Cato Policy Analysis No. 210 June 13, 1994

Policy Analysis

Nickles-Stearns Is Not
The Market Choice for Health Care Reform

by Tom Miller

Tom Miller is the senior policy analyst for the Competitive Enterprise Institute and director of its Economic Policy and Regulatory Reform program.


Executive Summary

The Consumer Choice Health Security Act, of which Sen. Don Nickles (R-Okla.) and Rep. Cliff Stearns (R-Fla.) are primary sponsors, is one of the leading proposals for health care reform. Unfortunately, it sets contradictory objectives: universal coverage and increased consumer choice, individual responsibility, and competition in health insurance markets. Absent a major overhaul, it will neither ensure that health care markets remain private and voluntary nor make them more competitive, efficient, and responsive to consumers' wishes.

The primary virtue of Nickles-Stearns is that it avoids the worst features of the Clinton administration's proposed Health Security Act. Nickles-Stearns would make health insurance more portable, avoid structural disruptions in coverage, and encourage individuals to choose health insurance in a more cost-conscious manner.

However, the legislation (as introduced last November) contains a number of serious flaws. It endorses the concept of compulsory universal insurance coverage and imposes a standardized "minimum" package of health insurance benefits. Its cost-sharing requirements would undercut the appeal of Medical Savings Accounts. Its efforts to eliminate risk selection in insurance markets are both futile and counter-productive. It provides inadequate incentives for restraining health care costs and hampers the use of more effective devices to do so.

By failing to provide a clear alternative based on market principles, Nickles-Stearns blurs opposition to Clinton-style health care legislation. By focusing the political debate on the wrong issues, it opens the door to extensive political interference in private health care decisions.

Introduction

Last September President Clinton's national address on health care formally launched a historic debate on the appropriate roles of government regulation and private markets in the U.S. health care system. On October 27, 1993, the Clinton administration unveiled the Health Security Act as its legislative blueprint for a government takeover of private-sector health care.

Many members of Congress, however, have offered competing visions of health care reform that range across the entire political spectrum. One of the leading proposals among congressional Republicans is the Consumer Choice Health Security Act (S. 1743, H.R. 3698). It was introduced on November 20, 1993, in the Senate by Don Nickles (R-Okla.) and in the House by Cliff Stearns (R-Fla.). The Senate version of the bill, which quickly gained 24 cosponsors, is viewed as a major alternative to the Clinton health care plan.(1)

The primary sponsors of the proposed legislation claim that it will meet both President Clinton's criteria for guaranteed universal coverage and their own goals of increased consumer choice, individual responsibility, and competition in health insurance markets.(2) Those broad objectives, however, are contradictory. Close scrutiny of many of the bill's provisions makes an urgent case for overhaul of Nickles-Stearns.

The first strategic objective in the current health care debate must be to derail the Clinton health care initiative. The equally important second objective is to ensure that necessary reform of health care markets will both keep them private and voluntary and make them more competitive, efficient, and responsive to consumers' wishes. In its present form, the Nickles-Stearns legislation is not up to either task.

Overview of the Consumer Choice Health Security Act

The Nickles-Stearns bill would require that every American receive a federally defined minimum level of health insurance coverage. That "family security benefits package" is quite comprehensive. It must cover "all medically necessary acute medical care." That encompasses the entire range of physicians' services; inpatient, outpatient, and emergency hospital services; appropriate alternatives to hospitalization; and inpatient and outpatient prescription drugs.

The legislation sets additional cost-sharing rules for the mandatory health insurance plans. Deductibles may be no greater than $1,000 per individual, or $2,000 per family, per calendar year before 1998. The limit on annual out-of- pocket expense is $5,000 for those years. (Both sets of limits are indexed for subsequent years.) The out-of-pocket limit applies to the total amount of deductibles, copayments, and coinsurance that is paid for items and services provided under the plans.

If individuals (or their families) fail to purchase the federally required coverage privately, they will automatically be enrolled in their state government's program designed to provide equivalent coverage. The Nickles-Stearns legislation apparently assumes that the state programs will operate as a last resort. The bill relies primarily on a pair of tax-based incentives to encourage voluntary compliance with its insurance purchase mandate. Taxpayers who fail to provide proof that they have purchased a federally qualified health plan will lose their ability to claim any personal tax exemptions for themselves or their similarly uninsured dependents. They will also pass up the opportunity to take advantage of the bill's scheme for health care- related tax relief.

The second major feature of the Nickles-Stearns bill is a new system of federal income tax credits to target finan- cial assistance to individuals and families enrolled in federally qualified health insurance plans. The credits are structured to provide higher levels of tax relief (in both absolute and relative terms) as eligible individuals spend more money on both health insurance premiums and unreim- bursed medical expenses.

The baseline tax credit level is 25 percent of such spending. Thus, a taxpayer whose annual insurance premiums and unreimbursed medical expenses total $1,000 would receive $250 in tax relief. A taxpayer whose expenses of that sort were twice as high ($2,000) would receive twice as much ($500) in tax relief.

The percentage provided in tax credit relief also increases on a sliding scale as a family's total health care expenses grow as a share of its annual adjusted gross income (AGI). The tax credit structure is given in Table 1.

Table 1
Tax Credits
Insurance Premiums and Unreimbursed Medical Expenses as a Proportion of AGI Tax Credit (percent)
Less than 10 percent 25
Between 10 and 20 percent 30
20 percent or more 75

The tax credits are also refundable. That means that if the total value of the credit were more than an individ- ual's annual income tax liability, she could either receive a government check for the difference at the end of the tax year or an earlier, equivalent adjustment in tax withholding levels.

To partly offset the budgetary cost of tax credits, the current income tax exclusion for employer-paid health care benefits would be eliminated. The Nickles-Stearns legislation would also end two other income tax deductions currently available under the Internal Revenue Code--the deduction for individual medical expenses above 7.5 percent of AGI and the deduction for 25 percent of health insurance costs incurred by self-employed individuals.

The Consumer Choice Health Security Act would provide an additional tax credit equal to 25 percent of annual cash contributions to a Medical Savings Account (MSA). Each household could have only one MSA. Each household's tax- advantaged aggregate contributions each year could not exceed the sum of $3,000 plus $500 for each dependent of the MSA beneficiary. Tax-free payments could be made from the MSA for any qualified health insurance plan premiums or unreimbursed medical expenses. Any other withdrawals or distributions from MSAs would be taxed as gross income and subject to an additional penalty of 10 percent.

The third major component of the Consumer Choice Health Security Act is limits on the scope and nature of insurance competition imposed by a number of pricing, underwriting, and marketing rules for insurance plans.

Modified Community Rating: Insurers can vary their premium rates only on the basis of age, sex, and geography. The same rates must be charged both existing policyholders and new applicants with the same age, sex, and geographic characteristics. Insurers could also offer individuals discounts to encourage participation in certain programs that (1) promote healthy behavior, (2) prevent or delay the onset of illness, or (3) check for illness at an early stage.

Guaranteed Issue: Insurance plans must provide guaranteed issue, at standard rates, to all applicants. Plans cannot exclude, or limit, coverage for any preexisting medical conditions of applicants who have been continuously insured for a period of at least one year prior to application. As of January 1, 1998, such exclusions or limits for other applicants can be no longer than the lesser of (1) the number of months that the applicant was uninsured between the date of first diagnosis and the date of application or (2) one year.

The Nickles-Stearns legislation also requires health insurance plans to participate in risk adjustment (or reinsurance) programs administered by state governments. Those programs would be designed to compensate for unusual distributions of "above-standard" and "below-standard" insured risks among such plans.

Therapeutic Benefits of Nickles-Stearns

It Is Not the Clinton Plan

The strongest argument for the Nickles-Stearns legislation is that it avoids the worst features of the Clinton administration's proposed Health Security Act. The Consumer Choice Health Security Act does not mandate that employers pay at least 80 percent of the cost of a standard benefits package for their employees. It does not set up a regional system of mandatory purchasing alliances for most Americans. There are neither global budgets for health care spending nor explicit price controls on the average level of health insurance premiums. The Nickles-Stearns legislation does not construct an elaborate bureaucracy to control key aspects of health care spending, medical education, and pharmaceutical pricing. It does not attempt to restrict consumers' choice of doctors.

On a more relative basis of comparison, the standardized minimum benefits package under Nickles-Stearns is less generous than the one under the Clinton plan. Community- rating requirements for health insurance plans under the Consumer Choice Health Security Act are less rigid because they can vary by age, sex, and geography.

Tax Reform and Individual Empowerment

The most positive aspect of Nickles-Stearns is its commitment to move away from the current system of private health care insurance financing that is built on the tax exclusion for employer-financed group policies. By proposing a system that uses individual-based tax credits to help finance health care spending, the legislation would make health insurance more portable and avoid structural disruptions in coverage whenever workers changed or lost jobs. The health insurance purchases of millions of consumers (the self-employed, the unemployed, and those employed by many small businesses) would no longer be treated less favorably under the tax code than those for workers who receive employer-sponsored group insurance. By eliminating the tax exclusion, the Nickles-Stearns legislation would encourage individuals to become more involved in choosing their own health insurance coverage and to do so in a more cost-conscious manner.

The tax exclusion treats both employer-paid group insurance and health insurance coverage provided by self- insured employers as nontaxable employee benefits. That distorts health care purchasing choices in a number of ways.

First, it encourages workers to think that their health care is paid for by someone else (their employer) and reduces their sensitivity to the cost of health insurance choices. The funds spent on health insurance and other health care benefits actually come out of the total compensation package that is paid to an employee. Spending more on such benefits reduces the amount of money available for wages. But since the cost of health benefits is not spelled out on workers' regular paychecks, employees are unlikely to be fully aware of the tradeoff.(3) They may even be tempted to believe that money spent on health benefits is their company's and not their own.(4) In addition, workers who receive employer-paid group insurance or coverage under self-insured employer plans are not directly rewarded for being cost conscious. When employees act as prudent buyers of health care, the immediate savings on health insurance costs go to their employers. The concealed nature of employer-provided health benefits also diminishes employees' awareness that, to the extent they are not concerned about the level of their own health care spending, their company-paid health insurance premiums will be higher.

Second, the substantial benefit of the tax exclusion (full deductibility) is available only for employer-financed insurance. The tax benefits for individually financed health insurance are quite limited.(5) Thus, the tax code's bias in favor of employer-financed health insurance means that most employees face strong incentives to turn over many decisions regarding the scope and terms of health insurance coverage to their employers, rather than make them on their own. Individuals who lack, or lose, access to employer- financed coverage find the after-tax dollar cost of health insurance increasingly unaffordable.

Third, the tax exclusion encourages individuals to obtain the most comprehensive versions available of employer-financed insurance coverage instead of using more of their own funds to pay for health care as they receive it. Employer-paid health insurance has distorted the whole concept of insurance. The financial structure of the tax exclusion provides the same percentage "discount" (the marginal income tax rate) on each additional dollar spent on employer-paid insurance. Thus, it imposes a penalty on workers who might prefer higher wages and greater cost sharing to more extensive medical benefits.(6)

Fourth, the tax exclusion encourages the private health insurance system to divide responsibility for health care purchases between individual consumers and third-party payers. Although individual consumers decide whether or not to receive particular medical treatments, they have relatively little to do with paying for them. That disconnection between consumption and payment decisions leads to conflicts of interest between insured workers and their employers. Workers covered by employer-paid health insurance are tempted to consume more health care than either their employer wants to underwrite or the workers would purchase with their own "cash" earnings. Employers who finance such insurance, on the other hand, tend to value the possible savings from restraint in additional units of marginally beneficial health care spending. The employers' viewpoint is more sensitive to the cost side of the cost vs. quality tradeoff. Thus, employers are likely to agree to more restrictions on insured parties' choices of physician and insurance coverage options than would individual employees who were free to purchase health insurance on their own, under the same circumstances (i.e., the equivalent amount of wages is available for health care consumption, and the tax treatment of such spending is the same as that for employer- financed insurance).

Finally, since employers get to call most of the shots regarding the nature of group insurance that they finance, such insurance policies have a short-term focus, are tied to a specific employment relationship, and are unlikely to reflect the particular preferences of individual workers. Employers who finance health insurance have to view insured employees as potential claimants of their company's funds. From their bottom-line standpoint, workers who are unlikely to keep producing more net revenue for a company than they absorb in employer-financed health benefits are not strong candidates for special insurance protection. Thus, the employer-based tax exclusion gives rise to structural disruptions in insurance coverage. Employer-paid insurance policies are likely to provide few, if any, continuing benefits for workers once they change or lose their jobs. Employers who sponsor health insurance are also less likely than individual workers to resist insurance policy exclusions for potentially costly preexisting conditions. In addition, most employers faced with the difficult task of structuring insurance options for a group of workers provide limited choices.(7)

Small Steps Forward

Apart from its elimination of many of the distortions created by the tax exclusion, the Nickles-Stearns legislation offers a number of other worthwhile, though limited, reforms of the health care system.

Preemption of state laws that mandate insurance benefits, restrict selective contracting by managed- care plans, and limit cost-sharing incentives would allow health insurers to better control costs and offer policies that were more affordable and flexible.(8) By mandating its own set of standard benefits on the federal level, however, the legislation negates some of the progress that it might achieve on this front.

The legislation's proposals for budgetary savings in the Medicare and Medicaid programs are thoughtful and well targeted.(9)

The provisions in Nickles-Stearns for administrative standardization of insurance forms and health care data can certainly help achieve some cost savings, but should be limited to voluntary compliance.(10) Mandatory standards run the dangers of squeezing out competitive innovation and cutting off critical feedback.

The Nickles-Stearns legislation dispenses some relief to health care providers from the most extreme applica- tions of the antitrust laws or the chilling effect of potential litigation.

The Nickles-Stearns legislation also targets new tax relief to encourage growth in the markets for two types of insurance products--long-term care insurance and accelerated death benefits. The first type of tax incentive provides a modest step toward encouraging greater reliance on private-sector saving to help underwrite long-term care expenses. The second tax measure removes a major barrier to letting terminally ill individuals gain access to their life insurance benefits before death. It clarifies the status of such accelerated payments so that they receive tax treatment that is comparable to that accorded the life insurance benefits from which they are derived.(11)

The Nickles-Stearns Prescription Needs a Warning Label

Despite those good ideas, there are serious flaws in the Nickles-Stearns bill that should deeply concern advocates of genuine free-market health care reform. The most troubling aspect of the Nickles-Stearns legislation, as introduced on November 20, is the mandate that it imposes on all Americans to purchase a standard package of health insurance benefits. By endorsing the concept of compulsory universal insurance coverage, Nickles-Stearns undermines the traditional principles of personal liberty and individual responsibility that provide essential bulwarks against all- intrusive governmental control of health care.

Political ratification of universal coverage erases a fundamental dividing line between the private sphere of civil society and the public arena of political conflict. Once we presume that government is ultimately responsible for guaranteeing that every American has health insurance, we also guarantee a permanent role for politicians in determining an accompanying set of issues. Once government mandates insurance coverage, it must define what constitutes "adequate" insurance coverage for each citizen. It then has to decide how insurance can be provided to everyone in an "affordable" manner. The process of dictating how much, at what cost, to whom, and in what manner would represent the triumph of politics over free markets. It would severely hamper, if not preclude, treating health care as an economic good that is distributed on the basis of individuals' different abilities to pay and desires to do so.

The Nickles-Stearns legislation relies on both positive and negative tax incentives to enforce its mandatory insurance requirement. The "carrot" is the stipulation that only individuals who purchase "federally qualified" health insurance plans would be eligible for tax credits. The "stick" is an explicit tax penalty on anyone who fails to do so; those people lose their right to claim a personal exemption for themselves (or any similarly "uninsured" dependents) on their personal income tax returns. (At the time this paper was written, the office of Representative Stearns had indicated that he intended to remove the tax penalty provision from H.R. 3698.) Even if one were comfortable with enforcng an insurance purchase mandate, the proposed approach to ensuring compliance would represent unwise and mistargeted tax policy that would set a dangerous precedent.

In the past, claiming tax relief under the personal exemption has required little specific, affirmative conduct on an individual taxpayer's part. Additional claims of personal exemptions for dependents have required little more than apparently sustainable claims of a legal marriage (for a spouse), parental rights (for children), or another blood relationship and, in some cases, additional proof that the taxpayer provided over half the support of the claimed dependent. (In recent years the Internal Revenue Service has imposed a small penalty on taxpayers who fail to provide the Social Security numbers of dependents claimed as personl exemptions.) Unlike tax relief provisions such as deductions for IRA contributions, child and dependent care expenses, mortgage interest, charitable contributions, or other itemized expenses, the personal exemption has been essentially available on the basis of one's status rather than one's conduct. Virtually everyone who files a tax return can either claim it or be claimed as a dependent on someone else's return.

In its clumsy attempt to enhance compliance with its health insurance mandate, Nickles-Stearns would open the legal door to future federal efforts to coerce specific conduct by conditioning the generally available personal exemption on appropriate behavior. Once the tax code is put into play in that manner, can penalties for smoking improper substances, engaging in risky activities, or acting in a politically incorrect manner be far behind? A more logical nexus between lack of health insurance and tax code penalties might be found in limits on the standard deduction (as a partial proxy for itemized deductions for medical expenses), but the most honest approach to enforcing the insurance mandate would involve a distinct civil penalty outside the tax laws.

As a final guarantee that every American would be insured according to its standards, the Nickles-Stearns legislation would set up state government programs to auto matically insure any individual who refused to "voluntarily" purchase a "private" version of federally required coverage. (At a Heritage Foundation meeting on March 24, 1994, Senator Nickles said that he planned to delete that provision. In addition, on May 4, Representative Stearns sent a "dear colleague" letter in which he also said he planned to delete the provision. However, as of this writing, the new language had not been submitted.) The premiums charged by states would be determined by what the states deemed approriate. If Medicare, Medicaid, and other government programs are any guide, the state program premiums will be underpriced and their claims costs will be underestimated.

Sweeping every American into a mandatory health insurance dragnet is not only offensive on philosophical grounds; it is also impossible to achieve. Prior experience with other universal insurance laws reveals that hitting the 100 percent coverage mark remains elusive. Despite Hawaii's requirement that all employers provide health insurance for their workers, approximately 6 to 11 percent of Hawaiians remain uninsured.(12) Even under Canada's system of national health insurance, an estimated 2 to 5 percent of the population in the province of British Columbia is uninsured. Despite 41 state laws that require motorists to purchase automobile liability insurance, one in seven automobile drivers remains uninsured.(13)

It is also rather difficult to enforce mandates on people who fall between the cracks of government databases. Not even heavy reliance on tax penalties can overcome the Internal Revenue Service's inability to track down millions of Americans who refuse, or fail, to file tax returns.(14) And every 10 years the Census Bureau demonstrates that it cannot locate several million citizens.

Thus, one can expect that any federal enforcement offensive to coerce the voluntarily uninsured into signing up for a mandatory coverage scheme will become both prohibitively onerous and politically pointless at the margin. When those costs are added to the havoc that further political control of the entire health care market would wreak, even subsidizing the full amount of uncompensated care with public funds looks like a better buy for American society.

The Cost-Shifting Cop-Out

The "conservative" argument for the Nickles-Stearns insurance mandate is that taxpayers and insured health care consumers must be protected from free riders who can afford, but fail to purchase, private insurance and subsequently receive "free" medical care.(15) Perhaps the most explicit example of officially sanctioned free riding can be found in the 1986 federal law that requires hospitals participating in the Medicare program to provide emergency services for people with an urgent need for care, regardless of their ability to pay. (Such patients also cannot be transferred unless their condition stabilizes or a doctor certifies that a move would be beneficial.) Hospitals cannot legally turn away acutely ill patients because they are unable to pay.

The cost of care of the uninsured can be imposed on third parties in various ways. It can be written off by providers as uncollectible debt, picked up by taxpayers, or passed on via cost shifting to the medical bills of other, insured payers. One recent estimate of the cost of all health care for the uninsured, excluding their out-of-pocket payments, is about $34 billion a year; it is assumed that about $26 billion is "shifted" to others who pay taxes and purchase private insurance.(16) More narrow estimates of the annual cost of uncompensated care in hospitals start around $11 billion.(17) Accurate estimates of the total cost of uncompensated care are limited by definitional vagaries and the lack of adequate data on unpaid services delivered by physicians.(18)

The issues of cost shifting and free riding become more murky when one breaks them down into their components. For example, the relative impact of uncompensated care in hospitals actually peaked in 1986 and has since fallen slightly as a percentage of hospital costs.(19) In any event, only about 20 percent of those uncompensated care costs was covered by subsidies from state and local governments.(20)

Hospital losses due to below-cost reimbursement from Medicaid nearly quadrupled from 1985 to 1989 and only stabilized after a substantial increase in taxpayer-financed "disproportionate share" payments. Additional losses attributable to undercompensation from Medicare surged after 1986 and may have equaled uncompensated care costs in 1992.(21)

According to 1991 statistics, Medicare covered 88 percent of its patients' treatment costs in hospitals, and state Medicaid programs paid 82 percent of their patients' costs. Hospitals collected almost 130 percent of the cost of treating privately insured patients.(22)

Cost shifting during the last half decade appears to have been fueled more by public programs than by uncompensated care of the uninsured. However, substantial skepticism remains about the purported magnitude of cost shifting. Since many hospitals already are under competitive pressure to price discriminate and charge their paying customers as much as the market (or the hospital's other objectives) will bear, additional losses on uncompensated care may simply reduce their total revenue.(23)

Even if mandating insurance coverage for the uninsured prevented free riding by uninsured patients and at least reduced the total amount of cost shifting, a good portion of the "savings" would be offset by a significant increase in the nation's overall bill for health care. Providing subsidized insurance coverage to the currently uninsured would increase their use of health care services--perhaps by as much as 50 percent. Hospitals no longer treating them "for free" would also have less reason to economize on their care.(24) The real effect of universal coverage would be to create a different form of cost shifting. Universal coverage would first increase the overall cost of health care consumed by those who were previously uninsured and underinsured and then distribute that higher cost to both taxpayers and current paying customers.(25)

In short, concerns about cost shifting provide little justification for mandating universal coverage. Moreover, the issue needs to be divided into two parts: giving safety net assistance to the poor and providing asset protection incentives to those who can afford to buy insurance. For potential health care consumers who are currently uninsured because they lack money, the political task involves setting income eligibility levels and then determining the necessarily modest level of "essential" care that taxpayers are willing to finance. Free riders who can afford to buy some level of health insurance but choose not to do so, on the other hand, should have their income and assets above the eligibility line (perhaps the poverty level) placed at risk and be held liable for unpaid health expenses. Until such funds were depleted, the health care bills of those free riders would not be eligible for any income-related public assistance.(26)

Standardized Benefits: Let the Bidding War Begin

Once the federal government is "empowered" to determine the terms of health insurance coverage that everyone must have, such universal coverage becomes a legal entitlement for people who cannot be required to pay for it on their own. Thus, the Nickles-Stearns insurance mandate provides the key that unlocks the door to the treasury vault. It also requires the resolution by political means of other interconnected issues: What is the scope of the entitlement? How will it be financed? Who will be forced to pay for more than they want? Who will be subsidized?

The history of broad-based entitlements is troubling. Our political system has tremendous difficulty setting limits on any desirable goods and services that it first offers people on a below-cost, open-ended basis. The Nickles-Stearns legislation demonstrates a similar lack of appetite for saying, "No more." In making its opening bid to define a standardized minimum benefits package for all Americans, Nickles-Stearns has already gone well beyond the true meaning of "catastrophic" insurance. Its mandated benefit package, for example, was "sweetened" to require more sweeping prescription drug benefits than are provided today under Medicare or many private insurance plans.

The public-choice implications of politically mandated insurance benefit packages strongly suggest that the initial boundaries set by Nickles-Stearns will inevitably yield to demands for expanded coverage. Special-interest providers of health care will lobby aggressively to bring their services within the coverage mandate. Ironically, even as Nickles-Stearns attempts to eliminate mandated benefit laws on the state level, it will provide new opportunities to recreate the same political dynamic on a national scale. Nickles-Stearns is launching an expensive auction, with little ability to control the final price tag.(27)

Putting MSAs on a Short Leash

The legislation's cost-sharing requirements for its mandatory "family security benefits package" create an additional problem. By setting maximum limits on deductibles (and setting them too low), Nickles-Stearns raises the minimum price of all private health insurance plans and seriously undercuts the appeal of MSAs.

Since it forces all private insurance consumers to purchase policies with deductibles no greater than $1,000 per individual, and no greater than $2,000 per family, the legislation essentially eliminates any possible buyers of more economical, "catastrophic" insurance with higher deductibles.(28) Insurers currently marketing high-deductible policies find that, as they increase the deductible amount, their costs of paying benefits and administering claims decline significantly. That allows them to offer substantial reductions in premiums.(29) The higher the deductible, the lower the premium.

The rationale for MSAs is the encouragement of more out-of-pocket health care spending by individual consumers and less third-party payment of medical bills via comprehensive insurance coverage. The MSA vehicle provides a tax deduction for savings that are earmarked for future health care expenses. But the primary source of those savings is supposed to be a switch by health care buyers to less costly high-deductible insurance policies. The synergy between MSAs and higher deductibles would allow health care consumers to first capture all the benefits of cost-conscious shopping with their own money and then preserve their full, pre-tax value on a long-term basis.

Because the current structure of the tax code encourages first-dollar insurance coverage, our experience with high-deductible policies is limited. Nevertheless, a handful of entrepreneurial insurance companies such as Golden Rule Insurance are already marketing policies that offer deductibles as high as $3,000 per family at comparatively low premiums.(30) The maximum deductible limits in Nickles- Stearns, however, would choke the competitive search for the most efficient mix of self-insurance and third-party insurance for different consumers. It also sets an artificial, legislative limit on future efforts to wean Americans away from comprehensive health insurance.(31)

Removing Risk from Insurance?

The federal mandates in Nickles-Stearns include specific restrictions on the prices for and terms under which federally qualified health insurance plans must be purchased. The legislation tries to have it both ways on risk- related pricing. It prohibits private insurers from varying premiums on the basis of factors other than age, sex, and geography but allows discounts for people who participate in programs to reduce the risk of illness. That position, which is not as rigid as pure community rating, allows some recognition of relevant risk factors. However, it redirects the incentives of private insurers to "risk select" (identify and attract good risks, drive away bad risks) within the prescribed categories in a more covert and targeted manner.

There is little dispute about the distorting misincentives of full-fledged community rating for health insurance. Setting a single price for all insureds regardless of their individual risk characteristics operates as a tax on healthy policyholders and a subsidy to high-risk ones. That tax is regressive as well, since lower risk insurance buyers will tend to be younger, earn less income, and hold fewer assets than higher risk insureds.

Under the cross-subsidies of community rating, healthier insureds are charged for more than the risk that they as individuals bring to an insurance pool. If that feature is combined with a mandatory, comprehensive benefit package, they are then forced to pay higher prices for more benefits than they would choose on their own. But if insurance buyers are left free to choose their own package of benefits, lower risk insureds will try to reduce their community-rating tax by buying less comprehensive insurance, or none at all. That "adverse selection" is compounded by the desire of higher risk insureds to obtain more extensive insurance coverage. Over time, the latter type of coverage will become increasingly expensive, as many low-cost risks stop buying it and insurers must raise their prices to cover the expensive claims of the predominantly high-cost risks that remain as paying customers.

To the extent that pure community rating cannot be evaded by adverse selection, charging every insured a single price will encourage free riding by insureds whose voluntary behavior is risky and increases the cost of health care claims within the insurance pool.(32) Besides reducing the incentives for risk reduction, community rating undermines standards of personal responsibility and unfairly penalizes individuals with good health habits.

Unfettered competition in insurance markets would avoid those problems by pricing risk accurately. That makes all kinds of risks equally attractive to insurers and eliminates incentives for selecting only lower risk customers. When risk-based pricing is prohibited or hampered, however, resource allocation is distorted. Higher risk insureds who benefit from pricing subsidies will demand more health insurance than they would purchase in a competitive market, and lower risk customers who pay above-market premiums will seek less insurance.

However, if insurers cannot openly adjust their prices to reflect the costs predicted by different risk factors, they will use risk selection and other nonprice competition to adjust their costs to regulated prices. That kind of competition will drive insurers to reduce the amount that they spend on care of the sick and increase their spending on lower risk customers.(33) A new "market" equilibrium will be established but on a less efficient basis.

Despite its bow to a limited set of risk-related factors, the Nickles-Stearns legislation attempts to prevent pricing based on all other possible risks, such as the medical history and claims experience of policyholders. The effort to eliminate risk selection is both futile and counterproductive.

To the extent that the permissible rating categories under Nickles-Stearns (age, sex, geography) fail to account for a significant amount of the variance of health care costs within a particular risk class, profit-maximizing health insurers will seek to "cream skim" within the boundary lines. Their incentives to recruit and retain low risks, while discouraging enrollment and renewal of high risks, will remain strong, since age, sex, and geography explain no more than 10 percent of the total explainable variance in health spending of a nonaged population.(34) Therefore, insurers will want to discover additional information about the risk profiles of potential and current customers that the statutory rating formula ignores.

That kind of information (well beyond directly observable variables such as an insured's health status and recent use of medical services) can be developed in a number of ways. Supplemental benefit packages above the statutory minimum can be designed to offer features that are more or less attractive to individuals with different risk profiles. High-risk insureds who expect to use particular medical services extensively will tend to seek more extensive benefits packages. Low-risk insureds will be more attracted to insurance plans with more cost sharing (e.g., a higher percentage copayment) and fewer benefits. Even if benefit features are standardized, different insurance plans can continue to compete for lower risks on the basis of their approach to managed care, utilization controls, limits on provider choice, claims service, and advertising.(35) Indeed, by simply offering new health plans with some managed-care aspects, insurers can improve their odds of attracting lower risk customers, since sicker patients are more attached to their present doctors and will be reluctant to switch plans to obtain price savings.(36) All the additional information can then be used to guide risk selection in a more targeted manner.

The unsuccessful effort of Nickles-Stearns to curb risk selection with modified community rating, however, would set off further rounds of tighter regulation of health insurance. As more creative forms of cream skimming by nonprice means developed, opponents of risk selection would insist on more extreme countermeasures. Unless and until the goal of zero risk selection was successfully challenged, the political market for a universal mandate to purchase uniform benefits from a monopoly insurer would grow. In the absence of a single-payer health care system that eliminates virtually all forms of competitive variety, it will be impossible to ensure solely by price regulation that all "private" health plans are equally attractive to individuals in different risk categories.

The Nickles-Stearns aversion to competitive pricing of risk-based insurance is further illustrated by several other requirements that it imposes on private insurers regarding guaranteed issue, exclusion of preexisting conditions, guaranteed renewability, and marketing practices.

Guaranteed issue provides insurance customers with a unilateral option to match their insurance plan to their health status, at their convenience. When healthy, custom- ers can choose cheaper, less extensive insurance with little risk that they will be caught short if their health status changes. (To the extent that universal coverage is not effectively enforced, healthy customers will also risk little by not buying any insurance at all.) With insurers required to offer coverage on demand, applicants can wait until they have more serious medical conditions before switching insurance plans and buying more extensive medical benefits. Although those benefits will be somewhat more expensive, opportunistic customers will really be purchasing them at a discount. The new insurer must charge them "standard" rates, despite their higher health risk profiles.

The combination of guaranteed-issue insurance and modified community rating under Nickles-Stearns prevents risk-based pricing and encourages healthier customers to buy less insurance. It also raises the price of health insur- ance for the majority of buyers who do not benefit from regulatory subsidies.(37) In a competitive free market, a private insurer would try to reduce those distortions by imposing some restrictions on the coverage available to new applicants with preexisting medical conditions.

However, the Nickles-Stearns legislation effectively precludes that alternative. It also prohibits virtually all exclusions of or limits for preexisting medical conditions, and its limited exception for previously uninsured applicants can be easily defeated.(38)

In a final effort to purge risk considerations from the sale of private insurance, Nickles-Stearns attempts to eliminate insurers' sales incentives to insurance agents that are based on applicants' age, sex, health status, claims experience, occupation, or geographic location. Even if those vague prohibitions can be superficially enforced without regulatory overkill, they will simply drive risk selection efforts toward more clever alternatives (e.g., targeted advertising; tie-in sales of "risk-sensitive" benefits such as health club memberships, exercise facilities, and frequent preventive services; subtle differences in location; selective outreach efforts; access to specialty care; and initial handling of patients).(39) The most likely development is that, once Nickles-Stearns opens the door to federal regulation of health insurance marketing practices, subsequent rounds of increasingly tighter controls will keep trying to catch up with the more subtle techniques of innovative private insurers.

Having largely detached individual risk considerations from private insurance underwriting, the Nickles-Stearns legislation will obviously distort risk distribution. It tries to compensate for that by requiring all insurance plans to participate in state government-run risk adjustment programs.(40)

The immediate problem with prospective risk adjustment is that it is very hard to accomplish with any accuracy. Harvard University health economist Joseph Newhouse estimates that an insurance plan might be able to predict at least 15 to 20 percent of the variance in annual spending across a random sample of the population. But the Nickles- Stearns rating categories of age, sex, and geography would account for only about 10 percent of that predictable variance figure. Even if one includes other observable factors such as health status in a risk adjustment formula, no more than 20 to 30 percent of the predictable variance can be accounted for by fully prospective characteristics.(41) Prospective risk adjustment is further complicated by variations in the way different insurance plans would treat apparently similar patients.(42) Attempts to risk adjust for health status would also create perverse incentives for health plans to manipulate diagnoses and provide either inferior or unnecessary treatment.(43) Since Nickles-Stearns encourages emphasis on both individual consumer choice and managed care, the limitations on risk adjustment will simply intensify the ensuing competition in covert risk selection.

Providing retrospective risk adjustment based on after- the-fact knowledge causes other problems. Reimbursing health plans on the basis of what they actually spend weakens incentives to contain costs.(44) That approach would ultimately eliminate the risk-bearing role of private insurers and create the functional equivalent of a single-payer health system.(45)

The Nickles-Stearns legislation is essentially trying to get insurers out of the business of pricing and managing risk and into the business of managing the delivery of health care. The unique function of health insurers, however, is to price and manage risk. They predict the likelihood of health care claims and base prices on those probabilities. True health insurance offers to protect an insured's assets against the costs of risky events. Pricing risks accurately can make bad risks as profitable to insurers as good ones.

The Nickles-Stearns version of private health insurance emphasizes efficiency in managing health care and paying medical bills. Instead of shopping for asset protection, insurance customers will seek out plans primarily as mechanisms to guarantee prepayment of their anticipated health care consumption decisions. Plans will compete for buyers on the basis of the medical benefits, rather than the insurance benefits, that they offer.(46)

By diluting the insurance role of private insurers, however, Nickles-Stearns will inspire a new form of competition to avoid sicker customers and attract healthier ones. Insurers who cannot adjust their prices to match risks will instead adjust their claims costs. The amount that they spend on insureds will inevitably move toward reflecting the true risk-adjusted price of the premiums that sicker insureds pay. Insurance plans will be discouraged from developing the reputation of paying for the best treatment and thereby attracting patients with the most expensive illnesses.(47) Nickles-Stearns creates a distorted tradeoff: either spend a great deal to provide benefits of little marginal value to low risks, or reduce the availability of necessary benefits to high risks.

There is a market-based alternative to those futile efforts to remove risk considerations based on health status from insurance decisions. It would allow insurers to charge actuarially fair prices that reflect the lower levels of risk for healthier people. In an insurance market where tax policy did not discourage individuals from choosing their own coverage, rather than relying on the group-based decisions of their employers, most customers would be concerned about long-term protection against a downturn in their health risk status and could negotiate for the enhanced protection of guaranteed renewability and limits on rate hikes.(48) (Such long-term contracts are common, for example, in the individual life insurance market.)(49)

There is obviously a price for such contractual protection, but it is best determined between willing buyers and sellers in competitive markets rather than hidden in indirect third-party subsidies, price distortions, and restricted availability of comprehensive health care benefits. Competitive insurance markets are best suited for finding the right prices for particular risks. The problems of customers with inadequate income or medically uninsurable risks can best be addressed as social problems by other, more targeted means (e.g., safety net subsidies, private charity, risk pools) that do not alter the relative prices of health insurance and medical care services.

Rearranging Tax Incentives and Playing the Progressive Redistribution Game

The Nickles-Stearns legislation makes two fundamental changes in the structure of tax benefits available for health care. It provides the same amount of tax relief for out-of-pocket medical expenses as it does for health insurance premium payments. Nickles-Stearns also makes tax financing more "progressive" by creating a three-tiered system of tax credits that provides more relief to individuals as they spend higher shares of their income on health care. Those features will, unfortunately, tend to diminish cost containment incentives, increase the likelihood of government price controls, raise marginal tax rates for many Americans, and redistribute income to create new economic winners and losers.

Whatever the other advantages to such an approach, it must be noted that extending tax-credit relief to medical bills that are not covered by insurance will dilute the cost containment effects of both deductibles and copayments in health insurance policies. If the applicable tax credit is 25 percent, for example, it will convert a $1,000 pre-tax deductible into a post-tax one of $750, and the credit will reduce the copayment rate for a particular treatment or procedure by one-quarter. When individuals become sick and face gaps in their insurance coverage, that tax credit "discount" could encourage them to spend further amounts on marginal health care items.

The Nickles-Stearns tax credit for unreimbursed health care expenses will also make the federal government a co- payer for every patient and every medical bill. As health care costs rise, so will the strong temptation for Washington to impose Medicare-style price controls on hospital procedures and physicians' services.(50)

By setting up three levels of tax relief that are based on the ratio of health care expenses to adjusted gross income, the Nickles-Stearns legislation will distort economic decisionmaking as consumers reach the margins of its various "bend point" tax thresholds. Individuals nearing those zones will be rewarded for reducing their income and boosting their medical spending.

Marginal tax calculations are further complicated by the fact that the Nickles-Stearns tax credits will apply only to the federal income tax. Unlike the tax exclusion that they "replace," the tax credits will leave an individual's wages subject to both the employer and employee shares of the payroll tax and state and local taxes.(51) That tax "windfall" for state and local governments, plus the under- financed trust funds of the Social Security and Medicare programs, will be particularly burdensome for the many middle-class Americans in income tax brackets of 28 percent or more.

Consider the case of a dual-earner couple in that tax bracket with two children and taxable income of $50,000 who spend $4,000 a year on a family health insurance plus an additional $1,000 in out-of-pocket medical expenses. Under Nickles-Stearns, they will receive a 25 percent credit against federal income taxes--$1,250 in tax relief. Assume that they used to receive $4,000 in employer-paid group health insurance, but after enactment of Nickles-Stearns, they receive that amount in higher wages. Those taxpayers are losing the benefit of the tax exclusion for the $4,000 that used to be "paid" out as nontaxable health insurance but is now wage income. Their higher wages are subject to a 28 percent federal income tax rate, a 15.3 percent payroll tax rate, and state income taxes that are likely to bring the total marginal tax rate close to 50 percent. So our couple ends up paying $750 more in total taxes ($2,000 minus the $1250 tax credit).

Although there are numerous variations that can push the before-and-after tax tradeoff in either direction (e.g., higher health care expenses relative to adjusted gross income, differences in coverage under employer-paid insurance and individually selected Nickles-Stearns-style insurance), the marginal tax rate effects are clear: taxpayers in federal income tax brackets above 15 percent (28 percent, 31 percent, 36 percent, 39.6 percent) are more likely than not to be net losers, particularly when they must also pay payroll and state taxes on their increased wage income.(52) To the extent that those marginal dollars (taken from what was previously employer-paid insurance) are spent on health care, the higher marginal tax bite imposed by the combination of all federal, state, and local taxes will not be fully offset by the 25 percent income tax credit for health care spending that is available under Nickles-Stearns.

Thus, in addition to its rules for modified community rating and other limits on risk-based insurance premiums, the Nickles-Stearns legislation rearranges the real price of health care via the tax code. The legislation's income redistribution efforts are saluted for reducing net spending on health care (after wage effects) by more than $1,000 a year for 39 percent of all families with a head under age 65, but they also increase health care expenses by more than $500 a year for one in four households.(53) Under Nickles- Stearns, net health care costs will increase by more than $500 a year for

27 percent of all families with annual incomes from $20,000 to $29,999, 29 percent of families with incomes from $30,000 to $39,999, 25 percent of families with incomes from $40,000 to $49,999, 30 percent of families with incomes from $50,000 to $74,999, 31 percent of families with incomes from $75,000 to $99,999, and 32 percent of families with incomes over $100,000.(54)

The game of deliberately creating economic winners and losers by progressive taxation strays far afield from the principles of both market-oriented health care reform and supply-side tax policy.

Administrative Complications

Nickles-Stearns relies on "refundable" tax credits to target maximum financial assistance to low-income individuals. In theory, taxpayers who were eligible for a tax credit that was larger than the amount of their federal income tax liability would receive a check for the difference from the federal government. In practice, delivering such tax assistance in a timely manner will be difficult-- particularly to unemployed or nonworking individuals who are not subject to regular income tax withholding. Problems of tax compliance and income volatility will be hard to overcome.

Many eligible individuals will simply fail to take advantage of federal assistance. Consider the fact that fewer than 1 percent of those eligible for the "earned income tax credit" exercise their right to get part of it in advance in their paychecks.(55) Yet Nickles-Stearns assumes that it will be possible to require a large number of currently nonfiling "taxpayers" to file tax returns in order to claim tax credits for health care expenses.

The Congressional Budget Office estimates that achieving that objective would involve the filing of almost 12 million new tax returns. The combination of those additional filings and the lure of refundable tax credits is likely to create substantial administrative burdens for the IRS and heavy compliance costs for low-income households. Generous sliding-scale tax credits would also encourage tax fraud (understating income and overstating health care expenses).(56)

Since the overall amount of the tax credit and its "refundable" status are both dependent on a taxpayer's current income level relative to eligible health care expenses, many individuals might be hard-pressed to determine the appropriate size of their tax credit (if it is based on current income) until after the end of a tax year. On the other hand, basing the percentage of such tax relief on one's income over the previous year would deliver the re- fundable money after the need for assistance had passed.(57)

Constraints on Cost Containment

The Nickles-Stearns legislation places much more emphasis on guaranteeing universal coverage, ensuring comprehensive benefits, and providing progressive tax relief than it does on establishing incentives to contain costs. Its advocates predict that it will moderate costs through "consumer choice in a competitive market."(58) Such claims are, at best, exaggerated. On the one hand, the legislation would encourage greater demand for health insurance and medical services, and, on the other hand, it would hamper the use of more effective devices for restraining health care costs.

By guaranteeing a substantial package of insurance benefits for the many Americans who currently are either uninsured or face gaps in their coverage, Nickles-Stearns would increase their overall spending on health care. Even if they originally might have preferred to make different consumption decisions on a cash basis, they would now have both the means (fairly comprehensive insurance) and the incentives (extracting the maximum cash value of coverage that has already been purchased, plus the leverage of very limited cost sharing) to incur higher medical bills that would be covered by insurance.

The tax benefits under Nickles-Stearns will also drive health care spending higher. The legislation's tax credits operate as price subsidies for both health insurance and out-of-pocket spending on health care. The tax credits will reduce the post-tax cost of health care for many, if not all, consumers (depending on their marginal tax brackets). The tax incentives will also encourage people to purchase more comprehensive insurance with fewer controls on costs (particularly cost containment strategies that increase the unsubsidized costs of time and inconvenience).(59) Extending the tax credit subsidy to out-of-pocket health care expenses may partly reduce the demand for health insurance, but it also has the potential to increase overall spending on health care.(60)

The Nickles-Stearns legislation's limits on cost sharing further tilt the odds against greater reliance on self- insurance and cost-conscious health care spending. As previously noted, its ceiling on insurance deductibles will cut short the potential cost savings from purchase of truly "catastrophic" insurance that would dampen demand for discretionary health care, reduce private insurers' administrative costs, and help individuals to fully fund their personal MSAs.(61)

After stimulating overall consumer demand for health care, and short-circuiting the full cost containment potential of the high-deductible insurance-MSA combination, Nickles-Stearns apparently hopes that greater individual choice in selecting private insurers and their respective benefit packages will help slow spending growth. The legislation's advocates tout the success of the Federal Employees Health Benefit Program (FEHBP) and claim that its health care reforms will operate in an even more efficient manner.(62) The FEHBP record on cost containment, however, is somewhat mixed and depends on when one sets the baseline period for measurement.(63) Political constraints keep a tight lid on patient cost sharing and encourage adverse selection by federal employees during annual "open season" enrollment periods.

Moreover, the Nickles-Stearns structure for health care purchasing is markedly different from the current FEHBP framework. One of the latter's strongest cost containment incentives comes from its equivalent of a tax subsidy cap. The federal government, as an employer, contributes a fixed amount for each employee's health insurance--75 percent of the average cost of the lowest cost health plans available under the FEHBP. As a result, federal workers and retirees must bear the full, unsubsidized cost of choosing more expensive health insurance, and they also get no tax assistance for out-of-pocket health care expenses. Tax benefits under Nickles-Stearns, on the other hand, are uncapped and are passed through directly to individuals.

Nickles-Stearns provides for a very limited form of competition among private insurers. Insurers may offer marginally differentiated packages of supplemental insurance benefits, but the legislation sets such a high standard for minimum benefits that the main effect of benefit competition will be to push spending upward, not downward. In short, benefit packages can be made more, but not less, comprehensive.

The Nickles-Stearns coverage standard for its mandatory benefits--"all medically necessary acute medical care"-- would further restrict competitive innovation in cost control.(64) If consumer choice is to move beyond superficial differences between competing health plans, insurers must be encouraged to craft enforceable private contracts that define real differences in the nature, intensity, and content of the health care services for which they will pay. The true market-based alternative to regulatory controls and political rationing of health care must center on greater use of private insurance contracts as economizing instruments.(65) Enshrining the traditional "medically necessary" standard in the Nickles-Stearns legislation, however, would lock in permanently an extensive line of court rulings that have been hostile to insurance coverage restrictions and essentially created open-ended entitlements to state-of-the- art medical treatment for insured patients.(66) It would lock the legal door and foreclose promising experiments in allowing individual consumers to make their own economizing choices and voluntary tradeoffs in health care via customized insurance contracts.(67)

Getting the Politics of Health Care Reform Right

The Nickles-Stearns legislation, of course, is a political document. It makes a number of assumptions about which policies will accomplish the primary goals of derailing passage of some version of the Clinton administration's health care plan and reforming the present health care system in a more market-oriented direction. The fundamental political premises of Nickles-Stearns include the following:

Any legislative alternative to the Clinton plan must guarantee universal coverage.

The minimum benefits package for all insured Americans must be fairly comprehensive, with specific limits on personal cost sharing under universal insurance coverage.

Risk selection and medical underwriting by private insurers are undesirable and should be kept to a minimum.

The federal government should ensure that the financing of health care is more "progressive" and that the distribution of health insurance coverage is more "equitable."

Those assumptions are not only flawed as matters of principle; they also very poorly frame key issues in this year's health care debate. Having chosen to fight on the wrong turf for only marginal advantages, Nickles-Stearns guarantees itself a losing political hand. By failing to provide a clear alternative based on market principles, the legislation blurs opposition to Clinton-style health care legislation and will help rather than hinder passage of the latter.

Acceptance of President Clinton's political bottom line--universal coverage--falls into a disastrous trap. Universal coverage can only be achieved at the price of federal control of "private" insurance markets. Once universal coverage is law, the government will dictate how we buy mandated insurance, determine what kind of benefits it provides, and inevitably regulate how it is priced. As Republican strategist William Kristol has pointed out, "Federally mandated insurance coverage requires either government control over nearly every detail of daily medical practice, or an aggressive enforcement regime that would be an affront to the idea of limited government, or both."(68)

Once one agrees to universal coverage, the next issue is who must pay for it. Nickles-Stearns insists on an individual mandate, as opposed to the employer mandate proposed in the Clinton administration's Health Security Act and its congressional offspring.(69) Either type of mandate is the equivalent of a tax. Assume that the only choice in this year's health care financing debate becomes: who pays the tax, you or your boss? Most voters will be a few credit hours short of their degree in labor economics and are likely to deal with the surface reality of politics. They will side with the president and conclude that their employers can, and should, pay the insurance bill. Republican members of Congress and other opponents will be caricatured once again as trying to protect business and put the burden on the little guy.

If support for an individual mandate makes clear that everyone agrees that some mandate is necessary, and the only question is who will pay for it, political dynamics will put the mandate on employers instead of their workers. If one's goal is to defeat an employer mandate, the only effective method is to oppose across the board the imposition of any new mandates and other hidden health taxes on everyone.

The larger issue, of course, is not to determine who will get stuck with the bill for new mandates, but how to first hold the line on, and then hopefully reduce, the burden that government health care policies already impose on all of us.

The next step beyond the mandate issue requires specifying the content of the insurance package that must be purchased. There can be no mandate for universal coverage without a single, uniform benefit plan for everyone to buy. The Nickles-Stearns acceptance of that provision sets the legislation up for further political ambushes.

The first drawback in defining standardized benefits is that the process sets off a political bidding war over what is covered by the mandatory insurance plan. In such a competition, the Clinton plan already has the upper hand over Nickles-Stearns in offering more generous benefits, and its supporters will have fewer inhibitions about promising to expand coverage for such items as home health care, long- term care, and mental health treatment. Once again, there is only one way to avoid coming off as a stingy loser in bidding over the scope of the health insurance entitlement: oppose the political specification of a single, uniform package of health care benefits in the first place and support full freedom of choice for consumers to determine what is in their own insurance plans. Nickles-Stearns has already demonstrated that, even before the bidding war commences and special interests bring all their potential pressure to bear, it cannot hold the line at minimal catastrophic insurance in setting its own initial benefit package.

The second drawback to standardizing benefits is its inherent resistance to cutting back costly reliance on third-party insurance payments to finance many routine health care services. The key political insight behind proposals to encourage greater use of MSAs and high-deductible insurance is that they restore incentives for consumers to seek lower priced, more efficient health care services. When they spend more of their own money and control a greater share of their health care purchases, consumers will also insist that providers make greater efforts to control costs and improve quality.

Unfortunately, however, Nickles-Stearns undermines the appeal of that politically attractive approach. It sets a restrictive ceiling on deductibles under its standardized insurance package. That arbitrarily cuts off opportunities to fully exploit the cost-saving potential of high-deductible insurance. Instead of viewing saving as the clear alternative to buying more comprehensive expensive third- party coverage, consumers may decide that the tax credit discount will allow them to cover modest gaps in insurance coverage with cash on hand.

The effect of the tax financing structure under Nick- les-Stearns will be to selectively penalize a substantial number of upper income and middle-income Americans with comprehensive employer-paid health benefits. The new revenue will then be redistributed to make health care cheaper for a somewhat larger fraction of middle-class taxpayers and to finance health insurance for the uninsured. That redistributionist scheme not only represents unwise tax policy; it also alienates another key constituency that is predisposed to oppose greater government control of health care.

Equally damaging is the manner in which Nickles-Stearns frames the issue of how health care reform will be financed. One of the Clinton plan's key vulnerabilities is the fact that it will cause a substantial number of people (at least 40 percent) to pay more for health care than they do today. The fact that a somewhat smaller, but still large, number of Americans will also have to pay more under a leading Republican alternative, such as Nickles-Stearns, provides political cover for the president's plan and lets it off the hook on the tax issue. The president can simply point to the sponsors and key supporters of the Nickles-Stearns legislation (such as the Heritage Foundation) and say, "Everyone on both sides of the aisle agrees that some will have to pay more for health care."

Advocates of limited government gain the high ground in political debates over across-the-board tax cuts. They lose their advantage when the issue becomes how to redistribute tax burdens among new sets of winners and losers.

Since Nickles-Stearns would launch a new entitlement program that would affect higher income people less favorably than those with lower incomes, it also risks creating the kind of political backlash that led to the speedy repeal of the Medicare catastrophic insurance program in 1989.(70)

Finally, the Nickles-Stearns legislation undermines the political case for consumer choice and competition in health insurance, because it remains ambivalent about risk-based underwriting and afraid of wide disparities in coverage. Nickles-Stearns treats competition among insurers as dangerous unless they are kept on a short leash. Therefore, it imposes modified community rating, standardizes basic benefits, limits cost sharing, prohibits certain marketing practices, and provides for risk adjustment. Since it will not allow private companies to be real insurers with risk- based underwriting and pricing, Nickles-Stearns fails to make a persuasive case for keeping health care insurance within the private sector.

Nickles-Stearns reveals lack of confidence in how the markets it purportedly promotes would actually work. Its political assumptions carry the underlying message that, without explicit federal regulation, vulnerable consumers cannot bargain in their own self-interest, resist exploitation by private insurers, and obtain the best value in health insurance.

Conclusion

The Nickles-Stearns legislation represents a significant effort by many conservative members of Congress to offer a less onerous alternative to the Clinton administration's proposal for health care overhaul. It certainly avoids the worst features of the latter's Health Security Act--employer mandates, purchasing alliances, global bud- gets, caps on insurance premiums, and generous guaranteed benefits. By replacing the tax exclusion with individual tax credits, Nickles-Stearns would encourage consumers to become more involved in choosing their own health insurance coverage and to do so in a more cost-conscious manner. The legislation's change in the way health care is financed would make private insurance more portable and avoid structural disruptions in coverage whenever workers changed or lost their jobs. The legislation also offers a number of modest, but useful, reforms in the areas of medical malpractice liability and antitrust law.

Unfortunately, Nickles-Stearns makes the fatal mistake of endorsing compulsory, government-defined, universal insurance coverage. That fundamental feature automatically leads to the requirement that all Americans purchase a standard package of health care benefits. It also opens the door wide to extensive political interference in private health care decisions.

Having already compromised its ability to deliver market-based health care reform, Nickles-Stearns then goes on to undercut the appeal of MSAs. It engages in a futile and counterproductive effort to stifle risk-based underwriting and pricing by private insurers. The legislation also distorts marginal tax rates and launches a dubious income redistribution scheme. Those measures diminish the ability of Nickles-Stearns to achieve one of the primary objectives of health care reform--cost containment.

Finally, Nickles-Stearns frames the political debate over health care on the wrong issues. That will hinder rather than help the groups that are trying to build opposition to the Clinton health plan and similar legislative initiatives.

The diagnosis is clear. Nickles-Stearns is well intentioned but in need of market-oriented corrective surgery.

Notes

(1) This paper is based on an analysis of Nickles-Stearns as introduced on November 20, 1993. Both Senator Nickles and Representative Stearns have indicated that they intend to introduce substantial revisions. Thus, the final version of the bill may differ from the version discussed here.

(2) Don Nickles, "A Health Plan Good Enough for Senators," Wall Street Journal, November 16, 1994; Don Nickles, "Health Care for Individuals, Not Bureaucrats, Washington Times, December 1, 1994; and Statement of Cliff Stearns, "Introduction of 'the Consumer Choice Health Security Act of 1993,'" November 18, 1994. See also Don Nickles, "Consumer Choice and Health Security Act Fact Sheet," November 18, 1994.

(3) "The provision of health insurance as a fringe benefit reduces the ability of employees to make efficient decisions because they may lack information. If employees are not aware of how much their insurance costs, they will make even less cost-conscious choices about health insurance than would be expected given the large tax subsidy." Congressio nal Budget Office, The Tax Treatment of Employment-Based Health Insurance (Washington: Congressional Budget Office, March 1994), p. 25.

(4) Grace-Marie Arnett, "It's Taxation, Stupid," Wall Street Journal, March 29, 1994.

(5) Self-employed individuals can deduct from their taxable income only 25 percent of their health insurance expenses. Individuals can also claim a much more limited itemized deduction--for the amount of their annual health care ex penses (including insurance premiums) that is greater than 7.5 percent of their adjusted gross income.

(6) "In the absence of a tax subsidy, the typical health insurance policy would probably contain more mechanisms to control costs and reduce premiums because consumers would reap the full benefit of premium reductions." Congressional Budget Office, The Tax Treatment of Employment-Based Health Insurance, p. 47.

(7) See ibid., p. 19.

(8) See John C. Goodman and Gerald L. Musgrave, "State Health Care Reform under the Clinton Administration," Na tional Center for Policy Analysis Policy Report no. 173, November 1992, p. 38; and Gail A. Jensen, "State Mandates on Private Insurance," Regulation, Fall 1992.

(9) Even if it becomes politically possible to obtain sig nificant funds from such reforms, the funds may well be needed just to deal with the rocky fiscal futures of both programs. Reprogramming those federal budget dollars into new spending initiatives will eliminate the net savings that are desperately needed to reduce future fiscal deficits. The board of trustees of the Medicare program estimates that the Federal Hospital Insurance Trust Fund, which pays inpatient hospital expenses for the elderly, will be able to continue paying benefits for only about seven years and is severely out of financial balance in the long run. Outlays for Medicare's Supplementary Medical Insurance program, which pays doctors' bills and other outpatient expenses, have increased 59 percent in the last five years and have grown 23 percent faster than the economy as a whole. 1994 Annual Report of the Board of Trustees of the Federal Hospi tal Insurance Trust Fund (Washington: FHITF, April 11, 1994); and 1994 Annual Report of the Board of Trustees of the Federal Supplementary Medical Insurance Trust Fund (Washington: SMITF, April 11, 1994). The federal govern ment's share of the Medicaid program's costs doubled from $31 billion in 1988 to $68 billion in 1992, and the federal costs for Medicaid are projected to reach $231 billion by 2003. "Health Care's Hour," Congressional Quarterly Special Report, September 25, 1993, p. 8.

(10) See, for example, The President's Comprehensive Health Reform Program (Washington: Government Printing Office, February 6, 1992), pp. 56-59.

(11) The legislation would ensure that accelerated benefits paid via a lien approach (benefits paid before the insured dies are deducted from the final death benefit) received the same tax treatment as those paid via the discounting ap proach (the death benefit and cash value of the life insur ance policy are reduced directly). The lien approach is less expensive to administer and does not require policy reductions and policy splits. It also preserves the policy holder's option to repay the loan and to restore the full death benefits for his beneficiaries. The Nickles-Stearns legislation, however, would put certain forms of "viatical" settlements (those involving people with less than one year to live) at a further competitive disadvantage relative to accelerated life insurance benefits. Such viaticals involve offers by commercial firms to provide a negotiated amount of cash to terminally ill people in exchange for being named the beneficiary of the clients' life insurance policies. Those more flexible versions of cash transfers before death would remain taxable income under Nickles-Stearns.

(12) Dick Armey, "Universal Coverage Is a Mirage," Dear colleague letter to the U.S. House of Representatives, March 17, 1994; and John Merline, "What Price Universal Coverage?" Investor's Business Daily, February 22, 1994, p. 2.

(13) Armey.

(14) In testimony before Subcommittee on Oversight of the House Ways and Means Committee on October 26, 1993, Internal Revenue Service commissioner Margaret Richardson indicated that approximately 7 million individuals had not filed income tax returns that they should have filed. C. Eugene Steurle, Economic Effects of Health Reform (Washington: American Enterprise Institute, 1994), p. 21.

(15) In insisting that every U.S. resident must, by law, be enrolled in a health care plan that covers major health care costs, Stuart Butler of the Heritage Foundation emphasizes that "Americans with sufficient means would no longer be able to be 'free riders' on society by avoiding sensible health insurance expenditures and relying on others to pay for care in an emergency or in retirement." A National Health System for America, ed. Stuart M. Butler and Edmund F. Haislmaier (Washington: Heritage Foundation, 1989), p. 51.

(16) That high-estimate figure includes both "uncompensated" care (which includes both private charity and bad debt) and care paid for by public funding (such as the Department of Veterans Affairs; the military; and state, city, or county programs but not Medicare or Medicaid). Steve Robinson, "Clinton's Bogus Diagnosis," Republican Study Committee, January 31, 1994, p. 8.

(17) Prospective Payment Assessment Commission, "Medicare and the American Health Care System," Washington, June 1993, p. 31; see also Congressional Budget Office, "Single-Payer and All-Payer Health Insurance Systems Using Medicare's Payment Rates," CBO Staff Memorandum, April 1993 (estimating total charges for uncompensated care for the uninsured at $26 billion in 1991, only a part of which would be included in private health insurance premiums); and Spencer Rich, "Higher Hospital Costs to Public Seen Rising from Medical Aid Programs," Washington Post, May 16, 1992 (citing a Lewin/ICF estimate of $11.9 billion for 1992).

(18) See Thomas J. Hoerger, "Who Pays for the Uninsured?" Paper presented at Cato Institute conference, Bitter Medi cine, Washington, April 30-May 1, 1992.

(19) Prospective Payment Assessment Commission, p. 29.

(20) Ibid, p. 31.

(21) Ibid., p. 30. The aforementioned Lewin ICF study esti mated that underpayments to hospitals on Medicare and Medicaid claims would total over $22 billion in 1992, more than twice as much as hospital losses on "free care" of the poor and uninsured. Rich, "Higher Hospital Costs to Public Seen Rising from Medical Aid Programs."

(22) Prospective Payment Assessment Commission, p. 31. The Physician Payment Review Commission recently reported that Medicare pays doctors only 59 percent of what private insur ance companies pay. That gap has widened in the last five years. In 1989 Medicare paid doctors 68 percent as much as private insurers did. Robert Pear, "Medicare Paying Doctors 59% of Insurers' Rate, Panel Finds," New York Times, April 5, 1994.

(23) "The empirical evidence on hospitals' ability to raise prices to make up for unsponsored care or lower payments by other payers is mixed at best. No study has concluded that hospitals have raised prices to fully adjust for such ac tions." "Hospital Pricing: Cost Shifting and Competition," Employee Benefit Research Institute Issue Brief no. 137, May 1993. See also Irwin Stelzer, "What Health Care Crisis?" Commentary, February 1994, p. 24. Indeed, to the extent that cost shifting occurs, it can operate to reduce the total amount of care consumed by the paying customers who face higher billing rates and fees. Norman B. Ture, "The Clintons' Health Care Reform Plan Would Increase Health Care Costs," Institute for Research on the Economics of Taxation Policy Bulletin no. 62, March 16, 1994, p. 6.

(24) Average health care expenses for the uninsured are less than two-thirds of costs for the insured. See Steven Robin son, "Republican Health Care Delusions," Washington Times, March 3, 1994.

(25) Ture, pp. 6, 7.

(26) Section 602 of the proposed Comprehensive Family Health Access and Savings Act (S. 1807) takes a tentative step toward adopting this approach: "No provision of Federal, State, or local law shall apply that prohibits the use of any statutory procedure for the collection of unpaid debts for medical expenses incurred by individuals." It may be necessary, however, to make the "payer of first resort" lien on assets more explicit.

(27) For estimates of the degree to which mandated benefits increase both the price of insurance premiums and the number of uninsured people, see John C. Goodman and Gerald L. Musgrave, "Freedom of Choice in Health Insurance," National Center for Policy Analysis Policy Report no. 134, November 1988; Goodman and Musgrave, "State Health Care Reform under the Clinton Administration, p. 38; Jensen, "State Mandates on Private Insurance," pp. 66-68; Gail A. Jensen, "Regulat ing the Content of Health Plans: A Review of the Evidence," Paper presented at American Enterprise Institute conference, American Health Policy: Critical Issues for Reform, Washing ton, October 3-4, 1991.

(28) The additional tax credit for unreimbursed medical expenses under Nickles-Stearns will be at least 25 percent, as well, and therefore reduce the real, post-tax amount of the deductibles to no more than $750 and $1,500, respectively.

(29) See "Medical Savings Accounts: Questions and Answers," Council for Affordable Health Insurance, Alexandria, Va., February 1993; John C. Goodman and Gerald L. Musgrave, "Controlling Health Care Costs with Medical Savings Accounts," National Center for Policy Analysis Policy Report no. 168, January 1992, pp. 9-13; John C. Goodman and Gerald L. Mus grave, "Personal Medical Savings Accounts (Medical IRAs): An Idea Whose Time Has Come," National Center for Policy Analy sis Policy Backgrounder no. 128, July 22, 1993.

(30) See "Medical Savings Accounts," pp. 12-15; John Merline, "Employees as Health Reformers," Investor's Business Daily, March 18, 1994; "Medical Savings Accounts: The Private Sector Already Has Them," National Center for Policy Analysis Brief Analysis no. 105, April 20, 1994.

(31) Even though copayment levels are not restricted under Nickles-Stearns, until the overall stop-loss limit of $5,000 is reached, that cost-sharing device operates less efficiently than high amounts of first-dollar, 100 percent deductibles. They would absorb higher administrative costs and, if past practice is any guide, insurance companies would be reluctant to use high-percentage copayments.

(32) Higher risk behavior might include overeating, smoking, alcohol and drug abuse, lack of exercise, unprotected sexual promiscuity, failure to wear seat belts, and engaging in high-risk activities such as skydiving and hang-gliding.

(33) John C. Goodman and Gerald L. Musgrave, "A Primer on Managed Competition," National Center for Policy Analysis Policy Report no. 183, April 1994, pp. 18-24.

(34) See Roger Feldman and Bryan Dowd, "Biased Selection: Fairness and Efficiency in Health Insurance Markets," Paper presented at American Enterprise Institute conference, American Health Policy: Critical Issues for Reform, Washington, October 3-4, 1991, p. 10; and Joseph P. Newhouse, Willard G. Manning, Emmett B. Keeler, and Elizabeth Sloss, "Adjusting Capitation Rates Using Objective Health Status Measures and Prior Utilization," Health Care Financing Review 10, no. 3 (Spring 1989): 41-54.

(35) White House Task Force on Health Risk Pooling, "Health Risk Pooling for Small-Group Health Insurance," January 1993, pp. 3, 4. See also Mark A. Hall, Is Community Rating Essential to Managed Competition? (Washington: American Enterprise Institute. 1994), p. 10.

(36) Ibid.

(37) See Stephen D. Brink, James C. Modaff, and Steven J. Sherman, "Variation by Duration in Small Group Medical Insurance Claims," Society of Actuaries Research Report, September 5, 1991; "Guaranteed Issue: Guaranteed to Make the Problems in the Small Group Market Worse," Council for Affordable Health Insurance, Alexandria, Va., November 1992, pp. 4-7; and John C. Goodman and Gerald L. Musgrave, "State Health Care Reform under the Clinton Administration," p. 20.

(38) Even if an applicant is uninsured and is suddenly sur prised to learn he has a serious medical condition, his insurance coverage under a new plan will be guaranteed with virtually no delay if he applies for such coverage immedi ately after the date of first diagnosis. Moreover, many medical conditions will not require immediate treatment, or treatment can be paid for out of pocket until new insurance coverage is obtained. "Guaranteed Issue," p. 5.

(39) Joseph P. Newhouse, "Patients at Risk: Health Reform and Risk Adjustment," Health Affairs, no. 1 (Spring 1994): 137, 138. Much alleged risk selection involves customers' pur suit of their own preferences. "Sick and healthy people will sort themselves into different plans, even without any effort at selective marketing by insurers." The President's Comprehensive Health Reform Program, pp. 19, 20.

(40) Nickles-Stearns also attempts to allow insurance plans to pass back to an applicant's previous insurer any claims relating to a preexisting condition already being treated at the time of application. However, the manner in which the cost of such claims can be reallocated, along with a portion of future insurance premiums, is not specified in the legis lation. The provision appears to invite contentious litiga tion and defy predictable execution.

(41) Newhouse, "Patients at Risk," pp. 139, 140. See also Goodman and Musgrave, "A Primer on Managed Competition," p. 36; and Hall, Is Community Rating Essential to Managed Competition? pp. 13, 14.

(42) Newhouse, "Patients at Risk," p. 141.

(43) Hall, Is Community Rating Essential to Managed Competi tion? p. 13.

(44) Ibid, p. 16; Goodman and Musgrave, "A Primer on Managed Competition," p. 36; and Newhouse, "Patients at Risk," pp. 142, 143.

(45) Hall, Is Community Rating Essential to Managed Competi tion? pp. 15, 16.

(46) SeeGoodman and Musgrave, "A Primer on Managed Competi tion," pp. 16-18.

(47) Ibid., pp. 17-19.

(48) See Congressional Budget Office, The Tax Treatment of Employment-Based Health Insurance, pp. 11, 15.

(49) Edmund F. Haislmaier, "A Policymaker's Guide to the Health Care Crisis, Part III: What's Wrong with America's Health Insurance Market," Heritage Foundation, Heritage Talking Points, August 14, 1992, p. 14; and Edmund F. Haisl maier, "A Policymaker's Guide to the Health Care Crisis, Part IV: The Right Road to Health Insurance Reform," Heri tage Foundation, Heritage Talking Points, November 5, 1992, p. 15.

(50) John C. Goodman, "The Best and Worst Ideas for Health Care Reform," National Center for Policy Analysis Policy Backgrounder no. 118, June 10, 1992, p. 9.

(51) The Nickles-Stearns legislation states that employers must hold their employees "harmless" for any increased employer share of payroll taxes due to the loss of the tax exclusion. But employers will simply shift the added burden back to employees, in the form of lower overall wages or reduced employment. The advocates of Nickles-Stearns under stand that principle of labor economics when they criticize the wage-dampening employer mandates under the Clinton administration's Health Security Act (see, for example, Stuart M. Butler, "How the Clinton and Nickles-Stearns Health Bills Would Affect American Workers," Heritage Foun dation Issue Bulletin no. 188, April 11, 1994, p. 2), but they apparently ignore its application in this case. From the standpoint of marginal tax rate analysis, the impact of the employer payroll tax rate of 7.65 percent is passed through to employees and included in their calculations of the after-tax gain from earning an additional dollar in wages.

(52) Even higher brackets are created by the phaseout of itemized deductions, application of the alternative minimum tax, and taxation of Social Security benefits above certain income thresholds.

(53) Butler, "How the Clinton and Nickles-Stearns Health Bills Would Affect American Workers," pp. 7, 10. A previous analysis of the tax credit plan by Lewin/ICF in 1992, howev er, estimated that 29.5 percent of families with a head under age 65 would pay at least an extra $500 a year for health care, and 43 percent of families with a head under age 65 would pay at least $100 a year more. Lewin/ICF, "The Individual Tax Credit Program: Estimated Cost and Impacts," Preliminary draft prepared for the Heritage Foundation, January 10, 1992, p. 24.

(54) Lewin-VHI, Inc., "The Effects of the Health Security Act on Employee Wages and a Comparison of the Effects of the Health Security Act and the Individual Tax Credit Program on Households," Fairfax, Va., March 9, 1994, p. 27. Although Stuart Butler of the Heritage Foundation contends that some of those increased costs are due to individuals' choosing to purchase more extensive health care benefits under the Nickles-Stearns approach, the portion of the specially commissioned Lewin study that has been made available by Heritage does not measure the amount of that effect. It might be noted that Butler does not make a similar adjust ment for more extensive benefit purchases under the even more generous Clinton plan when comparing the latter's costs to those of Nickles-Stearns. See Butler, "How the Clinton and Nickles-Stearns Health Bills Would Affect American Workers," Heritage Foundation Issue Bulletin no. 188, April 11, 19994, p.7. In any event, selective cost comparisons that argue over which legislation will do somewhat less harm in forcing a fraction of Americans to pay more for their health care give away more political points than they gain.

(55) "Tax Report," Wall Street Journal, April 13, 1994, p. 1.

(56) See Congressional Budget Office, The Tax Treatment of Employment-Based Health Insurance, p. 47.

(57) Ibid., p. 46.

(58) Stuart M. Butler and Edmund F. Haislmaier, "The Consumer Choice Health Security Act," Heritage Foundation Issue Bulletin no. 186, December 23, 1993, p. 27.

(59) Congressional Budget Office, The Tax Treatment of Em ployment-Based Health Insurance, pp. 19, 47.

(60) Ibid, p. 20; see also Bryan E. Dowd and Roger Feldman, "Voluntary Reduction in Health Insurance Coverage: A Theo retical Analysis," Eastern Economic Journal, July-September 1987.

(61) Although copayments are theoretically unrestricted under Nickles-Stearns (apart from the $5,000 annual stop-loss limit), their unbridled use to maximize cost sharing appears unlikely in the context of the entire legislation. One might imagine in the extreme a flat copayment requirement of 90 percent (rather than the more customary 25 percent), or widely varying copayments (10 percent, 25 percent, 50 per cent, 75 percent) for different medical procedures. The first approach would entail disproportionate administrative costs for most private insurers. Deductibles are a much more efficient tool. The second approach would be chal lenged for its potential to maximize risk selection in a narrowly targeted manner. In short, the political premises underlying the Nickles-Stearns legislation (universal cover age, standardized minimum benefits, reduction of risk selec tion, distributional equity) would tend to rule out such cost-sharing practices.

(62) Robert E. Moffit, "Consumer Choice in Health: Learning from the Federal Employees Health Benefits Program," Heri tage Foundation Backgrounder no. 878, February 6, 1992.

(63) See Goodman and Musgrave, "A Primer on Managed Competi tion," pp. 5, 6; and Janet P. Lundy, "The Federal Employees Health Benefits Program," Congressional Research Service Issue Brief, June 11, 1992.

(64) Ironically, that standard would also fail to achieve its purported objective of greater uniformity in insurance coverage. A recent General Accounting Office study of Medicare payment practices under the "medically necessary" standard, for example, revealed radical disparities in the claims approval and denial rates for Medicare beneficiaries across the nation. See Robert Pear, "Medicare Denials Vary Greatly State by State," New York Times, March 29, 1994; and Spencer Rich, "Medicare Claim Denials Vary Widely, GAO Says," Washington Post, March 29, 1994. Such regional disparities reflect the different interpretations of medical necessity used by different insurance carriers, as well as regional variations in medical practice, billing practices, and levels of fraud. By appearing to set a "uniform" cover age standard under federal law that, in practice, will operate much less consistently, however, Nickles-Stearns could fuel a second round of political demands for even tighter bureaucratic control of claims processing.

(65) The foremost proponent of the private contract approach to health care reform has been Duke Law School professor Clark C. Havighurst. His most recent work insists that "without private contracts articulating consumer choices and authorizing controversial economizing moves, health plans and health care providers will continue to be driven by legal fears to overspend on health care." Havighurst, Health Care Choices (Washington: American Enterprise Insti tute, forthcoming), p. 5.

(66) See Mark A. Hall and Gerald F. Anderson, "Health Insur ers' Assessment of Medical Necessity," University of Penn sylvania Law Review 140 (May 1992): 1637; Clark C. Havig hurst, Health Care Choices; Clark C. Havighurst, "Prospec tive Self-Denial: Can Consumers Contract Today to Accept Health Care Rationing Tomorrow?" University of Pennsylvania Law Review 140 (May 1992): 1755; Clark C. Havighurst, "Why Preserve Private Health Care Financing?" Paper presented at American Enterprise Institute conference, American Health Policy: Critical Issues for Reform, Washington, October 3-4, 1991; and Tom Miller, "Let Free Markets Dictate Coverage," Insight, January 3, 1994.

(67) Private health insurers should be free to use various contractual devices that would adjust the cost of the insur ance product they offer to match its value. The possible efforts might include incorporating different medical prac tice guidelines, delegating difficult decisions to an iden tified panel of medical experts, providing a specific proc ess for resolving disputes over medical appropriateness, offering different levels of access to technology based on cost/benefit assessments, inserting clauses that expressly waive the insured's right to have a policy construed liber ally against its drafter, varying standards for malpractice liability, and, most of all, adjusting premiums to reflect any of those choices. Incentives could also be added for signing a living will or choosing hospice care for a termi nal condition. The point is to allow flexibility to put more options to a market test.

(68) William Kristol, "Why Republicans Should Oppose 'Uni versal Coverage,'" Project for the Republican Future, Wash ington, April 7, 1994.

(69) The Clinton administration's Health Security Act would require employers (before they received any offsetting subsidies or reached percent of payroll premium caps) to pay 80 percent of their employees' insurance costs. Other leading health care bills sponsored by Reps. John Dingell and Pete Stark impose employer-financing mandates that apply different "cost-sharing" percentages or exempt certain small businesses on the basis of their total number of employees.

(70) See David S. Broder, "Health Care Reformers Dogged by 1988 'Catastrophic' Debacle," Washington Post, April 14, 1994.

1994 The Cato Institute
Please send comments to webmaster