Opponents of Medicare for All warn that the cost will scare people away. They speak in panicked tones about trillions of dollars in additional spending and flash alarmist chyrons across TV screens.

But the truth is that, because of the way the Congressional Budget Office (CBO) scores health care policy, any significant effort to reform health care in a progressive way is going to suffer from the exact same “big number problem” as Medicare for All.

The “big number problem” is the simple fact that the United States currently spends a lot of money on health care, some of which the CBO considers part of the federal budget and the rest of which it treats as private spending. Redirecting or reclassifying that private spending as public spending would technically result in a big increase in the federal budget.

Political operatives are so convinced this big number is going to be a political liability for Medicare for All that you have a libertarian think tank actively promoting their study showing Medicare for All would eliminate cost-sharing, cover everyone, and reduce health care spending. This is treated as bad news for the proposal simply because the study also finds that federal spending on health care would total $32 trillion over ten years.

For progressives and socialists, though, there is no way around this. It is impossible to design a health care system that both provides quality insurance coverage for everyone and doesn’t appear to vastly expand the US federal budget. Why? Because of the incoherent way the CBO decides what does and doesn’t constitute “private activity.”

The CBO doesn’t follow the simple logic of only considering something part of the federal budget if the federal government directly pays for it. Nor does the CBO follow the basic logic of considering private activity part of the federal budget if federal law requires it, like an individual/employer mandate. Instead, the CBO considers health care reform an “essentially government program” if it crosses some arbitrary line of regulation.

According to a 2009 CBO paper on the topic, “insurance purchased through exchanges or in the private market . . . should be classified as federal revenues if there is an individual mandate and tight government control of the insurance market,” but not part of the federal budget if “there is an individual mandate and an active, loosely restricted private market, and if premiums are paid through nongovernmental exchanges or directly to insurers.”

In effect, the CBO believes the government forcing you to pay premiums to insurance companies doesn’t make those premiums a tax. But if the government also mandates that private health insurers meet a basic quality floor for their plans, then it would be.

The CBO has provided only a general idea of what “tightly controlled” means, but even modest progressive reforms would cross the agency’s imaginary line. For example, requiring insurers to offer only one or two specific levels of benefits, as most countries with “managed competition” health systems do, would cross the threshold. Similarly, requiring all insurance to have an actuarial value of more than 80 percent would be a step too far, even though for an individual this would translate to a deductible of roughly $1,320 and out-of-pocket limit of $5,878. (Sixty-three percent of single-person households don’t have enough liquid assets to cover that out-of-pocket limit.)

The CBO’s quirky definitions have already had real-world consequences. During the drafting of the Affordable Care Act (ACA), senators initially wanted to mandate that insurers spend 90 percent of premium dollars on actual care (a medical-loss ratio that, again, is well below international norms). The CBO wrote them a letter warning that the regulation would push the ACA over the budget-scoring agency’s imaginary line. If they included the regulation, the CBO alerted them, it would consider the entire insurance market part of the federal budget. Legislators buckled, opting instead for a medical-loss ratio of 80 to 85 percent — a move that ended up costing the government significantly more.

It is important to look at this in an international context. There is simply no industrialized country whose health care system wouldn’t be considered a government program under the CBO’s definition. Even countries that are widely regarded as government-private hybrid systems or managed private insurance systems — such as Japan, Switzerland, the Netherlands, and Germany — have actuarial value requirements, plan standard rules, and/or medical-loss ratio regulations that would cause the CBO to score the entire health insurance market as federal spending.

While people often point to the Netherlands as an example of a regulated, fully private insurance system, benefit requirements are heavily standardized and the deductible/out-of-pocket limit is just 385 euros. In effect, the CBO considers all currently existing “private” universal health care systems as single-payer systems.

What this means is that copying any of the private systems others point to as alternatives to Medicare for All would produce the same “big number problem” as Medicare for All. The ACA is basically as far as you can go while staying within the CBO’s idea of what is private — and it clearly is insufficient. Even just modest improvements to a few existing regulations in the ACA, such as requiring that a larger percentage of insurance payments go to actual medical care, would cross the CBO’s line and end up being scored as de facto nationalization.

Strip away the complexities of budget scoring, and the upshot is clear: anyone who opposes Medicare for All because it will produce a big number from the CBO is effectively saying they are opposed to all significant health insurance reforms.