The problems of the health-care market begin with its reliance on third-party payments. Health providers are typically paid by insurance companies or governments, not directly by consumers. Thus, the typical consumer regards most of an additional dollar of health services as free. Not surprisingly, this environment promotes excessive demand and high costs.
Improvements to this structure are difficult partly because our society wants a system in which an individual's health care does not depend on an individual's income. Moreover, the quantity of care is not supposed to depend on costs. This makes it hard for the health sector to function in accordance with usual market principles.
Frankly, I do not understand why health care is viewed so differently from other consumer goods, including such essentials as food, clothing, and shelter. I accept that health services are difficult for consumers to evaluate, though the workings of cars and computers are equally incomprehensible to most people. The difficulties in evaluating medical treatment do suggest that people seek expert advice from third parties, which might look in some respects like HMOs or government agencies. But it is hard to see why this complexity means that incomes and prices should not play the role in the health-care market that they play in other markets.
Although the implementation of a free market in health care seems unattainable, there are less radical steps that could improve efficiency. One idea relates to health insurance coverage, which is often viewed as insufficiently widespread. The reality is the opposite for most people who have coverage -- the amount of their coverage is usually far too large. Typical health-insurance plans have modest deductibles and co-pays, even for relatively low expenses, such as office visits to doctors. As insurance, coverage should emphasize catastrophic situations that can represent major blows to a person's standard of living. For smaller outlays, the health-care market would work better if consumers paid for the services they received.
One important reason for excessive coverage is that health insurance benefits provided through employers are not taxable income. The government effectively subsidizes health insurance, which leads to too much coverage and, therefore, excessive demand for medical services. Thus, I have finally found something I would like to tax -- health insurance benefits. Most of these are fringe benefits paid for by employers, but small amounts that are now tax-favored are paid for by employees and the self-employed.
Similarly, Medicare has too much coverage, and the extension to include prescription drugs compounds the problem. To maintain incentives not to overconsume drugs, coverage should be subject to a substantial deductible, something like $5,000 per year. The only rationale for a more generous program through Medicare is that it redistributes income from the general taxpayer to the elderly, who benefit substantially from the availability of new drugs. I understand the political attractions of giving more money to the elderly. However, since the typical recipient of Medicare is not poor, I see no equity argument for this redistribution.
The one structural change to Medicare that the Administration had been pushing but is now deemphasizing involves the expanded use of private managed-care plans to provide prescription-drug coverage. Although this proposal is intended to achieve cost savings through expanded reliance on private providers, there's not much reason to think this plan would save money. Mark Duggan of the University of Chicago has studied the use of private providers in analogous programs -- state-run Medicaid systems that provide health-care coverage for low-income persons ("Does Contracting Out Increase the Efficiency of Government Programs? Evidence from Medicaid HMOs," National Bureau of Economic Research, August, 2002). Nationwide, the percentage of Medicaid recipients in managed-care plans rose from 10% in 1991 to 56% in 2000. Duggan found that in California managed care led to a substantial jump in government outlays. There were no discernible effects on health outcomes, though researchers typically have trouble isolating health effects for any kind of program or treatment. In any event, the Medicaid evidence suggests that managed-care plans will not be a big part of a solution for fixing Medicare. By Robert J. Barro