Does achieving “Medicare for All” mean mostly eliminating private health insurance? Single-payer proponents say yes: After all, if a public plan provides comprehensive, no-deductible coverage for everyone, nobody would want—much less be willing pay for—duplicative private coverage.

Yet candidates who previously embraced single-payer sometimes seem a bit unsure. For instance, Senator Cory Booker, who co-sponsored Senator Bernie Sanders’s single-payer plan back in 2017, was asked whether he would “do away with private health care” recently, and he responded, “Even countries that have vast access to publicly offered health care still have private health care, so no.”

There are actually two distinct questions wrapped into one here. First is whether we want a universal public plan for everyone, or a hodgepodge of public and private plans that cover different parts of the population according to age, income, workplace, disability, and so forth, but that together cover everyone. Last year in Dissent, I made the case that a nation like ours—with enormous unmet medical needs, an inadequate safety net, and galling inequality—is a poor fit with a multi-payer system that divides the population into a hierarchy of public and private plans with inequitable levels of access, varied copays and deductibles, and unequal benefits and provider networks. This would never achieve the equity, universality, or efficiency of a public plan that provides complete coverage to everyone.

But there’s another question. Let’s assume we agree on the need for a universal public-insurance plan that covers everyone, as in Canada, Great Britain, or France. Would there still be a role for private insurance? If so, what would it be?

In nations that have universal public-insurance programs, private health plans fall into three categories: “duplicative” plans, “supplementary” plans, and “complementary” plans. “The debate over eliminating [private] health insurance is actually offering a false choice,” says Sarah Kliff of Vox.

Let’s start with “duplicative” coverage, which refers to private plans that “duplicate” benefits of the public plan, like covering doctor visits or procedures that are also covered by the single-payer plan.

At first glance, it might seem odd that insurers would offer such plans, much less that anyone would pay for them. You wouldn’t, for instance, buy a private plan hawked by a company that promises you “access” to Central Park. You already have that. Obviously, such plans must offer some advantage to be viable.

And they do: In the single-payer context, they let individuals jump to the front of the line, gaining wider or quicker access to physicians’ services or other care covered by the public plan. Consider a case from the intensive-care unit where I work. Assume it’s a busy day, and the ICU is crowded. Should a scarce bed go to a less sick person over a sicker one who needs it more, just because the former has better-paying insurance? Most, I believe, would find that appalling.

But essentially, that’s what duplicative plans promise, albeit usually for non-emergency care. Now, some might argue that allowing people to have preferential access to office appointments or elective surgery is less problematic than when it involves an ICU bed. But such distinctions are arbitrary. Various types of care can be lifesaving, or limb-saving. Whether you’re talking about access to a primary-care doctor or a specialist, a psychiatrist or a hospital bed, health—not wealth—should be the factor determining access.

It is true, as Kliff describes, that countries with universal coverage handle this differently. Great Britain has retained a small private-insurance market that gives some people a leg up in seeing the doctor. But Canada prohibits duplicative coverage, and, in fact, so does the United States. It has long been illegal to sell duplicative individual coverage to Medicare beneficiaries. I’ve never heard any older adults complain about this fact and pine for a private, marketplace plan; in fact, I’m guessing few are even aware of the exclusion.

Duplicative plans, in other words, are not desirable, but they are also unnecessary. We should not embrace them.

But how about “supplementary” coverage, the private plans that provide benefits for services not covered by the public system? As Kliff notes, in Canada, the public system doesn’t offer universal drug benefits or dental care, so people need supplementary private plans to cover their medicines and their teeth. Similarly, in the United States, Medicare doesn’t cover dental-care benefits, and until 2003 didn’t cover prescription drugs.

The single-payer bills in Congress do not ban supplementary private coverage. However, because both the forthcoming bill in the House and (with the exception of long-term care) the bill in the Senate have comprehensive benefits—including dental care, prescription drugs, and vision care—there is not much left for supplementary plans to cover. Perhaps cosmetic surgery, or trips to Swiss medicinal spas?

The only way these bills could make way for supplemental insurance would be to strip coverage benefits for the sole purpose of creating business opportunities for the private insurance industry. Surely, we could do that: We could remove coverage for dental benefits or kidney care, for colonoscopies or elbow surgeries, and perhaps a private insurance market would emerge to cover such services. But why would we possibly want to?

Consider that Canada’s exclusion of drug coverage from its public system is a major problem—it’s the reason why Canada has higher rates of people not taking their medication because of cost relative to other high-income nations, apart from the United States.

When fashioning any new health program, we should pick and choose the best policies. For instance, the United Kingdom does have universal drug coverage (mostly without co-pays) and, consequently, basically everyone gets the medicine they need. That should be our model. The underlying question is simple: Do we offer comprehensive benefits in the universal public system, or do we drop benefits at random so as to give Aetna and Cigna something to do? The answer, to my mind, seems clear.

Finally, many nations have “complementary” private plans, which cover the co-pays and deductibles imposed by some (but not all) public systems.

For instance, many Medicare beneficiaries take out so-called “Medigap” plans today, which cover that program’s often substantial out-of-pocket expenses. In France, almost everyone has a complementary plan that covers the cost-sharing (e.g., co-pays) imposed by the single-payer program. The United Kingdom and Canada, in contrast, have no co-pays for physician care, diagnostic testing, emergency-room care, surgical procedures, or hospital care.

In order to preserve a role for private insurers under any new single-payer scheme, legislators would have to add co-pays for the purpose of accommodating a publicly subsidized private insurance bureaucracy. And even if the single-payer bills were rewritten along those lines, insurers would still fight them tooth-and-nail, and the program would still be branded as a Soviet death-panel scheme by the right.

But far more importantly, let’s not forget how bad co-pays and deductibles are. It’s not just that they are unnecessary for cost control: Canada and the United Kingdom provide no-deductible universal coverage and have lower overall health-care costs. And it’s not just that they squeeze family budgets, effectively worsening inequality: By deterring the use of needed care, they are also harmful to health itself, including for those with heart disease, lung disease, diabetes, and multiple sclerosis.

The presence of complementary private plans requires the erection of unnecessary financial barriers to care. Without the latter, we won’t need the former.

In other words, the only way to make room for a significant role for private insurance in the American context is to make the public system paltrier or skimpier, to impose onerous co-pays and deductibles, or to let the rich preferentially displace working-class people from hospital beds and doctors’ offices. But it doesn’t seem to make sense to punch holes in your own floor just to create work for a carpenter. That is particularly true if your floor is your health care—and your carpenter is an extractive insurance giant.