Within the broad Democratic coalition, it’s pretty clear that the discussion of health care has shifted to the left. Mainstream figures like Senator Kirsten Gillibrand, a potential presidential candidate in 2020, are embracing single payer. Representative John Conyers’s Medicare-for-All bill currently has 115 Democratic co-sponsors in the House. And Senate minority leader Chuck Schumer recently said that single payer is now “on the table.” Assuming we have free and fair elections in the future, and Democrats regain power at some point, this is all very good news for single-payer advocates.

But that momentum is tempered by the fact that the activist left, which has a ton of energy at the moment, has for the most part failed to grapple with the difficulties of transitioning to a single-payer system. A common view is that since every other advanced country has a single-payer system, and the advantages of these schemes are pretty clear, the only real obstacles are a lack of imagination, or feckless Democrats and their donors. But the reality is more complicated.

For one thing, a near-consensus has developed around using Medicare to achieve single-payer health care, but Medicare isn’t a single-payer system in the sense that people usually think of it. This year, around a third of all enrollees purchased a private plan under the Medicare Advantage program. These private policies have grown in popularity every year, in part because the field has been tilted against the traditional, government-run program. Medicare Advantage plans must have a cap on out-of-pocket costs, for example, while the public program does not. Around one-in-four Medicare enrollees also purchase some sort of “Medigap” policy to cover out-of-pocket costs and stuff that the program doesn’t cover, and then there are both public and private prescription drug plans.

The array of options can be bewildering—it’s a far cry from the simplicity that single-payer systems promise.

At the same time, Medicare-for-All is really smart politics. Medicare is not only popular, it’s also familiar. Many of us have parents or grandparents who are enrolled in the program. And polls show that a significant majority of Americans now believe that it’s the government’s “responsibility to provide health coverage for all.”

But from a policy standpoint, Medicare-for-All is probably the hardest way to get there. In fact, a number of experts who tout the benefits of single-payer systems say that the Medicare-for-All proposals currently on the table may be virtually impossible to enact. The timing alone would cause serious shocks to the system. Conyers’s House bill would move almost everyone in the country into Medicare within a single year. We don’t know exactly what Bernie Sanders will propose in the Senate, but his 2013 “American Health Security Act” had a two-year transition period. Radically restructuring a sixth of the economy in such short order would be like trying to stop a cruise ship on a dime.

Harold Pollack, a University of Chicago public-health researcher and liberal advocate for universal coverage, says, “There has not yet been a detailed single-payer bill that’s laid out the transitional issues about how to get from here to there. We’ve never actually seen that. Even if you believe everything people say about the cost savings that would result, there are still so many detailed questions about how we should finance this, how we can deal with the shock to the system, and so on.”

Achieving universal coverage—good coverage, not just “access” to emergency-room care—is a winnable fight if we sweat the details in a serious way. If we don’t, we’re just setting ourselves up for failure.

Centrist Democrats will no doubt be one obstacle to universal coverage, but a more fundamental problem is that compelling the entire population to move into Medicare, especially over a relatively short period of time, would invite a massive backlash.

The most important takeaway from recent efforts to reshape our health-care system is that “loss aversion” is probably the central force in health-care politics. That’s the well-established tendency of people to value something they have far more than they might value whatever they might gain if they give it up. This is one big reason that Democrats were shellacked after passing the Affordable Care Act (ACA) in 2010, and Republicans are now learning the hard way that this fear of loss cuts both ways.

“Remember how much trouble President Obama got into when he said that if you like your insurance you can keep it?” asks Pollack. “For something like 1.6 million people, that promise turned out to be hard to keep. And that created a firestorm.” Those 1.6 million people represented less than 1 percent of the non-elderly population, and most of them lost substandard McPlans which left them vulnerable if they got sick. The ACA extended coverage to almost 10 times as many people, but those who lost their policies nonetheless became the centerpiece of the right’s assault on the law. Trump and other Republicans are still talking about these “victims” of Obamacare to this day.

Under the current Medicare-for-All proposals, we would be forcing over 70 percent of the adult population—including tens of millions of people who have decent coverage from their employer or their union, or the Veteran’s Administration, or the Federal Employees Health Benefits Program—to give up their current insurance for Medicare. Many employer-provided policies cover more than Medicare does, so a lot of people would objectively lose out in the deal.

Some large companies skip the middle man and self-insure their employees—and many offer strong benefits. We’d be killing that form of coverage. If we were to turn Medicare into a single-payer program, as some advocates envision, then we’d also be asking a third of all seniors to give up the heavily subsidized Medicare Advantage plans that they chose to purchase. Consider the political ramifications of that move alone. And because some doctors would decline to participate in a single-payer scheme, which would come with a pay cut for many of them under Medicare reimbursement rates, we couldn’t even promise that if you like your physician you can keep seeing him or her.

Don’t be lulled into complacency by polls purporting to show that single payer is popular—forcing people to move into a new system is all but guaranteed to result in tons of resistance. And that’s not even considering the inevitable attacks from a conservative message machine that turned a little bit of money for voluntary end-of-life counseling into “death panels.” Public opinion is dubious given that nobody’s talking about the difficulties inherent in making such a transition.

It’s true that every other developed country has a universal health-care system, and we should too. But make no mistake: Moving the United States to national health care would be unprecedented, simply because we spend more on this sector than any other country ever has.

“For the most part, these countries were spending maybe 2 or 3 percent of [their economic output] on health care when they set up these systems after World War II,” says Dean Baker, co-director of the Center for Economic and Policy Research. “Most of them are spending 8 or 9 or maybe 10 percent of their [output] now, and this is 70 years later.”

In 2015, the United States spent 17.8 percent of its output on health care. The highest share ever for an advanced country establishing a universal system was the 9.2 percent that Switzerland spent in 1996, and they set up an Obamacare-like system of heavily regulated and subsidized private insurance. (They also spend more on health care today than anyone but us.)

There’s a common perception that because single-payer systems cost so much less than ours, passing such a scheme here would bring our spending in line with what the rest of the developed world shells out. But while there would be some savings on administrative costs, this gets the causal relationship wrong. Everyone else established their systems when they weren’t spending a lot on health care, and then kept prices down through aggressive cost-controls.

“Bringing costs down is a lot harder than starting low and keeping them from getting high,” says Baker. “We do waste money on [private] insurance, but we also pay basically twice as much for everything. We pay twice as much to doctors. Would single-payer get our doctors to accept half as much in wages? It could, but they won’t go there without a fight. This is a very powerful group. We have 900,000 doctors, all of whom are in the top 2 percent, and many are in the top 1 percent. We pay about twice as much for prescription drugs as other countries. Medical equipment, the whole list. You could get those costs down, but that’s not done magically by saying we’re switching to single payer. You’re going to have fights with all of these powerful interest groups.”

Baker is himself a single-payer advocate, and he’s worked with various groups that advocate for it, but, he says, “I don’t think you can get there overnight. I think you have to talk about doing it piecemeal, step-by-step.”

Single-payer advocates are mostly right about its benefits. These systems are simpler, they cut down on administrative costs, and they cover everyone.

But the term “single-payer” is itself misleading. The truth is that many of the systems we refer to as single-payer are a lot more complicated than we tend to think they are. Canada, for example, finances basic health care through six provincial payers. Its Medicare system provides good, basic coverage, but around two in three Canadians purchase supplemental insurance because it doesn’t cover things like prescription drugs, dental health, or vision care. About 30 percent of all Canadian health care is financed through the private sector.

Most countries have mixed funding schemes that vary in complexity, and the term “single-payer” may be giving some people a false promise. Conyers’s Medicare-for-All bill promises to cover virtually everything while banishing out-of-pocket costs, but no other health-care system offers such expansive benefits. Even people living in Scandinavian social democracies face out-of-pocket expenses: In 2015, the most recent year for OECD data, the Swedes covered 15 percent of their health costs out-of-pocket; in Norway, it was 14 percent and the Finns shelled out 20 percent out-of-pocket.

Most countries with “single-payer” systems rely on some combination of public insurance, various mixes of mandatory and voluntary private insurance (usually tightly regulated), and out-of-pocket expenditures (often with a cap). They offer free coverage for those who can’t afford it, but the exact benefits vary from country-to-country.

Germany’s “single-payer” system has 124 not-for-profit insurers participating in one national exchange. About 10 percent of Germans—the wealthiest ones—opt out of the national system and go fully private, and most of them buy plans from for-profit insurers.

The Dutch system is somewhat like Obamacare in that everyone must purchase insurance for basic services from private insurers. But the similarities end there: Insurers are barred from distributing profits to their shareholders, and a separate, entirely public scheme covers long-term care and other costly services. Premiums are subsidized, but most Dutch people purchase supplemental insurance to cover things like dental care, alternative medicine, contraceptives, and their co-payments.

The French system is often cited as the best in the world, and about a quarter of it is financed through the private sector. The French are mostly covered through nonprofit insurers in a single national pool, but most working people get their policies through their employers. Almost all French citizens either purchase government vouchers to cover things like vision and dental care, or are provided with them gratis if necessary. The system is financed through a complicated mix of general revenues, employer contributions, payroll taxes and taxes on drugs, tobacco, and alcohol.

So the United States isn’t unique because it uses a mix of public and private financing—the big difference, as these OECD data show, is that we rely much more heavily on private insurance than any other wealthy country.

Understanding that other countries’ schemes vary significantly in the details—and that in the United States, the cost of care would remain a serious challenge under any system—should lead to a different conversation among progressives. Rather than making Medicare-for-All a litmus test, we should start from the broader principle that comprehensive health care is a human right that should be guaranteed by the government—make that the litmus test—and then have an open debate about how best to get there. Maybe Medicaid is a better vehicle. Perhaps a long phase-in period to Medicare-for-All might help minimize the inevitable shocks. There are lots of ways to skin this cat.

At a minimum, it’s time to get past the idea that anyone who doesn’t embrace Medicare-for-All, as it’s currently defined, must be some kind of neoliberal hack.

An obvious alternative to moving everyone into Medicare is to simply open up the program and allow individuals and employers to buy into it. We could then subsidize the premiums on a sliding scale. But recent experience with the ACA suggests that this kind of voluntary buy-in won’t cover everyone, or spread out the risk over the entire population.

There are other alternatives, one of which was a popular progressive scheme before Barack Obama tried to tackle health care.

Progressive critics of the ACA are partially correct when they say that Democrats passed a plan that looked similar to “Romneycare” in Massachusetts. In the end, it was at least structurally similar to a scheme originally cooked up by the Heritage Foundation, thanks to a handful of Democrats—led by then-Senator Joseph Lieberman—who killed the public option.

But the sausage-making began well to the left of anything Heritage would ever countenance, modeled in part on Yale political scientist Jacob Hacker’s “Health Care for America” proposal. It would have left employment-based insurance—and Medicare coverage for the elderly—intact, and created a large new Medicare-like public insurance program that would have been far more robust than anything contemplated during the development of the ACA.

Hacker still thinks that, in broad terms, this is the best approach. He calls it “Medicare for More,” and hopes that it would do a better job at containing costs than employer-based insurance. Then, by creating a kind of virtuous cycle, there would be more buy-ins which would ultimately lead to “Medicare for Most.”

“In other countries, you’re basically guaranteed coverage and then they figure out how to pay for it,” he says. “Some of that money may come from you, some will come from your employer and some of it will come from general funds. We don’t have that approach. People who don’t have coverage from their employers have to figure out how to sign up—either for Medicaid, or through the exchanges. Yes, we have a penalty to encourage people to do that, but they still have to navigate this incredibly complex system.”

With Hacker’s program, perhaps to be called Medicare Part E, employers would have a choice of providing their employees with coverage as good as they would get in this big public insurance pool, or buying into the scheme. Premiums would vary based on workers’ incomes.

Hacker says he has various ideas for bringing people who aren’t attached to the labor force into the system. One possibility would be to automatically enroll everyone at birth, and cover them until they have a choice of switching to an employer-based plan. Call it Medicare-for-All-Who-Need-It.

“While the savings would be larger if everyone participated in a single pool, they’d still be significant,” says Hacker.

And maybe there’s a better way still that hasn’t yet been discussed. The fight for a universal health-care system in the United States is now in its 105th year, and if we don’t admit that financing any kind of universal system is going to be especially difficult given how much we spend, or acknowledge the role that loss aversion plays in the politics of reform, then we’re going to fail again the next time we get a shot at it.

Above all, progressives need to learn something from the Republicans’ effort to replace the ACA. They promised that facile slogans like “freedom” and “choice” would magically increase coverage and bring down costs. They were selling snake oil, and one way or the other, it’s going to come back and bite them.

We shouldn’t make promises that we aren’t going to be able to keep. “It’s not going to be easy to do,” Jacob Hacker says, “and anyone who tells you that the most expensive health-care system in the world is going to undergo a sudden shift to highly efficient and low-price medicine has not been studying American medicine.”

Editor’s Note: This article initially misrepresented the percentage of GDP spent by Switzerland on healthcare in 1996. Public spending on healthcare was 5 percent of GDP, as the article initially stated, but total spending was 9.2 percent. The article has been updated to correct the error.