Individual Health Accounts: An Alternative Health Care Financing Approach

Abstract

After examining the major determinants of inefficiency in health care markets and several recent proposals to correct these problems, this paper introduces a market-oriented alternative which could be highly efficient while meeting all the established goals of a national health plan. To achieve these objectives, traditional forms of insurance would be replaced by a system with the following characteristics: (1) Instead of buying insurance, individuals and their employers would be required to contribute into individual health accounts from which each family would pay for medical care; (2) Once accumulations attain a designated level, any excess accumulations are distributed to the individual; and (3) A national health fund is established to support those without regular accumulations or those whose accounts have been depleted.

This paper develops these principles to show how everyone would have access to care as well as the financial security normally associated with comprehensive insurance. But, by inducing many patients to behave as if they were paying for the full cost of care through reductions in potential earnings from their accounts, the paper explains how significant savings in total spending could also be achieved.

Public policymakers have been faced with a serious dilemma over the direction of national health care policy. On the one hand, there has been continuous support from various groups of constituents for measures which, at a minimum, would fill the “gaps” in the current mixture of private and public health insurance. To some, nothing short of a comprehensive national health insurance plan with universal coverage is acceptable. On the other hand, there has been great concern, most clearly manifested by the numerous legislative attempts at cost containment, with the rapid growth in health care spending. In 1979, per capita expenditures for personal health care were $838 and total expenditures accounted for 9 percent of the gross national product (Health Care Financing Administration, 1980). With the growing belief that more extensive third-party coverage is a principal force behind these rising costs, it is no wonder that public officials appear so indecisive.

This article suggests a method of restructuring the financing of health care to substantially reduce the conflicting pressures between rising costs and the established goals of eliminating financial hardships, while ensuring all with access to care. The plan achieves these results by providing patients with strong incentives to reduce their use of marginally beneficial services—behavior which is not encouraged with the nearly first-dollar coverage of conventional insurance. Such changes in consumer attitudes may also lead to greater provider competition and further reductions in spending. Unlike a number of insurance proposals which try to meet similar objectives by, among various means, increases in deductibles and coinsurance rates, this program rewards economy without requiring increases in consumers' out-of-pocket costs. These features are especially important if a proposal is to be both efficient and acceptable to broad segments of society.

To help readers understand and evaluate the plan, the first section of the paper develops a framework which distinguishes between the different kinds of inefficiencies which are thought to be reflected in our current system. Within this context, the second section examines a number of alternatives which have received wide attention in recent years. A discussion and analysis of our proposal follows.

Efficiency, Inefficiency, and Expenditures

Increasing academic concern with the economics of the medical marketplace has coincided with our nation's rapidly rising health care spending. Although this interest may have been precipitated and fueled by the ever larger share of resources taken up by health expenditures, the greater part of the scholarly literature is directed not so much at spending as such, but at potential inefficiencies of medical markets. Since it is widely believed that these inefficiencies have raised spending relative to competitive norms (although it is possible to have levels of consumption which are inefficiently too low), correcting these problems is seen as the means to control expenditures, while resources are freed for other socially more useful purposes.

For reasons easily understood, much of the recent research has focused on the problems created by insurance, especially the increasingly prevalent forms of first-dollar insurance. With comprehensive coverage for even minor risks, patients and their physicians are not penalized for using services. Treatment will be encouraged and readily accepted so long as there are any associated benefits, even though the expected benefits may not outweigh the real costs of those services. In addition, some researchers have argued that insurance has encouraged more expensive forms of treatment and has also led to reimbursement methods and other practices which have nullified the efficiency of market forces.

Less well understood, however, are the important distinctions between various classes of inefficiencies. Some inefficiencies, for example, may affect the quantity and kinds of resources which are used, while others may affect prices. Some may not be related to insurance per se but only to reimbursement practices which have been followed by insurers. Since any proposal is likely to have an uneven impact on each of these problems, it is important to understand the concept of efficiency and the various distortions which may exist in medical markets.

These distortions are illustrated in Figure 1. If D₁ represents the demand curve for a group of informed consumers and S₁ represents the long run competitive supply curve, the equilibrium unit price and quantity are 0P₁ and 0Q₁, respectively. At this efficient solution, the level of expenditures is represented by the rectangular area 0P₁EQ₁.

Figure 1. The Market for Medical Care.

For medical care, though, consumers may not be well informed, relying instead on the suggestions of providers; markets may not be highly competitive; and because the incidence of illness is often unpredictable and the costs of treatment potentially large, most individuals are now covered by some form of insurance. By typically reducing the patient's copayment, often to zero, insurance increases the demand for services. For simplicity, taking the limiting case where there are no deductibles or copayments, the demand curve of the well informed patients (the consequences of a relaxation of this assumption will also be examined) in Figure 1 would be rotated around Q₂ to D₂. Thus, although prices may remain at P₁, patients behave as if the good is free and consume the quantity 0Q₂. Expenditures are now indicated by the larger rectangular area 0P₁FQ₂. This additional quantity Q₁Q₂ resulting from a decrease in the cost to the patient is called “moral hazard,” and is commonly thought to represent inefficient use (benefits to patients that are not worth the price paid through insurance premiums or taxes).¹ A major goal of many proposals is to reduce the extent of moral hazard.

Insurance practices, however, have also introduced other distortions which could further raise expenditures. If reimbursement tends to be cost-related, which it is for many services, the intensity with which care is provided is not restrained. As noted by Newhouse (1978, p. 63), hospitals will add equipment and staff regardless of cost, so long as there is any possibility of increased production.

Research on the development of new products is similarly biased because the potential projects do not necessarily have to meet any market test in terms of expected benefits and costs (Newhouse, 1978, p. 64). Also, when cost reimbursement is combined with a minimum of cost sharing, consumers have little incentive to shop around. Without the discipline of the marketplace, competition is reduced and increasing amounts of what is called X-inefficiency (that is, wasteful and sloppy managerial practices) are tolerated.

Each of these forms of behavior raises the cost of care. In Figure 1, the combined effects of these phenomena are shown by an upward shift in the supply curve from S₁ to S₂ such that the price of a unit of care increases from 0P₁ to 0P₂. Total expenditures increase to 0P₂GQ₂.

The distortions in health care markets are not limited to those we have just described. An important characteristic of a competitive market, the significance of which is usually overlooked, is its tendency to produce the appropriate capacity level—that which is just necessary to meet the demands of consumers. In Figure 1, if patients want the quantity 0Q₂, which includes the moral hazard component, that quantity will generally be available. No extended periods of substantial overcapacity or undercapacity will occur.

With restrictions on entry, the health system may not be capable of providing the quantity of services patients would want or which physicians feel is in the best interests of their patients. Providers will have to impose some rationing mechanism to cut down on the number of patients they see or the intensity of care they render. The less than perfect information shared by consumers, and consequent reliance on providers, may make it possible for nonprice rationing through supplier control over demand to continue for long periods of time (as may have happened through most of the 1950's and 1960's).

Conversely, if there is excess capacity (that is, greater capacity than that necessary to provide 0Q₂ in Figure 1), competition would normally drive some providers out of the industry until the excess capacity is eliminated. But where consumers rely heavily on providers for information and advice, some evidence suggests that demand will be created through the recommendation and provision of services which a fully informed patient may not want. The demand curve D₂ would be shifted rightward to D₃, increasing spending to the rectangular area 0P₂HQ₃. For at least some specialties, and in some parts of the country with large concentrations of physicians, supplier induced demand of this type may be a significant problem (Fuchs, 1978).

Because of a combination of the phenomena previously described, the fundamental economic question is the difference between the actual inefficient level of spending as represented by 0P₂HQ₃ in Figure 1 and the socially efficient level as shown by the original rectangular area 0P₁EQ₁. If this excess spending (that is, the shaded area in Figure 1) adds no more than 5 to 10 percent to the original (unknown) amount, there would be relatively little interest with all the attendant dangers in radical changes to our present system. The concerns of many, however, are undoubtedly motivated by the belief that these and other problems likely account for much larger amounts, and that it is in our national interest to seek out and evaluate any promising alternatives.

Recently Proposed Solutions

Two broad strategies are being suggested for dealing with the cost problem and for reorganizing our health care system. One strategy involves increased regulation and planning, and includes the possibility that care may be rationed by reductions in the capacity of the health care system. The alternative is a strategy designed to increase competition in health care markets by such means as: 1) encouraging the development of health maintenance organizations; and 2) making patients more cost conscious, principally by increased cost sharing. While the first approach seems to have been favored, especially by health planners and legislators in the 1970s, there now appears to be a growing interest in options which would limit public intervention in the health sector. The costs and questionable success of existing regulation of the health (as well as other) industries justify a thorough search for the widest array of market-oriented options.² This section will briefly summarize and evaluate, within the context of the analytical framework we have established, several market solutions which have received wide public attention.

The simplest of these is Feldstein's (1977) major-risk approach. Under this plan, every family would receive major-risk coverage but would be responsible through deductibles, copayments, or both, for out-of-pocket costs which are relatively large (for example, up to 10 percent of a family's income). Existing tax subsidies for the purchase of supplemental first-dollar insurance coverage would be eliminated. By increasing the patient's share of total costs and eliminating incentives for shallow insurance coverage, the proposal is principally directed at the moral hazard component of use illustrated in Figure 1. However, even if this plan would be effective in curbing consumption of low benefit medical services, it fails to deal directly with any of the other inefficiencies previously described. Furthermore, despite the likelihood that total expenditures on health care would be decreased, thus reducing the average individual's true burden (consisting of insurance premiums, out-of-pocket costs, and tax related costs), the public may not understand the nature of these “savings” and is likely to fight against the higher out-of-pocket costs required by the plan. Other critics have also argued that the proposal is inequitable by introducing in their views excessive cost sharing for lower income groups.

Havighurst (1977, 1979) similarly would eliminate Federal tax subsidies for the purchase of insurance. The basic problem, he believes, lies in insurance with little control on use (moral hazard) or levels of reimbursement (the upward pressure on unit costs described in Figure 1). Without the existing tax subsidies, Havighurst argues, insurers would have adopted practices found elsewhere in the insurance industry which include: higher deductibles and copayments; limits on payments through indemnity schedules; utilization review; and exclusions of highly discretionary and experimental forms of treatment. However, his faith in the effectiveness of a voluntary solution which truly minimizes government involvement may not be widely shared. Also, the chances of eliminating tax subsidies without providing some other immediate rewards which are easily understood by the public seem very remote.

A more complex and fundamentally different approach has been advanced by Enthoven (1978, 1979). Because physicians are either directly or indirectly responsible for managing most resources for patient care, Enthoven believes that “The key issue in cost control is how to motivate physicians to use hospital and other health care resources economically.” (1979, p. 25). His Consumer-Choice Health Plan (CCHP) is intended to produce rewards for both patients and providers who do so.

The vehicle for implementing CCHP is a tax credit received by each family for the purchase of insurance. The credit will be some proportion of the actuarial costs in providing that family with care. These costs depend on the benefits included in the particular plan selected as well as each family's demographic characteristics and geographic location. Low income families would receive subsidies, which decrease with increasing incomes, for insurance premiums, and Medicare would be retained for the elderly.

The primary objective of CCHP is to increase competition and the alternatives available to the public by forcing providers to compete for consumers through more efficient forms of organization.³ This pressure on providers emerges from the incentives consumers have to search for lower cost but quality care because they pay a proportion of their health plan's expected total cost. Families which choose plans and providers that encourage economy will be able to reduce their out-of-pocket expenditures. Plans which are able to control costs and meet this market test will proliferate, driving down overall health care spending.

Specifically, in terms of Figure 1, Enthovan implies that through the self interests of both consumers and providers, each component of the inefficient excess level of expenditure will be reduced such that the outer rectangular area is squeezed in toward the efficient inner area. The uncertainty of whether the scenario Enthoven describes under CCHP would emerge is probably the proposal's major weakness. As Rushefsky (1981) explains, many critical assumptions are implicit in the plan, each of which would have to be realized if it is to be successful. In addition, he points out that although CCHP is proposed as a competitive alternative to regulation, an extraordinary level of government data gathering and monitoring would be required.

An Alternative Solution

Although we have been concentrating on efficiency criteria, any viable alternative program must also address certain equity and administrative considerations. Davis (1975) and Feldstein (1977), leading representatives of two interest groups with fundamentally different positions on many health issues, agree that an acceptable national health plan should meet the following objectives:

• Provide all with access to care;

• Eliminate financial hardships;

• Control or limit increases in costs;

• Be administratively simple; and

• Be widely accepted by the public and providers of care.

At first glance, it appears impossible to achieve the first two goals without extensive insurance coverage for all regardless of ability to pay. This in turn would aggravate the cost and efficiency problems unless substantially more cost sharing is included. Such measures, though, would be politically unpopular and meet strong public opposition.

However, the apparent need for insurance, and for first-dollar coverage if out-of-pocket costs are to be avoided, rests on the false premise that insurance (defined narrowly as third-party coverage) is the only device which can reduce the kinds of financial risk which most individuals clearly wish to avoid. Under current methods of insurance, no relationship exists between an individual's consumption of health services and premiums received by the insurer from that individual. Only for the group of insureds will premiums be sufficient to cover expected claims and other costs. This is an inevitable outcome of uncertainty and the pooling of risks in the conventional way. Consequently, there are large redistributions from healthy to sick persons, with no special financial rewards for avoiding the use of services, as well as the major flaws of overconsumption of health services and inefficiencies arising from a minimum of cost participation by patients.

It is possible, though, to provide the security associated with insurance, but with substantial reductions in insurance as measured by third-party payments while, at the same time, creating significant rewards for efficient and cost saving behavior by both patients and providers. To achieve these conditions, our proposal would abandon current methods of financing health care in favor of a system with the following three basic elements (specific details will be discussed later in this paper).

Individual Health Accounts

Rather than contributing premiums into an insurance pool, employers and individuals would be required to contribute, subject to both minimum and maximum amounts, into individual (interpreted as an appropriately defined family unit) health accounts handled by a qualified administrative organization. The functions of the administrator are: to credit the individual's account for contributions received; to invest the balance in suitable assets; and to debit the account for claims for services received by the individual at terms agreed to by the individual and the provider.

Distributions of Accumulations

The individual enjoys property rights to the accumulations and will receive distributions once they reach a minimum level (such as $4,000 for a family of four). Any excess accumulations are distributed to the individual on a regular basis.

National Health Fund

If, on the other hand, the fund is depleted by health care purchases, or for those who have no wage earnings, payments are made from a national health insurance fund (NHF).

The essence of the proposal is to induce individuals to act as if they were paying the full cost of medical care, (which in many instances would be true, since payments reduce their wealth and potential gains from distributions), but to avoid any possibility of individuals being denied care or being unable to pay for medical bills. Since health care would no longer be “free,” but would approach the true price, patients are financially motivated to economize on their use, especially of expensive forms of care with relatively low expected benefits. Among other efficiencies resulting from the plan, patients will have incentives to substitute less costly for more expensive forms of treatment and to search for lower cost providers. These changes in consumers' behavior will also put downward pressure on the prices of services by increasing competition in the marketplace, as providers will be forced to compete more strongly for cost conscious patients. Competitive forces will also require providers to become more efficient in their use of medical inputs, and to offer patients more options as the existing biases toward expensive care, encouraged by attitudes that cost does not matter, are replaced by significantly greater cost awareness.

Like Enthoven's proposal, the plan is designed to diminish each component of the inefficient area shown in Figure 1. As perceived prices increase and approach the true resource price, moral hazard should be much less of a problem. If increased consumer search and awareness of costs elicit increased provider competition and efficiency, the long run supply curve would be pushed toward the competitive supply curve S₁ in Figure 1. At higher prices, physicians may also find it more difficult to create demand as the patient now has a substantial financial interest in the treatment which has been recommended. If the plan is successful in each of these dimensions, the actual allocation of resources and corresponding expenditures will approach the efficient levels described in Figure 1. These results would also have been accomplished in a system which eliminates all out-of-pocket costs for covered benefits.

Of course the incentives are not uniform for all consumers. The relatively young and healthy, and those with higher earnings and accumulation rates will have greater prospects for receiving distributions in the not too distant future. Their perceived losses for consuming care will approach the costs of the care they receive. At the other extreme, members of low income groups and others whose accounts have been depleted will likely behave as fully insured persons in the conventional sense, unless additional controls or incentives are established. Others who fall between these extremes will receive a cost sharing ratio which lies between zero and one depending on expectations for receiving distributions. The choices of the minimum level for individual funds and contribution rates would have an important bearing on these expectations, and consequently on the effectiveness of the program.

In short, a limiting variant of incentive reimbursement is created by a form of self insurance which is backed up by a national health insurance plan. This plan would motivate patients to economize on services while ensuring access and eliminating or reducing (depending on coverage) out-of-pocket costs. Thus, all the characteristics associated with comprehensive first-dollar third-party coverage are preserved without as much insurance and the limitations associated with it.

To function successfully, however, the plan must be coordinated with certain other measures. Two measures are of particular importance. First, tax subsidies for the purchase of low deductible supplemental insurance must be eliminated; otherwise the purpose of the program would be defeated. Without existing subsidies, Keeler, et al. (1977, Table 2) estimated that it would cost, for example, $411 under relatively low group insurance rates to purchase family coverage for a $450 deductible. Their projections of the unsubsidized cost of supplemental insurance suggest that the demand for such policies is likely to be minimal.

Second, for those whose health care is financed by the NHF (for example, the poor, elderly, unemployed, and many with serious or chronic illnesses), alternative reimbursement methods must be introduced since there can no longer be meaningful bargaining on fees between provider and patient. Since these users also have little incentive to economize on the quantity of resources which they consume, the pressure on the NHF can be very severe without both price and use constraints. To the extent, though, that governments already finance over 40 percent of health care spending and much larger proportions for heavy users of care, these problems would not be new, and the existing constraints would represent one possible approach. Another possibility is Enthoven's voucher system, intended for the poor in his program, under which recipients receive a subsidy in the form of a voucher for insurance premiums only if they join a qualified organization which is an efficient provider of care.

There are also other important administrative issues, including: maximum and minimum contribution levels; the relative employer-employee shares; the ceiling before distributions are made; and the tax status of these distributions. Although the intent of this paper is to introduce the underlying principles of the plan, leaving administrative issues for subsequent discussion and clarification, we have the following suggestions on these matters. Maximum contribution levels are necessary to avoid turning individual accounts into potential tax shelters, especially if distributions are to receive favorable tax treatment. It seems practical to establish the maximum at a level which corresponds to the employer contributions under the most comprehensive existing group plans (approximately $200 a month for full family coverage—including dental and ophthalmological). Reasonable minimum amounts are also necessary to prevent too many individuals from either exhausting their accounts and drawing on the NHF, or not accumulating at a rate which provides reasonable hope for receiving distributions in the not too distant future. These conditions must be avoided if the program is to function effectively. Without detailed information on the distribution of health expenditures by income class, it would be hazardous to estimate what such a minimum should be, but for the sake of argument, we suggest an initial level which is about 50 percent of the maximum.⁴

Likewise, flexibility and respect for existing industry wage and fringe levels are needed to determine employer and employee shares. Employers should be permitted to pay the entire amount, but firms in industries which historically have had poor health fringe benefits may be unfairly burdened by making this into a requirement. A reasonable starting point might have employers pay at least 50 percent of the maximum contribution level, with business and individual income tax credits for contributions used to alleviate any inequities which may arise.

Turning to other tax issues, public support for the plan would probably be greatly reduced if employer contributions were to be considered as taxable income, and it would be politically foolish to attempt to change what would correspond to current policy. As for the earnings on accumulations and distributions, any additional tax advantages would further erode the tax base. On the other hand, the success of the plan depends on the size of the potential rewards to consumers, which is diminished with increasing tax rates. As a possible compromise, an exclusion similar to the interest and dividend exclusion (or combined with these) could be introduced on annual earnings and distributions could be treated at the favorable long term capital gains rates.

A final important decision concerns the ceiling that must be attained before distributions are made. Here, too, a compromise is needed. A low ceiling increases the probability and reduces the delay of receiving rewards. But it will also increase the numbers that deplete their funds and who, for at least some time, would be supported by the NHF. We suggest a level which would be adequate to cover most routine and all but the more serious episodes of illness, and which could also be attained within three years or so if a family has limited health expenditures. A ceiling of $4,000-$5,000 at current prices would appear to satisfy these criteria.

Further Discussion

Are there major flaws with the plan that have been overlooked? Can significant benefits and savings be produced at no apparent cost? If substantial inefficiencies do occur in society's use of health resources, then the community as a whole can gain from the proposed plan. Many alternative insurance schemes would probably secure at least some of these potential gains, but not enough for a plan to become viable. A viable plan must be relatively efficient, and the public must be willing to accept the constraints introduced by the proposal and the redistributions which are inevitable after any disruption of the status quo. We believe that our program meets the first criterion. However, the program may also produce substantial dislocations and distributional changes, though they need not conflict with the goals of national economic policy.

Because the distribution of certain medical expenditures is very unequal across families, a relatively small proportion of families accounts for a significant proportion of, for example, hospital use. When everyone is covered by conventional insurance, one person's expenses are made up by the insurance gains on others. In this plan, most of these costs would be paid from the NHF, which would be supported by tax revenues. Higher taxes reduce disposable incomes and consumption spending. However, so long as there are efficiency gains which reduce overall health spending, the declines in disposable incomes are more than offset by the savings in individual health funds.

The dislocation would be most severe in the first year, when few families have substantial accumulations, with decreased redistributions in subsequent years as the number of individual funds which become exhausted diminishes and as accumulations eventually are disbursed. As indicated earlier, though, governments already finance a significant proportion of health care spending, especially for heavy users of care. Thus, the magnitudes are not as great as they initially appear, and only the incremental amounts of public spending (which may be less than needed under several national health insurance bills pending in Congress) require additional revenues.

Despite these problems, the savings to society are potentially enormous. Recently, the national media (for example, Newsweek, 1980, p. 73) focused on a health plan administered by the Mendocino County, California Office of Education which partially incorporates some of the same underlying principles. Under this plan, individuals are provided with major medical insurance with a $500 deductible, and the $500 “savings” to the county is placed into a “side account” from which claims up to that amount are paid annually. Any unused annual balances accumulate and are distributed (without interest) when the employee leaves or retires. While no hard evidence exists yet on the effectiveness of the plan which is still in its infancy, the preliminary experience is encouraging, according to County officials. Almost half of the 218 employees covered have not had any medical bills in the first eleven months of the plan's operation, and only 22 had claims on the major medical insurance component.⁵

Our proposal, by establishing individual funds which earn income and offer the hope of imminent distributions, would be expected to reinforce this pattern of behavior. More significant, unlike the Mendocino plan, it completely eliminates conventional health insurance together with all its attendant inefficiencies. Our program creates a market-oriented system which provides universal comprehensive coverage on a uniform national basis.

Other Advantages

Apart from meeting the primary objectives of a national health system—accessibility, the elimination of financial deprivation, and cost containment—other important benefits would follow. These benefits are discussed briefly below.

Avoids excessive regulation

The primary function of the government's health care responsibilities will be to support and manage the NHF. While still a considerable task, it is far less demanding than those required by most other national health insurance proposals, and is similar in scope to managing the Medicare and Medicaid programs currently in effect.

Provides incentives for preventive health measures

Given the financial rewards for not using costly health services, preventive health measures and non-medical substitutions for medical care may become more prevalent and further decrease overall use.

Acceptability to the public and providers of care

With the exception of the insurance industry, although many existing firms would likely be willing to serve as fund administrators, there is no reason to believe that strong opposition from consumers and providers will develop. For providers in particular, fee-for-service could be retained and although the prospect of greater competition may not be especially welcome, this could hardly be made a public issue. Most providers would strongly prefer the flexibility offered by this plan to the stringent controls found in many alternatives.

Stimulates capital formation

The low rates of capital formation and decreasing productivity are subjects of nationwide concern. The increased flow of funds to the capital markets from the accumulations in individual accounts would be generally viewed as a desirable by-product of the plan.

Flexibility

Unlike other proposals which require much more government involvement, the plan is relatively easy to modify or even abandon if it does not live up to its expectations. Furthermore, it is compatible with other strategies for changes in health care delivery. For example, to encourage enrollments in qualified HMOs, the plan could permit individuals to withdraw any excess contribution over the HMO's premium immediately before the ceiling is reached, because no claims in this period would be made on the NHF. For low or moderate wage earners, whose contribution levels fall short of the premiums charged by qualified HMOs, this option may be impossible without government subsidies.

Coverage

The sharp divisions which currently exist on the scope of coverage under national health insurance become less important as individuals will realize that they are using their own money to buy services, and inefficiently high levels of consumption will be discouraged.

Equity

Redistributions other than the transfer of resources from consumption to saving will take place. Income will be redistributed toward the more healthy, to two wage earner families which currently have double insurance coverage, and toward low wage and other families which lack comprehensive or have no insurance coverage. The overall distribution of health costs is also likely to be more progressive if Federal income taxes constitute the primary revenue source for the NHF. Although questions of equity rest on value judgments, it is likely that strong support could be generated for at least some of the redistributive effects of the proposal.

Disadvantages

Implementation and administration

While the system should be no more difficult to administer than other national health insurance programs, it would be unrealistic to minimize the magnitude of the task. In addition to those questions previously discussed, administrative decisions are required on such issues as: (1) the start-up date and the creation of individual accounts, with the possibility of initially accelerating contributions to reduce the number of early claims on the NHF; (2) the regulation of firms administering the accounts; (3) the nature of the property rights to the accumulations, and divisions or distributions upon death or other changes in family status; (4) the reimbursement mechanism for the NHF; (5) covered benefits under the NHF; (6) the definition of the family unit; (7) adjustments for inflation; (8) the sources of the NHF's revenues; and (9) the changes which would be required in the current regulation of the health industry to avoid conflicts with the plan.

Inadequate incentives for some

If, despite all attempts to the contrary, too many families are supported by the NHF or have poor prospects of receiving distributions in the foreseeable future, the expected economies may never be realized.

Necessary care avoided

Some critics may argue that many families, for monetary gain, will postpone or completely eliminate medically necessary care. Along the same lines, it has been suggested that health insurance has become so popular because individuals want to avoid moral decisions involving trade-offs between health care and other goods. Clearly, these positions reject the principle of consumer sovereignty—a concept which is central to market-oriented strategies.

Uncertainty

Will the proposal be as effective as the analysis suggests? Certainly, before it can be seriously considered as a viable policy option, more definitive empirical support is needed for those basic theories we have relied upon to predict consumer and provider behavior under different economic environments.

Conclusion

Even though market-oriented solutions to the health care dilemma are receiving increased public attention, the range of possibilities offered by this strategy is just beginning to be explored. That these strategies may vary in terms of their distributional effects and efficiency gains is less well understood. Those who strongly favor a highly regulated system to ensure access and quality care for all instinctively assume that more competition and greater financial penalties for consuming services will inevitably produce further maldistribution of medical resources. While these fears may have been created by some proposals which include increased cost sharing and other restrictions to limit use, especially by lower income groups, other market-oriented schemes could be devised to meet all the commonly accepted goals of a national health system. We believe that this article develops the principles for one such plan.