The most challenging problem in health care policy is how to deal with the very tip of the cost curve — the 10 percent of the population who account for two-thirds of all personal health care spending; or among them, the 5 percent of the population who account for half of all spending; or among them, costliest of all, the 1 percent who account for a fifth of all spending.

The challenge is made harder still by the fact that insurance — in the traditional meaning of the term — is not an option. A large percentage of cases at the upper end of the curve fail to meet two standards of insurability.

One is that an insurable risk must be the result of unpredictable chance. In reality, though, many individuals suffer from chronic conditions like diabetes that make them certain to require costly care for the rest of their lives. Others have genetic markers that make them medical time bombs from the point of view of private insurers.

A second standard of insurability is that the actuarially fair premium — one high enough to cover the expected value of claims — must be affordable. However, an actuarially fair premium for many people with costly chronic conditions would exceed their entire income.

There are several partial solutions to the noninsurability of high-end health care risks. Guaranteed renewal requires insurers to continue to issue policies to those who become ill, provided there is no break in coverage. Guaranteed issue, which requires insurers to accept any applicants, regardless of pre-existing health conditions, is an even stronger step in the same direction. Community rating requires insurers to charge the same premium, based on average claims, to everyone in a general category regardless of their health status.

The Affordable Care Act uses a combination of these requirements to ensure that people can buy health insurance at a standard price regardless of pre-existing conditions. However, doing so creates problems of its own. For one thing, these requirements make the system vulnerable to adverse selection since healthy people can remain uninsured and buy into the system only when they become ill. Also, even with community rating, spreading health care costs evenly over an entire population can mean unaffordably high premiums for people with low incomes.

That brings us to the subject of this commentary — policies that aim to cut off the top end of the cost curve in order to make health care more affordable and accessible for everyone else. High-risk pools and reinsurance are two ways of doing this. After reviewing the way these approaches work, we will explain how their benefits can be realized through a policy of universal catastrophic coverage (UCC).

How high-risk pools and reinsurance work

In a high-risk pool, people whose health conditions make them uninsurable are placed in a separate category for which claims are paid by government. Prior to the ACA, according to a summary by Karen Pollitz of the Kaiser Family Foundation, 35 states had some type of high risk pool. All of the pools, by design, paid out more in claims than they brought in as premiums. The shortfall was made up, directly or indirectly, from general tax revenue.

The hope was that if insurance companies were relieved of the obligation to cover the highest-risk patients, they would lower premiums for the patients they continued to cover. However, because the pools were often underfunded, that hope was not always realized. In many states, high premiums, high deductibles, and waiting periods discouraged enrollment. As a result, the pools covered only 1 to 10 percent of the population that had insurance in the individual market — well below the 27 percent that Pollitz estimates should theoretically have been eligible. The state-run high-risk pools have now been made redundant by the guaranteed issue and community rating requirements of the ACA.

Reinsurance takes a different approach. Instead of segregating high-risk consumers into a separate pool, reinsurance leaves everyone in the general risk pool, while a government reinsurance fund reimburses insurers for claims above an attachment point and up to a cap. For example, suppose the attachment point is $50,000 in annual claims and the cap is $1 million. An insurance company would then get a reimbursement of $40,000 for a patient with $90,000 in claims and a maximum reimbursement of $950,000 for patients with claims of $1 million or more.

Under reinsurance, private insurers continue to cover the high-risk consumers, but the reimbursements make it possible for them to offer lower premiums to all their customers. As in the case of high-risk pools, the actual reduction in premiums would depend on the level of funding for the reinsurance program.

The traditional form of reinsurance is retrospective, in that no effort is made to identify high-risk consumers in advance. Under retrospective reinsurance, a large share of reimbursements would normally represent the claims of people with identifiable chronic conditions or genetic predispositions, but those claims would be treated no differently than those of healthy individuals who experienced high expenses because of unforeseeable accidents or illnesses.

A newer variant, prospective reinsurance, operates somewhat differently. Private insurers continue to issue policies to all applicants, regardless of health condition. However, when particular applicants are identified as being at high risk, the primary insurer “cedes” a part of the risk to the reinsurer, along with part of the premium. For example, the primary insurer might cede risks over $50,000 per year to the reinsurer in exchange for 90 percent of the premium. The reinsurer would then cover claims above the $50,000 attachment point, while the primary insurer would cover lesser claims in return for the remaining 10 percent of the premium.

Prospective reinsurance is, in effect, a hybrid. It resembles a high-risk pool in that it places consumers who are most likely to have high claims in a separate pool. Those high risks are covered by government funds, as under reinsurance. However, the whole process goes on out of sight. Because consumers do not know when a particular claim is being reimbursed by the government, prospective reinsurance schemes are sometimes called invisible high-risk pools.

The American Health Care Act of 2017, which passed the House but failed in the Senate, included a provision for prospective reinsurance, although that provision was criticized as underfunded. Despite the failure of the AHCA, several states are experimenting with the invisible risk-pool model, or are considering doing so.

Pros and cons

Reinsurance and high-risk pools in all their forms have pros and cons. On the positive side, all such schemes have the potential to handle costly chronic conditions that would otherwise be uninsurable. By taking the weight of the most expensive cases off the shoulders of private insurers, they offer the potential to hold down the premiums faced by other consumers who face only average risks. However, reinsurance and high-risk pools also have their drawbacks.

One obvious drawback is that they are expensive. Covering the costs of just the top 5 percent of health care spenders would cost something like $1.5 trillion as of 2018, or about 7 percent of GDP. If high-risk pools or reinsurance are not adequately funded, they do not lift enough of the burden from the broader individual market. In that case, premiums remain high for consumers with average risks. High premiums tempt healthy consumers to go without insurance, and that, in turn, exacerbates the problem of adverse selection, driving premiums higher still.

A second problem is that high-risk pools and prospective reinsurance require advance screening for health risks. Even if everyone is ultimately guaranteed coverage in one form or another, screening raises administrative costs. After the experience of guaranteed issue under the ACA, with no intrusive questionnaires, health histories, or physicals, a return to widespread screening for pre-existing conditions or genetic predispositions would probably encounter consumer resistance. Of the alternatives we have discussed, only retrospective reinsurance avoids the screening problem.

A third problem is that even if high-risk pools or reinsurance succeeded in lowering average premiums, many low- and middle-income consumers would be likely to find that coverage was still unaffordable. For example, if high-risk pools or reinsurance absorbed half of all personal health care expenditures, the remainder would still work out to an average close to $20,000 per year for a family of four. Families would have to pay those costs through a combination of premiums, deductibles, and copays. Health care expenses of $20,000 per year would be equivalent to some 40 percent of median household income — affordable for some, but by no means for all.

To overcome these problems, we need to lift the burden of the top tranche of health spending in a way that provides a reliable source of funding, that minimizes administrative complexities, and that takes household income into account. Fortunately, such an alternative is available — universal catastrophic coverage.

The UCC solution

The basic idea of universal catastrophic coverage is simple. UCC aims to protect all Americans against financially ruinous medical expenses while preserving the principle that those who can afford it should contribute toward the cost of their own care. For people below a specified low-income threshold, UCC would pay health care costs in full. Everyone else would get a UCC policy with a deductible scaled to their eligible income, that is, to the amount by which annual household income exceeds the low-income threshold.

For example, suppose the low-income threshold is set at $25,000 for a family of four, roughly equal to the official poverty line, and the deductible is set at 10 percent of eligible income. A family with income of $25,000 or less would then have no deductible; one with income of $75,000 would have a deductible of $5,000; and one with income of $400,000 would have a hefty deductible of $37,500. Regardless of income, once a household hit the deductible, their UCC policy would pay the rest of their medical expenses.

Don’t be surprised if this sounds a little like what we have been discussing all along. Universal catastrophic coverage belongs to the same general family of policies as high-risk pools and reinsurance. In its essence, UCC is simply retrospective reinsurance with an attachment point that is scaled to household income.

UCC would be funded using money that is already being spent on other federal health care programs, including Medicaid, ACA subsidies, and the tax expenditures that subsidize employer-sponsored insurance. For details, see my earlier post, “Could We Afford Universal Catastrophic Coverage?” As that post explains, there is a good chance that a properly designed UCC plan that included a range of cost-saving measures would actually reduce federal health care expenditures relative to their current level.

For administrative simplicity, UCC policies could be issued directly by the government, for example, as an extension of Medicare. Alternatively, the policies could be issued by private insurers, subject to coverage guidelines, as under Medicare Advantage. Those insurers would pay providers and then be reimbursed by a government reinsurance fund.

Because payments for catastrophic expenses are made retrospectively, UCC would avoid the need to screen for health risks. Everyone would be issued a UCC policy automatically. Most people would probably pay their share of noncatastrophic expenses out of pocket, possibly with the help of dedicated health savings accounts. Households with very high incomes and deductibles could, if they chose to do so, buy private supplemental insurance to cover expenses not reimbursed by UCC.

Finally, UCC would be affordable for everyone because of the way deductibles are scaled to income. People below the low-income threshold (very roughly, those now eligible for Medicaid) would get first-dollar coverage for all their health care needs. A package of basic preventive and primary care services would be exempt from deductibles for everyone, regardless of income. Middle-income households, would be responsible for their own fair share of noncatastrophic expenses, but in most cases, that share would be less than the premiums, deductibles, and copays they now pay for employer-sponsored insurance or policies bought on the ACA exchanges.

It is true that, in theory, all these issues of affordability and insurability could be addressed separately. You could use a combination of guaranteed issue and community rating to spread risks among all covered households. You could use a combination of premium subsidies and cost-sharing subsidies to limit the maximum cost to families with moderate incomes. You could use individual and employer mandates to fight adverse selection and enlarge the risk pool. You could run a completely separate program, like Medicaid, for households below the low-income threshold. You could do all that — but the result would be a cumbersome kludge that drove up administrative costs and still left millions without coverage. We know that because we tried it with the ACA, and look what we got.

So why not keep things simple by providing high quality, affordable health care for all with universal catastrophic coverage?