama and Trump administrations have proposed (price transparency, accountable care organizations, site-neutral payment, etc.) offer much hope of improving care quality while reducing costs?
Pauly: No. That’s impossible.
It’s not possible to reduce costs, first of all. You can slow the rate of growth of costs. In terms of aggregate impact, the only way to slow the growth of spending is to give up something. This might be called quality, but it would probably be things like convenience, the ability to go to the closest provider, or access to the newest technology as soon as it becomes available.
Nobody wants to sit next to economists at dinner parties because we say things like this. Sure, it’s going to hurt the quality of care a little bit, compared to what it otherwise would have been — but it’s going to save you so much more money that you’re better off with the lower spending; the quality wasn’t worth it.
I just published a paper with my colleague Rob Burns in the Millbank Quarterly, which is essentially a wet-blanket paper on transformation of the delivery system, saying that so far pretty much nothing has worked.
The only things that have a little bit of hope are bundled-payment systems [which consolidate reimbursement for a course of treatment, rather than paying for each service used] sometimes– although more recent studies show that perhaps it’s not so great. Maybe reference pricing [whereby insurers reimburse a flat amount for specific procedures, encouraging patients to shop around to avoid bearing additional cost] might work — the immediate response to it seems to work, but if you happen to have a condition where you really do need the expensive source of care because of convenience or because it is a little bit better for you, you will be on the hook for a higher percentage of the incremental cost. So it’s not clear it’s the optimal insurance coverage for everybody for everything, but it’s worth a shot.
So far my view is that we know an awful lot about the things that don’t work. But in terms of coming up with something that will reduce cost and improve quality — anyone who says “I have the answer to that,” I regard as a fraud.