A Guide to Tax Credits for the Uninsured

Report Health Care Reform

A Guide to Tax Credits for the Uninsured

May 4, 2000 About an hour read Download Report
James Frogue
Senior Fellow and Director of Government Finance Programs

Despite the fact that America is currently enjoying the longest uninterrupted economic boom in its history, a steadily increasing number of Americans have no health insurance coverage. This trend is likely to continue until Congress makes the necessary adjustments in current tax policy and creates tax credits that would allow the uninsured and their families to afford coverage.

There is growing consensus across the ideological spectrum that the 1950s model of employer-based health insurance still in place is no longer adequate. Typically, generations of Americans went to work for large firms and remained with them for life, and coverage was based on this model. But today's economy is characterized by increased job mobility and a proliferation of small businesses that cannot offer health benefits to their employees. Although many Americans are satisfied with their employment-based coverage, more and more individuals and families are being left out. Tax credits would create an alternative method of purchasing coverage. A number of bills now before Congress, some with bipartisan sponsorship, would create such a system .


The issue of the uninsured is prominent in health care policy debates today, and the estimates being cited are, on their face, quite stunning. According to the most recent Current Population Survey of the U.S. Bureau of the Census, 44.3 million Americans were without health insurance in March 1999--about 1 million more than were uninsured at the same point in 1998.1

The Center for Risk Management and Insurance Research at Georgia State University projects that the number of uninsured, non-elderly Americans will grow to 53 million, or about 21 percent of those who are not eligible for Medicare, by 2007. Should there be an economic downturn, or if health care costs rise faster than the Health Care Financing Administration (HCFA) projects, this same study estimates that the number of uninsured could reach 60 million--or nearly one in four Americans.2 Representative Fortney "Pete" Stark (D-CA), in testimony before the House Ways and Means Committee, cited a National Coalition on Health Care study with comparable findings: The uninsured population could soar to 61.4 million by 2009.3

The Washington-based Lewin Group, a health policy consulting firm, has issued similar estimates as well. Chart 1 shows the Lewin Group's projections, which predict a continual rise in the numbers of uninsured from 1980 to 2007, both in absolute numbers and as a percentage of the overall population. Lewin's projections are also consistent with the findings of other studies.

Clearly, the numbers of uninsured Americans has been climbing over the past decade and will continue to climb in the foreseeable future.

Who Are the Uninsured?

Perhaps surprisingly, 76.2 percent of the uninsured either are employed or are dependents of a worker.4 Uninsured Americans typically are low-income minorities who are employed in a small business in the service industry. Of the uninsured, 64 percent say the "main reason" they are uninsured is because coverage is "too expensive."5

It is instructive to note that the number of uninsured Americans is not quite as high as reported by the Census Bureau, for several reasons. For one, there appears to be a significant underreporting of Medicaid coverage. An estimated 5.2 million people claimed to be uninsured when in fact they were eligible for Medicaid. Additionally, approximately 2 million children have been covered by the Children's Health Insurance Program enacted in 1997.6 And about 3 million of those counted by the Census Bureau as uninsured were illegal aliens. This would bring the total estimate today of those uninsured down to about 33 million.7

Although the uninsured largely are lower-income workers, they are not exclusively low-income Americans. In most states, the poorest Americans--those below the poverty line--are eligible for Medicaid coverage, although not all of them are aware of it. The characteristics of the remainder of the uninsured are described in Chart 2.

Over 10 million people--workers and their dependents--are uninsured solely because they have declined their employer-provided coverage. Many of these workers are simply unable to afford the employees' share of the premiums--which is complicated by the fact that a rising percentage of policy premiums must be shouldered by the employee. In 1991, the average single worker paid 13 percent of an employer-provided plan. By 1996, 22 percent of the cost was shouldered by the employee. The employee's share of family coverage also increased from 23 percent in 1991 to 31.2 percent in 1996.8

The Congressional Budget Office (CBO) has estimated that for every 1 percent increase in premiums, 200,000 to 300,000 people lose their health insurance.9 A Lewin Group study estimated that number at 300,000 for every percentage point. Moreover, the CBO recently estimated that the House-passed Patients' Bill of Rights would cause premiums to rise an additional 4.1 percent.10

Thus, the relationship between health care costs and uninsurance is clear.

How the Current System Exacerbates the Problem

Under current law, the tax code gives considerable preference to workers whose employers have purchased their health insurance. This system became prominent during the 1950s, after wage controls during the World War II years forced employers to offer benefits such as health insurance to attract employees from the scarce numbers available. Several years after the war, the Internal Revenue Service ruled that the amount of money spent by an employer for an employee's health insurance can be excluded from an employee's taxable compensation.

This ruling means that workers do not pay any federal, state, or payroll taxes on the amount spent on their health insurance coverage. Additionally, the cost of providing health insurance to employees, like other forms of compensation, is a tax deduction for the employer. These two factors continue to provide a strong incentive for employers and employees to stay within the employment-based system. For individuals and families without access to an employer's plan, there are few or no tax breaks available today.

Cracks in the system began to appear in the late 1970s and 1980s as health insurance premiums and the cost of medical care consistently outpaced inflation and, in some cases, the willingness of employers to pay those increased costs. The predictable result was that fewer employers offered health insurance and the number of uninsured began to rise. That trend continues to this day and is expected to continue unabated. It can be attributed in large measure to the element of third-party payers who shelter the consumers of health care from its true costs.11

The Special Problems of Small Business

The problem of rising costs to employers is accentuated by the difficulty that small businesses (relative to large Fortune 500 firms) experience in affording and providing health coverage to employees. Small businesses employ 60 percent of America's workforce, yet they lack the advantages of large companies in designing and purchasing health care packages. They lack the ability, for example, to pool the risks of thousands of employees to reduce the cost per individual. In addition, the Employee Retirement Income Security Act (ERISA) allows businesses to self-insure, thereby escaping costly state mandates and regulations. Unfortunately, very few small businesses can afford to self-insure. Finally, big businesses almost universally set up sophisticated human resources departments to evaluate the available health insurance options and present the best alternatives to their employees.

Not surprisingly, in 1996, less than 50 percent of businesses with fewer than 50 employees each offered health insurance to their employees. In firms of this size that pay most employees less than $10,000, only 19 percent of workers were offered coverage.12

The way the tax code is structured for individuals is highly regressive. The more expensive a health insurance plan (the kind that tends to be offered to highly paid individuals), the greater the tax break. This is especially true the higher the marginal tax bracket of the worker becomes. Eugene Steuerle of the Urban Institute estimates that families in the top income quintile enjoy nearly six times the subsidy of families in the lowest quintile ($1,560 per year, compared with $270).13

The Lewin Group estimates that the federal government "spends" (in foregone income and payroll taxes) almost $126 billion per year to subsidize health insurance through the exclusion. States "spend" an additional $15 billion. (See Chart 3.)

Like the Steuerle findings, the Lewin Group's conclusions demonstrate the regressivity of the present tax code. The average federal tax "expenditure" per family is $1,155 annually. However, for families with incomes below $15,000 in 2000, the average federal "expenditure" is only $79, while for families with annual incomes exceeding $100,000, it is $2,638. (See Chart 4.)

This analysis illustrates the tendency of federal tax subsidies to accrue heavily to the rich while offering minimal assistance to those who need it most. It is no wonder that people with lower incomes cannot afford health insurance: They have less disposable income and are effectively discriminated against by the tax code.


There is a growing bipartisan consensus on using tax policy to improve the health care system. Both presidential candidates in the Democratic primary, former Senator Bill Bradley and Vice President Albert Gore, offered credit proposals. Bradley, in particular, offered a sweeping proposal that not only would apply to middle-income Americans, but also would replace the outdated and bureaucratic Medicaid system for low-income persons with new private health insurance options, thus mainstreaming low-income workers into America's commercial health insurance markets.

Members of Congress also see the value of allowing tax credits for the purchase of health insurance. Policymakers from across the political spectrum now recognize that the employer-based system no longer effectively serves all Americans.

In June 1999, House Majority Leader Richard Armey (R-TX) and Representative Pete Stark--the self-described "ultimate congressional odd couple"--agreed in a joint article published in The Washington Post that the problem of the uninsured is the "biggest health problem facing the country."14 They also agreed on the root causes of uninsurance: a workforce that is "increasingly mobile and part-time" and a perverse tax code that "discriminates against not only insurance purchased outside the workplace but also lower paid, part-time and small-business workers."15 They promoted the idea of refundable tax credits as "a bipartisan remedy," and identified three characteristics of a successful tax credit for health insurance:

  1. "The credit must be sufficiently generous to buy a decent policy."

  2. The credits would have to be "available to those who owe no tax liability."

  3. "To prevent fraud, (the credits would have to be) paid directly to insurers or other entities, not to individuals."16

Representatives Armey and Stark are correct: Refundable tax credits represent the best way to address the problem of uninsurance, provided these three criteria are met.

There are a number of advantages to the tax credit approach.

Advantage #1. It would restore equity to the tax code.
Because the current tax code is skewed heavily in favor of both the rich and the employer-based system, refundable tax credits, depending on their design, can target subsidies to where they are needed most--to lower-income Americans who lack coverage. They would help end the tax discrimination against 8.3 million Americans who currently purchase non-group policies.17

Advantage #2. It would promote consumer choice of health plans.
Today, only 17 percent of American employers offer their employees a choice of health plans.18 The resulting frustration felt by workers is the major reason Congress is embroiled in debate over the right to sue health plans. If individuals were allowed to choose and switch plans, as they are in the Federal Employees Health Benefits Program (FEHBP) for federal workers and Members of Congress, rates of dissatisfaction would diminish. Choice largely quells the desire to sue.

Advantage #3. It would shift control over health plans to consumers and automatically give patients a right to sue.
Tax credit policies would enable an individual to sign a contract directly with a health insurer. The insurer then becomes an agent of the individual and his family, who no longer are simply the recipients of terms and conditions that are decided by the employer and insurer. Workers and their families thus would become the principals in the contractual relationship--an arrangement radically different from today's employer-based insurance coverage in which employers are signatories to the contract. Under this plan, if the insurer violated the terms of the contract, the individual would be free to sue that insurer for breach of contract without the uncertainty of an employer's liability. Real accountability would be established in the insurance market.

Advantage #4. It would provide an alternative to the employer-based system.
As costs continue to rise, more and more employers are unable or unwilling to purchase health insurance for their employees. This trend will be exacerbated by any expansion of liability for health plans. According to a recent survey of 600 U.S. companies by Hewitt Associates, 36 percent would reconsider offering health insurance to their employees if the companies themselves were subject to possible litigation; 40 percent of that sample favored changing the tax code to allow for defined contributions, whereby employers would give employees an established amount of money so that they could go out and purchase their own insurance.19 Under current law, individuals and families wanting to purchase coverage must do so with after-tax dollars. This is prohibitively expensive for many families who lack coverage from their employer, and most simply choose to go without coverage. Tax credits would help level the playing field.

Advantage #5. It would stimulate groups other than employment-based pools to sponsor their own plans.
There exist other large, stable groups that could act as employers do and sponsor health insurance plans. Unions, for example, which in some instances have over a million members nationwide, could offer plans to their members. Some unions, such as the Mailhandlers and others, organize plans within the FEHBP. Outside of the federal government, however, union-sponsored plans are rare. Union-sponsored plans are alternatives for blue-collar workers. These workers may move around between jobs, but their union affiliation and health coverage would remain intact. Large church organizations in African-American or Hispanic communities, for example, also could sponsor plans. It is considerably more common these days for an individual in such communities to be a lifetime member of a church than a lifetime employee of a particular firm. Tax credits would facilitate the emergence of these groupings.

Advantage #6. Health insurers would be more directly responsive to the wants and needs of families.
Under the current system, in most cases, the purchaser and owner of health insurance is the employer, not the individuals and families actually covered. Therefore, insurance companies market their products to employers, whose primary concern is cost and not benefits-package design, rather than to families. Large Fortune 500 companies have employee benefit managers who take the time to study the benefit design for purposes of attracting and retaining members of the workforce, and not just to control the cost of a plan. But for a small employer under the tight pressures of running a business on a day-to-day basis, there is often neither the time nor the expertise to study the various needs of an often diverse group of workers to determine what plan will best accommodate them all. For many workers in smaller firms, it would be better for them to purchase health insurance through an affinity group or an association with a much larger risk pool, such as a union, a religious group, a fraternal organization, or perhaps a college alumni association.

Advantage #7. Consumers would have real portability of health insurance.
Tax credits would allow people to maintain their coverage even as they change jobs. The choice of doctor and insurance plan would no longer be at the mercy of an employer or a worker's employment status. Professor Mark Pauly, a leading health economist at the University of Pennsylvania's Wharton School of Business, considers this among the "main advantages" of individual insurance over group insurance.20 This holds particular appeal for younger Americans who are more apt to change jobs and start new companies. Addressing the issue of portability is also a tremendous advantage for individuals with pre-existing conditions. A person's medical history can be a barrier to employment, especially with smaller employers and employers that self-insure.


There remains a series of objections commonly raised against proposals offering tax credits so that families can have the resources to purchase health insurance outside of the employment-based system. Explanations to satisfy each one are readily available.

Objection #1. People are not capable of making decisions about something as highly complex as health insurance.
This is a common but patronizing argument. People choose a bank to handle their mortgages, colleges to educate their children, and mutual funds in which to invest, as well as automobile, fire, and life insurance. Ordinary Americans are just as capable of purchasing a health insurance plan as federal employees are of picking their own plan through the FEHBP. People's interest in picking a suitable plan for themselves and/or their families is far greater than that of any employer.21 This self-interest also suggests that they would spend substantially more time and effort in selecting appropriate coverage.

Objection #2. The cost of individually purchased plans would be significantly higher than the cost of group-purchased plans, and therefore unaffordable.
A recent study done by Professor Pauly found that the administrative cost differential between individual and group policies has been shrinking steadily since 1970. In a world of tax neutrality--where tax credits are available to individual purchasers--Pauly predicts that this difference would shrink even further. Mass-marketed, non-group health insurance purchased over the Internet, for example, likely would experience changes that mirror what happened to automobile insurance in the 1970s, when new policies permitted people to shop for the lowest rates over the telephone.22 In addition, as discussed above, tax credits would facilitate the purchase of health insurance by groupings other than employers, such as unions and church groups, so families would not necessarily be restricted to individually purchased policies.

Objection #3. The market for individual insurance is too small to handle the influx of millions of people.
The reason the individual insurance market is small is the lack of demand for such insurance. Less that 7 percent of the population has non-group insurance.23 This can be attributed to the fact that the tax code strongly favors employer-provided coverage. If tax credits were available, individuals and their families would have the resources with which to purchase plans. The market would respond accordingly. Demand inevitably creates supply.

Objection #4. Tax credits would ruin the employer-based system; younger and healthier workers would leave to buy their own (cheaper) plans, leaving businesses to insure an older and sicker (more costly) pool of workers.
The purpose of the tax credit approach need not be to undermine the employer-based system, but to offer an alternative to a system that has been shown to be increasingly inadequate. Design elements could be included such as allowing the credit only in cases where the worker is not offered insurance from an employer. This would erect a "wall of separation" and ensure that successful employer plans are not threatened.

Objection #5. Lower-income people would not utilize a tax credit, and therefore would not gain coverage, because they cannot afford the luxury of waiting until the following year for a tax refund.
Incorporated into the tax credit bills introduced by Representatives Armey and Stark is language creating pre-payment of the credit. A person signing up for health insurance could have the value of the credit transferred directly to the insurance provider from the Treasury Department. This need not be complicated. It could simply be factored into the withholding of the insurance company. This important feature would maximize the effectiveness of any credit by ensuring that it is readily available to those most in need.

Objection #6. Much of the credit would simply go to people who already are buying coverage, thereby costing the government lost revenue without reducing the number of uninsured.
As many as 8.3 million Americans who currently purchase non-group coverage could benefit from a tax credit, depending on eligibility restrictions.24 These individuals currently are discriminated against by the tax code. Allowing them tax treatment similar to that enjoyed by those in employer-sponsored plans is a simple matter of fairness, and ending the inequity is one of the major goals of the tax credit approach. While there are various design proposals for tax credits, even the least generous (a 30 percent credit for workers without access to employer coverage) would still add an estimated 1.5 million people to the ranks of the insured.25 Thus, not only would credits help to alleviate tax code discrimination for over millions of Americans, but the tax change also would reverse the steady increase in the number of uninsured.

Objection #7. Tax credits will not insure everyone.
This is a fallacy. Any government program that is voluntary will not have full participation by all people that are eligible. Regardless of the tax credit approach taken, and short of making health insurance a federal requirement, there always will be those who remain uncovered for any number of reasons. Persuading all young, healthy, lower-income single men, for example, to spend scarce money on health insurance because of what "might" happen will be impossible. But just because tax credits will not cover everyone does not mean that the approach should be discarded. The fact is that tax credits could add millions of Americans to the insurance rolls, reversing the current trend, and in the process restore equity to the tax code.


While there is broad agreement that tax credits can help the uninsured, the impact of any particular proposal will depend on the credit design, which in turn depends on a number of considerations.

The first decision is whether or not to make the credit refundable. A refundable credit means that even a taxpayer with no federal tax liability would receive a "refund" from the government in the amount of the credit for the purchase of health insurance. If a single taxpayer, for example, owed $600 in federal income taxes but had a $1,000 tax credit, then instead of having to send the federal government $600, that taxpayer would receive $400 from the government.

It is widely accepted that making a credit refundable is critical if it is to have a significant impact on reducing the number of uninsured. This is because 45 percent of the uninsured today are not liable for federal income tax.26 If a credit were not refundable, it would be of zero value for nearly half of the uninsured population.

The second decision is how to structure a tax credit. There are basically two ways:

  1. A fixed dollar credit typically would involve establishing set amounts for single taxpayers, married couples, and families (depending on the number of children). Each type of household would be assigned a maximum fixed dollar credit that could be used to offset the cost of a health insurance plan. (Rarely, if ever, would an approved plan cost less than the credit amount; but if it did, the credit would only match the amount spent on the plan.) A flat credit would likely be of more value to lower-income people who have less money to put out "up front" for a policy. If a family was eligible for a $3,000 tax credit, for example, a properly designed credit would allow them to have that money sent directly to an insurer to receive, depending upon availability, a policy at no or little cost. A fixed dollar credit has greater potential to limit the inflationary impact that tax subsidies have on health care costs.

  2. A percentage credit would offset the cost of a plan at whatever percent was made law. If a family purchased a $4,000 plan and had a 50 percent credit, they would subtract $2,000 from the amount of tax they owed to the federal government. A percentage credit would be fairer to taxpayers with higher incomes that wished to purchase a plan with a more expensive benefits package.

Some credit proposals would target the credits only to taxpayers below certain income limits. This is meant to ensure that resources are dedicated exclusively to lower-income people, who make up a disproportionate number of the uninsured. While this has certain appeal, it raises effective tax rates significantly for people whose income progresses through the income limit and past the phaseout.

For example, let us say that single taxpayers with up to $30,000 in adjusted gross income (AGI) were eligible for a $1,000 tax credit and that this credit was phased out completely by $40,000 in AGI. If a single individual earned $30,000 in AGI and took the $1,000 tax credit, his effective tax rate would be 8.1 percent.27 If he were suddenly given a raise to $40,000, his effective tax rate would skyrocket to 14.4 percent.28 This policy would be thought of as unfair to the affected taxpayer.

Another consideration is whether to restrict the credit or allow it for employees who opt out of their employer plans. Several proposals would restrict the tax credits to people who are not offered insurance at work. This restriction is designed to avoid offering incentives for healthier and presumably less costly workers to exit the employer-based system. The fear expressed by some, especially in the business community, is that if these younger, healthier employees took advantage of a credit, employers would be left insuring the more costly workers. This potential result could be mitigated by making the credit available only to people who are not offered insurance at work.

The downside to limiting the use of the credit to those not offered insurance at work is that it would not be available to assist those Americans who have to pay a portion of their premiums. Today, 10.2 million Americans--nearly 25 percent of the uninsured population--are workers (or dependents of workers) who have declined insurance that was offered by an employer.29 Most do so as a result of being unable to afford their portion of the cost-sharing. They would receive no help if the credit is limited to people who are not offered insurance at work.

A tax credit also could include a pre-payment mechanism, whereby resources are transferred directly from the U.S. Treasury to the health insurer on behalf of the individual or family purchasing coverage. This need not be complicated and could be done simply by the insurer's changing its tax withholding. Similar to making the credit refundable, this feature is critical if the credit is to be utilized by lower-income Americans. In many cases, lower-income taxpayers are unable to pay out the money for health coverage up front and recoup it a year later in the form of a tax refund. Making the credit pre-payable would help taxpayers to afford coverage when they really need it.


The estimated take-up rate of a proposed tax credit is another major factor to consider. This refers to the number of Americans who would become "newly insured" as a result of a tax credit policy. The Lewin Group has analyzed the cost and take-up rate of six proposals and the Heritage plan to provide alternative tax treatment for the purchase of health insurance.30

  1. A tax deduction:
    Before analyzing credits, Lewin looked at the effects of allowing a tax deduction for the full premium costs of non-group individually purchased policies. The deduction modeled was above-the-line (not itemized), and therefore was available to all taxpayers regardless of income. With this new deduction, premium payments would no longer count in determining health expenses above 7.5 percent of adjusted gross income, the threshold for the itemized health deduction today, because they would have been deducted already.31

Approximately 51.7 million Americans would qualify for this new deduction, of which 43.4 million are currently uninsured and 8.3 million currently have non-group coverage.32 The Lewin Group assumes that all 8.3 million people currently buying non-group coverage would take advantage of the deduction, while 3.9 million previously uninsured people would be likely to purchase coverage.

Such a deduction would "cost" the federal government $6.3 billion in reduced tax revenue, with $2.7 billion going to the newly insured and the remainder to those already buying insurance. The average annual cost per newly insured person would be $1,599.

  1. A credit of $500 per individual and $1,000 per family:
    In the design studied, availability would be restricted to people who are not participating in an employer-sponsored plan. As with the deduction proposal, 51.7 million persons could be eligible for the credits. Of those, all of the 8.3 million currently buying coverage would take the credit. An additional 4 million uninsured individuals would obtain coverage as well. Under this plan, the government would lose $5 billion in revenue; $1.7 billion would go to the newly insured and $3.3 billion to those previously with coverage. The average cost of the newly insured would be $1,247.

  2. A 30 percent credit for workers without access to employer coverage, workers who make below $35,000 (AGI), and married couples who make below $50,000 (AGI):
    In this scenario, the full 30 percent credit would be allowed for single taxpayers with an AGI below $25,000, phasing out at $35,000. For married couples and families, it would be $40,000 and $50,000, respectively. The Lewin Group estimates that 15 million workers and their dependents would be eligible for this credit; 2.8 million would already have coverage, and 12.2 million would be uninsured. It is assumed that all of the 2.8 million with insurance would utilize the credit, in addition to 1.5 million of the uninsured. The lower take-up rate is due to the modesty of the credit. Many low-income Americans cannot afford to pay 70 percent of the cost of a typical health insurance policy. The cost for the government is estimated to be $3.3 billion, with $2 billion of that going to those already with coverage and the rest dedicated to the newly insured, which would mean a cost of $2,121 per newly insured person per year.

  3. A 30 percent tax credit to all workers and non-workers, with the same income restrictions as in the previous model:
    This credit would be available to workers who do not have access to employer-provided insurance, those who contribute a portion to the cost of their employer-provided plan and those who do not work.33 This option would insure more people because it also would be available to many of the 10.2 million workers and dependents who currently decline employer coverage. About 6.1 million Americans falling under these income limits decline employer coverage. Lewin estimates 1.8 million of these people would use this credit to purchase coverage. Another 1.5 million (as discussed above) who do not have access to an employer plan also would gain coverage, while 1.1 million persons in non-worker families also would buy coverage. That brings the total of newly insured to 4.5 million at a cost to the government of $11.3 billion. The average yearly cost of the newly insured would be $2,530.

If the percentage of the credit were increased to 50 percent or 80 percent, the take-up rate would be significantly higher. A 50 percent credit available (with the same income restrictions as above) to both workers and non-workers, who could use the credit whether or not an employer contributes to their plan, would add 7.1 million workers to the ranks of the insured at a cost to the government of $21.9 billion. An 80 percent credit would pick up 14.6 million persons at a cost of $50.2 billion.

  1. A fixed dollar credit of $800 per taxpayer, $400 per child, and $2,400 maximum per family, which would replace today's exclusion: In this scenario, all individuals and families are eligible for the same refundable credit amounts. The credit could be used for group or non-group policies. It would replace the exemption currently enjoyed by employees when they receive compensation in the form of health benefits. Workers thus would pay income tax on the money that currently is spent by an employer on their health insurance.34

All 210 million Americans not covered by Medicare, Medicaid, or some other public program would be eligible for the credit. This includes both the 43.4 million uninsured and the 166.4 million with private coverage. It is estimated that 9.8 million of the currently uninsured would gain coverage, but 5.2 million currently with coverage would become uncovered due to a net reduction in their tax subsidies. That leaves a net increase in the insured population of 4.6 million persons at a cost to the government of $48.6 billion. The net cost would be $10,541 per newly insured person.

The advantage of this approach is that it makes more equitable use of federal tax exclusions and deductions. Under the system now in place, families earning $15,000 per year or less receive on average $79 in federal subsidies, whereas families earning over $100,000 receive $2,638. With this plan in place, subsidies to the poorest Americans would increase over 400 percent to $402 annually. Subsidies to the richest Americans would increase only 6.8 percent to $2,821. The increase for the average American family would be nearly 45 percent--from $1,155 to $1,670. All income groups would see an increase in federal subsidies for the purchase of health insurance, on average.

  1. A choice of credit or exemption:
    Another model looked at by Lewin examined the impact of allowing taxpayers a choice between maintaining their current employer exemption and taking the fixed dollar credits in the amounts above. For 45 percent of families, taking the credit would be most advantageous. The remaining 55 percent would stay with their current exclusion. Under this proposal, 9.8 million Americans would be added to the rolls of the insured. The government would lose $53.2 billion in revenues, making the cost of each newly insured person $5,429 annually.

The Heritage Foundation Proposal

The Heritage proposal calls for an elimination of the current tax exclusions for employer health benefits, including employer and employee contributions to Section 125 plans. It also would eliminate the deduction for medical expenditures in excess of 7.5 percent of adjusted gross income. The revenue from eliminating these tax breaks would be used to partially fund an identical refundable tax credit available to all Americans to cover health insurance premiums and out-of-pocket expenses. The credit would reimburse 25 percent of health care spending below 10 percent of AGI, 50 percent between 10 percent and 20 percent of AGI, and 75 percent for any spending above 20 percent of AGI.

Under this plan, employers would be required to cash out their health plans and individuals would be required to purchase coverage in the individual market. State-mandated insurance benefits and restrictions on managed care plans would be preempted nationwide by federal law. A federally determined minimum benefits package would be established, and premiums would be permitted to vary only by age, sex, geography, and family size--not by health status.

Due to the requirement that all individuals purchase health insurance, the Heritage plan would cover all 43.4 million uninsured Americans. The net cost of the plan would be $55.3 billion, for an average cost per newly insured of $1,274. Significantly, the Heritage plan would be most beneficial to families with incomes below $15,000. As mentioned above, the average current federal tax subsidy for families in this income bracket is $79. Under the Heritage plan, the average subsidy for this group would increase to $2,064. For the highest-income grouping, families with incomes over $100,000, the average subsidy would decline slightly, from $2,638 to $2,170.

Comparing Tax Incentives.
Jonathan Gruber, a professor of economics at the Massachusetts Institute of Technology, and Larry Levitt, the director of the Changing Health Care Marketplace Project of the Henry J. Kaiser Family Foundation, also performed a study using various tax incentives. Their base model was a refundable tax credit of a fixed $1,000 per individual and $2,000 per family. The credit would be available to single filers with adjusted gross incomes below $45,000. The credit would phase out at $60,000. For joint filers, the figures would be $75,000 and $100,000, respectively. The tax credit could not be used to offset employee contributions to an employer's plan, but individuals who opt out of employer coverage could use it.35

Gruber and Levitt found that the new credit would cost the federal government $13.3 billion per year. Those taking the credit would include 18.4 million persons, of which 4.7 million would be from the previously uninsured, 8.6 million who already purchase non-group insurance, 4.7 million who have employer coverage, and 400,000 on Medicaid. Just over 4 million persons on net would gain insurance, for an average annual cost per newly insured of nearly $3,300. However, the Gruber and Levitt estimates are based on the present cost of non-group coverage.36 If the average cost per policy were lower, more people would gain insurance. In a world in which credits of this size were available, it is entirely possible that an insurance product could emerge that is equal to the value of the credit amount.37 This would not be a comprehensive policy, but it would be "free" to the person utilizing the credit, thereby increasing take-up rates significantly, decreasing the average cost per newly insured, and providing coverage where it did not previously exist.


There is an ever-increasing number of health care tax credit bills in Congress. Several more are in development and will be introduced in the near future.

H.R. 1136: The Affordable Health Care Act of 1999

H.R. 1136, sponsored by Representative Charles Norwood (R-GA), includes several refundable fixed dollar tax credits. The legislation offers a $1,200 credit to each adult and $600 per child, up to a family maximum of $3,600 if insurance is not offered at work. If an individual declines employer coverage, the credit amounts are reduced to $400, $200, and $1,200, respectively. The purpose of these differing credit amounts is to minimize disruption of the employer-based system by offering credits so large that even those people insured at work might be tempted to leave the system to purchase coverage on their own. The credits in H.R. 1136 are available to anyone regardless of income level.

This legislation contains other provisions to expand access to health insurance. It includes language to help create "Healthmarts," or health insurance supermarkets that would serve a defined geographical region. An employer, if it chooses, could give a voucher to an employee in an amount equivalent to what it would contribute to the employer-provided plan. The employee could take that voucher, which would remain excluded from taxable compensation, down to the Healthmart and purchase a plan from among all available policies.

H.R. 1136 also would help create association health plans (AHPs) by amending the Employee Retirement Income Security Act to enable small-business trade associations to band together across state lines to purchase health insurance policies just as Fortune 500 companies do. Small-business owners and their employees who are members of the National Federation of Independent Business (NFIB), for example, could access the NFIB plan regardless of their state of residence. These alternative pooling mechanisms would be advantageous to small-business owners and their employees because they would increase the number of available insurance options. Individuals and their families could use the tax credit toward the cost of this coverage.

Finally, H.R. 1136 would lift the cap on the number of medical savings accounts (MSAs) by repealing the artificial limit on the number of MSAs set at 750,000 by the Health Insurance Portability and Accountability Act of 1996. All employers would be permitted to offer MSAs.

H.R. 1177: The Health Insurance Affordability Act

H.R. 1177, sponsored by Representative Steve Chabot (R-OH), would allow taxpayers, whether or not they itemize deductions on their returns, to deduct what they spend on insurance premiums in addition to a portion of their uncompensated medical expenses that exceed 7.5 percent of their adjusted gross income. The fact that H.R. 1177 contains no language stipulating whether this deduction is available to workers who contribute to an employer-provided plan, however, creates some confusion. Under current law, a taxpayer can deduct the cost of health insurance premiums and out-of-pocket expenditures only if the sum exceeds 7.5 percent of his or her adjusted gross income.

Approximately 51.7 million Americans would qualify for this deduction, including all of those currently without insurance and the 8.3 million who purchase non-group coverage. The Lewin Group assumed that all of these 8.3 million would take advantage of the deduction, in addition to 3.9 million of the uninsured. Because deductions are useful only to individuals who are liable for federal income tax, the benefits of H.R. 1177 would be of little assistance to lower-income taxpayers.38

H.R. 1687: The Patients' Health Care Choice Act of 1999

H.R. 1687, sponsored by Representative John Shadegg (R-AZ), would create refundable tax credits in the amounts of $500 per individual and $1,000 per family for the purchase of health insurance. The credits would be available to taxpayers when insurance is not obtained through an employer. Opting out of an employer plan and using these credits would be permissible.

Lewin examined this scenario and estimated that 4 million persons would become covered that were not previously covered. All of the 8.3 million persons currently purchasing non-group insurance would use the credits as well. The annual cost to the federal government would be $5 billion, for an average cost per newly insured of $1,247.

H.R. 1687 is similar to the Norwood H.R. 1136 bill in that it would create Healthmarts, allow for AHPs, and lift restrictions on MSAs. These measures, in addition to the tax credits, would simultaneously expand health insurance options for Americans while allowing them greater resources with which to purchase coverage.

H.R. 1819: The Working Uninsured Tax Equity Act of 1999

H.R. 1819, introduced by Representatives Jim McDermott (D-WA) and James Rogan (R-CA), would allow taxpayers a credit equivalent to 30 percent of the value of a health insurance policy. A taxpayer could use this credit if he or she is not eligible to participate in an employer's plan. This could occur if they are part-time or temporary workers or because their employer does not offer insurance at all. Use of the credit would be restricted to individuals who make under $30,000 in adjusted gross income and married couples with under $50,000 in AGI. The credit would phase out to zero at $10,000 above each of these amounts. This credit is unique in that it would be partially refundable. It could exceed the amount of income tax paid, but it could not exceed the sum of income tax and Social Security taxes.

The Lewin Group model examined that most closely resembles the proposal in H.R. 1819 estimates that approximately 1.5 million uninsured Americans would be added to the ranks of the insured with a tax credit of this scope.39 If the credit were also made available to workers who pay a portion of their employer-sponsored plan as well as non-workers, 4.5 million could gain insurance. Maintaining this expanded eligibility and increasing the percentage of the credit to 50 percent or even 80 percent of the cost of premiums would reduce the number of uninsured Americans by 7.1 million and 14.6 million, respectively.

H.R. 2185: The Health Insurance for Americans Act of 1999

H.R. 2185, sponsored by Representative Pete Stark, would allow a $1,200 credit per taxpayer, $600 per child, and a maximum of $3,600 per family per year. These amounts would be adjusted upward annually by the rate of inflation. The credits are fully refundable, so they would be of assistance to lower-income individuals and families by helping them to afford health insurance. Use of the credits is permitted for anyone who does not participate in an employer-subsidized plan. This would allow workers to select from plans beyond those offered by their employers. Additionally, this legislation has an advanced payment mechanism so lower-income Americans would have immediate means with which to purchase a policy. Title I of H.R. 2185, except for its slightly larger credit amounts, is similar to Title I of H.R. 2362, the tax credit bill introduced by Representative Richard Armey.

The Lewin model that most closely resembles H.R. 2185 is the credit/exemption model explained above. This proposal offers taxpayers a choice between taking their current exemption or the tax credits (in the Lewin analysis, the tax credits were $800 per adult and $400 per child, with a family maximum of $2,400, which is 33 percent less than the Stark bill's credit amounts). The Lewin analysis estimated that this model would add 9.8 million Americans to the ranks of the insured. It would cost the government $53.2 billion, for an average cost per newly insured of $5,429. Because the credit amounts in the Stark bill are larger, the total cost would be greater, but more people would become insured. While the average cost per newly insured is high, this approach, as discussed in the previous section, makes more equitable use of federal tax exemptions and deductions. In short, lower-income Americans would benefit more than upper-income Americans. The higher credit amounts in the Stark bill would be even more advantageous to lower-income Americans and likely would add more people to the ranks of the insured.

Title II of the Stark bill defines qualified health insurance. As a condition for an insurance carrier's participation in the FEHBP, an insurer would have to make available the same (or actuarially equivalent) plans to individuals eligible for this tax credit. Premiums charged would have to be community-rated.40 Title II also would create an Office of Health Insurance in the U.S. Department of Health and Human Services that would be responsible for determining whether a plan is a "qualified health insurance" and therefore eligible for the tax credit.

H.R. 2261: The Health Insurance Affordability and Equity Act of 1999

H.R. 2261, sponsored by Representative Nancy Johnson (R-CT), offers a tax credit equal to 60 percent of the cost of a health insurance premium. It would be available to persons with COBRA coverage41 and those who have not been eligible for an employer plan for at least one year. The maximum amount of the credit is $1,200 for an individual and $2,400 for a married couple and/or family. The credits could be used by single taxpayers with adjusted gross incomes of up to $30,000 and married couples with AGIs of up to $60,000. Like the credit in the bill sponsored by Representatives McDermott and Rogan, the credit in this legislation would phase out to zero at $10,000 above these limits. This credit would not be refundable. For other individuals, H.R. 2261 would allow a tax deduction to help defray the cost of health insurance premiums, to be phased in as follows: 60 percent in 2000, 70 percent in 2001, 80 percent in 2002, 90 percent in 2003, 100 percent in 2004 and thereafter. The deduction could be used only if the employee paid 50 percent or more of his or her premiums.

The 60 percent tax credit is twice that proposed in the McDermott-Rogan bill, but the fact that it is not refundable means that it would be of little or no value to lower-income taxpayers. Making the uninsured wait a year before they can use the credit to help purchase coverage should be eliminated. Including a prepayment mechanism would make it more helpful to poorer persons. The allowed deduction is also of little use to those with lower incomes.

H.R. 2362: The Fair Care for the Uninsured Act of 1999

H.R. 2362, sponsored by House Majority Leader Richard Armey, offers a $1,000 per adult and $500 per child tax credit up to a family maximum of $3,000. Like those in H.R. 2185, the amounts would be adjusted upward for inflation annually. Also like those in H.R. 2185, the credits would be fully refundable and could be prepaid to the insurance provider from the Treasury on behalf of the individual or family purchasing the coverage. The credits would be available to workers who opt out of an employer's plan and to taxpayers of any income level.

The Lewin model closest to the tax credit proposed in the Armey bill is the credit/exemption model. Like H.R. 2185's proposals, it would allow taxpayers to choose between taking their current exemption or the tax credits. In the Lewin model, the tax credits were $800 per adult and $400 per child, with a family maximum of $2,400, 20 percent less than the credits in H.R. 2362. In the Lewin scenario, 9.8 million Americans would be added to the ranks of the insured. It would cost the government $53.2 billion, for an average cost per newly insured of $5,429. H.R. 2362, like the Stark bill and the Lewin model, would make more equitable use of federal tax subsides. This is relatively more beneficial for lower-income Americans than it would be for upper-income Americans.

H.R. 2990: The Quality Care for the Uninsured Act

H.R. 2990, introduced by Representative James Talent (R-MO), was passed by the House on October 6, 1999, and contains a wide array of provisions to assist the uninsured. Like H.R. 1687 and H.R. 1136, it would allow the creation of association health plans and Healthmarts. It also would expand access to MSAs and accelerate to 2001 the point at which the self-employed could deduct 100 percent of their health insurance premiums.

For individuals, this proposal would phase in an above-the-line deduction (available to taxpayers whether or not they itemize) for the cost of health insurance premiums if the taxpayer pays at least 50 percent of the cost of his insurance. It would be phased in as follows: 25 percent from 2002 to 2004, 35 percent in 2005, 65 percent in 2006, and 100 percent in 2007 and thereafter.

Even if this deduction were fully phased in seven years from now, it would add far fewer than 4 million people to the ranks of the insured, because the deduction could be used only for those who pay over 50 percent of their policy premium. Fewer than half those in employer-provided plans do so now.

S. 2320: The Health Coverage, Access, Relief, and Equity Act

Senator James Jeffords (R-VT) introduced S. 2320 on March 29, 2000. Its original cosponsors form a bipartisan group: Senators John Breaux (D-LA), Bill Frist (R-TN), Blanche Lambert-Lincoln (D-AR), and Olympia Snowe (R-ME).

The Health Coverage, Access, Relief, and Equity (CARE) Act offers refundable tax credits in the amount of $1,000 for individuals and $2,000 for families for the purchase of health insurance. The credits could not be used by individuals who are eligible to participate in a plan subsidized by an employer. Only individuals with adjusted gross incomes below $35,000 and families with AGIs under $55,000 could use the entire credit amounts. The credits would phase out to zero at increments of $10,000 above these amounts.

Gruber and Levitt analyzed a scenario with the same credit amounts and in which employees could not use the credit to offset their portion of an employer-subsidized plan, but their scenario had slightly higher income limits and permitted employees to opt out of employer-sponsored coverage. In that study, it was estimated that 4 million new people would be added to the ranks of the insured. S. 2320 likely would add roughly the same number of people because few people would exit the employer plans for these credit amounts, and the different income restrictions would not account for significant variation.

The CARE Act is designed to avoid disrupting the employer-based system while simultaneously offering an alternative for those without access to an employer plan. If eligibility could be extended to assist the 10.2 million Americans and their dependents who have access to an employer plan but are not covered because they cannot afford to pay their portion of the premiums, the estimated take-up rate would be much higher.


To reduce the numbers of uninsured Americans, Congress should first address the inherent unfairness of the tax code. Not only is it highly regressive, giving the greatest benefits to those who make over $100,000 per year and those who participate in expensive employer-sponsored health plans, but it also fails to assist those who cannot afford or have no access to employment-based plans. The predictable result is that many lower-income Americans remain without health insurance.

The number of uninsured Americans is growing, despite an economy that continues to set records for productivity and job growth. Evidence that tax credits, such as those proposed in the various bills discussed here, could restore equity to the tax code and reverse this trend is also growing. In the process, such credits would empower more individuals and families to make their own health care decisions--the best way to assure health care security.

James Frogue is a former Health Care Policy Analyst at The Heritage Foundation.

1. Amy Goldstein, "Ranks of Uninsured Americans Swelling," The Washington Post, October 4, 1999, p. A1. See also /static/reportimages/738F09710E0A56F8E1E4681ABCDC96C1.pdf

2. William S. Custer, Ph.D., "Health Insurance Coverage and the Uninsured," prepared for Health Insurance Association of America, 1998. See http://www.hiaa.org/news/news-state/custer.htm

4. John Shiels, testimony before Subcommittee on Health, Committee on Ways and Means, U.S. House of Representatives, 106th Cong., 1st Sess., June 15, 1999, at http://waysandmeans.house.gov/health/106cong/6-15-99/6-15shei.htm

5. See Kaiser Family Foundation, "Uninsured in America, Chart Book," p. 32, at http://www.kff.org/content/archive/1407

6. CHIP (or S-CHIP) was enacted as Title XXI of the Social Security Act in the 1997 Balanced Budget Act.

7. John F. Shiels, "Changing the Tax Treatment of Health Benefits," paper presented to the Congressional Health Care Tax Reform Retreat, sponsored by The Heritage Foundation, Annapolis, Maryland, December 1, 1999.

8. Shiels, testimony before Subcommittee on Health.

9. Congressional Budget Office, "CBO's Estimate of the Impact on Employers of the Mental Health Parity Amendment in H.R. 3103," May 13, 1996.

10. "Legislation Would Raise Health Care Premiums," The Wall Street Journal, February 11, 2000, p. A16.

11. For a more detailed discussion of the current system and the problems inherent with third-party payers, see Michael Tanner, "What's Wrong with the Present System," in Grace-Marie Arnett, ed., Empowering Health Care Consumers Through Tax Reform (Ann Arbor: University of Michigan Press, 1999), pp. 27-35. See also Regina Herzlinger, Market Driven Health Care (Reading, Mass.: Addison-Wesley Publishing Co., 1997), pp. 249-253.

12. See U.S. General Accounting Office, Employment-Based Health Insurance: Medium and Large Employers Can Purchase Coverage, But Some Workers Are Not Eligible, GAO/HEHS-98-184, July 1998.

13. Arnett, ed., Empowering Health Care Consumers Through Tax Reform, p. xxxix.

14. Dick Armey and Pete Stark, "Medical Coverage for All," The Washington Post, June 18, 1999, p. A41.

15. Ibid.

16. Ibid.

17. John Shiels, Paul Hogan, and Randall Haught, "Health Insurance and Taxes: The Impact of Proposed Changes in Current Federal Policy," The Lewin Group, report prepared for the National Coalition on Health Care (NCHC), October 18, 1999, p. 11.

18. Steven Long and M. Susan Marquis, "How Widespread Is Managed Competition," Center for Studying Health System Change, Data Bulletin No. 12, Summer 1998, p. 1.

20. Mark Pauly, Ph.D., et al., "Individual Versus Job-Based Health Insurance: Weighing the Pros and Cons," Health Affairs, Vol. 18, No. 6 (November/December 1999), pp. 28-44.

21. For a discussion of the sophistication of the American health care consumer, see Herzlinger, Market Driven Health Care, Chapters 1, 2, 3, 4, and 11. See also Gina Kolata, "Web Research Transforms Visit to the Doctor," The New York Times, March 6, 2000, p. A1.

22. Pauly et al., "Individual Versus Job Based Health Insurance," pp. 33-39.

23. Ibid., p. 37.

24. Shiels, Hogan, and Haught, "Health Insurance and Taxes," p. 14.

25. Ibid., p. 22.

26. Jonathan Gruber and Larry Levitt, "Tax Subsidies for Health Insurance: Costs and Benefits," Health Affairs, Vol. 19, No. 1 (January/February 2000), p. 79.

27. The effective tax rate is the total tax liability divided by gross income. An individual earning $30,000 would get an estimated standard deduction and personal exemption totaling $7,200 for tax year 2000. The taxable income would be $22,800. A 15 percent tax rate applied to $22,800 yields a tax liability of $3,420. After subtracting $1,000 for the credit, the individual owes $2,420 to the federal government, which when divided by the gross income yields an effective tax rate of 8.1 percent ($2,420 / $30,000).

28. The $40,000 single earner has the same standard deduction and personal exemption as the $30,000 earner. Taxable income is $32,800, with a tax liability of $5,771.50 (only the first $26,250 is taxed at the 15 percent rate; the remainder is taxed at the 28 percent rate). Since there is no health insurance tax credit, the effective tax rate is 14.4 percent ($5,771.50 / $40,000). Note: The number of tax dollars going to the federal government increases nearly 140 percent with this $10,000 raise.

29. Shiels, testimony before Subcommittee on Health.

30. These examples are from Shiels, Hogan, and Haught, "Health Insurance and Taxes."

31. Under current law, individuals can only deduct health expenses above 7.5 percent of AGI.

32. Since release of the Lewin report, the figure of 43.4 million uninsured has been updated by the Bureau of the Census to 44.3 million.

33. The credit could not be used to offset tax-free contributions to Section 125 "cafeteria" plans.

34. The exemption would be lifted for the employer's share of health benefits, employee contributions to Section 125 plans, and the deduction for the self-employed. It would be retained for retirees participating in Medicare or Medicaid.

35. Gruber and Levitt, "Tax Subsidies for Health Insurance: Costs and Benefits," pp. 75-76.

36. From a conversation with Dr. Jonathan Gruber, February 25, 2000.

37. It would be made difficult, however, by expensive state mandates, which vary from state to state.

38. The same can be said of H.R. 145, introduced by Representative Gene Green (D-TX), which allows a deduction for the purchase of health insurance but is phased in over eight years, starting at 45 percent of premiums in 1999 and increasing to 100 percent in 2007.

39. In the Lewin model, individuals making under $25,000 are eligible for the full credit, as are families under $40,000. For both, eligibility phases out at $10,000 above each of these amounts. Because H.R. 1819 has slightly higher limits, it can be expected to "cost" the federal government more and will insure more people.

40. Community rating refers to the policy of charging the same premium to all persons in a defined region without regard to age, sex, health status, or address.

41. The Consolidated Omnibus Reconciliation Act of 1985 permits an employee to keep his or her employer coverage for up to 18 months after leaving a job, provided the employee pays the full premium cost.


James Frogue

Senior Fellow and Director of Government Finance Programs