The American Health Care Act has taken a beating since Speaker Paul Ryan and House Republicans introduced it in March. An early Congressional Budget Office estimate found that 24 million more Americans would be uninsured under the GOP’s bill. Medicaid spending would be slashed by more than $800 billion.
The reaction among the conservative intelligentsia to the original Republican bill released in early March was harsh. “GOP's Obamacare Replacement Will Make Coverage Unaffordable For Millions — Otherwise, It's Great,” Avik Roy wrote for Forbes.
But those stances have softened in the past two months, as the bill has changed. Many conservatives have come to see the AHCA as a vessel, albeit a still imperfect one, for the kind of health reforms they have waited for years to achieve. Now that it has passed the House, they are one step closer to that goal.
So I spoke by phone on Friday with Roy about how he thinks the bill would improve on current policy under Obamacare — and what still needs to be addressed as the Senate takes it up.
The transcript has been edited for clarity and length and to clean up one moment where Roy ordered tableside guacamole.
Dylan Scott
I don’t have to tell you that the bill, since it was first released, has taken a lot of hits from a lot of different angles, whether it’s the Medicaid cuts or the projected coverage losses. I’m curious, looking at it from the way you approach health policy philosophically, what pieces of this bill do you think are good things and could be an improvement over the system that we have right now under Obamacare?
Avik Roy
The first is the regulatory reform in the individual market. What the bill does to reform the way in which insurance is bought and sold in the individual market — so for people who don’t get insurance through their employer, who don’t get insurance through the government, Medicare and Medicaid, etc.
For that slice of the market, people who shop for insurance on their own, which is a big part of what Obamacare tried to change, the regulations that Obamacare imposed at the federal level on the individual insurance market have driven up premiums considerably, doubled or tripled in many cases. While some people who qualify for subsidies, there are a lot of people who are priced out of the market because their premiums have increased so much. That’s because the bill basically overcharges healthy people and younger people in order to partially offset the cost of covering sicker people.
The end result is if you reform that, if you’re able to attract more younger and healthier people into the insurance pool in the individual market, you should bring premiums down for everybody and you should actually be able to cover more people. That part of the bill is pretty good. It repeals and reforms the right regulations in order to ensure younger and healthier people aren’t priced out of the market.
Dylan Scott
And when you say the right regulations...
Avik Roy
There are three provisions of the ACA that have been most responsible for premium increases. The first is the fact that the bill allows to charge younger people only one-third of what they charge older people. In the real world, older people — say, 64-year-olds — consume about six times as much health care in dollar value [as] 18-year-olds. That’s been a big driver of why so many people are priced out of the Obamacare insurance market. The bill does fix that, moving the ratio back to five to one.
Piece No. 2 is actuarial value mandates. The actuarial value of an insurance plan is basically the proportion of the plan that is paid out by the insurance company versus what is paid out by the policyholder in the form of copays, deductibles, and the like. [In] the old individual market, prior to Obamacare, the typical actuarial value of a plan was about 40 percent. Obamacare drives that up effectively to 70 percent. That has a corresponding effect on premiums; it makes premiums a lot more expensive. In the AHCA, those actuarial value mandates are repealed. Which should provide a lot more opportunity for plans to design more affordable insurance policies for individuals.
Piece No. 3 is essential health benefits. That’s not to say we should go back to a system where there are no definitions of what health insurance should cover. But there are specific pieces where instead of putting them directly into health insurance, we should find alternative mechanisms of financing coverage of those particular services for people who need them.
For example, today we don’t have vision and dental services covered in health insurance. Those are covered either out of pocket or through separate insurance coverage. Similarly, we could do something like that for pregnancy coverage or maternity coverage. Maternity coverage is something that a lot of actuaries tell me has done a lot to drive people out of the Obamacare markets because it creates this big bulb of costs for women of childbearing age that, due to other Obamacare regulations, ends up also being borne by people who are women not of childbearing age and men of childbearing age.
That’s not to say we shouldn’t try to make sure that everybody has maternity coverage. But we should probably do so through a rider or through a separate policy that can be subsidized separately for that population rather than putting that into the entire individual market. The fact that states now have the flexibility to seek waivers from the federal Obamacare essential health benefits mandate allows you that flexibility.
It’s important finally to then talk about the pieces of Obamacare’s insurance regulations AHCA does not change. Most importantly, it does not change guaranteed issue, the requirement that insurers offer coverage to people regardless of their health status. The bill keeps the regulation requiring insurers to charge men and women the same rates, regardless of gender. In my view, the regulatory reform in AHCA strike the right balance of still providing important consumer protections while also doing things that will lower premiums.
Dylan Scott
So on the question of the waivers, obviously that’s been a big part of the discussion and you referred to the ability of states under the bill, with some strings attached, to waive the essential health benefits mandate as well as the requirement that insurers not charge anybody more or less based on their health status. Those, along with guaranteed issue, were the core consumer protections in Obamacare.
I’m interested in this idea of siphoning off some of these expensive services that not a lot of people will use but are important to the people that do need them, like maternity coverage. Is there actually a mechanism in this bill to do that?
Avik Roy
A couple points. It’s very important to remind your readers that every state in America has essential benefits mandates and had them before Obamacare was ever passed. So it’s not like if you were to repeal Obamacare’s benefit mandates, you’d go back to some Wild West. Many states have benefit mandates that are far more extensive than Obamacare’s.
The AHCA does not repeal benefits mandates. It simply allows states to seek waivers from Obamacare’s mandates, so long as that waiver serves the purpose of covering more people, lowering premiums, or expanding access to care. This is a pretty modest and incremental reform, and there’s been a lot of misinformed and sometimes disinformed commentary about this particular provision.
I think there are other opportunities to provide funding for people who need those services that you might want to shave off of essential health benefits. That’s why the AHCA has more than $100 billion in funding for state innovation grants, high-risk pools, and the like: precisely so if you don’t have those kinds of services in your essential health benefits, you still have the money to finance coverage of those particular conditions. That’s explicitly what that money is for.
The problem with the AHCA is not how it handles coverage for sick people. The AHCA has robust financing and robust policies to assist people in affordable coverage if they have poor health status. The part of the bill that’s not strong, that needs to be improved, is how you provide financial assistance for poor people.
Dylan Scott
The ACA has income- and geography-based tax credits for people to afford their insurance. The AHCA, as currently written, envisions transitioning to an age-based system.
What do you think the motivation was for the AHCA’s tax credits as they’re currently designed? Because it seems like a very foreseeable problem that this would end up not providing enough assistance for the lowest-income people. So why set it up this way in the first place?
Avik Roy
It really comes down to a long-held view of Paul Ryan, the House speaker, that a means-tested tax credit creates a disincentive to work, because as you make more money, as your income goes up, the amount of assistance you get goes down.
What he fails to consider is that by having a flat tax credit that isn’t adjusted by income, he actually makes the problem of work disincentives far worse than it is under Obamacare. Because what happens is, particularly for people who are in their 50s and 60s, if you make enough money to cross the poverty line and get off Medicaid and into the individual market, the health insurance subsidy that you get under Ryancare is $3,500 or $4,000 and your underlying premium might be $10,000 or $12,000. You go from having basically no premium under Medicaid to having a $9,000 premium in the individual market. That’s just unaffordable. You’re basically telling people whose incomes are just below the poverty line, “Don’t work that extra hour, don’t get that second job, don’t get a raise.” Because if you do, you’re gonna be stuck with $9,000 in premium costs.
There’s an additional rationale that you’ll hear from advocates of a fixed tax credit, that by making the dollar value fixed, you’re incentivizing the health care system to deliver an insurance product that is equivalent to the value of the tax credit. If you’re getting a $3,000 tax credit, the insurance companies have an incentive to create an insurance plan that costs $3,000.
That objection I take more seriously, but I think the right structure of a reformed tax credit can address those concerns while still covering the poor.
Dylan Scott
From what early talks we’ve heard from the Senate, we’re going to see some kind of means testing reintroduced. Is there a way to do it that address the concerns you just laid out? Or is it more important to provide the lowest-income people, the most vulnerable people, more financial support to make sure they can get insurance?
Avik Roy
It is still important exactly how that means-tested tax credit is designed. If you simply have a fixed tax credit of $3,000 and maybe you make that $6,000 for someone whose income is closer to the poverty line, that sounds like an improvement, and it is an improvement relative to the AHCA, but it still won’t work for a lot of people because there’s a lot of variation in health care costs under the country. If you have a fixed tax credit that doesn’t change, you’re going to price a lot of people out of the market.
Dylan Scott
So summarizing the private insurance side of this, in your view: The AHCA repeals some of the right Obamacare regulations to bring premiums down, it provides enough funding to address any rollbacks of Obamacare’s protections, but the financial assistance piece still needs work.
There’s been a lot of contention about whether the bill really has enough money to do what people supporting the bill say it will. Then you have the number that I think shocked a lot of people: The Congressional Budget Office estimated that 24 million more people would be uninsured.
Let’s come back to those bottom lines: Is it really enough funding, and what about the coverage losses?
Avik Roy
Josh Archambault, who writes for me at Forbes, has a good piece that looks at the Maine experience, where they have this invisible risk pool, and looks at the funding they needed to support that. It has been a very successful model. If you extrapolate that to the country as a whole, the amount of money that the Republican bill provides to cover people with poor health status, they actually overshoot it.
I find that analysis plausible, and the reason why is if you look at Obamacare, the mechanisms in Obamacare’s exchanges that served as a way to fund coverage for sick people, they were spending $8 billion a year on that program. If you look at it that way, if $8 billion was enough under Obamacare, then maybe $15 billion a year is enough. I really don’t think that’s the problem with this bill.
Dylan Scott
The coverage loss numbers, though, is what I think stuck with a lot of people. It comes from CBO, the official scorekeepers. 24 million uninsured people sounds like a lot.
I can maybe understand the philosophical rationale that that’s a worthwhile trade-off if some of these people are choosing to forgo insurance. But it seems like with that many people becoming uninsured, there would be people who want to be insured and can’t get coverage.
Avik Roy
When the first CBO score came out, I wrote a lengthy write-up where I went through it line by line. Where I came out is that the CBO estimate is significantly off. The CBO had projected that the difference between AHCA and current law in 2026 would be 24 million more people being uninsured. I put the number at more like 5 million. You might say, “Oh boy, Avik is being pretty cocky to disagree so strongly with CBO.” But my estimates are, I think, grounded in the empirical data.
Dylan Scott
In brief and for the layman, how did you get to where you got to?
Avik Roy
There are a couple of key elements where I disagree with the CBO. CBO used an outdated baseline with rosy estimates of future Obamacare Medicaid and exchange enrollment relative to what we now know to be the case.
The other big piece where I disagree with CBO, and I’m not the only one: The CBO way overstates the power of Obamacare’s individual mandate to drive people to buy health insurance. CBO believes that if you repeal the individual mandate, that alone would lead to 18 million fewer people having health insurance. Which is a very, very large number, and it’s unlikely to be that high in reality.
But your broader question was: Is that acceptable? It’s still going in the wrong direction, that’s still not a good result, that’s a lot of people that will not have health insurance under the Republican bill. That’s what I think needs to be fixed by the Senate. I don’t think they’re ever going to get to the point where CBO scores their bill as covering the same number of people as Obamacare, because the CBO places such magical powers in the individual mandate. However, they can get in the real world to where the number of people who have health insurance under their plan is comparable or even higher than Obamacare.
Dylan Scott
I want to ask you about Medicaid. I’ll actually set aside Obamacare’s Medicaid expansion and the way the bill deals with that because it at least makes logical sense to me [that] if the goal of your bill is to repeal and replace Obamacare, then you have to address Medicaid expansion, because that was one of the primary tools Obamacare used to expand health coverage.
But the bill goes beyond that. It reimagines the entire funding structure of the Medicaid program, for everyone, not just the people who were covered by the expanded Medicaid in Obamacare.
There’s a good argument to be made that the bill didn’t have to do that. So why overhaul Medicaid, the entire program, as part of this bill? Why have we merged that with Obamacare repeal and replace in the AHCA?
Avik Roy
I was one of the earliest advocates of this approach, of saying that we should include broader Medicaid reform in the Obamacare replacement bill. Because it’s something that is very straightforward to do through the Senate reconciliation process. Broadly speaking, Republicans are constrained in what they can do to repeal and replace Obamacare through reconciliation, but they have the opportunity to do a fair amount to reform the Medicaid program.
What Republicans have talked about for a long time with health reform is repeal and replace Obamacare, reform Medicaid, reform Medicare, and ideally reform the unlimited tax break for employer-based health coverage. Medicare is out; the president is opposed to Medicare reform. Medicaid reform was not, and I think it’s very much to the House’s credit that they decided to include Medicaid reform in this bill, because Medicaid reform is very important, fiscally and from a coverage standpoint.
Dylan Scott
So is it fair to say that Medicaid reform has been a longtime goal, [that] this was a moving vehicle and it works under the procedural rules they’re using? That’s the rationale for including it? There’s not something that’s necessarily inherent to repealing and replacing Obamacare that requires Medicaid reform. But looking at it holistically, from a conservative health reform standpoint, then it does?
Avik Roy
It’s important to note that Paul Ryan’s A Better Way proposal articulated, “Let’s repeal and replace Obamacare, let’s reform Medicaid through per capita caps, and let’s reform Medicare through premium support.” That these items were part of the Republican health reform agenda should not be surprising to anybody who’s been following what Republicans have said about health reform for the last few years.
The House could have chosen not to include Medicaid reform in this bill, but I’m glad they did because Medicaid reform is extremely important. The reforms in this bill will actually do a lot to improve the quality of coverage and care that people enrolled in Medicaid receive today.
Dylan Scott
Why does moving to a capped spending model improve the program over what currently exists?
Avik Roy
I would recommend to your readers to take a look at my treatise “How Medicaid Fails the Poor.” You start with the fact that access to care under Medicaid and health outcomes under Medicaid are very poor, far underperforming other health insurance programs and certainly way underperforming private insurance.
Why does that problem exist? It exists because states have very little flexibility in how they managed their Medicaid costs. They’re basically not able to do anything to keep Medicaid costs under control, except pay doctors and hospitals less money for the same amount of care. As a result of that, people have poor access.
By moving to a system in which you put Medicaid on a clear budget and you give states more flexibility in how they manage their Medicaid costs, you actually can end up with much better access to care and much better coverage. For example, right now, the Medicaid program doesn’t allow you to charge meaningful copays. As a result, an enrollee on Medicaid who has a cold can show up to the emergency room and get his cold managed in the emergency room. Now, that’s a very costly episode of care. If you’re able to charge significant copays for having your cold treated in the emergency room, and the savings from that go to fund higher reimbursement rates for primary care physicians, you can improve Medicaid access to care while also managing the program’s overall costs.
That kind of flexibility does not exist today. You can actually cover people better, with slightly less money, if you structure the benefit the right way.
A lot of the people who are crying bloody murder about this said the exact opposite when it came to Obamacare’s $850 billion-over-a-decade in Medicare cuts. A lot of people defended those cuts saying they would have no impact on access to care and coverage. If that’s correct, then surely Republicans are able to design Medicaid reforms that might achieve the same result.
Dylan Scott
I think on the Medicaid side, that is again where the numbers have shocked people. CBO projects an $800 billion cut in Medicaid spending, 14 million fewer people on Medicaid rolls. They hold that up against the repeal of investment taxes and industry taxes that were included in Obamacare.
In terms of the actual outcomes CBO is projecting for Medicaid, is this a case of CBO getting out over its skis? Is it a worthwhile trade-off for improving the program? What would you tell people about these numbers they’re hearing that I think are what’s stoked a lot of this anxiety?
Avik Roy
The CBO has done a disservice by not splitting out their assessment of how much of that reduction in Medicaid spending is coming from the repeal of Obamacare’s Medicaid expansion versus the per capita caps for the legacy Medicaid program. Without that, it’s very hard to answer your question.
My sense, from reading the CBO’s analysis, is that a significant chunk of that savings is coming from the repeal of the Medicaid expansion. At least half, though I don’t know that for a fact. Just my educated guess. Which is in part made up for by the tax credits, but, again, not enough.
What are the actual savings to the taxpayer that are coming from the per capita cap reform right now is unclear. I hope that as we get a final CBO score of this latest bill that actually passed the House, we’ll get more detail on that.
A big part of the spending reduction that the CBO estimates for the AHCA is driven by the coverage reduction that the CBO estimates. So if I’m right that CBO is significantly off in the decrease of coverage under the AHCA, then the spending reductions in the AHCA are also going to be a lot less in the real world than what the CBO has projected.
I have a lot of respect for the CBO and the people who work there, and I do think the CBO’s analysis should be taken seriously by policymakers and the public. But the experience with the ACA has shown they’re not omniscient, and I believe there are sound reasons to disagree with their analysis in this particular case.