Library / Publications
Tax Credits for Health Insurance
The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders.
Over 40 million Americans under age 65the overwhelming majority of them in working familieslack health insurance. They are less likely to obtain important preventive screenings while healthy, and they receive lower-quality care when sick.1 And, the public ultimately shoulders the burden of paying for the medical treatment of those lacking insurance, through either higher taxes or higher health care costs.
Moreover, health insurance costs more than it would in a perfect market, for several reasons. First, the very act of having insurance tends to increase utilization. People spend more when someone else is writing the check, but this causes insurance to be more expensive than it might be (a phenomenon known as moral hazard). Second, insurance tends to be most attractive to people who expect to benefit most from itsuch as those with chronic conditions and people who plan to have children. Since insurers can only imperfectly match premiums to expected utilization, they have to assume that purchasers have higher costs than the population average. That means that healthy people get a relatively bad deal from insuranceunless they can align themselves with a large group. (This feature of insurance is called adverse selection.) Third, the existence of freeeven if inadequateemergency health care for those with low incomes serves as a deterrent for purchasing health insurance, both because the free care provides a safety net and because uncompensated care tends to raise the cost of care for those with insurance. Finally, healthy peopleespecially in the non-group marketcan only imperfectly insure against the costs of developing chronic illnesses, because premiums for non-group health insurance tend to increase over time for sick people.
The government, in fact, intervenes heavily in the market for health insurance. Low-income households (and especially low-income children), those deemed "medically needy," military families and veterans, and the elderly all benefit from publicly provided insurance. Other working-age individuals and families receive substantial tax subsidies. Health insurance paid for by employers is a tax-free fringe benefitexempt from both income and payroll taxes. In addition, self-employed individuals can deduct the cost of health insurance premiums from their taxable income. These tax subsidies are worth over $140 billion a year.
The subsidies have worked in one sense: employer-sponsored insurance (ESI) covers more than two-thirds of workers and their families. Arguably, encouraging individuals to get insurance at work deals with the problem of adverse selection and also offers those who work for large firms a kind of renewable insurance (at least as long as they continue working and their employer continues offering insurance). However, the tax subsidies are poorly targeted. The value of a tax exclusion grows with income and is worth little or nothing to those with low incomes, even though they are most likely to be deterred by the cost of insurance.
The tax subsidies also tend to exacerbate the moral hazard problem mentioned above. Higher-income employees tend to value insurance very highly, in part because of the tax benefits. As a result, they tend to acquire relatively generous coverage. To address this problem, Congress enacted a provision in 2003 aimed at encouraging employees to purchase high-deductible health insurance, either directly or through their employers. Individuals with qualifying high-deductible health insurance can contribute pre-tax dollars in a health savings account (HSA), and withdrawals used to pay for medical care are also tax-free. Employer contributions to HSAs receive the same generous tax treatment as contributions to employer-sponsored insurance. This kind of turbocharged IRA is very valuable to higher-income (high tax bracket) employees, especially those who are healthier than average.
Numerous proposals would provide additional tax subsidies for health insurance. Most notably, for the past four years President Bush has proposed to provide a refundable tax credit for the purchase of health insurance by lower-income individuals not covered by employer-sponsored health insurance or a public insurance program. Although critics have complained the subsidy is far too small to substantially expand coverage among those who most need help, it would represent a major new expenditure on behalf of the poor.
Expanding health coverage through the tax system may not be the most efficient path, but tax subsidies appear the only game in town for expanding the federal role in the provision of health insurance. This paper examines the implications of major expansions in tax credits for health insurance, starting with the President's refundable tax credit proposal. Based on a microsimulation model, we examine the effects of the proposal on health insurance coverage generally, coverage by type of insurance (employer versus non-group), and the distribution of benefits by income level. At least in the short run, the President's proposal would modestly expand the number of people with health insurance on net, but it would also cause significant dislocationsmany people currently covered by health insurance at work would lose that coverage and would not be covered by alternative insurance. Yet, many currently uninsured people would gain coverage and many low-income people who already pay for their own insurance would see the cost of insurance reduced by the subsidy.
We also examine the impact of other, more generous tax subsidies in this paper. The basic model is a tax credit designed to mimic a voucher equal to the difference between the cost of insurance and 10 percent of a household's income. Importantly, this plan assumes that affordable health insurance would be available to individuals through the Federal Employees Health Benefits Plan (FEHBP), or something similar, whereas the President's proposal makes no such guarantee. We examine the effect of policies that would allow the credit only for individual non-group coverage (as in the President's plan), only for employer-sponsored insurance, or for insurance acquired in either market. All these options would reduce the number of uninsured by far more than the President's proposal, but at greater cost. The non-group-only credit would also cause millions of people to become newly uninsured, although two people would gain insurance for every one who loses it. The other two policies would not cause significant numbers of people to lose insurance, but their costs would be commensurately greater. All the reform options are much more progressive than the current tax subsidy.
The first section of this paper provides some background on the market for health insurance. A second section reviews who benefits from the current employer tax exclusion. The third section examines the four reform options. A concluding section sums up and lays out areas for future research.
Notes from this section of the paper
1. Hadley (2003) estimates that mortality declines by 4.5 to 7.0 percent for people when they gain health insurance.