urban institute nonprofit social and economic policy research

Health Insurance Tax Credits

Potential for Expanding Coverage

Read complete document: PDF


PrintPrint this page
Share:
Share on Facebook Share on Twitter Share on LinkedIn Share on Digg Share on Reddit
| Email this pageE-mail
Document date: August 01, 2001
Released online: August 01, 2001

Number 1 in Series, "Health Policy Briefs"

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.


Tax credits are being touted as possible mechanisms for expanding health insurance coverage in the United States. Analysts,1 members of Congress, and the Bush administration have all developed tax credit proposals in the past few years. However, although tax credit approaches are clearly appealing in certain respects, they are probably not the most effective tools for expanding health insurance coverage.

Tax credits for health insurance can be structured in an infinite number of ways. At the core of these proposals is the provision of a specified dollar amount, in the form of decreased tax liability, to those who obtain health insurance. Depending upon its design, the credit may be refundable (i.e., available to those with no or only limited tax liability).

Researchers agree that the only way to make significant inroads into solving the problems of the uninsured is through subsidization. The question is not whether coverage reforms should include subsidies, but rather how much subsidization is appropriate, who should receive the subsidies, and how they should be transferred. This brief assesses the promise and shortcomings of tax credits as a health insurance subsidy in light of a single primary objective: increasing health insurance coverage.

Tax Credit Structure

The conceptual underpinnings of tax credits and their potential application for use in the health insurance context have been described in detail elsewhere;2 consequently, the description here will be brief. Tax credits are subsidies that take the form of reduced end-of-year tax liability for those deemed eligible. In the case of refundable tax credits, those eligible who do not have a tax liability receive a cash transfer. Tax credit policies must define the eligible population, the value of the credit (including whether the value varies across different groups of eligibles), how much insurance coverage is required to qualify (and whether this varies by group of eligibles), and any offsets of existing tax subsidies that might apply.

Tax credits can be structured in a number of ways. Credits can be set at a fixed amount that does not vary by income (e.g., every filing unit receives a $500 credit), they can be structured to be more generous for some income groups than others, or they can even exclude some income groups altogether. For example, nonrefundable tax credits only benefit those with at least some tax liability, and their value cannot exceed the amount of taxes owed. Tax credits must be made refundable (i.e., the tax subsidy can exceed the individual's tax liability) in order to make them available to low-income families. Credit amounts could also, theoretically, be related to the cost of an available premium.

Tax credits for health insurance can be made available to all those purchasing health insurance, or they can be available only to those purchasing a minimum amount of coverage (defined by a benefit package or actuarial value), or to those purchasing through a particular source (e.g., the nongroup market or the employer-based market).

Timing of the credit is another design issue. One choice would be to make the credit available only at the end of the tax year, through the tax filing system. Another option would be to reduce the taxes paid throughout the year, by decreasing the amount of income taxes withheld by employers or by reducing estimated taxes for those who would ordinarily pay them. For those low-income persons who are not attached or only irregularly attached to the workforce, it would be considerably more difficult to pay out credits during the course of a tax year, so an alternative mechanism for delivering subsidies to this group would be necessary should they be targeted for the credit.

It is also important to consider how tax credits would interact with existing public insurance programs. Would those enrolled in Medicaid receive credits, for example? Would those insuring their children through the State Children's Health Insurance Program (SCHIP) be eligible? Given that SCHIP does require some family premium contribution in some states, while the Medicaid program does not, perhaps SCHIP might be considered qualified coverage, whereas Medicaid might not. Another option might be to allow individuals eligible for multiple programs to choose the subsidy they prefer. Depending upon the generosity of the credit relative to the public insurance, incentives for participation could vary significantly with design.

Although there are many ways to design tax credits for health insurance coverage, their political appeal is largely related to versions that offer administrative efficiency (e.g., fixed dollar credits) and the horizontal equity and income verification accuracy associated with the tax system.3 An additional advantage of fixed dollar credits is that those that do not vary by income do not increase marginal tax rates as income-related subsidies do. Furthermore, for those concerned with the inequities resulting from the current tax subsidy for employer-sponsored insurance coverage, it is also natural to view the tax system as the mechanism for a policy redesign.

How Well Can Tax Credits Expand Coverage?

Clearly, policymakers must balance competing objectives when designing public policies related to health insurance. Three main objectives are expansion of coverage, target efficiency, and horizontal equity—all worthwhile, but difficult to perfectly satisfy simultaneously. When tax credits' ability to expand coverage is assessed, which is the focus of this brief, several factors should be considered.

First, what will the individual be able to afford with the subsidy? Subsidy amounts that are small relative to an available premium will lead to lower new coverage rates. When fully phased in, the Bush administration's proposal would provide a maximum credit of $1,000 for individuals and $2,000 for families purchasing coverage in the nongroup market (U.S. Department of the Treasury 2001).4 Credit amounts must be compared with premiums of available policies in order to get a sense of the likelihood that such a subsidy would induce currently uninsured individuals to purchase coverage. According to a recent survey, average employer-based premiums in the United States were $6,348 for family policies and $2,424 for single policies in the year 2000 (Gabel et al. 2000). We know that 80 percent of uninsured workers in the United States are employed by firms that do not offer them health insurance (Garrett and Nichols 2001).

To illustrate the relative inadequacy of this degree of subsidization, let us assume that a worker can purchase health insurance coverage at employer group premium levels in the nongroup market.5 These premiums should be considered relevant for persons of average health risk—those with above-average health risk would face higher premiums, those with below-average health risk would face lower premiums. A family with income of $10,000 in 2001, prior to implementation of the credit, would have to pay 70 percent of their income in order to purchase a family insurance policy (table 1). As a reference point, the federal poverty level for a family of two is $11,610 in 2001. In 2002 that same family would have to pay 57 percent of their income for that policy even after applying their tax credit.6 In 2005 the family would still have to pay 61 percent of their income after the credit in order to purchase that coverage. A family making $15,000 in 2001 (approximately poverty level for a family of three) faces a premium of 46 percent of their income before the credit. By 2005, with a fully phased-in credit, that same family faces a cost of 41 percent of their income. Even a family with $35,000 in income in 2001 (approximately 200 percent of poverty level for a family of four) would have to pay 19 percent of their income in 2005 in order to purchase this insurance policy, only 1 percentage point lower than currently. For low-income families struggling to make ends meet, even these discounted premiums costs are a tremendous barrier to their ability to purchase coverage. Research evidence bears this out: Requiring low-income individuals to make significant out-of-pocket premium contributions leads to sizable decreases in insurance enrollment (Marquis and Long 1995).


TABLE 1. Family Premium Cost as Percentage of Income

Income in 2001 ($) 2001 before credit
is available
(%)
2002 maximum
credit = $1500
(%)
2005 maximum
credit = $200
(%)
10,000 70 57 61
15,000 46 38 41
35,000 20 17 19

Source: Urban Institute analysis.
Note: Assumes 7 percent premium growth, 3.6 to 3.9 percent income growth, and 2.5 to 2.6 percent growth in CPI-U annually.

Second, it is also important to recognize that the workers and nonworkers who do not have employer coverage available are unlikely to be well served by the individual insurance market, structured as it is today (Blumberg and Nichols 1995). Tax credit proposals usually do not vary the available subsidy with the health status of the recipient; doing so is widely considered to be too administratively difficult for the IRS. However, because an individual's health expenditures do vary with health status, insurance premium prices vary accordingly, except in the few states with community rating laws. Therefore, a credit that might cover a significant share of a premium for a healthy young person would likely cover a much smaller share for an individual with a current or past health problem (either his or her own or that of a family member).

In the nongroup market, only 13 states currently have guaranteed issue of health insurance of any kind, and 8 prohibit the use of health status for premium rating (Blue Cross Blue Shield 1999). Consequently, risk-pool issues can become dominant, with individuals potentially unable to access this market at all and others potentially unable to find an affordable premium. In addition, large administrative loads (35 to 40 percent of benefits or more) can consume a significant portion of an available credit. Consequently, expanding coverage to workers in firms that do not offer insurance coverage requires that policymakers consider making market reforms. Another way to make acceptable-quality insurance available to such workers would be opening access to publicly administered insurance policies (e.g., through SCHIP, Medicaid, or state high-risk pools). To be most effective, the credit amount would be tied to a premium available in a broad-based risk pool, something akin to a low-cost plan of acceptable quality. This would, however, make the administration of the credit more difficult, as the IRS would be required to coordinate information about insurance options in a given area with individuals' tax returns.

Third, the impact of a health insurance tax credit on coverage will depend on whether those eligible for a subsidy are provided with the liquidity that they need as their premiums come due, and whether the amount of their subsidy will be reconciled with their income as determined through their tax returns at the end of the year. If low-income people are to buy health insurance, they must have access to tax credit or subsidy prepayment throughout the insurance year. Clearly, tax credits can be designed to allow just that—the Earned Income Tax Credit (EITC) is an example. The trouble arises, however, when we attempt to reconcile the subsidy prepayment, most likely based on a prediction of income, with the actual subsidy owed to a taxpayer, which depends on the year's actual income.7

Low-income persons who need advance payments to purchase coverage will be less likely to take advantage of a credit because of uncertainty over what the size of the actual subsidy will be at the end of the year. The U.S. General Accounting Office (GAO) determined that fears of owing the IRS money at the end of the year were important in explaining the extremely low take-up rate of EITC advanced payments (U.S. GAO 1992). Second, EITC has taught us that for those who do participate, it is extremely difficult to reconcile prepayment of subsidies at the end of the year, and the costs of correcting the errors may substantially exceed the benefits of doing so.

According to the same GAO study, of those individuals receiving advanced EITC payment, a significant proportion do not file tax returns at the end of the year, making it impossible to determine if they have been over- or underpaid with respect to the credit owed them.8 In addition, a significant percentage who do file returns do not report their advance payments.9 Although many of these errors may be attributable to lack of understanding rather than intentional fraud, the costs associated with accurately reconciling these returns are large. Given that the size of errors in the credits are small by IRS standards, these errors' costs may be dwarfed by the costs associated with rectifying them.10 However, to low-income credit recipients, errors of any size could be very meaningful, and many may find it extraordinarily difficult to repay amounts owed to the Treasury.

If we were to agree that it is not worthwhile to reconcile, then we could design subsidy eligibility criteria using any of an infinite number of different income measures, such as the past three months' or last year's income. In a notable departure from previous proposals, the Bush plan does not reconcile advance payments. However, if we are not going to reconcile, the use of a tax credit to subsidize coverage loses much of its practical appeal. The tax system is attractive in large part because of its ability to accurately determine income. Without reconciliation, direct subsidization to individuals becomes more attractive, as it allows us to avoid relying upon employers and the IRS for eligibility determination and delivery of the subsidy.

Maximizing New Coverage with a Health Insurance Tax Credit

It is important to remember that strategies for expanding coverage may be inconsistent, at least to some extent, with other potential program objectives, such as target efficiency (particularly if such an objective were defined as spending as high a percentage as possible of new program dollars on previously uninsured persons). Not all program objectives can be met simultaneously, and priorities must be set. However, if the expansion of health insurance is the highest priority, how should tax credits be designed? Obviously, the desire for coverage ought to be balanced with other considerations, especially budget constraints. In summary, the following points provide guidelines for pursuing coverage expansions within a general tax credit framework.

The full tax credit amount should approximate the cost of an available plan of acceptable quality. The smaller the amount of the tax subsidy relative to the cost of an available plan, the lower the participation rate of the previously uninsured. Related design issues follow.

  • Eligibility for the credit should be limited to low- and moderate-income individuals. With budget constraints, inclusion of high-income individuals would lower the credit amount available to each eligible family. Also, higher-income individuals are less likely to be uninsured, so targeting the subsidy to those with low income, thus keeping it as large as is financially feasible, will lead to greater coverage.
  • The income range over which the full credit begins to phase out should be short. The currently uninsured will take up partial subsidies (those in the phase-out range) at a considerably lower rate than full subsidies. However, the currently covered, if eligible, will tend to take up partial subsidies. This means that long phase-out ranges can be costly without significantly reducing the number of uninsured. The trade-off of creating "cliffs" is that they produce high marginal tax rates for those with incomes at the top of the phase-out range. However, recent research indicates that such cliffs may have far less severe work disincentives than previously thought (Gruber and Saez 2000).

The credit should be refundable. More than half of the uninsured have incomes so low that they would either receive no credit or have their credit limited to some extent by a nonrefundable credit (Gruber and Levitt 2000). Excluding these individuals from a new benefit significantly inhibits a policy's ability to target the uninsured—so credits should be refundable.

Tax credit dollars should be made available when premium payments are due. Without the liquidity to purchase coverage, a tax credit will have limited value to low-income individuals and families. Those most in need of the benefit would be significantly less able to take advantage of it without advance payments.

Advance payments of the tax credit should not (and cannot) be perfectly reconciled at the end of the tax year. Uncertainty about the final annual credit amount is likely to dissuade low-income persons from using the credit to purchase coverage throughout the year. In addition, although the size of the errors in payments (due to unexpected fluctuations in income throughout the year) can be substantial from the individual's perspective, they are small from the perspective of the IRS, making the costs of collection likely to out-weigh its benefits.

Eligibility for a tax credit should not be a function of employer behavior, or else incentives will be distorted. If, for example, only those working for employers not currently offering employer-sponsored insurance (ESI) were eligible for the credit, some workers would have an incentive to seek out employers who do not offer coverage. This could lead to employers dropping coverage or to workers choosing jobs that are not the best fit for their particular skills. Allowing workers to use credits for purchasing ESI would eliminate any credit-related incentives for employers to stop offering ESI, and would give workers a potential source for purchasing stable, affordable coverage.

Likewise, eligibility for a credit should not be contingent upon past insurance status, in order to avoid horizontal inequities and distorted incentives. If, for example, only previously uninsured persons are eligible for a credit, a financial incentive to become uninsured will have been created. Public policies should not encourage individuals to create gaps in insurance coverage. In addition, such a policy effectively punishes low-income individuals who have sacrificed wages and other disposable income to purchase coverage.

Those without access to an employer-sponsored insurance policy must be given a source for purchasing insurance coverage—this is especially important for those who are at above-average risk for significant health care expenses. Without an affordable, stable policy of acceptable quality to purchase, a tax credit loses its value. The individual insurance market is, in most states, extremely difficult to navigate and does not well serve those with the highest health risks. There are a number of options that can be considered in this regard. Some of these are

  • Public contracting with private plans (as is the case in many states under SCHIP);
  • Reforming individual insurance market rules, including guaranteed issue and premium rating restrictions, and developing organized purchasing for individual products; and
  • Allowing individuals to purchase actuarially fair coverage through public programs such as Medicaid, state or federal employee systems, and state high-risk pools (individual purchasers could face a premium independent of the current enrollees in these existing programs and would not receive subsidization through the public program; they would use their tax credit to purchase coverage through the program).

Will addressing these important design issues satisfy the original supporters of tax credit approaches or leave them feeling as if the final product does not fit their conceptualization of a tax credit? In either event, ignoring these features would result in policy that could not be honestly touted as a program to significantly expand the number of insured.


Endnotes

1. Some examples are Mark V. Pauly, "Extending Health Insurance through Insurance Credits," in Expert Proposals to Expand Health Insurance Coverage for Children and Families, Henry J. Kaiser Family Foundation Project on Incremental Health Reform, draft, February 1999; Mark V. Pauly and John C. Goodman, "Tax Credits for Health Insurance and Medical Savings Accounts," Health Affairs 14(1), 126-139, 1995; C. Eugene Steuerle, "The Search for Adaptable Health Policy through Finance-Based Reform," in R. Helms, ed. American Health Policy: Critical Issues for Reform (Washington, D.C.: AEI Press), 334-361, 1993; Sue A. Blevins, "Restoring Health Freedom: The Case for a Universal Tax Credit for Health Insurance," Policy Analysis, No. 290, Cato Institute, December 1997; and Grace-Marie Arnett, "The Top Eight Reasons Why Employment-Based Health Insurance Is Trouble," Galen Institute Policy Paper, 1998.

2. Mark V. Pauly, "Extending Health Insurance through Insurance Credits," in Expert Proposals to Expand Health Insurance Coverage for Children and Families, Henry J. Kaiser Family Foundation Project on Incremental Health Reform, draft, February 1999; Mark V. Pauly, "How Can We Get Responsible National Health Insurance: What Constitutes A Good Plan? What Present Proposals Lack," The American Enterprise 3(4): 60-70, July/August 1992.

3. It should be noted, however, that using the tax system does not necessarily lead to horizontal equity. For example, a credit that is made available only to those purchasing in the nongroup market may create inequities for similarly situated individuals purchasing coverage through their employer. In addition, the tax system is not well-suited to adjust for geographic differences in costs. This means that similar individuals using tax credits of the same amount in the Northeast and the South may have very different degrees of health insurance purchasing power as a consequence.

4. The credit would be available only to those not enrolling in employer-sponsored or public insurance. The $1,000/$2,000 amounts are maximums. The credit amount phases down to zero for incomes between $15,000 and $30,000 for singles without dependents, for incomes between $30,000 and $45,000 for those with dependents buying a single policy, and for incomes between $30,000 and $60,000 for those buying family policies. See U.S. Department of the Treasury 2001, http://www.treas.gov/taxpolicy/library/blue-bk01.pdf.

5. Assume for discussion that the individual in the nongroup market pays a higher administrative cost than would be the case in the employer market, but compensates by purchasing a less-rich benefit package than what the average employer plan offers.

6. Family income is assumed to grow at the administration's projected rate of increase for federal civilian employees: 3.6 percent in 2002, and 3.9 percent annually after that (table 1-1, U.S. Office of Management and Budget 2001). Income eligibility cut-offs are designated by the President's plan to grow at the rate of CPI-U: 2.6 percent in 2002 and 2003 and 2.5 percent annually after that (same table). Premiums are assumed to grow at 7 percent annually, an estimate based upon recent annual growth in per capita national health expenditures (Heffler et al. 2001). Premiums are likely to grow faster than 7 percent per year between 2001 and 2005, however. Heffler et al. Estimate premium growth at 10.5 percent in 2001; consequently, the premium burdens relative to income presented in table 1 are likely to be somewhat low.

7. Of course, reconciliation is not an issue if the subsidy is not income-related, but budget constraints and a desire for significant expansion of coverage require limiting subsidies to the low/moderate income. In that way, the largest possible subsidy (one that has the best chance of approximating the cost of an available plan of acceptable quality) is offered to those most likely to be uninsured or vulnerable to losing insurance.

8. "GAO estimated that about 45 percent of those who, according to IRS records, might have received the advance payment never filed a tax return." This figure is based upon 1989 returns, and reforms implemented since that time may have improved that rate somewhat.

9. "GAO estimated that about 49 percent of the workers who clearly received advance payments in 1989 and filed a tax return did not report receiving the credit." Again, reforms since 1989 may have improved this rate to some extent.

10. In fact, 44 percent of all audits are now attributable to the working poor who apply for the EITC (New York Times, February 16, 2001). Congress ordered the IRS to redirect its resources in this way. As the audit rate for the low income soared, the audit rates for high-income filers and corporations fell.


References

Blumberg, Linda J., and Len M. Nichols. 1995. "Health Insurance Market Reforms: What They Can and Cannot Do," Urban Institute Monograph. Washington, D.C.: The Urban Institute.

Blue Cross Blue Shield Association. 1999. "State Legislative Health Care and Insurance Issues: 1999 Survey of Plans." Washington, D.C.

Gabel, Jon, Larry Levitt, Jeremy Pickering, Heidi Whitmore, Erin Holve, Samantha Hawkins, and Nick Miller. 2000. "Job-Based Health Insurance in 2000: Premiums Rise Sharply While Coverage Grows." Health Affairs 19(5): 144-151.

Garrett, Bowen, and Len Nichols. 2001. "Workers without Health Insurance: Who Are They and How Can Policy Reach Them?" Urban Institute Report to the W.K. Kellogg Foundation. Washington, D.C.: The Urban Institute.

Gruber, Jon, and Larry Levitt. 2000. "Tax Subsidies for Health Insurance: Costs and Benefits." Health Affairs 19(1): 7-22.

Gruber, Jon, and Emmanuel Saez. 2000. "The Elasticity of Taxable Income: Evidence and Implications." Cambridge, MA: NBER Working Paper 7512, January.

Heffler, Stephen, Katharine Levit, Sheila Smith, Cynthia Smith, Cathy Cowan, Helen Lazenby, and Mark Freeland. 2001. "Health Spending Growth Up in 1999; Faster Growth Expected in the Future." Health Affairs 20(2): 193-203.

Marquis, M. Susan, and Stephen H. Long. 1995. "Worker Demand for Health Insurance in the Non-Group Market." Journal of Health Economics 14: 47-63.

U.S. Department of the Treasury. 2001. "General Explanations of the Administration's Fiscal Year 2002 Tax Relief Proposals," http://www.treas.gov/taxpolicy/library/blue-bk01.pdf. (Accessed April 2001.)

U.S. General Accounting Office. 1992. "Earned Income Tax Credit: Advance Payment Option Is Not Widely Known or Understood by the Public." GAO/GGD-92-26. Washington, D.C.: U.S. Government Printing Office. February.

U.S. Office of Management and Budget, Executive Office of the President. 2001. "Analytical Perspectives, Budget of the United States Government, Fiscal Year 2002." U.S. Government Printing Office.


About the Author

Linda J. Blumberg is an economist and senior research associate at the Urban Institute. She is currently working on a variety of projects related to private health insurance and health care financing that include estimating the coverage and risk pool impacts of tax credit proposals, estimating price elasticities of employers offering and workers taking-up health insurance, the effects of insurance market reforms on the risk pool of the privately insured, and a series of analyses of the working uninsured. From August 1993 through October 1994, Dr. Blumberg served as health policy advisor to the Clinton administration during its initial health care reform effort.


The Urban Institute's Health Policy Center (HPC) was established in 1981 to study the public policy issues surrounding the dynamics of the health care market and health care financing, costs, and access. Research topics include health insurance coverage and costs, incentives for public and private provider reimbursement, reform of the long-term care system, and malpractice tort law and insurance. HPC researchers also examine Medicare and Medicaid benefits and proposals, assess proposed reforms in the private medical market, and study ways to expand health insurance coverage for children, among other issues.

The Health Policy Briefs series is intended to provide analysis and commentary on key health policy issues facing the nation. Topics will include Medicare and Medicaid policy, changes in private health care markets, strategies for expanding health insurance and the rising costs of health care. The series will include both data briefs and perspectives on national debates.



Topics/Tags: | Economy/Taxes | Health/Healthcare


Usage and reprints: Most publications may be downloaded free of charge from the web site and may be used and copies made for research, academic, policy or other non-commercial purposes. Proper attribution is required. Posting UI research papers on other websites is permitted subject to prior approval from the Urban Institute—contact publicaffairs@urban.org.

If you are unable to access or print the PDF document please contact us or call the Publications Office at (202) 261-5687.

Disclaimer: 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. Copyright of the written materials contained within the Urban Institute website is owned or controlled by the Urban Institute.

Email this Page