After seven years, Republicans have finally released their health care legislation. Now their bill can be compared side-by-side with the Affordable Care Act based on how it affects enrollees’ pocketbooks — including both premiums and out-of-pocket costs for care.

The bill does not measure up well: For all but the youngest individuals, it increases both overall costs and the risk of a financially devastating event.

Republicans will argue that by cutting back insurance standards, their bill would lower premiums —  because people can choose to buy less coverage than the ACA currently requires. But premiums are only one expense that enrollees have to worry about. First, the level of tax credits affects how much enrollees would actually pay for those premiums, out of pocket.

Second, the degree of insurance protection affects how much enrollees would pay in deductibles, co-pays, and other forms of cost-sharing. (“Cost-sharing” refers to how the insurance company and the enrollee split the costs of health services.)

We’re presenting an analysis here of the net financial impact of the Republican bill on premiums, after tax credits, plus cost-sharing. We estimate that the bill would increase costs for the average enrollee by $1,542, for the year, if the bill were in effect today. In 2020, the bill would increase costs for the average enrollee by $2,409.

We provide the figure for 2020 because that’s when the Republican tax credits would go into effect; we provide a figure for this year so that readers can get a sense of how the plan might affect their situation were it implemented today. Importantly, the gap between costs under the ACA and under the Republican bill would grow over time.

In general, the impact of the Republican bill would be particularly severe for older individuals, ages 55 to 64. Their costs would increase by $5,269 if the bill went into effect today and by $6,971 in 2020. Individuals with income below 250 percent of the federal poverty line would see their costs increase by $2,945 today and by $4,061 in 2020.

The bill’s key alterations to the framework created by the ACA, explained

The ACA includes three elements that affect the generosity of insurance. First, it requires insurers to cover a minimum share of costs, on average. This minimum “actuarial value” of the plans is set by law at 60 percent. Second, the act provides premium tax credits that are linked to the cost of a plan that covers 70 percent of costs, on average (the so-called silver plans). As the price of those plans rise, so does the tax credit.

Third, the ACA reduces cost-sharing levels — that is, it provides another kind of subsidy — for lower-income individuals with income from 100 percent to 250 percent of the federal poverty level. The combined effect of these elements is that, for non-group policies under the ACA, the weighted average share of costs covered by insurers is about 75 percent.

The Republican bill unveiled last night would remove these elements. It would eliminate the minimum required actuarial value; eliminate cost-sharing reductions for lower-income individuals; and provide flat tax credits by age unrelated to any plan’s cost. These tax credits are for the most part also unrelated to people’s income, but they start to phase out for individuals with income above $75,000. The combined effect of these changes is that the bill would dramatically reduce the generosity of insurance and sharply increase deductibles and other out-of-pocket costs.

The Republican bill would also relax the limit on how much insurers can charge older individuals (a practice known as “age rating”). Currently, the ACA prohibits insurers from charging older individuals premiums that are more than three times greater than premiums for younger individuals. Under the Republican bill, insurers could charge premiums for older individuals that are as much as five times greater. Therefore, obviously, premiums for older individuals would go up, those for younger individuals would go down.

Lastly, the Republican bill would eliminate the individual mandate. Instead, if individuals have a gap in coverage greater than 63 days, they would face a penalty equal to 30 percent of premiums for 12 months.

How we estimated costs

We previously published an analysis of the Republican plan using the tax credit levels prescribed in legislation sponsored by Secretary of Health and Human Services Tom Price when he was a member of Congress. For this analysis, we used the same methodology and altered parameters and assumptions as explained below.

We used the tax credit levels specified in the Republican bill: $2,000 for individuals up to age 30; $2,500 for individuals age 30 to 39; $3,000 for individuals age 40 to 49; $3,500 for individuals age 50 to 59; and $4,000 for individuals over age 60. These tax credits phase out for individuals with income above $75,000 and for those filing a joint return with income above $150,000. (In our analysis, we treated households with at least two adults as joint filers.)  

As stated above, we first applied these tax credits to premiums in 2017 and then to premiums in 2020, the year the credits would take effect. To estimate premiums in 2020, we inflated premiums using National Health Expenditure (NHE) projections. For comparison with the ACA, we adjusted the ACA’s subsidy amounts using Congressional Budget Office estimates and NHE projections.

Over time, the value of the new tax credits under the Republican bill would erode significantly. That’s because they are indexed to grow with consumer inflation plus 1 percentage point, which is much slower than the rate medical costs are actually growing. Under the ACA, by contrast, tax credits grow in lockstep with premiums for “silver” plans, and costs are capped as a percentage of income.

The ACA’s tax credit structure would help protect enrollees from premium increases from 2017 to 2020; under the Republican plan, enrollees would pay a greater share of premiums each year.

Whenever tax credit levels exceed premiums, we assumed that the excess could be deposited in health savings accounts to reduce cost-sharing levels, as specified in the Republican bill.

We assume that under the Republican bill the average actuarial value would fall to 50 percent; all things considered, plans would cover only half of health care costs. That’s because the Republican bill would encourage enrollment in “catastrophic” plans with high deductibles. The insurance industry pegs the actuarial value of such catastrophic plans at 50 percent. In addition, a recent analysis by the Urban Institute estimated that about half of people, under the Republican system, could use their tax credits to buy a plan with an actuarial value of a mere 47 percent.

Our assumption of a 50 percent actuarial value is actually generous to the Republican bill. Even though the bill repeals the minimum actuarial value, it does not repeal limits on out-of-pocket costs or essential health benefits. Using the 2018 actuarial value calculator, mathematically plans could not have an actuarial value much below 58 percent with these constraints.

The plan would lead to a sicker pool of enrollees 

We assume that the Republican bill’s replacement for the individual mandate would cause some degree of adverse selection. In other words, the change in incentives would likely lead disproportionately sicker individuals to enroll in coverage, driving up premiums.

Younger, healthier individuals may not be sufficiently motivated by the “continuous coverage” requirement, and modest penalty for violating it, to sign up for coverage to avoid an abstract penalty if they happen to get sick sometime in the unforeseeable future. If they do miss the open enrollment period, they would effectively be locked out of the market. Sicker individuals, on the other hand, would do absolutely everything they could to sign up and avoid a gap in coverage.

What’s more, the late enrollment penalty would be far lower than the additional premium insurers would charge to many sicker individuals if they were able to do so. So it would be a bargain for many individuals to wait until they get sick to enroll in coverage.

In the absence of the individual mandate, CBO estimates that adverse selection would increase premiums by 20 percent. We assume that the Republican bill’s replacement for the individual mandate would have at least some effect, but that it would not be as effective as the mandate. We assumed the shift to a continuous coverage model would increase premiums by 10 percent.

We found significant cost increases both for individuals and for families

Although premiums would be lower under the Republican bill, this decrease would be offset by an increase in cost-sharing. Once the differences in tax credits are accounted for, the bill would increase costs significantly. On top of this, shifting consumer spending from premiums to cost-sharing would greatly increase financial risk. (If you’re now paying 50 percent of your costs, instead of 25 percent, a big hospital bill could be devastating.)

We reported above the results for individuals. For families, the Republican bill would increase costs by $2,243 if the bill were in effect today. For families with a head of household age 55 to 64, the bill would increase costs by $7,604. For families with income below 250 percent of poverty, the bill would increase costs by $6,228.

These cost increases would explode by 2020. We estimate that the Republican bill would increase costs for families by $4,274. For families with a head of household age 55 to 64, the bill would increase costs by $10,591. For families with income below 250% percent of poverty, the bill would increase costs by $9,024.

There may be a reason why Republicans took so many years to reveal their health care plan: It is easier to attack an imperfect existing system than to improve it. Under examination, the Republican bill clearly does not compare favorably with the ACA. As many have pointed out, and as the CBO score will clearly show, the Republican bill would cause millions of people to lose their coverage — but it would also increase costs significantly for those who remain insured.

David Cutler is the Otto Eckstein Professor of Applied Economics at Harvard University. John Bertko is the chief actuary for Covered California, the state’s health insurance exchange. Topher Spiro is the vice president for health policy at the Center for American Progress. Emily Gee is a health economist at the Center for American Progress.

The Big Idea is Vox’s home for smart, often scholarly excursions into the most important issues and ideas in politics, science, and culture — typically written by outside contributors. If you have an idea for a piece, pitch us at thebigidea@vox.com. 

Last week, we presented the first analysis of the net financial impact of Republican plans to replace the Affordable Care Act.

The upshot was that while Republicans focus on rising premiums for plans sold through the ACA exchanges, their replacement proposals would hit consumers’ wallets hard. That’s because most consumers would end up paying a greater share of their health care costs — meaning premiums after tax credits plus cost-sharing in the form of deductibles and copays. We found that the average cost for an individual covered by the Affordable Care Act would go up by $1,744 per year. For families, the increase would be even greater: $6,089.

That analysis was based on common elements of the Republican plans, which remove minimum coverage for health plans — no longer would every plan cover prescription drugs, for example — and provide much lower tax credits to help low-income Americans pay their premiums. In our calculations, we used the tax credit levels prescribed in legislation sponsored by Secretary of Health and Human Services Tom Price when he was a member of Congress.

Shortly after our analysis was posted, legislation that House Republicans are working on was leaked. We have updated our estimates based on that version of the Republican plan, using the same methodology as in our original analysis, altering parameters and assumptions as explained below. (The leaked bill would appear to be the latest known version of the bill in progress that Sen. Rand Paul charged was being held “under lock & key, in a secure location,” as he put it on Twitter.)

To be as fair as possible given uncertain effects, we present a range of estimates. But in all scenarios the leaked bill would increase total costs — premiums after tax credits, plus cost-sharing — significantly for the average ACA enrollee.

Our central estimate is that the leaked bill would increase costs for the average individual by $1,380 per year — slightly less than under Secretary Price’s original plan, but still substantial. The impact would be particularly severe for older individuals ages 55 to 64, whose costs would increase by $5,118 per year. Lower income individuals with income below 250 percent of the federal poverty level would see their costs increase by $2,945 per year.

Although the tax credit levels specified in the leaked bill are higher than those prescribed by the Price bill, they are still not sufficient to avoid significant cost increases for most enrollees.

What we adjusted, based on the leaked bill

The biggest difference between the Republican plan that we modeled in our original analysis and the leaked bill is the level of tax credits. The tax credit levels specified in the leaked bill are more generous than those prescribed by Secretary Price, although they still do not take income into account like the ACA’s do. The new proposed tax credits are as follows: $2,000 for individuals up to age 30; $2,500 for individuals ages 30 to 39; $3,000 for individuals ages 40 to 49; $3,500 for individuals ages 50 to 59; and $4,000 for individuals over age 60.

Again, regardless of income.

Although these new tax credits would start in 2020, we applied them to premiums in 2017 so people can understand what the impact on them would be today. Doing so yields extremely conservative estimates, in part because over time the value of the new tax credits will erode significantly. That’s because they are indexed to grow with consumer inflation plus 1 percentage point, whereas medical costs rise far faster than that. (Under the ACA, tax credits grow along with premiums — if “silver plan” premiums rise, so do the credits. Moreover, costs are capped as a percentage of income.)

Because premiums will rise markedly between now and 2020, consumers would pay more in premiums out of pocket under the leaked bill in 2020 than they would now. By contrast, the ACA’s tax credit structure would help protect enrollees from premium increases from 2017 to 2020.

Whenever tax credit levels exceed premiums, we assumed that the excess could be deposited in health savings accounts to reduce cost-sharing levels (an option specified in the leaked bill).

In our original analysis, we assumed that insurers would cover only 50 percent of total medical costs, on average, under the Republican plan. (The ACA sets the minimum “actuarial value” at 60 percent, although the popular silver plans cover fully 70 percent.) This assumption was based on the fact that Republican plans generally eliminate the requirement on insurers to cover a minimum share of costs.

The Trump administration recently proposed a regulation to lower the actuarial floor from 60 percent to 56 percent. As a result, most plans would likely still have an actuarial value of at least 56 percent — unless and until courts strike down the Trump administration regulation, a possibility, given that it conflicts with the ACA statute.

However, the leaked bill contains several features that would poke holes in this floor:

• It allows younger individuals to use tax credits to buy “catastrophic” plans with an actuarial value below 60 percent.
• It allows insurers to continue to offer plans that existed before the ACA and — this is the key distinction — to enroll new individuals in these plans. Although insurers offer such pre-ACA plans now, they were never able to enroll new individuals and they were set to expire
• It allows individuals to use tax credits to buy pre-ACA plans as well as short-term plans. Insurers can offer short-term plans with shoddy coverage for 364 days as a way to evade ACA standards.

Because plans with an actuarial value below 60 percent would newly qualify for tax credits under the leaked bill, enrollment in these plans would surge.

The distribution of enrollees among plans with an actuarial value of 56 percent and plans with a lower actuarial value would be uncertain. Given this uncertainty, we present a range of estimates for actuarial values of 50 percent and 56 percent. Such plans lack key consumer protections and would not provide nearly the same degree of financial protection as ACA plans.

The pool of enrollees under the Republican plan is likelier to be sicker

In our original analysis, we assumed that replacing the individual mandate with a requirement to maintain “continuous coverage” (or else face higher premiums based on health status) would not cause the mix of people in plans to get sicker — a process called “adverse selection.”

This was a very conservative assumption. It assumes that younger, healthier individuals would be sufficiently motivated to sign up for coverage to avoid an abstract penalty if they happen to get sick sometime in the unforeseeable future. (Human psychology was not built to judge such risks accurately.) But if this assumption does not hold, these younger, healthier individuals would be locked out of the market.     

There is no evidence that a continuous coverage requirement would actually work in the real world. Nonetheless, in our original analysis, we gave the policy the benefit of the doubt. In evaluating the leaked bill, however, we are more certain that serious adverse would occur.

First, the leaked bill provides more detail about how the continuous coverage rule would work. If individuals have a gap in coverage greater than 63 days, they would face a penalty equal to 30 percent of premiums. But that penalty would be far lower than the additional premium insurers would charge to many sicker individuals if they were able to discriminate based on health status. In other words, it would be a bargain for many individuals to wait until they get sick to enroll in coverage.

Second, because the tax credits under the leaked bill are much lower than under the ACA, many healthy low-income individuals would find insurance to be unaffordable. (This effect may be partially offset because the leaked bill allows insurers to charge younger individuals lower premiums than under the ACA and allows them to offer less comprehensive plans with fewer benefits.)

Third, the leaked bill repeals the Medicaid expansion. Individuals with income between 100 percent and 138 percent of the federal poverty level — who are sicker on average — would move from Medicaid to the private insurance market. Previous research has estimated that this shift could increase premiums by 7 percent.

Fourth, the leaked bill would in effect create a separate market of sub-standard plans: pre-ACA plans and “short-term” plans. This separate market (with a separate risk pool) exists now, but it is very small. As discussed above, enrollment in these plans would surge because they would newly qualify for tax credits. These cheaper, sub-standard plans would attract the youngest and healthiest individuals, who would leave the pool of plans that meet ACA standards and no longer help balance out older and sicker enrollees.

On the other hand, the leaked bill includes $100 billion in funding for “state innovation grants” that could theoretically help offset adverse selection. This effect would be possible if states used the money to fund “reinsurance” that subsidizes insurers for their high-cost enrollees.

As a result of these factors, we expect that the leaked bill would cause serious adverse selection — but we’re not sure how much. In the absence of the individual mandate, CBO estimates that adverse selection would increase premiums by 20 percent. We present a range of estimates for adverse selection that would increase premiums by 0, 10, and 20 percent.

Under every scenario, we found significant cost increases under the Republican plan, relative to the ACA

Table 1 displays the results for select groups of individuals if the average actuarial value for insurance plans available to current ACA enrollees is 50 percent. Table 2 displays the results if the average actuarial value is 56 percent. Tables 3 and 4 display the results for families, in the same scenarios. Not surprisingly, cost increases are higher with greater degrees of adverse selection: The more sick individuals in the system, relatively speaking, the costlier it would be.

For our central, intermediate estimate, we assume moderate adverse selection that marks up premiums by 10 percent and an average actuarial value of 50 percent.

To reiterate, we estimate that the leaked bill would increase average costs for individuals by $1,380 per year. For older individuals ages 55 to 64, the bill would increase costs by $5,118 per year. For lower-income individuals with income below 250 percent of the federal poverty level, the bill would increase costs by $2,945 per year. And of course, all of these numbers would be much higher for families.

And to repeat, these estimates are extremely conservative. We applied the leaked bill’s 2020 tax credit levels to 2017 premiums — a generous move because premiums will rise from 2017 to 2020 but Republican tax credits will not. If we inflate 2017 premiums to 2020, our central estimate of the cost increase, $1,380, would increase by 63 percent, to $2,249.

As under our original analysis, these estimates are only base cost increases. Because the leaked bill also repeals essential health benefits that insurers are required to cover under the ACA, costs would increase by even more for some individuals. People who need expensive medicine, mental health care, or even maternity care might find themselves paying thousands more

For individuals who are enrolled in such plans and need these services, the cost increases reported above would increase by an additional $1,000 to $8,500 per year. Table 3 from our original analysis provides more detail about these increases.

Where the Republican proposal leaves us, in the end

The newly leaked Republican health care plan makes some things better and some things worse than earlier Republican proposals. Premium credits are more generous — yet it is also more likely that the insured population will be sicker, driving up premiums.

Overall, Republicans are working within a ballpark in which the same general result will hold. Some enrollees will see lower premiums, but their overall costs will rise.

Our original conclusion stands: The Republican approach is deeply flawed and would result in significant cost increases for American consumers. 

David Cutler is the Otto Eckstein Professor of Applied Economics at Harvard University. John Bertko is the chief actuary for Covered California, the state’s health insurance exchange. Topher Spiro is the vice president for health policy at the Center for American Progress. Emily Gee is a health economist at the Center for American Progress.

The Big Idea is Vox’s home for smart, often scholarly excursions into the most important issues and ideas in politics, science, and culture — typically written by outside contributors. If you have an idea for a piece, pitch us at thebigidea@vox.com.