The American Health Care Act — what it would mean

Republican Congressional leadership has released their healthcare plan, and I thought some folks might find a summary of the likely impacts helpful. The Congressional Budget Office released its analysis of the bill Monday. I think it provides a useful framework for thinking about the effects, but I’ll provide a lot of my own commentary.

I realize some people may be mistrustful of the CBO. I think it’s useful to look at their recent track record. The US government spent $50 billion on Medicare Part D in 2013. When the Medicare Modernization Act was passed in 2003, the CBO projected costs of $99 billion in 2013. Costs were literally half of the CBO’s projections! The reasons were (a) lots of high priced drugs went off patent (i.e., became available as low cost generics) after 2003 and few new high priced drugs came to market and (b) fewer people signed up for Medicare Part D than expected. More recently, the CBO recently updated its projections of the costs of the Affordable Care Act from 2015–2019 to be $101 billion per year compared to their original estimate of $142 billion. That’s a pretty big difference, too!

So the CBO’s estimates are often pretty off in terms of magnitude, which is understandable. Healthcare markets are very complex and difficult to project. However, the CBO’s estimates aren’t off in terms of direction. They predicted the MMA would cost billions, and it did. They predicted the ACA would extend coverage to millions and reduce the deficit through its new taxes, and it did. They modeled lots of different funds flows, and they were right on every one on the direction of the impact. That’s why I think it’s a useful framework, though we should treat the actual numbers as predictions rather than “facts”.

So let’s talk about the American Health Care Act! First, I think it’s important to realize the intent of the bill: to (a) reduce taxes and (b) reduce the deficit. Paul Ryan, the chief architect of the bill, has been very clear about those goals. You can see them on his web site: “We want all Americans, when they look at Washington, to see spending going down, taxes going down, debt going down.”

That’s the point of the bill — to provide a big tax break, small deficit reduction in the near term, and large deficit reduction in the long term. The price they’re willing to pay is less healthcare coverage, and the structure of the bill is designed to do so in a way that minimizes the disruption on the middle class while lower income Americans experience the brunt of the impact.

So how does it do these things? First, the CBO estimates outside its coverage provisions, the AHCA cuts taxes $59 billion per year. The Committee for a Responsible Budget provides a nice breakdown of the numerous separate tax cuts. Households making over $200,000 a year would receive $28 billion of those tax cuts. The healthcare industry would receive $20 billion of those tax cuts. And employers and households more broadly would receive the remaining $11 billion in tax cuts. While the coverage numbers are tough to predict, we can have a lot of confidence in these numbers.

Moving on to goal two. The CBO estimates that the AHCA would cut federal budget deficits by $34 billion a year over the next 10 years. How does it do that with the above tax cuts? It makes substantial cuts to what the federal government is spending on health insurance for Americans (more on that later). For now I’ll note that the $34 billion number is a bit misleading. The AHCA starts the tax cuts early but it delays many of the spending cuts. The $34 billion average includes a projected increase in the budget deficit of $33 billion in 2019 and a decrease of $92 billion in 2026.[1]

The biggest portion of the savings by far, $88 billion per year, come from reductions to Medicaid, the federal government’s program to provide health insurance to the lowest income citizens. The primary changes are (a) restricting who is eligible and (b) setting per capita caps on how much the federal government will pay (right now, it will match what states choose to spend). This structure increases savings over time and results in a 25% reduction in federal Medicaid spending ($155 billion) by 2026.

The next chunk of savings, $67 billion per year, comes from eliminating the ACA subsidies for low income individuals to buy private insurance. However, it replaces these subsidies with age-based subsidies costing $36 billion. So the actual reduction would only be $31 billion.

Finally, the AHCA eliminates the penalty for middle and upper class choosing not to buy insurance (individual mandate) and the penalty for employers with over 50 employees choosing not to offer coverage (employer mandate). Employers and individuals have been paying those penalties, so that’s another $21 billion in annual revenue cuts.

There are lots of other provisions, but they’re relatively small compared to these. Collectively, the CBO estimates they would increase the deficit by $5 billion per year.

For those following along at home, here’s what we’ve got:

· $88 billion in Medicaid spending cuts

· $31 billion in net spending cuts to individual insurance subsidies

· ($59 billion) in tax cuts unrelated to coverage

· ($21 billion) in tax cuts from repealing the employer and individual mandates

· ($5 billion) from other provisions

· $34 billion in deficit reduction

Again, let’s not get caught up in the specific numbers, but I think these estimates are helpful in understanding what the AHCA is trying to do. It is fundamentally shifting more responsibility for healthcare costs from the government to citizens. It is reducing support in getting insurance for low income Americans. Some of the federal budget savings are being passed on to taxpayers, particularly the wealthy and the healthcare industry. And some are left for deficit reduction.

Beyond the budget impact, it’s worth taking a moment to understand what the AHCA would do to the insurance market. The most direct and obvious impact is reducing the number of Americans on Medicaid. The CBO estimates a reduction of 14 million enrollees by 2026, which seems plausible and is consistent with the intent of the law. That’s a big drop when compared with the 62 million enrollees today. Over time, the quality of the health insurance will likely decline as well, given the per capita caps.

The CBO also estimates a 7 million member reduction in the employer/group market and a 2 million member reduction in the individual market. For reference, the current group market is 178 million and the current individual market is 52 million. I’m somewhat skeptical of these numbers, though I certainly haven’t done my own analysis. My own belief is that the policy tools being repealed were never strong enough to materially impact employer decisions to offer insurance, and repealing them won’t have a sizable impact.

Part of the reason the CBO only sees a decrease in the individual market of 2 million is it assumes employers “dumping” (dropping coverage) for 7 million people, and many of those would then get individual coverage. A second reason for the 2 million assumption is a large influx of young, middle class Americans into insurance because it will be cheaper for them than it is today given the altered subsidy structure. While this may be the case, I think this is repeating the same mistake made in the ACA CBO estimate — assuming a greater effect of financial incentives on purchase decisions than actually happens in the real world. Depending on those dynamics, the actual reduction in individual market coverage could be much higher than 2 million.

Why? Remember, the AHCA cuts the funding for subsidies buy insurance by about half and it eliminates the individual mandate and replaces it with a much weaker provision (allowing insurance companies to charge 30% for breaks in coverage). The elimination of these provisions will almost certainly lead to many healthy people deciding that health insurance is too expensive and deciding not to purchase it. The remaining insurance customers will be sicker on average, so health insurance companies will need to raise prices to cover their costs. When the prices go up, even more healthy people will drop out of the system. And so on. This cycle is what the industry calls a “death spiral”, and we saw it happen before the ACA in a number of states that did not allow insurers to charge higher premiums to the less healthy. Though I don’t think the market will collapse entirely, the equilibrium point will likely be fewer enrollees.

One final note — the CBO actually expects individual market premiums to go down, not up, despite this phenomenon. Part of that is due to them believing more young people, who will have lower premiums, will enter the market, which would be a good thing if it happens. However, part of that is because they project the quality of the health insurance will go down. The ACA defines minimum essential coverage as including many features such as prescription drugs and maternity benefits. It also requires an “actuarial value” of at least 60%. This means that, if your healthcare costs are $100, your insurer is paying $60 and you’re paying $40 out of pocket on average. That’s not great, but that’s still meaningful.

Now suppose that the insurer cuts out drugs ($10) entirely and decides on an actuarial value of 50% for the remaining $90. So now the insurer is paying $45, and you’re paying $55 of your healthcare costs. The cost of healthcare hasn’t gone down — just who’s paying for it and how has shifted. Anyone claiming that premiums will go down under the AHCA because of increased competition is off. The market won’t get any more or less efficient. It’s already efficient. Average premiums will just go down if healthier people are insured and if the quality of the insurance product decreases.

[1] This is a classic trick of Washington budget making — put the goodies in early and the baddies in late, and look at the 10-year picture to make them offset. Then go back and change the law before the baddies kick in. I do think that Paul Ryan wants this full plan to take effect, but Washington realities mean that the out years are less likely to take effect than the early years.