Does the Free Lunch Live?So you say there are no free lunches? Princeton University professor and New York Times columnist Paul Krugman, referring to a recent McKinsey study, argues (February 16) that adoption of a single–payer (i.e., government) health insurance program for the U.S. would eliminate the "excess administrative costs" inherent in private health insurance, saving $98 billion annually, while "the cost of providing full medical care to all of America’s uninsured [would be] $77 billion" annually.
Benjamin Zycher, Ph.D.
Medical Progress Today
March 16, 2007
In short, there now is emerging a prominent argument to the effect that without the administrative and bureaucratic overhead burdens of the private health insurance industry, the U.S. could have universal health care "coverage" without the ruinous costs and ubiquitous rationing observed in other nations tempted by the siren song of "free" health care.
Would the U.S. experience be different? Much health care insurance coverage here is organized and provided by private insurers, who must allocate substantial resources toward such administrative tasks as the estimation of absolute and relative risks (i.e., the future expected costs) among patients and patient groups, and the enforcement of contractual limits on insurance coverage.
There is the further argument that private–sector insurers, so as to reduce coverage costs, add to administrative costs by attempting to exclude those prospectively unhealthy (an assertion, by the way, highly problematic). The insurers, in short, have powerful incentives to impose discipline on the consumption of health-care resources.
With universal coverage, on the other hand, a single payer (the federal government) would cover everyone without regard to current and prospective health status, and would cover either all health–care services (far less likely) or a list of such services deemed most cost–effective on average under some set of criteria (far more likely). And so there would remain no need to determine risks and no need to wear green eyeshades in eternal efforts to interpret the fine print in insurance contracts with respect to the coverage status of given medical procedures and goods and services. And so with this purported saving in administrative costs, a single–payer system could extend "coverage" to those now without it by eliminating profits and the interest groups that benefit from the added expenses of the current system.
Now, Professor Krugman ought to concede that the system of third–party payment (i.e., by other than the patients), encouraged strongly by the U.S. tax system, yields much of the excess cost and mini–me rationing for which everyone blames everyone else. After all, people respond to the costs and prices that they confront themselves; if someone else is picking up the tab, why not demand more and higher–quality care?
But there is little hint of this effect in his argument, other than the assertion that $77 billion annually will pay for the health care demands of those now uninsured. Annual U.S. health care spending now is about $2 trillion; since the common (but inflated) assumption about the population without insurance is well over 40 million, simple proportionality suggests that the attendant additional costs would be at least $250 billion, even assuming away price increases engendered by increased demand.
But let us shunt all that aside. Let us ignore the premise almost explicit in the Krugman argument that administrative costs incurred by private insurance programs—in pursuit of discipline in the consumption of health care resources and in avoidance of excess costs imposed upon some groups so as to subsidize others—are wasteful socially.
Let us defer to some other time examination of the curious premise that insurers seek to avoid the sick, even if insurance premiums are fair actuarially. Let us not dwell on the familiar denial of service by government insurance programs, utterly inevitable in a world in which resources are limited and demands are unconstrained by prices.
Let us focus instead upon possible factors that may bias the common assertion, upon which Professor Krugman relies, that Medicare administrative costs are about 2 percent of claims, while the comparable figure for private–sector health insurance programs is 20 percent or more. In short, several considerations allow us to hypothesize that administrative costs for Medicare in reality may not be lower than those for private insurance programs, and that the latter may be worth incurring even if they are higher than the former.
First, relative administrative costs often are measured as a percentage of claims (or average claims). Because average Medicare claims are higher than those for private insurance—Medicare beneficiaries, after all, are older and afflicted with more ailments than others taken as a group—administrative costs as a percentage of claims can be expected to be lower for the former than for the latter if such per–claim costs are independent of the size of individual claims to some nontrivial degree.
Second, some significant administrative costs for Medicare may not appear in the Medicare budget; instead, they are part of the budgets for the Social Security Administration, the Department of Health and Human Services, and other federal agencies. An example is the labor cost associated with federal management of the program, as well as the implicit rent for the attendant office space; another is the cost of computer and other such management information systems.
Third, various government policies may impose record–keeping and other such administrative requirements upon private insurers, yielding higher management costs; and other policies may shift Medicare administrative functions onto the health care providers. The former set of policies would have the effect of increasing private administrative costs as measured, while the latter would have the effect of reducing those for Medicare. The degree to which such policies are efficient is irrelevant in the context of relative administrative cost comparisons; the point here is that such policies would make simple comparisons misleading.
Finally, the economic cost of government programs is higher than the spending figures presented in formal budgets, for two reasons. First, the tax system creates economic distortions, as individuals and businesses adjust their consumption, production, and investment decisions in response to the incentives created by tax policies. In other words, the private sector gives up more than a dollar to send a dollar to the Beltway; that excess is the "marginal deadweight loss" caused by the federal tax system. Because administrative costs for Medicare must be financed with current or future tax revenues, the true economic cost of such administrative functions should reflect those economic distortion costs. Second, administrative costs for Medicare in part are financed with government debt, the interest costs of which are a current expense for the federal government not allocated by program; accordingly, the measurement of administrative costs properly should reflect some part of the interest on that debt.
The issue of relative administrative cost for federal and private health insurance is far more complex than may appear to be the case. Some substantial part of the costs borne by private programs clearly is efficient, as they are investments in the imposition of discipline in the consumption of health–care resources.
Other such costs are incurred in attempts to allocate prices (premiums) for individuals and groups in accordance with the cost of that consumption, a process inherent in most market processes but inconsistent with the central redistribution dynamic characteristic of government.
Benjamin Zycher is a senior fellow at the Manhattan Institute for Policy Research. Email: firstname.lastname@example.org.