The US Health Care System: Administrative Costs

Why are health care costs so high in the US? Some possible explanations include medical malpractice insurance, improving (but also more expensive) medical technology, the desire by people to simply spend more of their income on health care services as their income rises, and large and increasing administrative costs.

If he addresses the question of high medical spending at all in his State of the Union address tonight, Bush is likely to focus solely on the medical liability insurance explanation (though I’m willing to be pleasantly surprised on this). His proposed solution to this problem is tort reform. Unfortunately, as I’ve written about previously, non-partisan estimates suggest that tort reform is likely to produce only a small decline in health care costs, of perhaps $6 billion per year, out of total medical care spending of $1,878 billion in 2004. That’s not much of an impact.

So let’s consider a more substantial contributor to the US’s extraordinary health care spending: administrative costs.

How Big is the Bureaucratic Burden?

We have a couple of estimates of how high administrative costs are – i.e., expenses incurred by the health care system to do things other than to provide health care services. One prominent study that appeared in the New England Journal of Medicine in 2003 estimated that the cost of administering the US’s health care system was about $300bn in 1999. A more recent study in the International Journal of Health Services found that in 2003, administration costs in the US health care system ate up about $400bn, or about 25% of total health care spending.

By comparison, national health care systems incur administrative costs of a few percent of total health expenditures: according to the NEJM study Canada’s national health insurance system spends just 1.3% on overhead, and the US’s Medicare and Medicaid programs have administrative costs of between 2-5%.

In addition to being large in an absolute sense, administrative costs have been rising faster than other health care costs in recent years. The Medicare and Medicaid Actuary’s Office estimates that the direct administrative costs of private health insurance plans (including profits) rose from about $40bn in the late 1990s to about $95bn in 2004. The following chart shows these costs represented in dollar terms and as a percent of total private health insurance payments.

Note that these estimates only measure the direct costs paid by the insurance plans themselves, however, and exclude the bureaucratic burdens faced by doctors, hospitals, etc. to deal with insurance paperwork. The NEJM study estimated that hospitals and doctors devote roughly one-fourth of their resources to dealing with insurance paperwork (no time series of these costs exists, unfortunately). Indirect administration costs in the US health care system are roughly four times the direct administrative expenses incurred by the insurance companies themselves.

All in all, the IJHS study concluded that the US health care system could probably eliminate 75% of its administrative costs by switching to a single-payer system (see this document for one good example of what such a system might look like). The excessive bureaucratic burden (i.e. over what it could have with a single-payer system) of the US’s health care system seems to be in the neighborhood of $300bn per year. That’s some real money. And it’s growing, not shrinking.

What Explains the Excessive Expenses?

Economic theory provides several fairly clear and convincing explanations for why private health insurance plans spend so much money on administration, including economies of scale (or lack thereof), burden-shifting, and selection.

First of all, private health insurance plans are far smaller than national health insurance plans, and thus have much less ability to reap economies of scale. For the same reason, the relative administrative costs of the Medicare system have steadily been falling over time.

Secondly, private health insurance plans have a strong financial incentive to try to shift as much of the costs of each insurance claim on to individuals, providers, and other health insurance providers. It therefore makes sense for them to devote substantial resources to the task of trying to avoid paying claims that are brought to them. One example of this effect is how insurance companies go over claims with a fine-toothed comb to try to deny them whenever possible. A national government-run insurance plan would have no such incentive, since there would be no one to try to shift the burden to.

Finally, private plans also have a strong financial incentive to try to exclude high-cost individuals from their plans. It therefore makes sense for them to devote a lot of resources toward vetting potential enrollees and screening out those that they guess will have large claims. Again, a national insurance plan has no such incentive, because by definition it is set up to insure everyone.

These are just some of the ways in which the market for health care is rife with what economists call “market imperfections” and “market failures”. And as every economist learns in their first year in graduate school, when there are market failures and imperfections, the private market outcome is not likely to be the most efficient outcome, and government intervention (or even provision of the good) will probably produce far better results. Unfortunately, the evidence from the US’s health care system seems to suggest that economic theory is exactly right in this case.


UPDATE: For more about some implications of this for the US, see my next post: “Moving to a Single-Payer System“.