Is there a right way to promote health insurance through the tax system?
The exclusion of employer contributions to health premiums has skewed the development of the insurance market, resulting in generous coverage for higher-income workers but leaving millions of others uninsured and facing rapidly rising health costs. The paper considers four recent reform proposals: capping the exclusion, tax credits for insurance, tax incentives for high-deductible insurance and health savings accounts, and full tax deductibility of out-of-pocket spending. Such proposals could promote greater efficiency and equity in the health market, but insurance market reforms are also needed to minimize potential disruption to employer risk pools.

Health insurance (Taxation)
Tax expenditures (Laws, regulations and rules)
Tax incentives (Laws, regulations and rules)
Tax reform (Laws, regulations and rules)
Antos, Joseph R.
Pub Date:
Name: National Tax Journal Publisher: National Tax Association Audience: Academic; Professional Format: Magazine/Journal Subject: Business; Economics; Law Copyright: COPYRIGHT 2006 National Tax Association ISSN: 0028-0283
Date: Sept, 2006 Source Volume: 59 Source Issue: 3
Event Code: 920 Taxes; 930 Government regulation; 940 Government regulation (cont); 980 Legal issues & crime Advertising Code: 94 Legal/Government Regulation Computer Subject: Government regulation
Product Code: 6322000 Medical Care Insurance; 6320000 Accident & Health Insurance; 9101147 Corporate Taxes-Savings Incentives NAICS Code: 524114 Direct Health and Medical Insurance Carriers; 5241 Insurance Carriers; 92113 Public Finance Activities SIC Code: 6324 Hospital and medical service plans
Jurisdiction: United States
Accession Number:
Full Text:

The private health insurance system in the U.S. has been erected on a foundation of tax incentives that promote employment-based coverage over the individual purchase of insurance. In 2006, persons taking advantage of tax breaks for health insurance will save about $150 billion in federal and state income taxes and an additional $75 billion in payroll tax contributions (Sheils and Haught, 2004; Sheils, 2006). (1) Most Americans--some 174 million people, or about 70 percent of those with insurance--are covered by employment-based health insurance (DeNavas-Walt, Proctor and Lee, 2005). Included in that count are about 12 million seniors covered by Medicare who also have supplementary retiree coverage through a former employer.

Although tax incentives have helped millions of people buy health insurance through their employers, this policy approach brings a host of problems. Millions of people do not have access to tax-favored employment-based insurance, and many go without coverage. Many who are offered such insurance turn it down because it is too expensive or does not meet a worker's individual financial and health needs. Employees may find themselves locked into their current jobs to retain coverage, especially if someone in their family develops a serious medical condition. Even then, there is no guarantee that the employer will not reduce benefits or drop coverage in the future.

Current tax incentives for health insurance also fail on equity and efficiency grounds. The tax expenditure is regressive, providing a greater subsidy to those with good jobs and high incomes and much less to the unemployed and disadvantaged. In addition, the tax system promotes the purchase of excessive insurance coverage that blunts the incentive for efficiency in the production and use of health services. The resulting cost escalation in our health system affects everyone, but its greatest impact is arguably on the uninsured, many of whom do not have the option to take advantage of current tax benefits.

Those flaws in our current system of subsidizing employment-based health insurance are well known. Experts inside and outside government have advanced a variety of policy reforms intended to improve the performance of tax incentives for health insurance. Recent proposals include capping the tax exclusion for employment-based health insurance, tax credits for the purchase of private insurance, tax subsidies for the purchase of high-deductible insurance and health savings accounts (HSAs), and expanding tax subsidies for out-of-pocket health spending. The proposals address different problems in our current system of tax incentives for health spending, and they represent only part of broader health system reform.

In this paper, I discuss how such proposals could help (or hinder) the purchase of private health insurance and how the health system might react to changes in tax incentives. Following a critique of the current tax provisions affecting health spending, I discuss issues raised by the recent reform proposals, including their potential impact on insurance markets. The final section attempts to answer the question, is there a right way to promote health insurance through the tax system?


The third largest federal subsidy program for health care is operated by the Internal Revenue Service, not the Department of Health and Human Services. The two largest programs are Medicare, with outlays of $380 billion in FY 2006, and Medicaid, with federal outlays of $190 billion (CBO, 2006b). Federal tax expenditures for private health insurance or other spending for health services will total about $143 billion in 2006 (OMB, 2006a). (2)

The three subsidy programs address different beneficiary populations. Medicare pays for a substantial part of the health care for the elderly and disabled. Medicaid covers the cost of care for the poor. The tax code subsidizes private health spending, primarily the cost of insurance premiums, and benefits accrue primarily to working-age people with substantial incomes.

All three health subsidy programs are entitlements, in the sense that spending proceeds automatically without any necessary intervention by Congress. Reflecting rapid growth in the health sector, spending in the three programs has been expanding at high rates for decades, substantially faster than growth in the economy or other major federal programs.

However, Congress takes an active interest in Medicare and Medicaid, enacting in most years some legislation intended to modify how money is spent under those programs (not always to good effect). In contrast, Congress rarely debates and even more rarely adjusts the major tax provisions for health spending.

The institutional bias in favor of the health tax provisions is illustrated by legislative actions taken in 2005. After heated argument about the wisdom of cutting Medicare and Medicaid spending, Congress reduced program outlays by $11 billion over the next five years in the Deficit Reduction Act of 2005 (CBO, 2006a). At the same time, no action was taken on tax provisions affecting private health spending. Those provisions are expected to grow by more than $16 billion annually over the same time period (OMB, 2006a).

Tax subsidies favor employer-sponsored health insurance rather than insurance purchased in the non-group market. Employer contributions to health insurance premiums are excluded from the employee's income subject to income and payroll taxation. The exclusion is worth $132 billion in foregone federal income tax receipts for 2006, by far the largest tax subsidy for health spending (OMB, 2006a). In addition, Section 125 plans permit employees to pay their share of premiums using pre-tax dollars, making the entire premium tax free to participating employees. These two subsidies account for more than 95 percent of total health-related tax expenditures.

The tax code gives far less benefit to people who purchase their health care directly rather than through insurance. Employees who contribute to a flexible spending account use pre-tax dollars to cover out-of-pocket health spending. Tax filers who itemize may deduct out-of-pocket health spending that exceeds 7.5 percent of their adjusted gross incomes. People with HSAs may contribute to those accounts on a pre-tax basis. Less than five percent of health-related tax expenditures flow through these provisions (OMB, 2006a).


The current set of preferences dominated by the tax exclusion has serious limitations as a way of promoting the purchase of private health insurance. Although the tax exclusion helps many workers obtain coverage through their employers, the incentives established by this policy have also contributed to the runaway growth, inefficiency, and inequity of the health system.

Promoting One Type of Risk Pool

The tax exclusion promotes the workplace as the primary source of health insurance in the U.S., although the association of health care with employment is an even older idea (Glied, 1994). Even at the inception of the federal income tax in 1913, health and other benefits provided by employers were excluded from taxation on the grounds that they benefited the employer and they were of minimal size. Some fringe benefits, but not health care, became subject to taxation as they grew in prevalence and cost.

Employer-sponsored health insurance began to expand rapidly during wage and price controls during World War II, but that growth did not cease with the end of controls. (3) The spread of employer-sponsored insurance during the 1950s came about as employers and labor unions recognized that, on the margin, employer contributions to health insurance premiums were more valuable to the average worker than an equivalent increase in their taxable wages. (4)

There are a number of advantages to organizing health insurance around the firm. Since workers seek employment for reasons other than simply gaining health insurance, the firm can pool health risks across a diverse group of individuals who are healthier, on average, than non-workers. Employers also can reduce the insurers' cost of marketing and administration by taking on some of those responsibilities through their human resources departments. Both factors help to lower the average premium, particularly for large employers.

Employer-sponsored insurance has grown not only because of lower insurance costs and the tax subsidy. In addition, the employer contribution to premiums fosters the illusion among many workers that their health insurance is less expensive than it actually is.

For example, the premium for family coverage under employer plans averaged $10,880 in 2005, with employers contributing about $8,167 of that amount (Kaiser Family Foundation and Health Research and Educational Trust, 2005). Most workers probably focused on their direct payment for that coverage, which averaged a little more than $2,700 (or about $225 per month), even though they paid the full cost of the insurance by accepting wages lower than they otherwise would have been without the coverage. While this premium illusion has resulted in more workers accepting their employer's benefit offer, most workers are unaware of the true cost of their health coverage and are likely to buy more generous insurance than they would otherwise.

Despite both the real and illusory advantages of employer-sponsored insurance, several million workers refuse their employer's offer of coverage (Thorpe and Florence, 1999). This is partly because few employers are able to offer a wide choice of insurance products. The health benefit offered by employers is often generous, meeting the needs of a middle-aged worker rather than someone just starting a career out of school. This "one size fits all" approach does not provide the right balance of coverage and cost for many workers. Not surprisingly, such workers tend to be young, lower-income, and working for small firms (Blumberg and Nichols, 2001).

The lack of choice manifests itself in other ways that can disadvantage employees. The employer may unilaterally change the terms of coverage (such as the employee's share of the premium, cost-sharing requirements, coverage of specific services, and breadth of the provider network) or drop it altogether, with the worker having little or no say in such changes (Gabel, Claxton, Gil, Pickreign, Whitmore, Holve, Finder, Hawkins and Rowland, 2004; Reschovsky, Strunk and Ginsburg, 2006).

Those who buy their health insurance through their employers may find themselves locked into their current employment for fear that they might lose coverage (Madrian, 1994). If the worker or a family member develops a serious health condition, a job change could mean the total loss of insurance or exclusion of that condition from coverage. The Health Insurance Portability and Accountability Act limited the use of pre-existing condition restrictions for group coverage, but such restrictions remain possible for non-group coverage.

The tax exclusion has had the effect of crowding out alternative pooling arrangements that are not eligible for the subsidy (Burman and Gruber, 2005). This is a particular problem for the millions of people who have no access to employer-sponsored coverage, including many who work in small firms that do not offer health insurance or who are not in the labor force. (5)

Consumers who must purchase coverage in the non-group market are disadvantaged in three ways. Their premiums are paid with after-tax dollars. They pay higher rates, reflecting larger costs of marketing insurance to individuals and greater health risks of people wishing to buy non-group insurance in the face of high premiums. They are often subject to state regulations and insurance mandates that increase the cost of coverage even further, while employees participating in a self-insured plan are exempt from those rules by Employee Retirement Income Security Act (ERISA). These factors contribute to the rising number of people without health insurance.

Exacerbating Health Spending Growth

Health insurance with lower cost-sharing requirements and higher premiums takes greater advantage of the tax preferences, but such coverage has adverse consequences for the health system. More generous health insurance coverage blunts the consumer's sensitivity to health care prices and encourages greater use of services.

Consumers directly paying a fraction of the cost of care are apt to use services worth less than the full cost, a phenomenon known as moral hazard (Pauly, 1968). The cost of additional care induced by this moral hazard effect of insurance is reflected eventually in higher insurance premiums. By promoting first-dollar coverage, tax incentives help fuel the escalation of health care costs and insurance premiums.

Numerous studies have confirmed that consumers are sensitive to the price of health care, and will use fewer services if the price they must pay increases. The RAND health insurance experiment, for example, found that the imposition of a $3,200 deductible (in 2004 dollars) for family coverage reduced health spending by 31 percent, compared to completely free care (Manning, Newhouse, Duan, Keeler and Leibowitz, 1987; Morrisey, 2005). Higher deductibles and other cost-sharing requirements are customarily imposed by insurers to limit covered health spending, which helps keep premiums down.

There is growing evidence that additional spending for health services may not yield full value in terms of improved health status (Fisher and Welch, 1999). However, people subject to high cost sharing are likely to forego both unnecessary and necessary services (Schoen, Doty, Collins and Holmgren, 2005).

Inequitable Treatment

The tax exclusion favors those with higher taxable incomes and discriminates among individuals based on their employment status. After income and payroll taxes, a high earner could save as much as 50 cents for every dollar spent on health insurance premiums, at the margin. (6) In contrast, a low earner might save as little as three cents on the dollar for employer-sponsored insurance (Burman and Gruber, 2001). People without access to employer-sponsored insurance are not helped by the exclusion, regardless of their income or health status.

Because high-income families tend to buy more insurance, their average tax savings rise faster than income. Approximately a quarter of federal tax expenditures for health spending accrue to families with incomes above $100,000, even though that group accounts for only 14 percent of the population (Sheils and Haught, 2004). Families with incomes below $50,000 also receive about a quarter of the total subsidy, but they represent nearly 60 percent of the population. As a tool for promoting the purchase of private insurance, the tax exclusion is poorly targeted, providing the least help for those with the most limited ability to pay for coverage.


The tax exclusion and other provisions have provided a strong incentive to expand employer-sponsored health insurance over the past six decades. Millions of workers have benefited from the reduction in their insurance premiums net of taxes, but millions of others have not. Those who gain the most from this system are least likely to be without coverage. The tax incentives promote excessive use of health services, in too many cases yielding only marginal value to the patient. This system of providing health insurance has slowed job mobility and imposed other inefficiencies on the economy.

There are certainly better ways to spend $143 billion in federal tax expenditures. A variety of reforms have been advanced in the recent past to ameliorate some of the problems with existing tax preferences for health spending. Those reforms include:

* Capping the exclusion,

* Tax credits for insurance,

* Tax incentives for HSAs, and

* Tax breaks for out-of-pocket health spending.

The discussion that follows focuses on the impact such proposals have on insurance markets. Although these proposals address different aspects of the tax structure, each of them intends to alter tax incentives that could lead to broader changes in the health care system.

An overarching concern for any tax reform is the potential disruption of existing employer risk pools. Tax credits can be used to target subsidies to the low-income uninsured, and are included in most reform proposals. In addition, state insurance regulations could be modified or eliminated to promote competition among insurers and reduce the cost of health insurance in individual and small group markets. Alternative risk pooling arrangements also could be developed, including association health plans (AHPs), state-sponsored purchasing groups, and high-risk pools or other arrangements to cover people who are "uninsurable."

Each of these risk pooling approaches has been considered or implemented by policymakers. Federal legislation has been proposed for several years to exempt AHPs from state benefit mandates. By giving such associations (which could include fraternal or religious organizations, clubs, civic groups, and others) the same exemptions from state benefit mandates available to large employers under ERISA, they have the potential to offer insurance at attractive rates (McClellan and Baicker, 2002).

Thirty states have created high-risk pools to cover persons who have been denied coverage by private insurers (Achman and Chollet, 2001). Individuals receiving pool coverage pay substantial premiums in most instances, and the states also provide significant subsidies. However, enrollment in high-risk pools has been low, totaling 105,000 in 1999. That is in part due to high premiums, limited benefits, ineffective outreach to prospective enrollees, and caps on enrollment imposed by some states to limit their budget outlays. An alternative to high-risk pools would provide a subsidy to help chronically ill persons with predictably high medical costs purchase private insurance (Cogan, Hubbard and Kessler, 2005). The total cost of the subsidy would probably be substantial if it was effective in buying most of the chronically ill into private insurance.

The most prominent recent example of a state purchasing group is the Massachusetts Health Care Reform Plan, signed into law by Governor Mitt Romney on April 12, 2006 (Kaiser Commission on Medicaid and the Uninsured, 2006). The plan will establish a purchasing organization known as the Connector, which will provide one-stop shopping for individuals, small employers, and insurers. In addition to simplifying the purchase of insurance, that arrangement is likely to reduce marketing and administration costs. It also creates a way for employees of firms that might not otherwise offer a health benefit to pay their premiums using pre-tax dollars.

Capping the Exclusion

Proposals to limit the amount of employer premium contributions that can be excluded from a worker's taxable income have surfaced from time to time over the last two decades. The Reagan administration proposed such a cap on the exclusion in its Health Incentives Reform Program, submitted to Congress in 1983 (Steinwald, 1983; Rubin, 1983). That year Sen. Robert Dole (R-Kansas) introduced S.640, the Health Care Cost Containment Tax Act of 1983, which would have limited that open-ended subsidy for employer-sponsored insurance. Most recently, the tax panel appointed by the Bush administration proposed a similar cap on the tax exclusion in their 2005 report (President's Advisory Panel on Federal Tax Reform, 2005).

A well-designed tax cap could be the key to finding solutions to problems plaguing the health system. Bias in the current tax structure toward inefficient first-dollar coverage would be blunted by capping the exclusion. Many workers would find that extra coverage above the tax cap is not worth its full cost after the taxpayer subsidy is reduced. Workers would seek more affordable insurance, and employers and insurers would be under pressure to offer more efficient health plans.

To minimize potential disruption of employer risk pools, the cap could be phased in by setting a high initial level and indexing that amount to general inflation rather than medical inflation. (7) A cap with that design becomes more binding over time as medical inflation outstrips general inflation. A longer phase-in would give the insurance system more time to adjust to a fundamental change in the financial calculus of health care. In addition, tax revenue from limiting the exclusion could be used to help fund tax credits or other subsidies better targeted to the low-income uninsured who wish to buy private health coverage.

Given enough time, even the complex health insurance system could accommodate the new demands that would arise. Alternative pooling arrangements and new insurance products would develop in the individual market as the value of the exclusion declined, resulting in more competition in the insurance market. There would be new consumer pressure to deregulate the individual insurance market and make more affordable types of coverage available.

Consumer demand can be a powerful inducement for changes that have long been recognized as necessary but which have been difficult to achieve by legislation or public cajoling. We are beginning to see that with the growth of consumer-driven health plans. Such plans increase consumer awareness of health costs through higher deductibles and personal accounts that allow the consumer to pay directly for services using tax-advantaged dollars. As consumer-driven plans have become a more significant part of the market, insurers have begun to experiment with ways to make information on the price of health services available to consumers (Butcher, 2006).

In contrast, there has been little demand for such information with conventional insurance. With that type of coverage, the consumer is only directly responsible for a small fraction of the price and generally cannot benefit by seeking to economize on the cost of care.

Employer-sponsored health insurance has not been reliable for many workers. Even without legislation to reduce tax subsidies, employers have been paring back or dropping coverage. Some policymakers have proposed additional subsidies as an inducement to employers who maintain their benefits. (8) Additional employer subsidies might slow the departure of employers from health Insurance, but eventually cost pressures are likely to unravel this system.

Tax Credits for Insurance

There are good arguments for increasing the number of people with health insurance. Uninsured individuals impose a cost on everyone else when they need health care and are unable to pay. Uncompensated care could amount to as much as $35 billion annually (Hadley and Holahan, 2003). If we can increase the number of people with insurance, the newly insured will begin to contribute to the cost of their care by paying premiums. Moreover, people without insurance tend to delay treatment, which can cause them to suffer needlessly, lead to complications, and require more aggressive and expensive medical interventions.

Subsidizing the purchase of health insurance can be an effective policy approach, but as we have seen, the form of the subsidy matters. An open-ended subsidy such as the tax exclusion promotes insurance but contributes to the rising cost of health care. That makes insurance less affordable and causes an increasing number of people to go without coverage.

Analysts have long recommended the use of refundable tax credits for the purchase of health insurance as a substitute for the tax exclusion (Arnett, 1999; Hoff and Pauly, 2002). The Bush administration has endorsed refundable tax credits for health insurance to supplement the tax exclusion, but the 2007 budget narrowed the availability of the proposed credit to those who purchase high-deductible insurance (OMB, 2005 and 2006b). The latter approach is discussed in the next section.

A health tax credit could be targeted to low-income people or made available to everyone, particularly if other tax subsidies for insurance were curtailed. Unlike the exclusion, which increases in value along with taxable income, a tax credit can be structured to provide the greatest benefit to those with the least financial means. A tax credit would also be more flexible than the exclusion, allowing individuals to purchase any coverage on the market rather than being restricted to one or a few plans offered by most employers.

To be effective for the low-income, the credit must be refundable, providing a benefit even to those who do not have any income tax liability. In addition, the credit must be available at the time health insurance is purchased rather than when income taxes are filed, a year or more after the first insurance premium had to be paid. The Health Care Tax Credit, a small scale program authorized by the Trade Adjustment Assistance Reform Act of 2002, has recently solved many of the technical challenges of implementing a refundable, advanceable tax credit (Dorn, 2004).

Sizeable tax credits may be needed to increase substantially the number of newly insured individuals, particularly if the available coverage is expensive (Pauly and Herring, 2001). Those with the lowest incomes would probably not purchase insurance unless nearly all of the premiums were subsidized. The working uninsured may be better able to pay some of the cost of insurance. However, the tax credit would have to exceed the value of the tax exclusion for a worker who has rejected the offer of coverage from his employer.

Offering a tax credit worth more than the exclusion raises the possibility that younger, healthier workers would opt for the credit and leave the employer's risk pool. That is more likely if there is a wider choice of plans in the non-group market than offered by the employer, allowing the worker to select a plan that better meets his financial and health needs. Migration out of employer plans is also more likely if the administrative cost savings of the employer plan are not very large relative to the non-group market.

Conditions could be such that an employer, particularly a small employer, would discontinue the offer of health insurance. Although such an outcome would be welfare-improving for some workers, older and sicker workers would be medically underwritten and charged higher premiums or able to purchase only limited coverage in the non-group market.

A frequently proposed way to minimize this problem would limit eligibility for the tax credit to people who are not currently enrolled in a group plan. (9) That would help preserve the risk pools of smaller employers by preventing out-migration.

However, restricting the tax credit to people without group coverage leaves intact the inequitable distribution of subsidies under the exclusion. Such a restriction reinforces the loss of consumer sovereignty and job mobility that results when health insurance is tied to the workplace. Moreover, lower-income workers who were not offered the option of a tax credit instead of the exclusion would be disadvantaged compared to others with equally low income who were offered the credit.

The restriction also disadvantages lower-wage employees of large firms without any compensating gain to the insurance system. Such a restriction would not substantially contribute to preserving the risk pools of large employers. Large firms can offer insurance at better rates than found in the individual and small group markets. Any employer contribution to the insurance premium further reduces the premium paid directly by the worker. Large firms may also offer a range of health insurance options that may closely match the preferences of workers. Consequently, the threat to the risk pool by introducing a choice of tax subsidy is minimal in the large group market.

A widely available refundable tax credit would reduce the inequity in our current system of subsidizing private health insurance. Such a credit would increase the number of people buying non-group insurance, which is likely to spur the development of more affordable insurance products. Premiums in the individual and small-group market are likely to decline in relative terms with the influx of lower--risk purchasers who previously did not have the means to buy insurance.

A well-designed tax credit would increase the number of newly insured, but that means setting the subsidy high enough so that low-income individuals can afford to pay their share of the premiums. Tax credits and other reforms of the current tax subsidy for insurance could cause some small firms to drop their health benefits. As discussed earlier, steps can be taken to improve the operation of the insurance market and ameliorate such problems. The continuing decline of employer-sponsored health insurance calls for action on such improvements in any event.

Tax Incentives for HSAs

The tilt of the tax exclusion toward conventional health insurance with low cost-sharing requirements and high premiums exacerbates the moral hazard problem. That leads to higher utilization of services that, on the margin, are not worth their full cost. Employers turned to managed care in the 1990s in the hope of slowing the rapid rise in the cost of health benefits. By imposing controls on the use of services through gatekeepers, utilization review, preadmission screening, and other mechanisms, managed care could in concept assure that care was appropriate but not excessive.

The managed care experiment ended in the economic boom of the late 1990s. It was brought down by employee dissatisfaction and a health plan business model that relied more on negotiating discounts from providers than on managing care. As health costs continued to rise, employers and individuals began to adopt insurance with higher deductibles as a way of keeping premiums more affordable. That movement gained momentum in 2003 when Congress created a new subsidy for HSAs--tax-favored savings accounts, which can be used only if the individual purchases high-deductible insurance. (10)

High-deductible insurance is designed to make consumers more aware of the cost of health services, which reduces both the utilization of services and cost of coverage. With more of their own money at stake, consumers have a greater incentive to seek higher value in the care they receive. Ultimately, heightened consumer awareness of cost and value will cause health care providers to adopt a more efficient practice style, focusing on services that are proven to be cost-effective.

HSAs could initiate a cultural revolution in our health system that may, over time, lead to greater efficiency and slower growth of spending. Many other changes are needed, including more consumer-friendly information on prices, quality, and effectiveness of care. Physicians and other providers would see their roles expand, placing more emphasis on advising the patient about treatment alternatives and serving as the patient's expert advocate. Such changes will not come quickly, and they probably will not come easily.

Under current law, individuals may contribute to HSAs on a pre-tax basis. The inside build-up is tax free, and withdrawals are also tax-free as long as that money is used to pay for out-of-pocket medical costs. This triple tax incentive has been criticized as creating a tax shelter for the wealthy, but the potential for abuse may be limited by two factors. First, there is a low limit on maximum annual contributions, currently $2,700 for individuals and $5,450 for families. Second, wealthy individuals are likely to have comprehensive health insurance and may not wish to have greater exposure to health costs by shifting to high-deductible insurance.

The Bush administration has advanced several proposals to expand the tax advantages of HSAs tied to high-deductible insurance (U.S. Department of the Treasury, 2006). Those provisions include:

* An above-the-line deduction and tax credit for the purchase of high-deductible insurance purchased in the non-group market,

* Increases in the maximum amounts that may be saved in an HSA plus a tax credit to offset payroll taxes on contributions made by the worker, and

* A refundable tax credit to low-income individuals who purchase high-deductible insurance in the non-group market.

The refundable low-income tax credit proposal is narrower than in previous years, when the administration supported such a tax credit for the purchase of both high-deductible and conventional insurance.

The long-standing bias in the tax structure in favor of first-dollar health coverage has created unrealistic expectations of what basic insurance must mean, particularly when skyrocketing health costs are absorbing a rising share of employee compensation and growing numbers of people forego insurance altogether (Weller, 2006). One might argue that it is necessary to counter those deep-seated expectations by tilting the tax structure in favor of more efficient insurance products.

The administration has attempted to strike a compromise. Rather than eliminating the tax exclusion (and risking disruption in the insurance arrangements of most Americans), the administration has opted to add a new tax subsidy to promote coverage requiring greater out-of-pocket payments by consumers. This approach recognizes that alternatives to employer risk pools currently are limited, but it also creates a new distortion affecting the consumer's decision of what type of insurance to purchase.

Tax Breaks for Out-of-Pocket Health Spending

Repealing the tax exclusion would eliminate the tax system's bias in favor of excessive and inefficient health insurance, but only if such a policy could be enacted. Repealing a tax break that helps millions of voters is improbable under the best of circumstances, but such an action is even more unlikely in the face of rapidly rising health care costs.

An alternative strategy would equalize the tax advantages of paying for health care through insurance and paying for care out of pocket. John Cogan, Glenn Hubbard, and Daniel Kessler propose to make out-of-pocket health spending fully deductible as long as the individual purchases insurance that at least covers catastrophic health expenses (Cogan, et al., 2005). They include this proposal in a broad agenda that includes refundable health tax credits for low-income individuals, insurance market reforms, expansion of health information, policies to promote health sector competition, and malpractice reform. The Bush administration adopted the concept of full deductibility, but limited the additional tax break to individuals purchasing high-deductible insurance in the non-group market.

Leveling the playing field between paying for health care through insurance and paying for care out of pocket eventually would cause consumers to re-evaluate the wisdom of paying routine expenses through insurance. This proposal preserves the consumer's ability to choose the type of insurance product that is most appropriate for his circumstances, and it largely eliminates the tax advantage of first-dollar coverage over leaner insurance products. This is not simply adding fuel (in the form of a new subsidy) to the fire of health spending. Full deductibility also reduces the incentives in the current system to purchase inefficient forms of health insurance.

However, as mentioned above, the tax exclusion has had a powerful influence on consumer attitudes toward health care and insurance that may not easily be reversed through indirect policy means. The additional infusion of resources through the tax system would be immediate, while changes in individual attitudes and institutional conventions would occur over time. The budgetary cost of this approach could be substantial in the near term as the health system adjusts to the full package of reforms proposed by Cogan, Hubbard, and Kessler.


The employment-based insurance system is under increasing pressure. Rising health costs threaten the bottom lines of even the largest firms, forcing reductions in promised benefits (Sloan, 2005). There are increasing numbers of uninsured and a growing sense of unease among insured workers that they might not be able to keep their health benefits.

Tax expenditures worth hundreds of billions of dollars have helped create this increasingly dysfunctional insurance system. If we hope to improve the system, we cannot simply add new subsidies on top of the existing structure. The open-ended tax exclusion has contributed to the moral hazard problem of insurance that leads to excessive coverage and excessive use of services. The tax subsidy coupled with employer contributions disguises the true cost of health insurance, causing workers to buy more coverage than they might otherwise.

Lynn Etheredge, a well-known health policy expert, observed that "the aver age working family wouldn't go out and spend [$10,000] to buy insurance if they had to buy it in the individual market. They'd be shocked at the sticker price and they would look for something less expensive" (Cunningham, 2002).

The irony is that the average working family unknowingly is paying every penny of that full premium. That payment takes three routes: the worker's share of the premium (paid directly), the employer's share of the premium (paid through lower wages), and the tax subsidy (paid through higher income taxes or lower government services). There is no free lunch, nor even a reduced price lunch, in this system.

Redirecting current tax expenditures could promote the purchase of insurance and encourage more efficient use of health services, but any reform risks upsetting the insurance arrangements of millions of workers. Capping the exclusion to finance tax credits for those most in need is a conceptually straightforward approach that could only be accepted if those with higher incomes were prepared to pay more for their own health insurance. The right tax reform recognizes that political reality and balances the need for institutional improvements in health insurance with the need to maintain some stability in the insurance market. Such a reform is essential if we hope to resolve the larger problems of the health sector.


I am grateful to Bob Helms, Grace-Marie Turner, and participants at the National Tax Association Spring 2006 Symposium for helpful comments.


Achman, Lori, and Deborah Chollet. Insuring the Uninsurable: An Overview of State High-Risk Health Insurance Pools. New York: Commonwealth Fund, August, 2001.

Arnett, Grace-Marie (ed.). Empowering Health Care Consumers Through Tax Reform. Ann Arbor: University of Michigan Press, 1999.

Blumberg, Linda J., and Len M. Nichols. "The Health Status of Workers Who Decline Employer-Sponsored Insurance." Health Affairs 20 No. 6 (November/December, 2001): 180-7.

Burman, Leonard E., and Amelia Gruber. "First, Do No Harm: Designing Tax Incentives for Health Insurance." National Tax Journal 54 No. 3 (September, 2001): 473-93.

Burman, Leonard E., and Jonathan Gruber. Tax Credits for Health Insurance. Urban-Brookings Tax Policy Center Discussion Paper No. 19. Washington, D.C.: Urban Institute, June, 2005.

Butcher, Lola. "Plans Put Provider Prices Out For Their Enrollees to Inspect." Managed Care 15 No. 2 (February, 2006): 60-2.

Cogan, John F., R. Glenn Hubbard, and Daniel P. Kessler. Healthy, Wealthy, and Wise: Five Steps to a Better Health Care System. Washington, D.C. and Stanford, CA: The AEI Press and the Hoover Institution, 2005.

Congressional Budget Office (CBO). Cost Estimate: S. 1932, Deficit Reduction Act of 2005. Washington, D.C., January 27, 2006a.

CBO. Fact Sheet for CBO's March 2006 Baseline: Medicare. Washington, D.C., March 2006b.

Cunningham, Robert. "Joint Custody: Bipartisan Interest Expands Scope Of Tax-Credit Proposals." Health Affairs Web Exclusive (September 18, 2002): W290-W298.

DeNavas-Walt, Carmen, Bernadette D. Proctor, and Cheryl Hill Lee. Income, Poverty, and Health Insurance Coverage in the United States: 2004. U.S. Census Bureau, Current Population Reports, P60-229. Washington, D.C.: U.S. Government Printing Office, 2005.

Dorn, Stan. How Can National Policymakers Improve Health Coverage Tax Credits Provided under the Trade Act of 2002? Washington, D.C.: Economic and Social Research Institute, May 2004.

Fisher, Elliott S., and H. Gilbert Welch. "Avoiding the Unintended Consequences of Growth in Medical Care: How Might More Be Worse?" Journal of the American Medical Association 281 No. 5 (February 3, 1999): 446-53.

Gabel, Jon, Gary Claxton, Isadora Gil, Jeremy Pickreign, Heidi Whitmore, Erin Holve, Benjamin Finder, Samantha Hawkins, and Diane Rowland. "Health Benefits in 2004: Four Years of Double-Digit Premium Increases Take Their Toll on Coverage." Health Affairs 23 No. 5 (September/October, 2004): 200-9.

Glied, Sherry. Revising the Tax Treatment of Employer-Provided Health Insurance. Washington, D.C.: The AEI Press, 1994.

Hadley, Jack, and John Holahan. "Covering The Uninsured: How Much Would It Cost?" Health Affairs Web Exclusive (June 4, 2003): W3-66-W3-81.

Helms, Robert B. "The Tax Treatment of Health Insurance: Early History and Evidence, 1940-1970." In Empowering Health Care Consumers Through Tax Reform, edited by Grace-Marie Arnett, 1-26. Ann Arbor: University of Michigan Press, 1999.

Hoff, John S., and Mark V. Pauly. Responsible Tax Credits for Health Insurance. Washington, D.C.: The AEI Press, 2002.

Kaiser Commission on Medicaid and the Uninsured. Key Facts: Massachusetts Health Care Reform Plan. Washington, D.C.:The Henry J. Kaiser Family Foundation, April, 2006.

Kaiser Family Foundation and Health Research and Educational Trust. Employer Health Benefits 2005. Publication 7315. Menlo Park: The Henry J. Kaiser Family Foundation, 2005.

Madrian, Bridget C. "Employment Based Health Insurance and Job Mobility: Is There Evidence of Job Lock?" Quarterly Journal of Economics 109 No. 1 (February, 1994): 27-54.

Manning, Willard G., Joseph P. Newhouse, Naihua Duan, Emmett B. Keeler, and Arleen Leibowitz. "Health Insurance and the Demand for Medical Care: Evidence from a Randomized Experiment." American Economic Review 77 No. 3 (June, 1987): 251-77.

McClellan, Mark, and Katherine Baicker. "Perspective: Reducing Uninsurance Through the Nongroup Market: Health Insurance Credits and Purchasing Groups." Health Affairs Web Exclusive (October 23, 2002): W363-W366.

Morrisey, Michael A. Price Sensitivity in Health Care: Implications for Health Care Policy. Washington, D.C.: NFIB Research Foundation, 2005.

Office of Management and Budget (OMB). Budget of the U.S. Government, FY 2006. Washington, D.C.: U.S. Government Printing Office, 2005.

OMB. Analytical Perspectives: Budget of the U.S. Government, FY 2007. Washington, D.C.: U.S. Government Printing Office, 2006a.

OMB. Budget of the U.S. Government, FY 2007. Washington, D.C.: U.S. Government Printing Office, 2006b.

Pauly, Mark V. "The Economics of Moral Hazard: Comment." American Economic Review 58 No. 3 (June, 1968): 531-7.

Pauly, Mark, and Bradley Herring. "Expanding Coverage Via Tax Credits: Trade-Offs and Outcomes." Health Affairs 20 No. 1 (January/February, 2001): 9-26.

President's Advisory Panel on Federal Tax Reform. Simple, Fair, & Pro-Growth: Proposals to Fix America's Tax System: Report of the President's Advisory Panel on Federal Tax Reform. Washington, D.C.: President's Advisory Panel on Federal Tax Reform, 2005.

Reschovsky, James D., Bradley C. Strunk, and Paul Ginsburg. "Why Employer-Sponsored Insurance Coverage Changed, 1997-2003." Health Affairs 25 No. 3 (May/June, 2006): 774-82.

Rubin, Robert J. Statement on the Administration's Proposal to Limit the Exclusion of Employer-Paid Health Insurance from Employee Taxable Family Income. Washington D.C., Committee on Finance, U.S. Senate, June 22, 1983.

Schoen, Cathy, Michelle M. Doty, Sara R. Collins, and Alyssa L. Holmgren. "Insured But Not Protected: How Many Adults Are Underinsured?" Health Affairs Web Exclusive (June 14, 2005): W5-289-W5-302.

Sheils, John. Personal communication. February 2, 2006.

Sheils, John, and Randall Haught. "The Cost of Tax-Exempt Health Benefits in 2004." Health Affairs Web Exclusive (February 25, 2004): W4-106-W4-112.

Sloan, Allan. "General Motors Getting Eaten Alive by a Free Lunch." Washington Post (April 19, 2005): E03.

Steinwald, Bruce. "The Case for the Tax Cap." Office of the Assistant Secretary for Planning and Evaluation, U.S. Department of Health and Human Services. Mimeo June, 1983.

Thorpe, Kenneth E., and Curtis S. Florence. "Why Are Workers Uninsured? Employer-Sponsored Health Insurance in 1997." Health Affairs 18 No. 2 (March/April, 1999): 213-8.

U.S. Department of the Treasury. General Explanations of the Administration's Fiscal Year 2007 Revenue Proposals. Washington, D.C.: Department of the Treasury, 2006.

Weller, Christian E. Statistical Snapshot. Washington, D.C.: Center for American Progress, May 8, 2006.

Joseph R. Antes

American Enterprise

Institute, Washington, D.C. 20036

(1) In general, reducing Medicare payroll taxes does not lower the benefit a worker will receive once he reaches age 65 and enrolls in the program. Reducing Social Security payroll taxes lowers the amount that program eventually pays to retirees, although that is not likely to have a measurable effect on their decision to buy employer-sponsored health insurance. Sheils' estimate of payroll tax savings includes all payroll tax reductions.

(2) That figure represents foregone federal personal and corporate income tax revenue associated with the purchase of health insurance, contributions to health savings accounts, and out-of-pocket health spending. The estimates reported here and later in this report exclude another $6.7 billion in tax expenditures for activities not directly tied to the consumption of health services (which include hospital construction, charitable contributions, and research). They also exclude reductions in payroll taxes or state and local income taxes. One estimate of the total reduction in tax collections at all levels of government is $237 billion in 2006 (Shells, 2006).

(3) A 1943 ruling by the Internal Revenue Service excluded employer payments for health insurance premiums from the taxable income of employees (Helms, 1999). An attempt by the IRS to make such payments taxable was overruled by Congress in 1954.

(4) One analyst asserts that the rise of employer coverage did not occur because of the tax exclusion, citing the fact that employers did not typically contribute to such benefits during the 1940s (Cunningham, 2002). That ignores the 50-year record since then.

(5) In 2005, 98 percent of firms with 200 or more workers offered health insurance benefits compared to 59 percent of firms with 3 to 199 workers (Kaiser Family Foundation and Health Research and Education Trust, 2005).

(6) The top income tax bracket is 35 percent. Medicare and Social Security payroll taxes are levied at a total rate of 15.3 percent. Many states and some localities levy an additional income tax, often with a top marginal rate around seven percent. Note, however, that the 12.4 percent Social Security tax is paid on incomes up to $94,200 in 2006; above that income, people pay no additional amount. The marginal tax rate for high earners is, thus, about 45 percent. We follow the standard economic convention that workers pay the employer's share of payroll taxes by receiving lower wages than they otherwise would.

(7) The initial dollar amount of the cap could be set a variety of ways. One might use the 90th percentile of group insurance premiums, for example, or the average cost of providing a basic package of health benefits (such as hospitalization, physician services, and other services deemed "essential"), both adjusted to reflect the typical employer contribution to the cost of the full premium. A cap based on a basic benefit package would have less impact on older, sicker workers than a dollar-denominated cap (Glied, 1994).

(8) Sen. John Kerry offered such a proposal in his 2004 run for president. Medicare's prescription drug program offers a subsidy to employers who maintain their retiree drug benefit.

(9) Another approach is to offer credits whose value is lower for people who currently have employer-sponsored insurance. That reduces but does not eliminate the incentive to leave the employer risk pool.

(10) Medical savings accounts (MSAs) are the precursors of HSAs, and also tie tax-preferred savings to the purchase of high-deductible insurance. Restrictions on the scope of the MSA market (such as the limitation to employers with 50 or fewer employees) made MSAs unattractive to insurers, who largely ignored this product. Flexible spending accounts and health reimbursement arrangements, which also provide tax advantages for health spending, preceded the enactment of HSAs.
Gale Copyright:
Copyright 2006 Gale, Cengage Learning. All rights reserved.