Price controls: on the one hand ... and on the other.

Controlling health care costs requires that limits be placed either on prices, quantities of services, or both. Prices are measurable and more easily controlled than is quantity and, consequently, health care cost containment has frequently focused on mechanisms for controlling prices. Regulatory approaches, however, may create market distortions and change access patterns. An alternative approach to controlling prices is to restructure the market for health services to encourage greater price competition among providers. Because this type of health reform has not previously been attempted, there is much more uncertainty about the outcome of market-oriented approaches than for direct regulatory control over prices.

Introduction

Spending on health care in the United States is very high, compared with other industrialized Nations, and is increasing more rapidly than national income. A number of policies have been proposed— and some implemented—with the goal of reducing the level of health spending and its rate of growth.

Controlling the prices of services would be a straightforward means to reduce spending—relative to the more complex task of reducing the quantity of services or developing policies that would directly regulate quality of care and future technological change. Prices are easier to measure and to monitor, whereas decisions that affect quantity and quality of care require more complex processes and subjective judgments about the value of specific services to individuals and to society. Thus, policies that would affect the price of services are perceived by many as the most manageable approach to achieving control over health spending in the short term.

Cost containment using mechanisms that affect prices could be achieved through several differing policy approaches. Although direct regulation of prices through government intervention has most often been the focus of policy deliberations, market-oriented health reform proposals also would, if successful, affect prices charged by providers. For example, under managed competition, which involves considerable government intervention in the health insurance and health care markets, insurers would have greater market power to negotiate with providers over price and quantity. Those who favor market-based approaches argue that relying on the market to determine fees would allow for greater flexibility in pricing and provide for variations in quality of care.

Proponents of direct price controls believe that the market for health services is irretrievably flawed. The presence of extensive health insurance renders consumers insensitive to the price of services. And the inability of consumers, under many circumstances, to make informed decisions leads to delegation of decisionmaking to providers who have little incentive to consider costs.

At present, there is much more evidence with which to assess the potential effects of direct price controls than the potential outcome of market-oriented approaches to control of health care prices. The prices of health services were frozen during the Economic Stabilization Program in the early 1970s, and the Medicare program imposed a freeze on physicians' fees during the mid-1980s. All-payer rate setting for hospital services has been used in several States. Under the Medicare program, hospitals have been paid a prospectively determined amount per admission based on diagnosis since 1983, and physicians are now paid under a fee schedule based on a relative value system. In addition, many States severely limit payments to providers who serve Medicaid recipients.

In contrast, market-oriented approaches would restructure the health insurance and health care markets in ways that have never before been attempted, either in the United States or other countries. Some evidence is available on some components of this strategy (for example, savings associated with enrollment in a staff model health maintenance organization). The full responses of insurers, providers, and consumers to such a major change in the financing and delivery of health services, however, would be very difficult to predict in advance.

Potential Effects of Price Controls

In the absence of any changes in the quantity or mix of services, reducing the prices of services should lead to lower total expenditures. Studies of the effects of fee freezes or price controls suggest, however, that changes in the quantity or mix of services typically do occur. For example:

More services may be provided or billing practices may change to offset the reduction in providers' revenues when prices are reduced.

Providers may substitute other services for those whose prices are controlled.

Price controls implemented for a specific population group (such as Medicare or Medicaid enrollees) may result in higher prices charged or an increase in services provided to other population groups.

When prices are controlled for only some groups, those groups may have reduced access to health care or receive lower quality services than others.

Adopting a nationwide system of price controls, such as the use of Medicare payment rules under an all-payer system (which is the topic of this issue of the Review), would avoid some, but not all, of these potential effects. In particular, price controls would have greater potential for reducing health care costs when applied uniformly to the whole health care system, because cost shifting among services and payers is less likely to be an issue. Thus, a Medicare-based all-payer system that extends to all consumers and covers both hospital and physician services would probably reduce spending by more than an all-payer system for physicians only. In addition, access to care would be less likely to be differentially affected if price controls were applied uniformly within a geographic region. Volume responses, however, would still be possible, even within a national system of controlled prices. Further, such responses could be large enough to reduce potential savings substantially, unless price controls were combined with systematic utilization monitoring and review of all providers.

A recent study by Sandra Christensen (1993) of the U.S. Congressional Budget Office (CBO) suggests that, if providers offset 55 percent of the potential savings under a Medicare-based all-payer system for hospitals and physicians, significant savings in health spending for these services could still be achieved. Had such a policy been implemented in 1991 as a stand-alone policy, with no expansion of insurance coverage to the currently uninsured, payments to physicians and hospitals would have been nearly $13 billion less and national health expenditures would have been about $10 billion lower in that year. If a Medicare-based all-payer system had been implemented in 1991 along with universal health insurance coverage, then payments to providers would have increased about $17 billion and national health expenditures would have been about $23 billion higher. Given that the cost of covering the uninsured under the current system has been estimated to be approximately $30-$50 billion, an all-payer system would generate some savings relative to expanding coverage under current payment policies. These savings, however, would be relatively modest in comparison with 1991 national health expenditures of $748 billion.

The estimates presented by CBO assume the Medicare's rates for hospitals would have been increased above current levels to cover hospitals' current average costs. These adjusted rates would be 111 percent of current Medicare rates, 78 percent of private insurance rates, and 125 percent of Medicaid rates. The net result, on average, for all patients, would be to increase hospital rates per service by about one-half of 1 percent. No adjustment to Medicare's rates for physicians is assumed and, on average, for all patients, payments per service would be reduced by about 13 percent. The estimates also assume that an increase in the volume of services provided would offset 55 percent of the potential reduction in physician revenues that would otherwise result from reducing payment rates. No volume offset is assumed for hospital services because most hospitals would experience an increase in revenues under the options considered.

The CBO analysis estimates the potential savings for a single year in which a Medicare-based all-payer system was implemented. Over time, however, adopting an all-payer system could improve control over health care costs, thereby achieving even greater savings, for several reasons.

First, if prices were regulated by the government, then the rate of increase of prices would also be controlled. Consequently, the rate of increase in health spending could be lower—assuming that the allowed price increases each year were less than the uncontrolled price increases that would have occurred and that increases in volume would offset less than the full amount of potential savings.

Second, under an all-payer system with uniform rates, it could be easier to develop a data system to examine the practice profiles of individual providers, which would then be used to identify those whose practice patterns deviated from the norm. Doing so would permit uniform monitoring of utilization on a provider basis, rather than the present utilization-management methods that require case-by-case review to determine appropriateness. Such data development could also provide an effective mechanism for limiting the growth in quantity of services under price controls. In addition, an all-payer claims data base could also provide useful detailed information on treatment patterns and their variance across geographic areas, thereby facilitating the development of practice guidelines.

Third, there would be some potential for administrative savings under an all-payer system, because providers would face only one set of rates and could use uniform claim forms. CBO estimates that, in 1991, providers' overhead would have been about $4 billion lower under an all-payer system that did not include provisions to cover the uninsured. In addition, if utilization monitoring shifted from a case-by-case review approach to profiling of providers, there could be even greater reduction in overhead expenses, as providers would spend less time justifying individual decisionmaking.

There are a number of arguments against adopting this approach, despite its potential benefits. Although total payments to hospitals and physicians by private sector payers would fall, a Medicare-based all-payer system that set hospital rates at a level consistent with observed hospital costs would result in higher spending by Federal and State governments on behalf of Medicare and Medicaid patients. If payment rates were not adjusted upward to a level that covered hospitals' costs, it is possible that some hospitals would become more efficient in their provision of services. Hospitals that were unable to reduce their costs, however, would either reduce quality of care or close down, potentially reducing access to care for some consumers.

Another argument against government-controlled prices for health services is that variations in quality of care, in regional patterns of practice, and in the geographic distribution of providers would make it difficult, if not impossible, to design a system of prices that would not distort the provision of health services. If some services were significantly underpriced, relative to the costs of providing them, access to these services would probably become difficult. Other services could be overpriced, relative to costs, giving providers incentives to offer these services even when there would be little benefit for the patient.

Even if payment rates were set at a level that would, on average, cover costs, access to care could become more difficult for some patients who, for whatever reason, were expected to cost more than average to treat. Also, providers vary in their experience, training, talents, and the quality of care they render. Consumers who wanted to pay more in order to obtain higher quality services or particular practice styles or amenities would not be able to do so. The result could be significant queuing to obtain the services of providers recognized as offering exceptionally high quality of care.

Finally, if an all-payer system were adopted as a stand-alone policy and a significant number of uninsured people remained, provisions would be needed to ensure that hospitals and other providers continued to serve those without insurance even though these providers would no longer be able to recover uncompensated costs by charging higher prices to other payers. A recent study by Harriet Komisar (1993) of the CBO found that in 1989, hospitals on average, were able to recover about 95 percent of uncompensated and unreimbursed costs through a combination of non-patient revenues, State and local government subsidies, and higher charges to private payers. If hospitals were no longer able to raise charges to private payers, access to care for those without insurance could become more difficult.

Discussion

A Medicare-based all-payer system could reduce the level of health spending and offer great potential for controlling future price increases. At the same time, establishing the correct level and rate of increase in prices to avoid distorting decisionmaking in this market would be difficult and uncertain. For some patients, access to care could worsen; for others, there could be a reduction in the choice of provider, treatment patterns, and quality of care available.

Market-oriented approaches to controlling health care costs offer less certainty about the outcomes that would be achieved, as there is little experience from which to predict the effects on prices and how these effects would differ across patients, regions, and providers. Certainly, if consumers were forced to become more price-conscious and were provided with more information, downward pressure would be exerted on both insurance premiums and on providers' prices. The speed with which these market forces would operate, and the potential reductions in prices that would occur, are difficult to assess. If market-oriented approaches to cost control were successful, however, consumers would probably retain greater ability to choose among alternative treatments, amenities, and quality of care than under direct price controls.

Thus, direct government-established price controls offer greater certainty of achieving savings, while market-oriented approaches designed to negotiate lower prices would retain greater flexibility in the system. If either approach were successful in controlling the level and rate of increase of national health expenditures, however, there would be consequences affecting all participants in the health care system. Providers' revenues would fall. Some patients would experience reduced access to some services. Fewer new jobs would be created in the health sector in the future, and employment might actually fall.

Moreover, successful cost containment would mean that some of the desirable features of the current health care system would be restricted. In particular, effective cost controls of any type could impede research and development, reduce access to new technology, and limit consumers' choices about providers and treatment alternatives. Whether these tradeoffs would be acceptable depends on whether the Nation places a higher priority on controlling costs or on maintaining other desirable characteristics of the current health care system.