“What would you do? Just let people die?”

“Well then, what would you do? Just let people die?”

Those words were thrown at me the first time I debated a national healthcare program for America, way back in the 1990s. Through all the years since then, I have been hearing some version of them at regular intervals.

During the debate over the Patient Protection and Affordable Care Act (ACA), that question was the ultimate resort of anyone arguing in favor of the law: whatever its problems, it was better than letting tens of thousands of Americans die each year.

There are a couple of shaky assumptions underlying the question. The first is that health insurance does a great deal to increase health and reduce mortality. This seems obvious enough, but it’s surprisingly hard to tease out of the data.

For example, the creation of Medicare, which vastly expanded utilization of health care, seems to have produced no measurable impact on mortality among the elderly in its first 10 years of existence. There is ample evidence that health insurance protects people from financial risk — which isn’t surprising, because that’s what insurance is for. The evidence that it protects people from premature death is less compelling.

Of course, the financial risk is a real problem — a health-related financial disaster can be devastating for families that go through it. But even if we also assume that there are real, and large, health benefits from providing insurance to people, that still wouldn’t mean that the ACA was better than nothing. This is the fallacious syllogism that led America into the Iraq War:

1. Something must be done.

2. This is something.

3. Therefore, this must be done.

And thus we got a bloated, complicated law that still isn’t quite working as planned. Fewer people are insured than projected, the insurance is less generous than be expected, the exchanges are in financial trouble, the federal back end to pay insurers still hasn’t been built.

Worst of all, we’ve locked in most of the features that people hated about the old system: the lack of transparency, the endless battles with insurers over what is covered and what isn’t, the feeling that you are captive to behemoth government and corporate bureaucracies that are more interested in the numbers on their spreadsheets than in what you want out of your health care.

I do think that something should have been done. But not this something. What we should have done is created a system that focused on protecting people from the risk we know they face — catastrophic medical bills — and that sought to preserve the best of the American system rather than the worst — that is, to preserve our endless talent for innovation through markets rather than our decidedly lesser talent for creating and managing massive regulatory bureaucracies.

Government as the Insurer of Last Resort

How could a government program have freed up markets to innovate? Simple: by getting the government to do something it already does decently well, which is to function as the insurer of last resort. Deposit insurance, pioneered by the United States, has basically halted bank runs. Pension benefit guarantees have made sure seniors don’t end up in penury. (The Pension Benefit Guaranty Corporation could be better financed, but that doesn’t mean the idea itself is bad.) FEMA essentially functions as an insurer of last resort for people struck by natural disasters.

These programs introduce a certain amount of moral hazard, as people take more risks and underinsure themselves in the expectation that the government will pick up the tab. But when you look at the devastation these programs have mitigated, it is hard to call them anything but an overwhelming success.

How would a similar program work for health care? The government would pick up 100 percent of the tab for health care over a certain percentage of adjusted gross income — the number would have to be negotiated through the political process, but I have suggested between 15 and 20 percent.

There could be special treatment for people living at or near the poverty line, and for people who have medical bills that exceed the set percentage of their income for five years in a row, so that the poor and people with chronic illness are not disadvantaged by the system.

In exchange, we would get rid of the tax deduction for employer-sponsored health insurance, and all the other government health insurance programs, with the exception of the military’s system, which for obvious reasons does need to be run by the government.

People would be free to insure the gap if they wanted, and such insurance would be relatively cheap, because the insurers would see their losses strictly limited. Or people could choose to save money in a tax-deductible health savings account to cover the eventual likelihood of a serious medical problem.

Advantages of the Insurer-of-Last-Resort Alternative

Of course, anyone proposing an alternative to the ACA, or to the previous status quo, has to be able to say why the alternative is better. In this case, there are three answers to that challenge. First of all, it is dead simple, and the simpler a government program is, the better it works. The ideal government program can be explained to a third grader on a postcard, and this one comes close.

The second reason this is better is that it protects people from actual catastrophic costs better than the existing system, while also being more progressive. Warren Buffett will get nothing out of the system; someone with very little income will have all medical bills paid. No one will have to worry about being slapped with an unpayable bill if, say, an anesthesiologist turns out to be out of network.

But the third and most important reason this alternative is better is that it introduces a key element that has gone missing from health care since third-party payers started to take responsibility for the bills: transparent prices, and consumers who make decisions based on them.

Milton Friedman famously divided spending into four categories, which P. J. O’Rourke summarized thus:

1. You spend your money on yourself. You’re motivated to get the thing you want most at the best price. This is the way middle-aged men haggle with Porsche dealers.

2. You spend your money on other people. You still want a bargain, but you’re less interested in pleasing the recipient of your largesse. This is why children get underwear at Christmas.

3. You spend other people’s money on yourself. You get what you want but price no longer matters. The second wives who ride around with the middle-aged men in the Porsches do this kind of spending at Neiman Marcus.

4. You spend other people’s money on other people. And in this case, who gives a [damn]?

The first category is what produces market efficiency. Unfortunately, almost no one in the system does that. Instead, we have insurers spending their money on someone else, consumers spending someone else’s money on themselves, and the government spending other people’s money on someone else. No one gets what he or she wants, vast oceans of times are wasted fighting over what to buy, and it all costs too much.

Of course, some things are too expensive to get price discipline from this system: organ transplants, very early preemies, many forms of cancer. But there’s little price discipline in those areas now, so we wouldn’t be any worse off. Meanwhile, lots of areas, from hospital beds to the details of knee surgery, would for the first time in decades be subject to the decisions of consumers who care both about getting what they need and about how much they’re spending.

Conclusion

I spent years uninsured in my twenties, and remarkably, I got the best health care of my life, because doctors stopped performing tests and procedures “just in case” and thought hard about what was necessary. I was obviously taking an enormous financial risk, and the government can usefully mitigate that. But I was also an empowered consumer rather than a number in our vast healthcare bureaucracy. A better future for American health care would be one where more people are uninsured and fewer of them are at risk.

Â