Peter Singer has an article in the New York Times Magazine, arguing that healthcare should be rationed in order to keep costs down. Peter Suderman at Reason Magazine disagrees:

[T]he QALY standard results in an essentially command-and-control approach to health-care distribution: Rather than let individual preferences and agreements work out prices and reach an equilibrium, the government simply sets the value of a year of good life for all people, without differentiating between them, and extrapolates from there. I agree that, in the end, we do have to make economic decisions about the value of life. But shouldn’t those be decisions made by individuals, their families, and their doctors? Do we really want bureaucrats in Washington handing down indiscriminate dictates on what a year of productive, healthy life is worth? Must everyone be blindly herded into the same pen?

I’m naturally inclined to agree with Suderman over Singer. But I’m not sure this argument makes sense.

As I understand it, there are two fundamental problems with healthcare. First, it is extremely expensive. Second, individual needs are arbitrary and unpredictable. Maybe you’ll get hit by a car in your early 50s and die without ever needing expensive medical care; maybe you’ll develop kidney cancer and need that $54,000 Sutent treatment just to stay alive. One way to deal with this problem would be to let each person bear the cost of his or her individual needs. But this is where the libertarian framework runs into problems. Because this might be the freedom-maximising option, in the sense that no individual is compelled to pay for somebody else’s medical problems, and each person is allowed buy as much medical care as they want. But it’s hard to square this with any abstract conception of justice, because the extremely high cost of many medical treatments makes them prohibitively expensive for even middle-class income earners who are unfortunate enough to require them. Essentially, this would be a system in which life and death decisions would be too heavily decided by luck.

As a result, every healthcare system in the world makes use of some form of risk-pooling. This can either be done through private insurance companies, government institutions like the NHS, or some combination of the two. Everyone in the pool pays money, and the car crash victims subsidise the kidney cancer sufferers. The problem, as Singer points out, is that the moment you have a risk pool, you create a situation in which the cost of each individual treatment effects everyone in the pool, not just the person who is being treated. This is why Suderman’s argument - that cost-benefit calculations should be taken by “individuals, their families, and their doctors” rather than soulless bureaucrats - seems unrealistic to me. Individuals (and their families) will always place an extremely high value on their own life. Moreover, the benefits of longer life accrue solely to them, where as the costs of treatment are spread throughout the risk pool. This creates a prisoner’s dilemma: each individual, acting rationally, will try to maximise the amount of healthcare they receive, even if the care is very expensive and buys only a small increase in life expectancy. But if everyone in the pool does this, the overall cost will become unsustainable, and the risk pool will collapse.

The unpleasant conclusion that we can draw from this is that a risk pool only works by taking away some degree of individual choice, especially at the margins. This can either be done by a government-instituted QALY system, or by an insurance company deciding that it will cover some treatments and not others. This is not an endorsement of Singer’s vision of government-provided healthcare, nor a defence of the current American healthcare system, which is problematic in all sorts of ways. My point is simply that as long as people are part of a risk pool, these decisions are never going to be entirely left to individuals.