Why is U.S. Health Care so Expensive?

The U.S. spends far more on health care than other rich countries. At 13.9 percent, our health spending to GDP is the highest among wealthy countries that I have been able to find data on by a substantial margin. Germany is a distant second at 10.7 percent, followed by France at 9.6 percent and Canada at 9.2 percent. The U.K. brings up the rear at 6.8 percent. Why is U.S. health care so expensive? Three perfectly simple reasons suggest themselves:

1. You get what you pay for. It is common to criticize the U.S. for its lack of “universal health care,” by which is meant politics as the ultimate decision procedure for deciding who gets what kind of health care. And, it is said, the deficiencies of the system are apparent by looking at U.S. life expectancy, which is lower than that in most OECD countries, and U.S. infant mortality, which is higher. But these outcomes are the results of choices made both by physicians and patients. Americans make many poor lifestyle choices, and attempt to save premature babies whose rescue would be politically infeasible in many single-payer countries because it would be too costly. These attempts count as U.S. infant mortality (and this variable too is affected by lifestyle choices), while the stillbirths in other countries do not. We make those desperate efforts because we can, whereas political rationing in other countries means they can’t. So the information in these statistics is questionable.

But what happens if you actually get sick? As I noted elsewhere, U.S. cancer survival rates generally exceed those of single-payer countries. (A report outlining those differences is here). Mortality rates contingent on getting a heart attack are harder to track down. According to Table 6-11 of this report from the UK, the U.S. is tied for second-lowest in mortality rates per 100,000 population, but what I am interested in is mortality once you get a heart attack, not as a percentage of the whole population. But according to Tables 6.4 and 6.14 in the same report the U.S. has a much higher rate of cardiac surgeries, from which I infer that we have many more heart attacks. To have a low death rate with what looks like so much more heart disease again testifies to the ability of the U.S. system to provide state-of-the-art care if you actually fall ill.

This is because U.S. care is in part still consumer-driven. Health-care producers respond to the desires of consumers, although consumers don’t typically seek them out until something is wrong. And so I wouldn’t expect the U.S. system to do particularly well on preventive measures, and indeed many U.S. lifestyle behaviors are much worse than in other rich countries. (One, smoking, is much better.) But once you get to the doctor’s office his incentive to provide you with quality care and a nicer experience is more compelling in a private system than a state-powered one. And so U.S. doctor’s offices might be more common and nicer on the inside, and the willingness of U.S. doctors to treat patients with respect greater.

But this is all expensive. And expensive is not necessarily bad. Cars are much more expensive than thirty years ago, but they have many more features and last much longer. Costs in the health industry are going up substantially because we are getting much better health care. Diseases that were much harder to treat now can be treated effectively (AIDS being an obvious example), and so consumers make bids on resources through the price system, and, if the opportunity cost of those new treatments is sufficiently low, they get their wish. Previously unsolvable problems like acid reflux and erectile dysfunction become treatable with breakthrough medicines. High U.S. costs are to a significant extent a function of high U.S. quality. A quick glance at the number of Canadian license plates in hospitals on the U.S. side will confirm this, as will the medical tourists we draw (particularly wealthy ones) from all over the world. On the amenities side, the U.S. system also rates very highly on overall satisfaction. In OECD surveys of how well citizens are satisfied with the quality of the system, the U.S. ranks first out of roughly 20 countries.

2, I use, you pay…higher premiums. Like all advanced countries, the U.S. relies substantially on third-party payment. If I go to the doctor, my insurance (or the government) picks up most of the tab. And third-party payment – health insurance or Medicare or Medicaid – is expensive. It is subject to moral-hazard problems, where the (un-policeable) best choice for the patient is frequently not the best one for the payer. Since other people – other insurance company customers in particular – are paying most of the cost of my health visit, I have little incentive to economize. I go to the doctor on the slightest provocation, and I want (and often get, because monitoring costs for the insurance company are so expensive) the best treatment available (e.g., branded instead of chemically identical generic medicine), even if of dubious medical marginal utility.
And this moral-hazard problem is inextricably related to…

2. I use, you pay...higher taxes. There is no obvious reason why employers should offer health insurance as a benefit, and if they do there is no obvious reason why it should be of the low-deductible variety, the kind that most encourages overuse of scarce health-care resources. But the reason they do is apparently because of a historical quirk. During World War II the U.S. government imposed wage controls, and to attract workers companies offered fringe benefits not subject to these controls, one of the most popular of which was health insurance. To this day health insurance is not taxable income in the way wages are, and so employers therefore provide versions that are far more generous than ordinary insurance schemes.

Insurance is supposed to pool risk for severe events that will not affect most people. This is how private firms can afford to insure against house fires, for example. But most health care – doctor’s visits and non-catastrophic care – is not rare, it is common. It should be budgeted for like every other expense that adults incur. Milton Friedman is said to have said that if U.S. homeowner’s insurance were run like U.S. health insurance (with the employer picking up some or all of the tab directly, and with taxpayers in turn subsidizing that purchase), it would look just as gold-plated. We would be demanding that our homeowner’s insurance pay every time the kitchen sink got clogged. The hard truth of the matter is that health care is an important good, but it does not therefore follow that we should be relying on insurance or a public single-payer system to take care of ordinary medical fees, even for the poor. Food and clothing are every bit as important as medical care, but we properly expect that the poor and everyone else should be primarily responsible for feeding and clothing themselves, rather than “insuring” against their disappearance. Medical care is no different. Indeed, even now many doctors and hospitals are willing to work out feasible repayment plans for their poorer customers, and it used to be true that doctors felt they had an ethical obligation to treat the poor regardless of willingness to pay, an obligation that has been eroding since the introduction of Medicaid in the 1960s. (The AMA was once a vigorous opponent of “socialized medicine,” and now advocates universal coverage, albeit via a “market-based” system, which allows them to work collectively and leverage their collective power when they negotiate with the government. In this they are no better or worse than every other pressure group.) To the obvious rejoinder that the poor cannot afford quality care because health care is so expensive, it can only be said that one of the reasons it is so expensive is because consumers rich, poor and in-between have so little incentive to make choices – to not get the extra medical test with almost no chance of providing useful information, to not use the expensive medicine when the cheap one will do fine.

I have refrained from mentioning two of the usual suspects – malpractice litigation and drug costs. Undoubtedly these contribute, but I suspect not as much as people think. In any case the drug problem would be easy to solve by repealing the prohibition on importing drugs sold in other advanced countries such as Canada. U.S. prices would come down, theirs would go up, and they would meet in the middle. Currently the power of the U.S. market means that its desires dominate which drugs get made, and other countries free-ride by imposing price controls in their single-payer systems. We could do that too of course, but only at the cost of further cutting the link between consumer desires and drug-company incentives. If we allow imports of drugs other countries would quickly find that their drugs were ending up in the hands of Americans, who are paying more. If they controlled drug exports, smuggling would surely follow. More probably, drug companies would negotiate a harder bargain, and the prices that foreigners pay for various drugs would more closely approximate the more market-driven ones that Americans pay. This is turn would improve the informational value of the price signals drug companies are getting.

All of this leads to a radical proposal for significantly improving the performance of the U.S. health-care system – end the deductibility of employer-paid health insurance. Consumers would immediately notice the higher out-of-pocket costs of care, but would correspondingly choose more wisely. They would also in short ordersee higher wages in compensation, which they could spend on health care or on antyhing else. They would probably be driven to choose health insurance that looks like their car insurance, which covers thefts or major accidents but not oil changes. (Think of how common it is for people not to report minor car accidents because they don’t want their rates to go up.) Such policies would have high deductibles, giving people compelling reasons to save for expenses such as sprained ankles that are not predictable at any moment but can be expected to occur from time to time, while covering most of the expense of heart surgery (which only a minority of people have over the course of their lives). The ability of the health industry to provide more value at less cost would be enhanced without the subsidy of third-party payment. Employers might still find employee demand for health insurance to take advantage of the benefits of risk pooling, but the insurance would be more rational.

But proposing this is probably political suicide, and some thought would have to be given to what would happen to people who are known to be in line for higher health costs. (A cystic fibrosis patient or a diabetic, for example.) In the meantime any policy that encourages people to take account of the costs they impose on others through their health choices is a step forward. Health savings accounts are not ideal, in that they add one tax distortion on top of another, but it is possible (though far from assured) that their better incentive structure may make up for this.