Yuri Gripas / Reuters

Whether they’re running for president or just hoping to hold on to their seats, Democratic lawmakers face growing pressure to endorse one of Bernie Sanders’s signature causes. “Doc, they keep coming—pressing me to sign onto Medicare for all,” a somewhat hesitant and confused congressman told me recently. “Should I?”

“It all depends what you mean by ‘Medicare for all,’” I said. He was hoping for a better answer than I had. About 70 percent of Americans say they support the idea—under which Medicare, the federal program that now provides health coverage for about 60 million seniors and disabled individuals, would expand to cover millions more people.

Yet Medicare for all is a messy concept. At least four different approaches to health reform could truthfully carry the Medicare for all label. Sanders’s plan is the best known, but it’s also the most politically impractical. It ignores the brutal history of repeated defeats for all Democratic health-reform proposals that try to abolish private health insurers.

In the senator from Vermont’s proposal, the federal government—Medicare—would become the primary payer for health-care services. Private insurance companies would be relegated to selling small, specialized plans for medical services not covered by Medicare. According to some bad-faith criticisms of his plan, Sanders is proposing socialized medicine—a government takeover of the whole health-care system. That’s not so. Hospitals, physician and dental practices, pharmacies, home-health-care agencies, hospices, and other providers would remain privately owned and operated under Sanders’s plan.

Other ideologically driven critics of Sanders-style Medicare for all have been stoking fears about its cost. The conservative Mercatus Center, a George Mason University think tank funded in part by the Koch brothers, has estimated that from 2022 to 2031, total health-care spending in Medicare for all would be a whopping $57.6 trillion and would add $32.6 trillion to the federal budget. Without context, that sounds astronomical. But it leaves out that the United States is already slated to spend nearly $60 trillion in those 10 years on health care—which, it turns out, is actually $2.1 trillion more than under Sanders’s Medicare for all plan. (The Mercatus Center argues those savings would be illusory, because hospitals and doctors would demand higher Medicare rates to make up for lost revenue from private-insurance payments.)

The real obstacle to Sanders’s plan is the public’s expectations. As much as Americans hate insurance companies in general, they want the right to have a love-hate relationship with their own insurer. During the battle over the Affordable Care Act, President Barack Obama promised, “If you like your plan, you can keep it.” When a handful of Americans lost their plans, the backlash was tremendous—even when the cancellations had nothing to do with the new law. The polling data today are clear: When Americans are told they might have to give up their current insurer, fewer than 40 percent support Medicare for all. That’s nowhere near enough to override the entrenched interests in health care.

More important, Sanders’s Medicare for all would put almost all private insurance companies out of business—or diminish them to runts. These companies manage more than $1 trillion in revenue per year and cover more than 175 million Americans. For Sanders and his supporters, the prospect of putting nearly all those companies out of business is a major attraction of Medicare for all. But no trillion-dollar industry has ever just rolled over and died. Insurers have experience fighting far less ambitious health reforms—and winning.

If you don’t believe insurers would effectively thwart Sanders’s plan, go to YouTube and look at the old Harry and Louise ads that ran in 1993 and 1994—the last time health insurers felt an existential threat from Washington. Those ads did the impossible, turning widely despised insurance companies into lovable teddy bears that Americans suddenly thought they would miss if Bill Clinton’s health-care-reform plan passed. Ultimately, no congressional committee ever even voted on his proposal. Sanders’s Medicare for all plan could easily suffer the same fate.

The genius of “Medicare for all” is that it explicitly links one of today’s most complicated and challenging policy issues—securing health coverage for all Americans—to the most widely celebrated achievements of Lyndon Johnson’s Great Society. Yet some versions of Medicare for all are hardly ambitious at all.

At the minimalist end of the Medicare for all spectrum is an approach known as “buy-in.” Under this plan, Medicare becomes a safety net for Americans who cannot otherwise get health insurance. One variant would allow 50-to-64-year-old Americans to pay premiums to buy into Medicare. Another flavor offers Medicare enrollment as a public option on insurance exchanges to fill the gaps in counties that don’t have enough private insurers. This type of buy-in is, frankly, a trivial plan. In a country of 325 million people, it addresses a problem of affordable health coverage for a small percentage of them. It doesn’t make health care more affordable for the 175 million Americans with private health insurance. A Medicare buy-in for all isn’t necessarily a bad idea, but it won’t energize voters, much less solve the nation’s most pressing health-care challenges.

Between Bernie Sanders and a buy-in are two more practical and politically appealing plans. One is Medicare for America, a proposal drafted by the Center for American Progress. (Full disclosure: I helped design it. I’ve also received speaking fees from groups representing insurers, hospitals, doctors, and employers.) Now before Congress in a bill by Democratic Representatives Rosa DeLauro of Connecticut and Jan Schakowsky of Illinois, this plan creates two systems—one driven by Medicare and another run by private insurance. Employers could keep offering private insurance as a fringe benefit. However, all other Americans—those with Medicaid, individual insurance, and the uninsured—would be automatically enrolled in an expanded version of Medicare. Workers who do not like their employer’s insurance could switch over to Medicare. The new Medicare would have more generous benefits, and co-pays would be on a sliding scale, with households earning under roughly $50,000 paying nothing.          

Similar to Medicare for America is a fourth approach, which could be called Medicare Advantage for all. This plan builds on the wildly popular Medicare Part C, also known as Medicare Advantage, in which people choose between a managed-care plan offered by private insurers and the traditional fee-for-service Medicare. Employers could either continue to offer private insurance or pay most of the premium of a Medicare Advantage plan chosen by their workers.

By allowing businesses to continue to offer private insurance and allowing insurers to compete in Medicare Part C, these proposals expand coverage—and the government’s role in health insurance—without threatening private insurance companies with extinction. In fact, a number of insurers might see an opportunity under Medicare Advantage for all to enroll new customers.

There are also good policy rationales to preserve a role for private insurers. While progressives often claim these companies do nothing for the health-care system but add paperwork and extract profits, this view is anything but universal. Medicare Advantage plans offered by private insurers currently enroll about a third of seniors and are the fastest-growing part of Medicare. The evidence—only 2 percent switch back to regular Medicare— suggests that seniors like these plans and, by implication, the private insurers that offer them. In addition, having multiple payers adds competition, which can improve performance and prevent the government’s health plan from ossifying. The health systems of the Netherlands, Germany, and Switzerland all include multiple competing private insurers and sickness funds.

As the 2020 campaign approaches, Democrats risk letting the future role of private insurance become a major fault line that divides the party. Advocates of Sanders’s Medicare for all dislike alternative versions that leave private insurance companies in place. But the general public’s disdain for insurers only goes so far, and to pretend otherwise is to build a health-reform plan upon a fantasy. When asked to give up their private insurance, Americans demur. A rhetorical attack on insurers might generate public applause, but that doesn’t mean Americans will support a Medicare for all plan that abolishes them.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.