The public discussion of prospective reform of the U.S. health-care system has focused in substantial part on the question of how to extend insurance coverage to those now uninsured, and on how to deal with the attendant increased costs for the system as a whole. Some argue that a single-payer system similar to Medicare would realize savings in administrative costs sufficient to extend insurance coverage to all of the uninsured. The central objective of this study is a comparison of the administrative and other important non-benefit costs of private health-insurance plans with those of Medicare, which is used as a prototype for a large single-payer (that is, government-financed) insurance system. The central findings can be summarized as follows:
Administrative costs for private health insurance, defined broadly, are in the range of 11-14 percent of total premiums.
Administrative costs reported directly in the Medicare budget, combined with a proportional allocation of the costs of other federal government administrative functions, yield a finding of 6 percent of Medicare outlays as the total reported administrative costs for Medicare. This more complete estimate is twice as high as a proportion of Medicare outlays as commonly asserted.
A shift to a single-payer system would yield net savings of about $99.6 billion (as of 2006) annually in reported administrative costs, or about $2100 in potential health-care benefits for each of the 47 million individuals currently uninsured.
Under a single-payer system, the increase (from about $2262) in average health-care consumption by those currently uninsured would be in the range of about $1700 to $3400; this results in an annual impact on government costs, as measured, between a saving of about $19 billion to a funding shortfall of about $61 billion. The midpoint estimate thus is an approximate funding shortfall of $21 billion annually.
Accordingly, the argument that the administrative cost savings yielded by a shift to a single-payer system would be sufficient to cover the uninsured is highly problematic.
These estimates of the fiscal effect of covering those currently uninsured in a single-payer system are likely to be biased downward because not all the current health-care consumption by the uninsured is funded by the public sector; moreover, we ignore any increases in the prices of medical goods and services attendant upon an increase in demand engendered by a doubling of the population eligible for Medicare or a similar single-payer program.
In addition, the federal government must acquire revenues through a tax system that creates economic distortions, that is, that imposes economic costs upon the economy in addition to the revenues generated. The lowest plausible assumption about the magnitude of that “excess burden” of the tax system raises the true cost of delivering Medicare benefits to 24-25 percent of Medicare outlays, or about double the net cost of private health insurance.
The lower reported administrative costs for Medicare are unsurprising, in that Medicare spends substantially less on such functions as marketing, risk evaluation, claims scrutiny, and compliance with the regulatory requirements of the individual states. This does not mean that the higher reported administrative costs of private health insurance are “wasteful.” Instead, they serve the interests of consumers by reducing the extent to which insurance creates cross-subsidies among consumer classes; such cross-subsidies reduce the economic benefits of risk-pooling. Private administrative functions also impose discipline on the consumption of health-care resources, thus reducing upward pressure on insurance premiums.
In contrast to private insurance, single-payer systems must have the effect of creating and increasing cross-subsidies among patient and voter groups, because eligibility, tax burdens, and premiums are not based on health status, and the tax system prevents competition on the basis of price. Accordingly, the deeper question underlying the issue of relative administrative costs is far more fundamental: Should a health-insurance market be viewed as an institution with which risk-averse individuals and groups can pool risks efficiently? Or should it be viewed as a vehicle with which to redistribute wealth?
Note that the effort of individuals and groups to avoid the costs of subsidizing others is not a phenomenon limited to the private sector. Such competition for lower costs is a prominent feature of public finance as well, as various groups prefer—strongly—to enjoy increases in their preferred programs at the expense of others’ programs, and to shift the tax burdens necessary for public spending programs onto others. Neither private nor public health insurance in the context of allocating costs is a charitable endeavor.
No health-insurance system, whether private or public, can “cover” all individuals or all medical services because resources are limited always and everywhere. This means that both private and public health-insurance systems must impose limits on the consumption of health care: Some classes of services will be denied to patients, and some classes of patients will be denied given services. “Universal coverage” is an unattainable goal; private and public insurance programs are likely to use different criteria with which to impose those limits.
The empirical analysis presented in this study suggests strongly that the real economic cost of delivering health-insurance benefits under a single-payer system would be substantially greater—at a minimum, roughly double—than that under the current private system. Moreover, the administrative and other net costs of private health-insurance programs are very likely to be efficient in terms of satisfying the preferences of consumers. Such benefits of market institutions should not be discarded lightly.
It is clear that a social consensus—perhaps even near-unanimity—exists with respect to the proposition that to some substantial degree, health-care services ought to be made available to those who cannot afford to pay market prices. Efforts in the private sector to reduce the size and cost of cross-subsidies are efficient economically; but that says only that elimination of such efforts would not yield a free lunch in the form of lower administrative costs without adverse effects. This is not to say that subsidies for the consumption of health-care services necessarily are inefficient; again, such subsidies for, say, the poor are supported widely and thus may be efficient in terms of the preferences of consumers/voters even if delivered through the tax system rather than through private charity. The issue to be addressed is the relative virtues of alternative vehicles with which to deliver such subsidies if it is deemed appropriate to do so. Detailed examination of that question lies outside the scope of this study.
Single-payer systems inexorably must ration care and impose various types of price controls on providers, as the budget pressures attendant upon “free” (or low-cost) health care grows. A deregulated system not tied to employment, on the other hand, would resemble the markets for life insurance or long-term-care insurance: individuals would have powerful incentives to purchase such policies when young, paying actuarially fair premiums, with efficient risk-pooling in the insurance market yielding coverage for whatever level of health-care expenditures for which individual consumers are willing to pay. The problem of the poor can be addressed in such a system in several straightforward ways, among them the provision of vouchers for the purchase of private insurance plans. An alternative approach is the subsidization of private-sector competition in the provision of insurance services, an example of which is the Medicare Part D drug benefit. A reorientation of the current public debate toward such kinds of reform would be salutary.