Order Code RL30762
Report for Congress
Received through the CRS Web
Tax Subsidies for Health Insurance for the
Uninsured: An Economic Analysis of Selected
Policy Issues for Congress
Updated January 16, 2001
Gary Guenther
Analyst in Business Taxation and Finance
Government and Finance Division
Congressional Research Service ˜ The Library of Congress

---

Tax Subsidies for Health Insurance for the Uninsured:
An Economic Analysis of Selected Policy Issues for
Congress
Summary
In spite of strong sustained growth in U.S. output of goods and services,
personal income, and employment in the past decade, the number of Americans
without health insurance has increased. An estimated 42.2 million people under the
age 65 were uninsured in 1999, up from 33.6 million in 1988. A major factor behind
this rise has been sharp increases in health insurance premiums, driven by the rising
cost of health care.
The increases in the uninsured population have sparked a renewed interest in
finding ways to achieve substantial reductions in this population. While a variety of
approaches are possible, in recent years Congress has shown keen interest in the use
of tax subsidies to expand health insurance coverage. Tax-based approaches have
considerable appeal for some policymakers. Proponents of these approaches believe
that they would improve the fairness of the tax code by expanding the proportion of
taxpayers eligible for tax subsidies for the purchase of health insurance; they would
rely on the private insurance market rather than federal programs to expand coverage;
and they would constitute a tax cut, which for some is more desirable than launching
a new federal spending or entitlement program. A host of bills to offer tax credits and
deductions for the purchase of health insurance to those who are uninsured or insured
but ineligible for existing tax subsidies for health insurance were considered in the
106th Congress. Similar proposals are expected to resurface in the 107th Congress.
These proposals raise the issue of whether additional government intervention
in the market for health insurance is justified on economic grounds. In the absence
of serious problems in the health insurance market, proposals to supplement or scrap
existing tax subsidies for health insurance and supplant them with more generous tax
subsidies could end up making consumers as a whole worse off. The problems of
adverse selection and moral hazard in the health insurance market and the existence
of free riders in the health care market arguably provide an economic rationale for
government intervention in the health insurance market. They also suggest that
optimal public policy might attempt to encourage all individuals to purchase health
insurance that avoids promoting the excessive consumption of health care.
Proposals to create tax subsidies for the purchase of health insurance by the
uninsured (along with those ineligible for current tax subsidies) raise a host of
important policy issues, including their potential efficacy and their impact on social
welfare, equity, and the administrative burden of complying with the tax code. Recent
studies indicate that these effects hinge on the design of such subsidies. In designing
a tax subsidy for the purchase of health insurance, the following issues appear critical
in shaping its net effects: (1) whether it is a deduction or a credit; (2) if it is a credit,
whether it is refundable; (3) whether the subsidy is unlimited or capped; (4) whether
it is targeted to particular income groups; (5) whether it includes insurance market
reforms; and (6) whether it addresses important administrative issues like the timing
of subsidy transfers and variations in health insurance premiums tied to geographic
location, age, and health status.

---

Contents
Profile of the U.S. Uninsured Population . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Existing Tax Subsidies for the Purchase of Health Insurance . . . . . . . . . . . . . . . 4
Exclusion of Employer Payments for Employee Health Insurance . . . . . . . . 5
Itemized Deduction for Medical Expenses . . . . . . . . . . . . . . . . . . . . . . . . . 6
Deduction for Health Insurance Purchased by the Self-Employed . . . . . . . . 7
Medical Savings Accounts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Economic Rationale for Government Intervention in the Health Insurance
Market . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Economic Effects of Current Federal Tax Subsidies for Health Insurance
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Recent Congressional Proposals to Extend Tax Subsidies for the Purchase of Health
Insurance to the Uninsured . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Potential Effectiveness and Cost of Selected Proposed Tax Subsidies for Health
Insurance for the Uninsured . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Lewin Group and Gruber Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Lewin Group Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Gruber Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Implications of the Gruber and Lewin Group Studies for the Potential Cost and
Effectiveness of Proposed Tax Subsidies for Health Insurance for the
Uninsured . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

---

Tax Subsidies for Health Insurance for the
Uninsured: An Economic Analysis of
Selected Policy Issues for Congress
At the close of the 106th Congress, there was widespread concern about the
number of Americans without health insurance, most of whom are under age of 65.1
An estimated 42.2 million nonelderly people were uninsured in 1999, down from 43.9
million in 1998 but up from 33.6 million in 1988. What many find especially
worrisome is that the uninsured population has increased during a period of sustained
and at times robust growth in national and personal incomes and employment.
Historically, more than 60% of nonelderly Americans have obtained health insurance
through their employers. If, in the next few years, projected increases in health
insurance premiums materialize, or the economy enters a period of slower growth or
falls into a recession, the rate of growth in the nonelderly uninsured population could
accelerate, assuming, of course, no changes in federal policy.2
Concern about the size of the uninsured population appears rooted in the
disadvantages the uninsured face in receiving adequate health care and the social costs
being uninsured entails. Compared with the insured, the uninsured are less likely to
obtain preventive and routine health care, less likely to receive health care when they
feel they need it, less likely to have a regular source of primary health care, more
likely to be hospitalized for preventable or avoidable conditions, and more likely to
have trouble paying medical bills.3 In addition, when the uninsured receive medical
care but do not pay for it, the costs are borne by the insured population through
1According to the results of a recent survey funded by the NewsHour with Jim Lehrer and the
Henry J. Kaiser Family Foundation, a majority of Americans know someone who is uninsured
and eight out of 10 Americans regard the persistent growth in the ranks of the uninsured as
a serious national problem. (Complete survey results and background information on the
u n i n s u r e d a r e a v a i l a b l e a t t h e N e w s H o u r ’ s w e b s i t e :
[http://www.pbs.org/newshour/health/uninsured/index.html].) In addition, a survey of
registered voters conducted by Princeton Survey Research Associates and released on October
16, 2000 found that health care ranked as the top issue that the next Congress and president
should address. It also found that of the voters who considered health care to be an important
issue in both the 2000 congressional and presidential elections, helping the uninsured get
health insurance rated as the top health care policy issue. (A copy of the survey report can
be obtained via the Internet at [http://www.pbs.org/criticalcondition].)
2See “Poll of Small Businesses Predicts Number of Uninsured May Rise in 2001,” Daily
Report for Executives (Washington: Bureau of National Affairs, July 19, 2000), p. A-18;
Joel E. Miller, Deja Vu All Over Again: The Soaring Cost of Private Health Insurance and
Its Impact on Consumers and Employers (Washington: National Coalition on Health Care,
May 2000), pp. 7-8.
3Catherine Hoffman and Alan Schlobohm, Uninsured in America: A Chart Book, 2nd Edition,
Kaiser Commission on Medicaid and the Uninsured (May 2000), pp. 56-81.

---

CRS-2
higher health insurance premiums, local, state, and federal governments through
higher subsidies for uncompensated care, and health care providers through reduced
incomes.
Yet, though most Americans believe that health care should be provided to
everyone, there is a notable lack of consensus on how to solve the problem of a large
and growing uninsured population. Respondents to a recent survey on health care
issues by the NewsHour with Jim Lehrer and the Henry J. Kaiser Family Foundation
seemed divided on the issue of paying more in taxes to provide health insurance to the
uninsured and expressed equal support for four policy options to address the problem:
21% were in favor of expanding existing public programs like Medicaid and the State
Children’s Health Insurance Program (SCHIP); 21% backed a proposal to require
firms to offer private health insurance to their employees; 21% liked the idea of
creating a national health insurance system to cover all Americans; and 20% favored
extending tax deductions and credits to help uninsured individuals purchase health
insurance in the non-group market.4
The persistent growth of the uninsured population at a time of sustained
economic expansion has prompted policymakers to search for ways to expand access
to adequate and affordable health care. While a variety of approaches have been
considered, recent Congresses have shown a strong interest in tax-based strategies for
expanding health insurance coverage. Numerous bills to create tax deductions or tax
credits for the purchase of health insurance by those not covered by employment-
based health insurance were introduced in the 106th Congress, and both houses passed
measures that would have granted a tax deduction for health insurance expenditures
to those who do not itemize yet pay for a large share of their health insurance
premiums. Similar proposals are likely to emerge in the 107th Congress.
Proposals to create tax subsidies for the purchase of health insurance by the
uninsured raise numerous policy issues. This report discusses several of these issues.
Specifically, it examines the potential impact of these proposals on the size of the
uninsured population, the revenue cost of these proposals, their distributional effects
among major income groups, their potential revenue cost relative to the potential
gains in the insured population, and some of the administrative issues raised by the
proposals. To lend some needed context to the analysis, the report begins with a
profile of the U.S. uninsured population, a description of current federal tax subsidies
for health insurance, and a discussion of the economic justification for government
intervention in the health insurance market.
Profile of the U.S. Uninsured Population
This much is certain about the U.S. uninsured population: it is sizable and has
grown substantially in the past 10 to 15 years. According to U.S. Census Bureau
estimates, 42.5 million Americans (or one of every six citizens) lacked health
4See footnote # 1.

---

CRS-3
insurance in 1999.5 This figure excludes the many persons who gained and lost health
insurance during the course of that year. Nearly all the uninsured in1999 were under
age 65 (42.2 million or 99.3% of the total), and of those, 10 million (or 24% of
nonelderly uninsured) were under age 18. The nonelderly uninsured population
expanded from 31.8 million (or 14.8 % of nonelderly Americans) in 1987 to 43.9
million in 1998 (or 18.4%), before dipping to 42.2 million (or 17.5%) in 1999.
Many factors have been cited to explain the recent rise in the uninsured
population, including increases in the number of employers who do not offer health
insurance and in the number of employees who choose not to participate in employer-
provided health plans because their out-of-pocket costs, and the recent transition of
former welfare recipients to low-paying jobs without health insurance.6 But perhaps
the most important economic factor has been faster growth in per-capita spending on
health care than per-capita personal income through much of the 1990s.7 The cost of
medical services plays a major role in the pricing of health insurance in that premiums
are set to cover expected claims for these services – along with the selling and
administrative costs of policies – and still yield profits for insurers. Recent surveys
indicate that most nonelderly uninsured persons lack health insurance because they
find it too expensive, and that only a small share of the uninsured say that they do not
want or need it.8 As the cost of medical care (and by extension, health insurance)
rises faster than personal incomes, individuals are forced to spend more of their
incomes on a given set of health benefits; at some point, the expected benefits of the
insurance policy are not worth the cost.
Health insurance coverage is closely tied to household income. Not surprisingly,
most uninsured households fall in the bottom half of the personal income scale. In
1998, according to Census data, 27% of nonelderly uninsured households had pre-tax
incomes below the federal poverty level (which for a family of three was $13,650 in
1998), and 29% of nonelderly uninsured households earned pre-tax incomes between
100% and 200% of that level.9 In addition, it is estimated that 85% of uninsured
households have incomes that are below the median household income for their family
structure, 45% of the uninsured have no federal income tax liability, and over 60% of
the uninsured have tax liabilities of less than $1,000.10 Yet not all the uninsured can
5See U.S. Department of Commerce, Bureau of the Census, Health Insurance Coverage:
1999 (Washington: September 2000), 15 p.
6Jack A. Meyer, Sharon Silow-Carroll, and Elliot K. Wicks, Tax Reform to Expand Health
Coverage: Administrative Issues and Challenges (Washington: Economic and Social
Research Institute, January 2000), pp. 4-5.
7See Richard Kronick and Todd Gilmer, “Explaining the Decline in Health Insurance
Coverage, 1979-1995," Health Affairs, vol. 18, March/April 1999, pp. 30-47.
8Catherine Hoffman, Uninsured in America: A Chart Book, Kaiser Commission on Medicaid
and the Uninsured (June 1998), p. 3.
9U.S. Library of Congress, Congressional Research Service, Health Insurance Coverage:
Characteristics of the Insured and Uninsured Populations in 1998, by Madeleine Smith,
CRS report 96-891 EPW (Washington: October 6, 1999), p. 6.
10Jonathan Gruber, Tax Subsidies for Health Insurance: Evaluating the Costs and Benefits,
(continued...)

---

CRS-4
be considered poor: in 1998, about 8% of the uninsured lived in households with pre-
tax incomes of $75,000 or above, and 27% of the uninsured lived in households with
pre-tax incomes equal to 300% or more of the federal poverty level.
Contrary to the conventional image of the uninsured as persons who are
chronically unemployed and living on the margin of society, many are the working
poor and their dependents. In 1998, 83% of the nonelderly uninsured lived in
households headed by workers, 61% of the heads of uninsured households worked
full-time throughout the year, and 56% of the nonelderly uninsured were adults who
worked part-time or full-time.11 Employees of small firms are much more likely to
be uninsured than employees of large firms: in 1998, 34% of workers in firms with
fewer than 10 employees were uninsured compared with 13% of workers in firms
with 1,000 or more employees, and 47% of nonelderly uninsured workers were
employed by firms with fewer than 100 employees. Presumably, the link between firm
size and health insurance coverage stems from the fact that the cost of insurance
depends in part on the size of the risk pool it serves. Assume a single firm constitutes
such a pool. As the number of employees rises, the cost per employee of marketing
and administering the insurance plan falls, and the risk of claims for costly health
problems is spread over more individuals, reducing the average risk per employee.

While the uninsured exhibit considerable diversity in age, household income,
racial or ethnic status, and employment status, a profile of the typical uninsured
individual can be sketched from the tangle of available data on the U.S. uninsured
population. Such an individual is as likely to be male as female, likely to be white,
likely to work for a small firm, likely to be under the age of 35, likely to be single with
children, likely to reside in a household with at least one full-time worker, and likely
to reside in a household with an income above the federal poverty level.
Existing Tax Subsidies for the Purchase of Health
Insurance
Many recent congressional proposals to establish tax subsidies for the purchase
of health insurance by uninsured individuals sought to build on or supplement existing
tax subsidies for health insurance. Thus, in order to understand the proposals’
possible equity, efficiency, and administrative effects, it is helpful to have a clear
picture of how the current federal tax code subsidizes the purchase of health
insurance. Most nonelderly Americans benefit from them to one extent or another.
Jonathan Gruber, an economist at the Massachusetts Institute of Technology, has
estimated that, in 1996, 84% of nonelderly Americans were eligible for a tax subsidy
10(...continued)
Working Paper 7553 (Cambridge, MA: National Bureau of Economic Research, February
2000), p. 7; and Jonathan Gruber and Larry Levitt, “Tax Subsidies for Health Insurance:
Costs and Benefits,” Health Affairs, vol. 19, January/February 2000, p. 74, 78.
11Paul Fronstin, Sources of Health Insurance and Characteristics of the Uninsured: Analysis
of the March 1999 Current Population Survey (Washington: Employee Benefit Research
Institute, January 2000), p. 15.

---

CRS-5
for the purchase of health insurance and the tax code subsidized 81% of U.S.
spending on private health insurance.12
The federal tax code contains several provisions that subsidize the purchase of
health insurance in the private market. They are in the form of deductions or
exclusions from income. As the value of a deduction or exclusion hinges on an
individual’s income tax bracket, they become more valuable as personal income rises.
Each provision is described briefly below.
Exclusion of Employer Payments for Employee Health
Insurance
Under section 106 of the Internal Revenue code (IRC), employees pay no
income or payroll taxes on compensation received in the form of employer
contributions for health insurance premiums. And employers may deduct these
payments as ordinary and necessary business expenses. There is no dollar limit on the
exclusion, and it applies whether the employer self-insures or contracts with private
insurers to provide individual or group health insurance plans for its employees.
Moreover, employees who participate in so-called cafeteria plans established by their
employers under IRC section 125 may exclude their payments for employer-provided
health insurance premiums from taxable income. Cafeteria plans are benefit packages
that give employees a choice between receiving qualified benefits – such as accident
and health insurance – or taking the equivalent amount in cash. Employees who opt
for the benefits may exclude them from their taxable income, but those who receive
cash instead must include it in their taxable income. Many firms also offer health
benefits to employees through flexible spending accounts (FSA). Under a health FSA,
an employee chooses a benefit limit at the beginning of a plan year and makes claims
against the account as he or she incurs medical expenses; a FSA may be used to pay
for annual health insurance deductibles and copayments and other expenses not
covered by the employer’s health plan. FSAs are funded either through salary
reductions or employer contributions, both of which are exempt from federal income
and employment taxes.13
The exclusion for employer-paid health insurance premiums has had a major
impact on the evolution of the U.S. health insurance market since the early 1940s.
In 1942, a ruling by the War Labor Board permitted employers to offer generous
fringe benefits in a bid to attract scarce workers without violating the wage and price
controls then in effect because of U.S. involvement in World War II; and in 1943, the
Internal Revenue Service (IRS) ruled that employers’ payments to insurance
companies for group medical and hospital insurance premiums were not taxable as
employee income.14 The two rulings set in motion a process that culminated in the
12Gruber, Tax Subsidies for Health Insurance: Evaluating the Costs and Benefits, p. 4.
13Meyer, Silow-Carroll, Wicks, Tax Reform to Expand Health Coverage: Administrative
Issues and Challenges, p. 8.
14Melissa A.Thomasson, The Importance of Group Coverage: How Tax Policy Shaped U.S.
Health Insurance (Cambridge, MA: National Bureau of Economic Research, February
(continued...)

---

CRS-6
widespread coverage of nonelderly Americans under employment-based group health
plans by the late 1950s. It can be argued that the exclusion played a pivotal role in
this process because it improved access to group health insurance, which is easier to
obtain and generally cheaper than comparable individual or non-group insurance.
Nowadays, the vast majority of nonelderly Americans (65% in 1998, including
dependents) and nonelderly American workers (73% in 1998) receive health insurance
through their employers.15 While it is not clear what share of this coverage can be
attributed to the exclusion, there is no doubt that it gives employees a robust incentive
to receive part of their compensation in the form of employer-provided health
insurance. To illustrate this incentive effect, assume that a worker’s wages are equal
to his or her taxable income. If that individual is paid an annual wage of $50,000 and
faces a combined income and employment tax rate of 20%, then his or her after-tax
income equals $40,000; but if the same worker receives $47,000 in wages and $3,000
in employer contributions to health insurance, then his or her after-tax income is
$37,600 but his or her total after-tax compensation is worth $40,600. As this
example shows, the exclusion for employer-provided health insurance makes
compensation as health insurance more valuable than compensation as wages. In
effect, it reduces the after-tax cost of health insurance by a factor equal to an
employee’s combined income and employment tax rate. Economists Jonathan Gruber
and James Poterba have estimated that the exclusion lowers the after-tax cost of
health insurance by 32% relative to the after-tax cost of other goods and services.16
Such a generous subsidy is bound to be costly when it is widely used. The Lewin
Group has estimated that the exclusion could cost the federal government $74.5
billion in forgone tax revenue in 2000.17
Itemized Deduction for Medical Expenses
Under IRC section 213, an individual is allowed a deduction for qualified medical
expenses for himself or herself and his or her spouse and dependents that exceed 7.5%
of adjusted gross income (AGI). The deduction is available only to taxpayers who
itemize on their tax returns and is mainly limited to unreimbursed medical expenses
and health insurance premiums paid out-of-pocket. The income threshold for the use
of the deduction limits the number of taxpayers claiming it. For example, only 4.4%
of the 124.7 million federal individual income tax returns filed for 1998 claimed the
14(...continued)
2000), p. 2-3.
15Fronstin, Sources of Health Insurance and Characteristics of the Uninsured, Table 1 (p.
4) and Chart 4 (p. 6); and U.S. Office of Management and Budget, Analytical Perspectives:
Budget of the United States Government, Fiscal Year 2001 (Washington: 2000), p. 114.
16Jonathan Gruber and James Poterba, “Fundamental Tax Reform and Employer-Provided
Health Insurance,” in Economic Effects of Fundamental Tax Reform, Henry J Aaron and
William G. Gale, eds. (Washington: Brookings Institution Press, 1996), p. 135.
17John Sheils, Paul Hogan, and Randall Haught, Health Insurance and Taxes: The Impact of
Proposed Changes in Current Federal Policy (Washington: National Coalition for health
Care, October 1999), p. 6.

---

CRS-7
deduction. Consequently, its revenue cost to the federal government is relatively
small: the Lewin Group estimates that this cost could total $4.8 billion in 2000, or
about 4% of the revenue cost of the exclusion.18
It seems that the deduction’s impact on the decision to purchase health insurance
is ambiguous. On the one hand, it subsidizes the purchase of medical care and health
insurance by lowering the after-tax cost of medical expenses and out-of-pocket health
insurance premiums above 7.5% of AGI for taxpayers who itemize. On the other
hand, the deduction gives the same group of taxpayers an incentive not to buy health
insurance by reducing their after-tax cost of major medical expenses.19 In reality, it
is unclear which effect predominates. If anything, the deduction probably has little
influence on the decision to purchase health insurance because most taxpayers are
unlikely to be able to claim the deduction in any given tax year.
Deduction for Health Insurance Purchased by the Self-
Employed
Yet another tax subsidy for the purchase of health insurance is the deduction for
health insurance purchased by the self-employed for themselves and their immediate
families allowed under IRC section 162. In 2000 and 2001, the self-employed may
deduct 60% of eligible health insurance expenses, and this share is scheduled to rise
to 100% in 2003 and thereafter. Self-employed individuals are defined as sole
proprietors, working partners in a partnership, and employees of subchapter S
corporations who own more than 2% of the stock. The deduction is taken “above-
the-line,” meaning the deduction is available to all eligible self-employed individuals,
not just those who itemize. It lowers the after-tax cost of health insurance by a factor
equal to 60% of a self-employed individual’s marginal income tax rate in 2000 and
2001. Use of the deduction is limited in the following ways: (1) self-employed
persons may not deduct their health insurance expenses in computing their self-
employment tax liability; (2) the deduction cannot exceed the taxpayer’s net earned
income from the trade or business in which the health insurance plan was established,
minus the deduction for 50% of the self-employment tax and the deduction for
contributions to pension plans; (3) the deduction is not available for any month in
which a self-employed person is eligible to participate in a health insurance plan
maintained by his or her employer or his or her spouse’s employer; and (4) any
deduction claimed cannot be included in the expenses used to compute the itemized
deduction for medical expenses.
In 1998, an estimated 12.5 million workers were self-employed, or 9% of the
nonelderly labor force. Of these, 53% received employment-based health insurance,
19 % purchased their own insurance, and 25% were uninsured. Thus, it appeared that
44% of the self-employed (or 5.5 million individuals) were eligible to claim the
deduction for that year. According to IRS figures, 3.2 million individuals (or 58% of
18Ibid., p. 11.
19Jonathan Gruber and James Poterba, Tax Subsidies to Employer-Provided Health
Insurance, Working Paper 5147 (Cambridge, MA: National Bureau of Economic Research,
June 1995), p. 9.

---

CRS-8
those eligible) claimed the deduction on their income tax returns for 1998. The
revenue cost to the federal government of the deduction is expected to amount to
$1.2 billion in 2000.20
Medical Savings Accounts
The federal tax code also subsidizes enrollment in high-deductible health
insurance plans through the tax treatment of contributions to medical savings accounts
(MSAs). MSAs are personal savings accounts established for the express purpose of
paying for health care costs not covered by health insurance, including insurance
copayments and deductibles.21 Under IRC section 220, an employer or an individual
(but not both) may contribute to a MSA up to an annual amount of 65% of the
deductible for an insurance plan covering a single person (which may range from
$1,550 to $2,350 in 2000) and 75% of the deductible for a plan covering more than
one person (which may range from $3,100 to $4,650 in 2000) to a MSA. Only the
self-employed and employees of firms with fewer than 50 employees are eligible to
establish MSAs. Employer contributions are exempt from federal income and
employment taxes, and individual contributions are deductible from gross income.
Earnings on account balances are not taxed. To be eligible for these tax benefits, an
individual must be covered by a high-deductible health plan and no other insurance
(with some exceptions). Withdrawals from MSAs are tax-exempt if they are used to
pay for medical expenses not covered by the high-deductible insurance plans
(including those needed to satisfy the deductible), health insurance continuation
coverage required under federal law, or health insurance when an individual is
unemployed. There are no limits on withdrawals for qualified expenses. Withdrawals
for other purposes are included in a person’s gross income and subject to a 15%
penalty (except in the cases of disability, death, or reaching age 65).
The law establishing the tax benefits for MSAs, the Health Insurance Portability
and Accountability Act of 1996 (P.L. 104-191), authorized the creation of 750,000
MSAs in 1997 through 2000 as a demonstration project; this amount did not include
MSAs opened by those who previously were uninsured. Eligibility for the same
number of MSAs has been extended to the end of 2002 as a result of the enactment
of the last piece of tax legislation considered by the 106th Congress: the Community
Renewal Tax Relief Act of 2000 (P.L. 106-554). MSAs have been much less popular
than originally hoped: by one estimate, about 100,000 MSAs had been established
as of early 2000.22
20U.S. Office of Management and Budget, Analytical Perspectives: Budget of the U.S.
Government, Fiscal Year 2001, p. 114.
21For a comparison of MSAs and flexible spending accounts, see U.S. Library of Congress,
Congressional Research Service, Flexible Spending Accounts and Medical Savings Accounts:
A Comparison, by Bob Lyke, CRS report 96-500 EPW (Washington: October 6, 2000).
22Jennifer Steinhauer,“A Spending Account in Need of a Cure,” New York Times, Sept. 3,
2000, sec. BU, p. 11. (No source for the estimate is cited in the article.)

---

CRS-9
Economic Rationale for Government Intervention in the Health
Insurance Market
The tax subsidies described above represent a form of government intervention
in the market for health insurance. In addition, the federal government takes on the
role of an insurer by financing (at a cost of $317 billion in fiscal year 1999) the private
purchases of health care by the elderly, a large segment of the poor, and many
disabled persons through the Medicare and Medicaid programs. In general,
economists counsel against government intervention in private markets unless they fail
to achieve efficient or equitable outcomes. These market failures are likely to arise
under the following conditions: a lack of perfect competition; the existence of public
goods; the existence of positive or negative externalities; the absence of markets when
they should exist; the presence of imperfect information; episodes of inflation,
unemployment, and other macroeconomic disequilibria; inequality in the distribution
of income; and the existence of merit goods, which are goods and services that
governments provide, induce, or compel people to consume in an effort to enhance
social welfare.23
Is there reason to believe that the private health insurance market would fail to
achieve efficient or equitable outcomes in the absence of government involvement?
Answering this question is difficult because it entails making counterfactual
judgments about how the insurance market would operate without federal support for
health insurance coverage. Nonetheless, it is possible that without government
intervention too little health insurance would be supplied relative to its potential
efficiency gains. The presence of more than 40 million Americans without health
insurance, many of whom reside in low-income households, suggests that such a
scenario is more than just a theoretical possibility.
Mainstream economic theory holds that social welfare is enhanced by the
presence of actuarially fair health insurance.24 This is because risk-averse individuals
are thought to be better off when they can reduce their financial risks and transfer
them to entities that have a comparative advantage in risk bearing because of their
capacity for risk pooling and diversification. Accordingly, most individuals
presumably would want to buy health insurance to protect themselves against the
catastrophic financial losses that serious illnesses or injuries can entail. Health
insurance is most likely to yield efficiency gains when risks are large and neither
individuals nor insurers know the probability that these same individuals will
experience specific costly health problems.
Yet because of fundamental problems in the U.S. markets for health insurance
and health care, an inefficient amount of health insurance might be provided and its
23For more details, see Joseph E. Stiglitz, Economics of the Public Sector, 3rd Edition (New
York: W.W. Norton & Co., 2000), pp. 77-85.
24James M. Poterba, “Government Intervention in the Markets for Education and Health Care:
How and Why,” in Individual and Social Responsibility: Child Care, Education, Medical
Care, and Long-Term Care in America, Victor R. Fuchs, ed. (Chicago: University of
Chicago Press, 1996), p. 282.

---

CRS-10
distribution may be inequitable without government intervention. The chief problems
are adverse selection and moral hazard in the health insurance market and the
presence of free riders in the medical care market.
Adverse selection refers to the condition that arises when a majority of buyers
of a given health insurance plan have above-average risks of suffering costly health
problems relative to the group or segment of the population for which the plan was
developed and to which it is being marketed. Such a condition is likely to develop in
health insurance markets where participation is voluntary and there are many insurers
offering consumers a range of competing plans with varied benefits. Adverse
selection stems from two defining features of the health care market: the risk of ill
health and the propensity to use health care vary considerably in the population at
large (with a small percentage accounting for a large percentage of medical
expenditures); and individuals know more about their health status and preferences
for health care than do insurers, despite the steps taken by insurers to gather this
information. In theory, adverse selection can trigger an upward spiral in premiums
for generous policies that drive many individuals out of health insurance risk pools.
Moral hazard refers to the efficiency losses that stem from the impact of health
insurance on the demand for health care. Economists view moral hazard as a problem
of incentives rather than a matter of immoral or unethical behavior. Health insurance
lowers the prices that insured individuals pay out-of-pocket for most health care
services, making them less sensitive to the marginal costs and benefits of these
services and encouraging excessive consumption of health care. Health insurers, of
course, are keenly aware of this problem and try to control it by limiting coverage for
specific services, requiring individuals to share the cost of the care through the
payment of deductibles and copayments, excluding certain services from coverage,
and closely monitoring the use of care. The presence of moral hazard suggests that
there is an inescapable trade-off between the risk reduction and spreading insurance
offers and the overutilization of health care it fosters.
In addition, the health care market is rife with imperfections, one of which is the
existence of free riders. In every state, it is unlawful for health care providers –
especially publicly funded hospitals – to deny treatment to someone with life-
threatening or serious health problems. This legal sanction lessens the incentive to
purchase health insurance – especially for individuals with few financial assets to
protect – and endows health care with a principal trait of public goods: the
impossibility of excluding anybody from consuming such a good. Since in practice
nobody can be excluded from consuming urgent medical care, consumers have less
of an incentive to pay for it or to purchase health insurance, giving rise to a free rider
problem.25 The cost of uncompensated care – which totaled an estimated $37.9
billion for community hospitals and private physicians in 199426 – is passed on to
25The problem can be defined as the incentive to allow other individuals pay for a public good
while you enjoy the benefits it provides. For more details, see Harvey S. Rosen, Public
Finance, 5th Edition (New York: Irwin/McGraw Hill,1999), p. 69-70.
26Peter J. Cunningham and Ha T. Tu, “A Changing Picture of Uncompensated Care,” Health
(continued...)

---

CRS-11
consumers through higher prices for health care services and higher health insurance
premiums, to health care providers through reduced incomes, and to local, state, and
federal governments through increased subsidies for charity care (the cost of which
ultimately is borne by taxpayers).
The problems of adverse selection, moral hazard, and free riders have important
implications for the health insurance market, as the Congressional Budget Office
noted in a 1994 report27. On the one hand, they can raise the price of health
insurance, causing too few individuals to be covered by it. On the other hand, those
who are insured are likely to consume an inefficient amount of health care. Both
outcomes produce efficiency losses, and although it is unknown how large these
losses are, the fact that they could occur arguably provides an economic rationale for
government intervention in the health insurance market.
To say that in theory there is an economic justification for federal subsidies for
the purchase of health insurance is not to suggest that current federal involvement in
the health insurance market leads to optimal outcomes. Some would contend that the
existence of over 40 million uninsured is sufficient proof that current federal policy
toward health insurance is far from optimal. But the problems that invite government
intervention in the health insurance market also offer some guidance about how the
intervention could be structured to achieve something closer to optimal results.
Among other things, they imply that federal policy should at once encourage all
individuals to purchase health insurance but discourage individuals from purchasing
insurance that promotes the excessive consumption of health care.
Economic Effects of Current Federal Tax Subsidies for Health
Insurance
The federal tax code subsidizes the purchase of health insurance by lowering its
after-tax price relative to the after-tax prices of most other goods and services. It
does this through a variety of deductions and exclusions, the most important of which
in revenue cost is the tax exclusion for employer contributions to employee health
insurance premiums. Tax subsidies of these types are more valuable to high-income
households than low-income ones because the subsidy rates reflect an individual’s
marginal tax rate. Through their price effects, the subsidies directly boost U.S. health
insurance coverage. And through their impact on health insurance coverage, they
indirectly boost the consumption of health care. As noted in the previous section,
insured individuals are likely to spend more on health care and be less sensitive to its
marginal costs and benefits than uninsured individuals because insurance dramatically
lowers the out-of-pocket cost of this care. The Lewin Group estimates that current
26(...continued)
Affairs, vol. 16, July/August 1997, pp. 167-175.
27U.S. Congressional Budget Office, The Tax Treatment of Employment-Based Health
Insurance (Washington: GPO, March 1994), pp. 13-16.

---

CRS-12
tax subsidies for the purchase of health insurance could cost the federal government
$125.6 billion in forgone income and employment tax revenues in 2000.28
Are these subsidies desirable from the perspective of economic efficiency and
equity? They appear to be something of a mixed blessing.
On the one hand, it can be argued that current federal tax subsidies for health
insurance improve social welfare by increasing health insurance coverage and
expanding access to health care among the nonelderly. In 1998, about 155 million
nonelderly Americans (or 65% of the nonelderly population) were covered by
employment-based health insurance. The deductibility of health benefits as a business
expense makes employers indifferent between paying workers in health benefits or
wages, and the exclusion of employer payments for health insurance from employees’
taxable income gives employees a robust incentive to seek compensation in the form
of generous health benefits. In addition, health insurance offered through employers
is less likely to fall prey to adverse selection. (See the previous section for an
explanation of adverse selection and its efficiency effects.) Employment-based groups
narrow the scope for adverse selection because individuals generally choose where to
work for reasons other than their health status and expected health care needs, and
large groups of employees typically encompass a wide range of health risks, making
it possible for insurers to use community rating in setting health insurance premiums.29
Group health insurance is much cheaper than individual or non-group insurance,
mainly because of its advantages in risk pooling and its economies of scale in
marketing and plan administration.
On the other hand, not all the efficiency and equity effects of existing tax
subsidies for the purchase of health insurance are desirable. Critics maintain that the
exclusion for employer health insurance premium payments contributes to excessive
consumption of health care by encouraging eligible individuals to seek comprehensive
health insurance with generous benefits and low copayments or deductibles. It will
be recalled from the discussion of moral hazard in the previous section that increases
in health benefits stimulate increases in the consumption of health care. Health
economist Charles Phelps has estimated that in the absence of the exclusion for
employer-provided health insurance, employment-based health insurance premiums
would be significantly lower, because the demand for generous health insurance
coverage would be significantly less.30
Furthermore, critics contend that current tax subsidies for health insurance are
inequitable on two grounds. One is that they benefit wealthy individuals more than
poor individuals, even though poor individuals not covered by Medicaid are more
likely to be uninsured and thus have a greater need for financial assistance in the
28Sheils, Hogan, and Haught, Health Insurance and Taxes: The Impact of Proposed Changes
in Current Federal Policy, p. 6.
29U.S. Congressional Budget Office, The Tax Treatment of Employment-Based Health
Insurance, p. 23.
30Charles E. Phelps, Health Economics, 2nd edition (Reading, MA: Addison-Wesley, 1997),
pp. 356-357.

---

CRS-13
purchase of health insurance. The Lewin Group has estimated that in 2000, the
average value of available tax subsidies for health insurance for families with incomes
over $100,000 will be $2,638, compared with $79 for families with incomes of less
than $15,000.31 The second reason why the tax subsidies are inequitable is that not
all taxpayers can claim them. In particular, two groups receive no tax benefits under
current law: those who work for firms that offer no health insurance and those who
are unemployed or retire before becoming eligible for Medicare. Together, they
represent up to 16% of the nonelderly population, according to estimates by Gruber.32
And there is some concern that the exclusion for employer contributions for
health insurance causes strains in the labor market that may result in efficiency losses.
Critics charge that it distorts the cost of labor by giving larger firms an advantage over
smaller firms in offering health insurance to prospective workers. The exclusion also
affects the decisions of where to work and whether to change jobs in ways that might
reduce individual welfare.33 While the average worker may be indifferent between
insurance coverage and additional wages in his or her compensation, workers with
pre-existing health problems or a strong preference for continuous health insurance
coverage are unlikely to have the same view. As a result, these workers will be less
likely to leave jobs with health insurance or take jobs without this coverage. In
addition, in periods of rising health care costs and stagnant or falling real incomes, not
all workers are willing to accept lower wages to pay for increases in the cost of health
insurance. This is especially true of low-income workers. Employers have responded
to this reluctance by increasing their use of contingent workers, temporary workers,
and part-time workers, for whom health benefits are not generally provided.34
Recent Congressional Proposals to Extend Tax
Subsidies for the Purchase of Health Insurance to
the Uninsured
Following the defeat of President Clinton’s controversial and complex health
care reform plan in 1994, policy discussions on expanding domestic health insurance
coverage have focused largely on incremental reforms. Congressional consideration
of the issue in recent years has taken a similar bent. Since the establishment of the
State Children’s Health Insurance Program under the Balanced Budget Act of 1997
(P.L. 105-33), which makes available matching federal funds for states and territories
to provide health insurance for children from low-income families, Congress in
31Sheils, Hogan, and Haught, Health Insurance and Taxes: The Impact of Proposed Changes
in Current Federal Policy, p. 8.
32Gruber, Tax Subsidies for Health Insurance: Evaluating the Costs and Benefits, p. 4.
33See U.S. Congressional Budget Office, The Tax Treatment of Employment-Based Health
Insurance, pp. 20-23; and Jonathan Gruber, Health Insurance and the Labor Market,
Working Paper 6762 (Cambridge, MA: National Bureau of Economic Research, October
1998).
34David M. Cutler, “Public Policy for Health Care,” in Fiscal Policy: Lessons from Economic
Research, Alan J. Auerbach, ed. (Cambridge, MA: MIT Press, 1997), p. 190.

---

CRS-14
particular has shown a strong interest in tax-based approaches to expanding health
insurance coverage.
The depth of this interest can be gauged by the numerous proposals in the 106th
Congress to create new tax subsidies for the purchase of health insurance and expand
existing ones. Numerous bills were introduced to accomplish one or more of the
following objectives: establish tax deductions or credits for the purchase of health
insurance by individuals with and without access to employment-based insurance,
accelerate the schedule for making health insurance spending by the self-employed
fully deductible, loosen eligibility for medical savings accounts, offer tax credits to
small firms that provide health insurance to employees, and permit unused balances
in cafeteria plans and flexible spending accounts to be carried over to the following
year without being subject to taxation.35 None of these measures was targeted
exclusively at uninsured individuals and their spouses and dependents. Rather, the
proposals that would have assisted the uninsured also would have benefitted in
varying degrees persons who already were insured through policies they purchased
in the individual or non-group market.
As a general principle, the effectiveness and equity effects of tax-based
approaches to expanding health insurance coverage hinge on their design. A crucial
consideration is the form taken by the subsidy: specifically, is it a deduction, a non-
refundable tax credit, or a refundable tax credit? Nearly half of the uninsured have no
tax liability and thus would not benefit from a deduction or non-refundable tax credit
for the purchase of health insurance. A deduction would allocate more of its benefits
to upper-income individuals than a credit would. An itemized deduction would have
a smaller impact than a deduction that is available to all taxpayers because only 15%
of taxpayers with adjusted gross incomes below $50,000 itemize deductions on their
tax returns. And only a refundable tax credit would be available to all taxpayers. But
even that may be of limited assistance to the uninsured unless it is tied to the cost of
group or non-group health insurance policies rather than an individual taxpayer’s
income tax liability. In a recent study of tax subsidies for health insurance, Jonathan
Gruber of the Massachusetts Institute of Technology found that in 1997, more than
60% of uninsured taxpayers have tax liabilities of under $1,000, while the average
cost of non-group health insurance was $2,542 for individuals and $6,740 for
families.36
Tax-based approaches to expanding health insurance coverage have considerable
appeal for many policymakers. They attempt to improve horizontal equity in the tax
code by enabling more individuals to claim tax subsidies for health insurance. They
rely on the private insurers rather than the federal government to extend coverage.
And they represent a tax cut, which for some is preferable to undertaking new and
possibly controversial federal spending and entitlement programs.
35For a description of proposals in the current Congress to modify current tax benefits for
health insurance, see U.S. Library of Congress, Congressional Research Service, Tax Benefits
for Health Insurance: Current Legislation, by Bob Lyke, CRS issue brief IB98037, updated
October 26, 2000 (Washington: updated continually), pp. 8-16.
36Gruber, Tax Subsidies for Health Insurance: Evaluating the Costs and Benefits, p. 37.

---

CRS-15
Potential Effectiveness and Cost of Selected
Proposed Tax Subsidies for Health Insurance for
the Uninsured
Recent congressional proposals to employ tax policy to reduce the uninsured
population raise the issues of what approach is likely to be most effective, efficient,
and equitable. Two recent studies of tax subsidies for health insurance shed some
light on this important policy issue. One is a study by the Lewin Group for the
National Coalition on Health Care (NCHC), which is a non-profit, non-partisan
organization based in Washington, D.C. that is supported by a variety of large and
small firms, labor unions, consumer advocacy groups, religious organizations, and
health care providers and seeks to raise public awareness of problems with the U.S.
health care system.37 The second study was done by Jonathan Gruber, an economist
at the Massachusetts Institute of Technology, with financial backing from the Kaiser
Family Foundation.38 Both examined the potential effectiveness and revenue cost of
a variety of proposed tax subsidies for the purchase of health insurance. In addition,
they also briefly discussed some of the key administrative issues presented by the
proposals, and the Gruber study tried to illuminate the equity effects of the subsidies.
The subsidies analyzed in both reports were similar to some of the proposals to create
tax subsidies for the purchase of health insurance considered in the 106th Congress.39
As a result, the studies’ findings might be of use to the 107th Congress if it focuses on
the desirability of various tax-based approaches to expanding health insurance
coverage among nonelderly Americans.
Lewin Group and Gruber Studies
Lewin Group Analysis.
A 1999 study by the Lewin Group assessed the effectiveness and revenue cost
of nine proposed tax subsidies for the purchase of health insurance. The proposals
ranged from something as simple as a tax deduction for the purchase of non-group
health insurance to something as complex as replacing the current tax subsidies for
health insurance with individual insurance mandates and refundable tax credits for the
37Sheils, Hogan, and Haught, Health Insurance and Taxes: The Impact of Proposed Changes
in Current Federal Policy, 54 p.
38Gruber, Tax Subsidies for Health Insurance: Evaluating the Costs and Benefits, 63 p.
39For instance, H.R. 4113 from the 106th Congress would have established an income-limited
refundable tax credit (capped at $1,000 for individuals and $2,000 for families) for the
purchase of non-group health insurance by any taxpayer without employment-based health
insurance and not covered under federal health insurance programs. Similarly, the Lewin
Group evaluated the cost and effectiveness of a refundable tax credit (limited to $500 for
individuals and $1,000 for families) for the purchase of non-group health insurance by
individuals without employment-based health insurance and not covered by Medicare or
Medicaid. And the base case in Gruber’s analysis was an income-limited refundable tax credit
for the purchase of non-group health insurance by all individuals without employment-based
insurance coverage and not covered by Medicare that was limited to $1,000 for single filers
and $2,000 for joint and head-of-household filers.

---

CRS-16
purchase of non-group health insurance. For each proposal, the Lewin Group
employed a computerized model of the U.S. health care system (known as the Lewin
Group Health Benefits Simulation Model) to estimate the number of persons who
would be eligible for it, the change in the number of persons with health insurance in
response to it, and its federal revenue cost in 2000. These estimates in turn were used
to compute the percentage change in the uninsured population relative to 1997 – the
most recent year for which figures on the uninsured population were available when
the study was done – and the revenue cost per newly insured person.
The simulation model generated these estimates on the basis of several key
assumptions. Two deserve mention here. One assumption addressed the sensitivity
of consumers to changes in the after-tax price of health insurance – or the price
elasticity of demand for health insurance. The proposed tax subsidies would lower
the after-tax price of health insurance by varying amounts for eligible individuals,
spurring an increase in health insurance coverage among these individuals. The
magnitude of the increase depends on the price elasticity of demand. The study used
an elasticity of -0.2, which meant that a 1.0 % decrease in the after-tax price of
insurance triggered a 0.2% rise in coverage nation-wide. There is some disagreement
among health economists over the actual price elasticity of demand for health
insurance; various analysts have derived estimates ranging from 0.0 to -2.7.40
Another important assumption concerned the number of individuals who would
claim the proposed tax subsidies. For subsidies that supplemented (rather than
supplanted) current federal tax subsidies for health insurance, the study assumed that
every eligible individual currently covered by non-group insurance would claim the
subsidies, whether a deduction or refundable tax credit. The rate at which a tax
subsidy is claimed or taken up matters because its revenue cost and efficacy depend
in part on how many individuals actually claim it. However, if the early response to
the tax deduction for health insurance purchased by the self-employed is any
indication, the initial take-up rate is likely to be significantly below 100%.41
40The coefficient for the price elasticity used in the Lewin Group study came from a 1998
analysis by the same organization that analyzed the effect of changes in employee
contributions to employer health insurance on the number of workers and dependents
purchasing employer-provided health insurance. Other studies have come up with different
estimates of the price elasticity of demand for health insurance, but in doing so they employed
different estimation methods, for the most part. For a discussion of these studies, see Gruber
and Poterba, “Fundamental Tax Reform and Employer-Provided Health Insurance,” pp. 159-
162.
41In a 1993 study of the response of self-employed individuals to the creation of a partial tax
deduction for their health insurance expenditures by the Tax Reform Act of 1986, economists
Jonathan Gruber and James Poterba found that from 1986 to 1989, between 15% and 20%
of the self-employed with incomes below $20,000 and about 50% of those with incomes above
$50,000 claimed the deduction. From these response rates, they concluded that a lack of
awareness of the presence of the subsidy undercut its effectiveness. See U.S. Congress,
House Committee on Ways and Means, Subcommittee on Health, Health Insurance Premium
Deductions for the Self-Employed, hearing, 104th Cong., 1st sess., Jan. 27, 1995 (Washington:
GPO,1996), pp. 42-43.

---

CRS-17
Several of the study’s findings have implications for the potential effectiveness,
cost, and design of tax subsidies for the purchase of health insurance by the uninsured
and thus are worth highlighting:
Lewin Groups’s Findings on Cost-Effectiveness.42
! The most cost-effective subsidy was a refundable tax credit for the purchase
of non-group insurance by taxpayers not covered by employment-based health
insurance, Medicare, or Medicaid that was capped at $500 for single persons
and $1,000 for families; the cost per newly insured person was $1,246 (2000
dollars).
! The least cost-effective subsidy was one that scrapped existing tax subsidies
for the purchase of health insurance – except for medical savings accounts –
and replaced them with a fixed refundable tax credit of $800 per adult and
$400 per child (capped at $2,400 per family) for the purchase of health
insurance by all taxpayers except those who are covered under Medicare or
Medicaid; the cost per newly insured person was $10,541 (2000 dollars).
! Refundable tax credits for the purchase of health insurance by low-income
working and non-working individuals were more effective but more costly per
newly insured person than an above-the-line tax deduction for the purchase of
non-group health insurance by taxpayers who were not covered by
employment-based insurance, Medicare, or Medicaid.
Lewin Group’s Findings on Effectiveness.43
! The most effective subsidy was a plan developed by the Heritage Foundation;
among other things, it abolished all existing tax subsidies for health insurance,
replaced them with a refundable tax credit for health insurance and medical
care expenditures, and required all individuals to purchase a minimum level of
health benefits and all employers to convert their health benefit plans to wages;
it eliminated the uninsured population, which was assumed to total 43.3 million
in 2000.
! The least effective subsidy was a 30% refundable tax credit for non-group
health insurance purchased by taxpayers without access to employment-based
health insurance and not covered under Medicare or Medicaid; the credit
phased out for single persons with adjusted gross incomes between $25,000
and $35,000 and for married couples with adjusted gross incomes between
$40,000 and $50,000; it led to a reduction in the uninsured population of 1.5
million.
42Cost-effectiveness in this context denotes the revenue cost per newly insured person.
43Effectiveness in this context refers to the total decrease in the uninsured population.

---

CRS-18
Lewin Group’s Findings on Cost.
! The most costly subsidy was the Heritage Foundation plan; its net revenue cost
was $55.3 billion (2000 dollars).
! The least costly subsidy was the 30% refundable tax credit for the purchase of
non-group health insurance by individuals without access to employment-based
health insurance and not covered by Medicare or Medicaid; its net revenue cost
was $3.3 billion (2000 dollars).
Gruber Analysis.
The Gruber study covered similar terrain: the potential effects of alternative tax
subsidies for health insurance on the number of uninsured and their federal revenue
cost. Like the Lewin Group study, many of the tax subsidies it evaluated were
modeled after legislative proposals considered in the 106th Congress. And the Gruber
study also employed a computerized model of the U.S. health care system to analyze
these effects. But the Gruber study went beyond the Lewin Group study in that it
addressed two important issues raised by tax-based approaches to expanding health
insurance coverage that were either not included or treated marginally in the Lewin
Group study: the extent to which different income groups benefitted from the
proposed subsidies, and the impact of the subsidies on the market for group health
insurance. The Gruber analysis also differed in its approach (but not in its basic
methodology). It simulated the effects of a “base case” along a number of
dimensions, including the federal revenue cost, the size of the uninsured population,
the number of individuals with employment-based health insurance, and the
distribution of the net cost of the subsidy among major income groups.44 Gruber then
simulated the effects of other proposed tax subsidies (e.g., a non-refundable tax credit
for the purchase of non-group insurance and an above-the-line tax deduction for the
same purpose) on the same variables and compared the results to those for the base
case. The insights for policy making lay in the differences between the base case and
each of the alternatives.
For each proposed tax subsidy, Gruber’s simulation model estimated the revenue
cost to the federal government, how many individuals would become insured, how the
benefits would be distributed among income groups, and how many individuals with
employment-based health insurance would drop or lose it in response to the subsidy.45
In estimating these effects, he made assumptions about a number of key behavioral
variables, including the extent to which those who were uninsured used the subsidies
to purchase non-group coverage, the extent to which those covered by non-group
health insurance claimed the proposed subsidies, and the extent to which firms reacted
44The base case involved a refundable tax credit for the purchase of non-group health
insurance by individuals not covered by Medicare or employment-based health insurance. The
credit was limited to $1,000 for single filers and $2,000 for joint filers and head-of-household
filers, and it phased out for single filers with adjusted gross incomes between $45,000 and
$60,000 and for joint and head-of-household filers with adjusted gross incomes between
$75,000 and $100,000.
45Gruber, Tax Subsidies for Health Insurance: Evaluating the Costs and Benefits, p. 9

---

CRS-19
to the subsidies by dropping or cutting health benefits for employees. Like the Lewin
Group study, the results of Gruber’s analysis hinged on the sensitivity of the demand
for health insurance coverage to declines in its after-tax price. Gruber used a higher
price elasticity than the Lewin Group study: -0.53 compared to -0.2. This was a
significant difference in that a 1% fall in the after-tax price of health insurance
stimulated an increase in insurance coverage among the uninsured that was 2.6 times
greater than the increase in the Lewin Group’s analysis. In assessing the validity of
Gruber’s assumed price elasticity, it should be kept in mind that it was closer to the
middle of the range of available elasticity estimates than the one used in the Lewin
Group study.
Like the Lewin Group study, several results of the Gruber analysis are worth
mentioning because of what they imply about the effectiveness, cost, and design of tax
subsidies for the purchase of health insurance by the uninsured:
Gruber’s Findings on Cost-Effectiveness.
! The most cost-effective subsidy was a refundable tax credit for the purchase
of non-group health insurance by individuals not covered by Medicare or
employment-based health insurance; it was limited to $500 for single filers and
$1,000 for joint filers and head-of-household filers and phased out for single
filers with adjusted gross incomes between $45,000 and $60,000 and for joint
and head-of-household filers with adjusted gross incomes between $75,000
and $100,000; the cost per newly insured person was $2,239 (1999 dollars).
! The least cost-effective subsidy was the refundable tax credit for all out-of-
pocket health insurance expenditures by individuals not covered by Medicare;
the cost per newly insured person was $5,003 (1999 dollars).
Gruber’s Findings on Effectiveness.
! The most effective tax subsidy was a refundable tax credit for all out-of-pocket
health insurance expenditures by individuals not covered by Medicare that was
limited to $1,000 for single filers and $2,000 for joint filers and head-of-
household filers; it reduced the uninsured population by 12.4 million.
! Running a close second in effectiveness was a refundable tax credit for the
purchase of non-group health insurance by individuals not covered by
Medicare or employment-based health insurance; the credit was limited to
$2,000 for single filers and $4,000 for joint and head-of-household filers and
paid directly to insurers when premium payments were due; it reduced the
uninsured population by 12.1 million.
! The least effective subsidy was an above-the-line tax deduction for the
purchase of non-group health insurance by individuals not covered by
Medicare or employment-based insurance; it lowered the uninsured population
by 0.25 million.

---

CRS-20
Gruber’s Findings on Cost.
! The most costly subsidy was a refundable tax credit for all out-of-pocket health
insurance expenditures by individuals not covered by Medicare that was limited
to $1,000 for single filers and $2,000 for joint filers and head-of-household
filers; its net revenue cost was $62.2 billion (1999 dollars).
! The least costly subsidy was the above-the-line tax deduction, which carried
a net revenue cost of $0.9 billion (1999 dollars).
Gruber’s Findings on Vertical Equity.
! The subsidy with the highest share of its benefits going to low-income
households was a refundable tax credit for the purchase of non-group health
insurance by individuals not covered by Medicare or employment-based health
insurance; the credit was limited to $1,000 for single filers and $2,000 for joint
filers and head-of-household filers and phased out for single filers with adjusted
gross incomes between $18,000 and $25,000 and for joint and head-of-
household filers with adjusted gross incomes between $30,000 and $50,000;
the share of its net revenue cost received by households with incomes below
200% of the federal poverty level in 1999 was 69%.
! The subsidy with the lowest share of its benefits going to low-income
households was a non-refundable tax credit for the purchase of non-group
health insurance by individuals not covered by Medicare or employment-based
health insurance; the credit was limited to $1,000 for single filers and $2,000
for joint filers and head-of-household filers; the share of its net revenue cost
received by households with incomes below 200% of the federal poverty level
in 1999 was 22%.
! On the whole, refundable tax credits for the purchase of non-group health
insurance by individuals not covered by Medicare or employment-based health
insurance were of much greater benefit to low-income households than either
a non-refundable tax credit or a tax deduction of the same design.
Gruber’s Findings on Employment-Based Health Insurance
Coverage.
! Only one subsidy resulted in an increase (6.6%) in the number of individuals
covered by employment-based health insurance: the refundable tax credit for
out-of-pocket health insurance expenditures by all individuals not covered by
Medicare.
! The subsidy that triggered the smallest decline (-0.9%) in the number of
individuals covered by employment-based insurance was the tax deduction for
the purchase of non-group health insurance.
! Two subsidies led to the largest decline (-9.6%) in the number of individuals
covered by employment-based insurance: (1) a refundable tax credit for the
purchase of non-group health insurance by individuals not covered by

---

CRS-21
Medicare or employment-based health insurance that was limited to $2,000 for
single filers and $4,000 for joint filers and head-of-household filers and phased
out for single filers with adjusted gross incomes between$45,000 and $60,000
and for joint and head-of-household filers with adjusted gross incomes between
$75,000 and $100,000; (2) the same credit with no mismatches between the
timing of health insurance premium payments and the timing of subsidy
transfers to recipients with minimal household savings.
Gruber’s Findings on Key Administrative Issues.
! Avoiding mismatches between the timing of health insurance premium
payments and the timing of subsidy transfers to eligible individuals with
minimal household savings had a significant impact on the efficacy of the base
case, which was a refundable tax credit for the purchase of non-group health
insurance by individuals not covered by Medicare or employment-based health
insurance that was limited to $1,000 for single filers and $2,000 for joint and
head-of-household filers and phased out for single filers with adjusted gross
incomes between $45,000 and $60,000 and for joint and head-of-household
filers with adjusted gross incomes between $75,000 and $100,000; the
reduction in the uninsured population was 37% greater when the base case was
simulated on the assumption that there were no such mismatches.
Implications of the Gruber and Lewin Group Studies for the
Potential Cost and Effectiveness of Proposed Tax Subsidies
for Health Insurance for the Uninsured
The results of both studies highlighted above have important implications for the
ability of tax policy to achieve substantial reductions in the uninsured population.
Specifically, they make clear that the effectiveness, net revenue cost, and equity
effects of tax subsidies for this purpose depend critically on their design. And the
results suggest that certain issues are critical to the design of cost-effective tax
subsidies for health insurance for the uninsured. Those issues are framed below as
questions:
! Is the subsidy a tax deduction or a tax credit? This is a key issue because
many uninsured individuals reside in low-income households, and, assuming
the head of such a household has a tax liability, a tax credit for the purchase of
health insurance is more valuable than a tax deduction for the same purpose.
Gruber has estimated that 90% of uninsured taxpayers with tax liabilities
belong to the 15% income tax bracket, and that 50% of uninsured taxpayers
have no federal income tax liability.
! If the subsidy is a tax credit, is it refundable? Non-refundable tax credits
are valuable only if a taxpayer has a tax liability against which it can be
claimed, whereas refundable tax credits are valuable to all taxpayers because
they can be claimed regardless of whether or not someone has a tax liability.
Gruber has estimated that half of uninsured taxpayers have no tax liability
against which a tax credit for the purchase of health insurance could be
claimed.

---

CRS-22
! If the subsidy is a refundable tax credit, what proportion of the cost of a
typical non-group health insurance policy does it cover? The results of the
Gruber and Lewin Group studies indicate that a key factor shaping the
effectiveness of a tax subsidy for the purchase of health insurance by the
uninsured is the proportion of the cost of insurance covered by the subsidy.
The greater the share of the cost met by the subsidy, the more effective it is
likely to be. This is to be expected as many uninsured individuals live in low-
income households, where spending on necessities such as food, clothing, and
shelter accounts for a huge share of disposable income.
! Is the subsidy targeted to low-income groups? The Gruber analysis
suggests that the efficacy and cost per newly insured person of tax subsidies
for the purchase of health insurance by the uninsured depend in part on the
extent to which they are targeted at low-income households. Targeting of this
sort is typically accomplished by specifying an income range over which a
subsidy is phased out.
! Who is eligible for the subsidy? Tax subsidies for the purchase of health
insurance by the uninsured can be limited to individuals without access to
Medicaid and employment-based insurance or to all uninsured individuals,
including those who choose not to buy insurance offered by their employers
because of the out-of-pocket cost. The results of both studies indicate that
who is eligible has a significant effect on a proposed subsidy’s total cost, cost
per newly insured person, and effectiveness.
! Does the subsidy address administrative issues it may raise? Among other
things, the results of Gruber’s analysis imply that measures to ease liquidity or
cash-flow constraints on the recipients of tax subsidies for the purchase of
health insurance enhance their effectiveness. This raises the broader question
of the administrative issues that could complicate the implementation of such
tax subsidies. The key issues cited by many analysts include how to prevent
mismatches between the payment of health insurance premiums and the
transfer of subsidies to eligible individuals, how to define health plans that are
eligible for the subsidy, how to verify that employers offer health insurance to
employees, and whether a subsidy should be uniform for all eligible individuals
or adjusted to reflect variations in premiums related to geographic location,
age, and health status.
Conclusions
Steady increases in the number of uninsured Americans through much of the
1990s, despite sustained economic growth and falling unemployment rates, have
instilled in many policymakers a renewed interest in finding ways to expand health
insurance coverage. Recent Congresses have exhibited a preference for using tax
policy to accomplish this goal. For example, numerous bills creating tax deductions,
non-refundable tax credits, and refundable tax credits for the purchase of non-group
health insurance by those without employment-based health insurance were
introduced in the 106th Congress. Two of their main objectives were to make health

---

CRS-23
insurance more affordable for the uninsured and to inject greater equity into the tax
treatment of health insurance. The proposals raised a number of interesting policy
issues, including their likely efficiency and equity effects, the cost and difficulty of
administering and complying with the proposed subsidies, and their likely
effectiveness in shrinking the uninsured population.
Several conclusions related to these issues can be drawn from the material
presented in this report. First, the cost, effectiveness, and equity and efficiency effects
of tax subsidies for the purchase of health insurance by the uninsured hinge on their
design. What is more, the critical design components are the type of tax incentive
(i.e., above-the-line deduction, itemized deduction, non-refundable credit or
refundable credit), the proportion of the cost of a typical non-group health insurance
policy covered by the subsidy, who is eligible for the subsidy (e.g., all individuals or
only low-income individuals or low-income workers), and administrative issues raised
by the subsidy (e.g., mismatches between the payment of health insurance premiums
and transfers of subsidy payments to eligible individuals with minimal savings).
Second, it is difficult to design a tax subsidy that delivers a substantial reduction
in the number of uninsured individuals at a modest cost per newly insured person
without profoundly altering the existing system of employment-based health
insurance. In the studies by the Lewin Group and Gruber, the only proposed subsidy
that led to universal coverage was the fundamental reform plan advocated by the
Heritage Foundation, which required every person to purchase insurance in the non-
group market and all employers to convert their health benefit plans to the equivalent
in wages, repealed current tax subsidies for health insurance, and replaced them with
a refundable tax credit for health insurance and medical expenditures. Under Gruber’s
base case – a refundable credit of $1,000 for single persons and $2,000 for families
not covered by Medicare or employment-based health insurance with income limits
on eligibility – the size of the uninsured population declined by only 10%; although
this subsidy was more generous than many of the proposals considered in the 106th
Congress, it still covered less than half the estimated cost of a non-group policy for
a single adult.46
Third, some proposed tax subsidies to expand insurance coverage entail
challenging tradeoffs among the goals of improved equity, increased efficiency,
superior efficacy, and administrative simplicity. For example, Gruber’s analysis
indicated that a non-refundable tax credit or an above-the-line tax deduction would
be simpler to administer than a refundable tax credit of the same design, but the non-
refundable credit and the deduction would spur smaller declines in the uninsured
population, would be of greater benefit to high-income individuals than low-income
individuals, and would cost more per newly insured individual.
Fourth, many recent proposals to extend tax subsidies for the purchase of non-
group health insurance to the uninsured could disrupt the market for group health
insurance, with potentially adverse consequences for social welfare. All but one of
the tax subsidies evaluated by Gruber resulted in declines in the number of individuals
covered by employment-based health insurance. And some analysts have pointed out
46Gruber, Tax Subsidies for Health Insurance: Evaluating the Costs and Benefits, pp. 29-30.

---

CRS-24
that, in the presence of rising health care costs and the threat of liability for inadequate
health care under managed care plans, some firms would be likely to discontinue or
curtail their employee health benefits and raise their salaries and wages by comparable
amounts, if new tax subsidies for the purchase of health insurance by the uninsured
are enacted.47 If this were to happen, some who are now covered by employment-
based group insurance would end up uninsured or covered by non-group insurance
plans that offer less desirable risk protection and benefits or the same benefits at a
higher price. As was noted earlier, group insurance has important advantages over
non-group insurance: lower premiums because of economies of scale in the
administration of benefits, reduced scope for adverse selection, and less variation in
premiums.
Fifth, in pondering whether to adopt tax subsidies for the purchase of health
insurance by the uninsured, policymakers should be mindful that the subsidies may end
up aggravating some of the economic problems they are intended to solve. Tax
subsidies are equivalent in practice to price subsidies. Generally, a key advantage of
price subsidies is that they preserve consumer choice in selecting the level of goods
or services to be consumed and the firms that supply them. Such choice enables
consumers to seek out goods and services that accommodate their needs, preferences,
and budgets. And if consumer tastes vary considerably, allowing consumer choice can
enhance social welfare.48 However, price subsidies that recognize differences in
consumer taste can end up exacerbating the market failures they are intended to
remedy or ameliorate. Some of the proposed tax subsidies for the purchase of health
insurance considered by the 106th Congress seemed to carry such a risk. To varying
degrees, these proposals would have done little (or nothing) to curb adverse selection
in the private health insurance market, offset inequities in the distribution of existing
tax benefits for health insurance, discourage excessive consumption of health care,
and strengthen the market for group health insurance.
Sixth, despite the research that has been done on tax subsidies to expand health
insurance coverage, it remains uncertain how much of a reduction in the uninsured
population any particular subsidy could stimulate. Tax deductions or credits lower
the after-tax price of health insurance, so their efficacy clearly depends largely on the
sensitivity of the demand for health insurance to price decreases. But considerable
uncertainty surrounds this sensitivity among the uninsured. Some economists have
noted that price subsidies may be less desirable than other kinds of government
intervention in the market for health insurance (e.g., government provision of health
insurance) if the price elasticity of demand is low, the target population differs widely
its responsiveness to changes in the price of health insurance, or it is unknown how
price-sensitive demand for health insurance coverage is among the target population.49
Further research on the price sensitivity of health insurance coverage among the
47Gail S. Keller, “Large Health Insurance Tax Credits Seen Destroying Employer-Provided
Coverage,” Daily Report for Executives (Washington: Bureau of National Affairs, June 9,
2000), p. G-1.
48Poterba, “Government Intervention in the Markets for Education and Health Care: How and
Why,” p. 289.
49Ibid., p. 288.

---

CRS-25
uninsured is needed to clarify the prospects for success of proposed tax subsidies to
expand this coverage.
Finally, making private health insurance more affordable for uninsured
households through tax subsidies may reduce the uninsured population, but there is
reason to think that such a step would be insufficient if the aim of policy is to achieve
universal health insurance coverage. Even with the financial assistance provided by
a generous tax subsidy, low-income individuals with chronic health problems or high
risks of developing such problems may be unable to purchase adequate health
insurance in the private market. In many states, insurance firms are permitted to
refuse to cover such individuals, charge them rates significantly above the rates
offered to healthy or low-risk persons, or deny coverage for certain pre-existing
health conditions. Barriers such as these suggest that measures to require insurers to
offer health insurance to anyone regardless of health status or to restrict the ability of
insurers to discriminate on the basis of health status in pricing their policies may be
needed in addition to price subsidies in order to achieve universal coverage. Gruber
has argued that the complex policy issues raised by efforts to shrink the uninsured
population mean that “tax policy can likely be most useful as one part of an overall
strategy to address uninsurance in the U.S., as opposed to a solution in and of itself.”

---