Statement of Iris Lav, Deputy Director, Center on Budget and Policy Priorities
Testimony Before the House Committee on Ways and Means
Hearing on Health Care Tax Credits to Decrease the Number of Uninsured
February 13, 2002
I appreciate the invitation to testify today. I am Iris Lav, deputy director of the Center on Budget and Policy Priorities. The Center is a nonprofit policy institute here in Washington that specializes both in fiscal policy and in programs and policies affecting low- and moderate-income families. The Center does not hold (and never has received) a grant or contract from any federal agency.
My testimony today largely focuses on the Bush Administration's health insurance coverage initiative in the fiscal year 2003 budget: a proposal to provide a refundable tax credit for the purchase of private health insurance to families and individuals not covered by employer-based coverage. This proposal would cost $89 billion over 10 years and would account for the vast majority of the new resources the Administration is proposing in the health insurance coverage area.
While we welcome the Administration's commitment of significant financial resources to provide assistance to the 39 million Americans without health insurance, we view a tax credit as the wrong approach for solving the problems of the uninsured. The Administration's tax credit proposal suffers from a number of significant flaws, including likelihood that it will materially weaken the employer-based health system through which the large majority of insured Americans currently obtain quality health insurance coverage. While it might be possible in theory to design a tax credit that would better address this and other concerns about the Administration's proposal, it is highly unlikely that such a tax credit would be politically viable at this time because it would require some combination of mandates on employers to offer insurance, states to reform individual insurance markets, and/or individuals to remain in employer-provided insurance.
This testimony also suggests that there is a superior alternative to a tax credit in covering the uninsured: an expansion of Medicaid and the State Children's Health Insurance Program (SCHIP).
In addition, a section of this testimony addresses approaches to helping the unemployed maintain health insurance during the current economic downturn. Subsidizing the purchase of COBRA insurance would help maintain coverage for substantial numbers of unemployed people who otherwise could not afford to pay the pay the COBRA premiums. A tax credit for purchase of insurance in the individual market would be of limited help to low-income or older and sicker unemployed workers. Moreover, if such a credit were extended beyond the unemployed, it would post the same risks as the Administration's plan.
Finally, the appendix to this testimony addresses another health initiative in the Administration budget that, like the health insurance tax credits, threatens to undermine employer-provided insurance. This is the proposal to expand Medical Savings Accounts.
Tax Credit for the Purchase of Health Insurance in the Individual Market
The Administration is proposing to provide a refundable tax credit to individuals and families not participating in employer-based health insurance or public health insurance. (1) Families with two or more children could receive a tax credit of up to $3,000 annually to pay for health insurance primarily in the individual market, so long as the subsidy does not exceed 90 percent of the premium cost. Individuals could receive a credit of $1,000. (The tax credit also could be used for individual health insurance purchased through private purchasing pools or state high-risk pools where such pools exist). The credit would not be available to families with incomes above $60,000, and the subsidy would begin to phase down once a family's income reached $25,000. (Similarly, individuals making $30,000 would not be eligible for the credit, with the subsidy beginning to phase out when an individual's income reached $15,000.)
Under the proposal, the credit could be issued in advance (rather than waiting until a family or individual filed a tax return after the year was over); insurers would reduce the premium cost by the size of a family's credit and be reimbursed by the federal government. States would also have the option of letting certain tax credit recipients purchase coverage in their Medicaid or SCHIP managed care plans (or through their state employees' health plan if no managed care plans are available), but there would be no requirement that states do so.
Likely Weakening of the Employer-Based Health Insurance System
The principal concern with the Administration's tax credit proposal is that the availability of the tax credit could lead some employers to cease providing coverage to their workers and induce new employers not to offer coverage.
Analysts from M.I.T., the Kaiser Family Foundation, and the Urban Institute all have written that enactment of a tax credit of this design (open to individuals currently eligible for employer-based coverage) could encourage firms not to offer health insurance coverage to their employees because firms would know their workers could now get a tax credit to purchase coverage in the individual market. (2) For example, new research by Professor Jonathan Gruber at M.I.T. shows that the Administration's proposal would draw four million people out of employer-provided insurance. Gruber's research shows that 2.4 million people would be dropped from group insurance by their employers -- one million of which will move to nongroup insurance and 1.4 million of which will become uninsured. Overall, the research shows that for every person gaining insurance under this proposal, two persons will be leaving the group insurance market.
Substituting the purchase of health insurance in the individual market for group coverage through an employer is particularly troublesome. It could seriously disadvantage older and less healthy workers, many of whom would not be able to obtain coverage or could obtain coverage only at exorbitant costs. In most states, insurers can vary premiums for health insurance policies offered in the individual market on the basis of age and medical history and can refuse to cover people entirely. If employers that otherwise would offer coverage decline to do so because of the availability of a tax credit of this nature, the consequences could be serious for many older and less healthy workers, who generally would have to pay far more than the tax credit would provide to secure coverage in the individual market. Moreover, the individual market often denies insurance entirely to people with certain health conditions.
Aggravating this problem is the fact that under the Administration's proposal, some workers whose employers do offer coverage and ask their employees to pay a share of the premium could opt out of employer-based coverage and use the tax credits instead to purchase insurance in the individual market. Such a move could be attractive to young, healthy employees. These young and healthy workers could have a double advantage. Because they are a low risk, the policy they could buy in the individual market may be cheaper than the average cost of the employer-provided coverage, especially if they choose more limited coverage. In addition, if the tax credit covers 90 percent of this cheaper coverage, the tax credit subsidy may be larger than the premium subsidy their employer provides. Thus these young and healthy workers could find it financially advantageous to opt out of employer coverage and move into the individual market. Professor Gruber's research indicates that approximately 1.5 million persons would voluntarily switch from their group policies to nongroup policies.
But if these workers -- largely those who would get the lowest cost policies in the individual market -- opt out of employer coverage, the pool of workers remaining in employer plans would become older and sicker on average, which in turn would drive up the costs of employer-based insurance. This phenomenon is known as "adverse selection." Once adverse selection starts and the cost of employer-based insurance begins to rise, additional younger, healthier workers would be induced to abandon employer-based coverage and use their tax credit instead, because they now could personally do better in the individual market using the tax credit.
In this way, a vicious cycle -- sometimes called an insurance death spiral -- could be set in motion. The increase in premiums for employer-based coverage that ultimately could occur could induce many employers either to cease offering health insurance or to increase substantially the amounts their employees must pay for insurance. The end result would likely be that many older and less healthy individuals would eventually lose their employer-based coverage and become uninsured or underinsured or have to pay exorbitant amounts for decent coverage.
Intensifying the risk that many firms might not offer coverage is the recent return of a high rate of inflation in health care costs, which are now rising at double-digit rates in many areas. Institution of the tax credit could provide a rationale for some employers seeking to cut costs to drop or not to institute coverage.
On balance, M.I.T. professor Jonathan Gruber finds that 10.5 million people would take up the credit. Of those, roughly one-third, 3.3 million people, would have been uninsured. But because the credit would cause a lot of churning in employer-provided insurance, some people become newly uninsured. The net reduction in the number of uninsured is only 1.9 million people. (3)
Some tax credit supporters have argued that additional changes to this type of tax credit could lessen the likelihood that the tax credit would weaken the employer-based health insurance system. For example, the credit could be limited only to those persons not currently eligible for employer-based coverage. Changing the design in this way would not, however, eliminate the incentive for employers to drop existing coverage. Employers that know their workers can turn to a tax credit to obtain coverage in the individual market may be more reluctant to offer coverage. In addition, it is difficult to imagine how a credit limited to those not eligible for employer-provided health insurance could be administered. It would require the Internal Revenue Service to determine whether a person is eligible for insurance through their employer or their spouse's employer. Administering such a credit would require massive new reporting requirements for employers and an entirely new -- and arguably inappropriate -- burden for the IRS.
Going in the opposite direction, one could say the credit could be used by employees to pay for their contribution to the cost of health insurance. Such a proposal could help uninsured employees who are offered insurance but cannot currently afford their premiums obtain coverage through their employers. However, these credits could also encourage firms to lower their contributions, and thereby substitute public money for employer contributions. Since a "maintenance of effort" requirement for employers would not be popular and would be impossible to administer, employers -- rather than the employees for whom insurance is not affordable -- could reap the benefit of such a credit.
There is a design of a tax credit that arguably could avoid the type of damage to the employer-based system discussed above. It would include a mandate on employers to offer coverage -- so employers would not be tempted to drop coverage -- and a mandate on individual workers with employer-based coverage to use their tax credit solely for the purchase of insurance through their employer's group -- so healthy employees would not opt out of employer coverage to save money. Such requirements certainly are not politically feasible. Moreover, they would result in the substitution of public funds for employer contributions and thus would constitute a much more expansive, and costly, credit than the Bush Administration has proposed.
Limited Access in the Individual Market
The Administration envisions that most tax credit recipients would primarily use the credit to purchase health insurance in the individual market. However, many of the uninsured face significant barriers to obtaining insurance in the individual market. More than one quarter of all uninsured adults suffer from serious medical conditions such as cancer, heart disease and diabetes and over half (53 percent) have a history of serious medical conditions, smoke, or are obese. (4) One quarter of non-elderly uninsured adults are over 45 and among lower-income uninsured adults above age 50, some 39 percent reported a limited disability and 66 percent had been diagnosed with a chronic condition. (5) All of these are people for whom insurance in the individual market is either expensive or unavailable. By contrast, only a small segment of the uninsured population, 15 percent, are young adults ages 19-34 who do not have children and lack problematic health conditions. (6)
As noted, these sicker and older individuals who constitute such a large percentage of the uninsured likely would be unable to access adequate health insurance in the individual market without paying exorbitant amounts. This is because the individual market is generally unregulated. The individual market generally permits individual medical underwriting, that is insurers can vary premiums based on age and medical history and can deny coverage entirely. For example, according to the Commonwealth Fund, only 16 states require that insurers provide a plan to most applicants -- and that does not necessarily mean an affordable plan. (7)
A recent Kaiser Family Foundation study used hypothetical families and individuals to apply for coverage in the individual health insurance market (the hypothetical applicants were structured to test the medical underwriting process through 60 applications in eight geographic markets). The study found, as expected, that older and sicker people are often unable to obtain coverage in the individual market. (8) This means that under the Administration's proposal, a family containing older or sick members could find itself excluded from coverage or charged premiums that are unaffordable, even with a tax credit. Alternatively, such a family could be offered a plan that is affordable but does not provide coverage for a variety of medical conditions. For example, the Kaiser study used the hypothetical "Crane" family, consisting of two adults and two children. In this family, while one child, "Cindy," is in excellent health, the older brother, "Colin," has asthma. The family received an offer of insurance under each application in every market but 15 percent of the offers excluded coverage of Colin entirely and more than half excluded coverage of Colin's asthma.
In addition, many plans in the individual market do not offer comprehensive coverage required by older and sicker families and individuals. For example, they may require high deductibles of $1,000 or more and may not cover maternity care, preventive benefits, and mental health services. Others may set limits on prescription drug coverage.
The Administration purports to respond to this concern by allowing tax credit recipients to buy coverage through high risk pools as well as other private purchasing pools. However, according to the Commonwealth Fund and other researchers, the success and scope of these mechanisms has been limited. (9) While more than half the states operate high-risk pools, participation is low -- only 105,000 people participated in 1999. Such pools also often impose high premiums, deductibles and other cost-sharing that limit affordability and may provide limited benefits (for example, excluding mental health and maternity care or capping prescription drug costs). Participants may also face a preexisting condition exclusion for some period of time.
The Administration also suggests that that certain low-income individuals would be permitted to use their tax credits to buy into comprehensive public coverage. It is uncertain how many states would elect this option and open their Medicaid and SCHIP managed care plans to tax-credit recipients. Furthermore, the persons most in need of these buy-ins to public coverage are sicker, high-risk individuals who cannot otherwise obtain coverage in the individual market. Adding these individuals to the current Medicaid and SCHIP managed care risk pools (which currently tend to include healthier families and children, rather than the elderly and disabled who cannot be enrolled on a mandatory basis in managed care), could raise Medicaid and SCHIP costs significantly.
Alternatively, reforms to the individual health insurance market could be added to the Administration's tax credit proposal. In other words, the federal government could mandate that states enact certain insurance reforms to ensure access for tax credit recipients including: guaranteed access (insurers have to offer coverage to all applicants), minimum benefits (all plans provide at least some standard comprehensive coverage), and community rating (premiums cannot vary by age, medical history, or both or can vary only within certain limits). While some states have adopted some of these market reforms, it seems politically unlikely that the federal government at this time would require all states to take these steps.
Inadequate Size of the Tax Credit
The tax credit is of inadequate size to make health insurance affordable for many low- and moderate-income families. Health insurance can be expensive. According to the General Accounting Office, the mid-range premium for family insurance in the non-group market exceeded $7,300 in 1998. (10) Nevertheless, even without factoring in the increases in health insurance premium costs since 1998 for the GAO estimate, a family of four with income of $25,000 that receives a $3,000 tax credit would have to spend $4,300 in health insurance premiums ($7,300 minus $3,000). That would constitute more than 17 percent of the family's gross income to purchase insurance at this price. The family would have additional out-of-pocket costs for deductibles and co-pays before they could receive any benefit from having the insurance.
Similarly, the Commonwealth Fund looked at premiums for relatively comprehensive individual health insurance with a $250 deductible for a single healthy adult age 60 in 15 cities. The median annual premium was $5,688. (11) Even with a tax credit of $1,000, a 60 year-old with income of $15,000 would still have to pay over 30 percent of his gross income to obtain insurance. Of course, a less healthy person is likely to pay even more, if they are not excluded entirely from the individual market. Furthermore, in some high-cost geographic areas, higher premiums could consume still-greater percentages of family income. Studies indicate that such expenditure levels are beyond what most low- and moderate-income families can afford.
Some supporters of the Administration's tax credit proposal may argue that family coverage in the individual market is far more affordable than GAO determined. According to a study by an online health insurance broker, the average premium cost was estimated to be between $3,600 and $4,500 for families of three who have actually obtained coverage in the individual market through the broker. (12) That average, however, is likely to have been skewed downward considerably by the better risks associated with younger, healthy individuals who are most able to access insurance in the non-group market as well as the potentially less generous benefits provided. For example, this average is not the average premium offer made; applicants that turned down the offer of health insurance because it was too expensive would not be factored into the average cited by the study. (And it does not indicate the number of applicants who were denied coverage entirely.)
One could argue that the problem could be solved by increasing the size of the credit. Increasing the size of the credit, however, involves a trade-off. While a larger credit could make health insurance in the individual market more affordable for some tax credit recipients, it also would intensify the likelihood that the adverse effects on the employer-based system would occur. This is because a larger credit would make it more attractive for employers to cease offering health insurance coverage, as well as increase the probability that young, healthy individuals would prefer to leave employer-based coverage. (13)
Continued Timing Problems for Advance Payment of the Credit
A number of studies have pointed out that to be effective -- especially for low- and moderate-income families -- the credit must be available at the time the insurance premiums are due rather than at the end of the year when taxes are filed. Low-income families on tight budgets would have difficulty paying health insurance premiums during the year and then waiting until the tax filing season in the following year to be reimbursed through a tax credit. (14) The Administration proposes to address this timing problem by permitting advance payment of the tax credit. Insurers would discount premiums paid by tax credit recipients and be reimbursed for the discount by the federal government. Eligibility for the advance credit would be based on the taxpayer's prior year tax return.
There is a drawback to basing eligibility for an advance credit on the prior year's tax return. The incomes of low- and moderate-income families fluctuate significantly during the course of a year due to changes in family situation, job losses or changes, overtime pay and other variables. Consider a taxpayer with prior year income that is too high to qualify for the credit this year. This year, his work hours are reduced considerably with the result that he no longer qualifies for health insurance coverage through his employer. His income has declined sufficiently to be eligible for the tax credit, but he is not eligible for the advance credit. Since he likely has many obligations to meet with his reduced income, he is unlikely to be able to purchase health insurance and wait for a reimbursement at the end of the year. He probably would remain uninsured.
Alternatively, one could base eligibility for advance payment of the credit on this year's income. This creates a different set of problems. A person could purchase insurance using an advance payment now, but later in the year find that increases in income have made him ineligible for the credit based on his annual income. If a reconciliation process is then required at tax time, this person would owe -- up to $3,000 -- to the IRS. Based on experience with advance payment of the Earned Income Tax Credit, fear of owing money to the IRS at tax time would likely deter most families from using the advance payment option. Without the financial resources necessary to pay for premiums up-front, such families would not benefit from the tax credit.
Lack of Cost Effectiveness
In addition to the significant concerns with the tax credit proposal, as discussed above, the tax credit proposal is not likely to be a cost-effective way to reduce the ranks of the uninsured, since the large majority of those who would use the credit are expected to already have insurance. Analysts from M.I.T. and the Kaiser Family Foundation have estimated that under somewhat similar tax credit proposals, more than two-thirds of those using the tax credit would be people who already were insured. (15)
For example, as noted above, Jonathan Gruber of M.I.T. projects that 10.5 million persons would take up the Administration's tax credit. On its face, this appears to be a large population receiving assistance through the tax credit. However, only 3.3 million people would have been previously uninsured; more than two-thirds of tax recipients would have already had insurance. In addition, under his estimates, employer dropping would cause 1.4 million people formerly with employer-based coverage to be unable to find coverage and become uninsured. The net reduction in the number of uninsured ends up being only 1.9 million (which constitutes only 18 percent of the total number of recipients). (16)
As a result, relatively little of the benefit of the credit would go to reducing the ranks of the uninsured. Instead, a large share of the credit's substantial cost would go either to provide people who already are insured in the individual market with a tax cut or to shift people from their current insurance arrangements (primarily through employer-sponsored coverage) to different insurance arrangements.
A Better Alternative: Expansion of Public Programs
As I have discussed, the tax credit proposed by the Administration could threaten the stability of the employer-based health insurance system through which the overwhelming majority of Americans obtain coverage. In addition, the tax credit would favor young, healthy families and individuals over older, sicker persons who most need comprehensive and affordable health insurance coverage.
An effective alternative that avoids these pitfalls is an expansion of the State Children's Health Insurance Program (SCHIP). For example, bipartisan legislation introduced in this Congress would provide $50 billion in new SCHIP funds to states to expand Medicaid and SCHIP coverage to the low-income parents of children eligible for those programs. In 2000, some 34 percent of parents -- 6.9 million -- in families with incomes below 200 percent of the poverty line ($29,260 for a family of three) were uninsured. This is partly the result of limited coverage within the Medicaid program; the Medicaid income eligibility level in the median state is only 69 percent of the federal poverty line (about $10,000). Just as SCHIP facilitated coverage expansions for low-income children, additional federal SCHIP funds could be provided for states to expand Medicaid and SCHIP coverage to parents. While suffering from a slow start, the SCHIP program is now highly successful. As the Department of Health and Human Services announced last week, a total of 4.6 million children were enrolled in SCHIP last year which constitutes a 38 percent increase from 2000 (3.3 million children were enrolled in fiscal year 2000). (17)
A number of states such as Arizona, California, New Jersey, Rhode Island, and Wisconsin have already used SCHIP to expand coverage to parents, as well as other adults. However, the ability of states to continue to expand coverage, as the Administration has urged states to do, is threatened by a lack of sufficient SCHIP funding. The Balanced Budget Act of1997 instituted a 26 percent reduction in federal SCHIP funding for the fiscal years 2002, 2003 and 2004 -- a reduction of over $1 billion each year. Because of this reduction, in the Administration's budget, OMB projects that SCHIP enrollment will decline by 900,000 people, mostly children, between 2003 and 2006. (18) While the Administration has appropriately proposed to extend SCHIP funds scheduled to expire in the next two years which will help delay or avert some of these enrollment declines, it is likely states will have insufficient funding over the long-term to further expand coverage. As a result, states need additional funding under the SCHIP program. This public expansion proposal would provide a number of advantages:
Insurance for the Unemployed
The House of Representatives late last year passed an economic stimulus proposal that included a tax credit certain unemployed individuals could use to purchase COBRA coverage or health insurance in the individual market. The tax credit would be equal to 60 percent of the cost of health insurance. Only workers currently eligible for unemployment insurance could receive the tax credit. If the tax credit is used in the individual market, the worker must have previously had insurance for 12 months.
Many workers who become unemployed do have difficulty affording the COBRA continuation coverage for which they are eligible. With the costs of a family health insurance policy under COBRA averaging more than $7,000, it is unlikely that many unemployed workers -- particularly those with low or moderate incomes -- could afford to maintain their insurance. A 60 percent tax credit subsidy, however, still leaves the necessity to pay 40 percent of the COBRA premium. A family getting by on unemployment insurance is quite likely to find an average after-credit premium cost of $2,800 (in addition to deductibles and co-pays) unaffordable. So this proposal is most likely to help families that are unemployed, receive unemployment compensation, but also have sufficient resources from savings or another worker in the family that can be used to pay these health insurance premiums.
The aspect of the plan that would make the tax credit available for the purchase of insurance in the individual market is even more problematic. As noted above, the individual market is largely unregulated and lacks the advantages of group insurance purchased through employers. Many plans sold on the individual market impose high deductibles and offer limited coverage. Furthermore, premiums in the individual market can vary based on risk factors such as age and medical history. The new proposal lacks substantive insurance-market reforms to ensure that individual health insurance policies that provide adequate coverage will be made available at affordable prices to unemployed workers in the individual health insurance market. Under the proposal, states would have to guarantee that some form of individual coverage is made available to laid-off workers who previously had employer-based coverage, but similar requirements are part of current law and most states comply with them simply by allowing individuals who otherwise cannot secure coverage to purchase insurance through high-risk pools. As described above, policies sold through high-risk pools generally are unaffordable and the coverage provided is usually limited; as a result, few individuals purchase insurance through these pools.
Only individuals who are receiving Unemployment Insurance payments would be eligible for the tax credit included in the House stimulus bill. This would leave out a substantial number of low-income unemployed workers, including most part-time workers and more recently employed workers, many of whom have recently left welfare for work.
A better proposal would provide a deeper COBRA subsidy -- delivered through a direct payment arrangement with insurers and employers so those without minimal up-front resources could use it -- coupled with an option for states to cover under Medicaid (with a 90 percent federal match) those unemployed persons who do not have COBRA coverage or who cannot afford their share of the COBRA premium. This is similar to the proposals included in the Senate Finance Committee stimulus package. This alternative proposal covers a significantly larger proportion of the unemployed, since it is not limited solely to those covered by Unemployment Insurance.
In addition, a COBRA plus Medicaid option assures that most older, sicker unemployed persons can access the comprehensive insurance they need. For example, a 55 year old construction worker who becomes unemployed may have worked for a company too small to be required to offer COBRA. If he has a history of heart problems or any one of a number of other serious conditions, he probably would not be able to access insurance in the individual market. But he could be eligible for comprehensive insurance through the Medicaid program.
The tax credit approach to helping the uninsured has still one more drawback. If the tax credit in the House stimulus bill is enacted, subsequent efforts almost surely will be made to broaden this credit into a general individual health insurance tax credit -- of the type included in the Administration's budget -- that can be used to purchase insurance in the individual market by employed as well as unemployed individuals. As described at length above, such a general individual health insurance credit could have deleterious effects on insurance coverage.
Conclusion
The tax credit that the Administration proposes poses a threat to the employer-based health insurance system. While this proposal may expand coverage to some currently uninsured Americans, many others who have insurance through their employers may lose their coverage and become uninsured. It is theoretically possible to design a tax credit that does not have as many negative effects, but doing so would require substantial reform of the individual insurance market and some combination of mandates on employers, states, and/or individuals -- all of which is not likely to be politically acceptable. A far better alternative to address the problem of the uninsured would be to provide states additional federal SCHIP funds to expand coverage to parents of children eligible for public programs. That has the side benefit of encouraging enrollment of eligible children who are not currently enrolled in those programs.
For the immediate problem of helping the unemployed maintain health insurance coverage, the House-passed proposal for a tax credit for people receiving unemployment insurance would reach only a limited group of the unemployed and, given the problems with the individual insurance market, would fail to provide many older and sicker workers with the insurance they need. A deeper COBRA subsidy delivered directly through insurers and employers coupled with a largely federally-paid option for states to cover unemployed workers under Medicaid could provide better coverage to more unemployed workers.
APPENDIX
Expansion of Medical Savings Accounts
The Administration's fiscal year 2003 budget proposal includes additional tax proposals related to health care, including a proposed expansion of Medical Savings Accounts. (22) In this appendix, I would likely to briefly address concerns with the MSA proposal, which like the tax credit, is likely to weaken the conventional employer-based health system.
Established under a limited demonstration project scheduled to expire at the end of this year, MSAs are tax-advantaged personal savings accounts available to the self-employed and employees at small businesses who are covered by high-deductible health insurance policies. Funds in MSAs may be used to pay for a wide range of out-of-pocket health care costs. They also may be retained in the MSA accounts and placed in investment vehicles such as stocks and bonds, with the investment earnings accumulating tax-free in the accounts. The funds may be withdrawn for non-medical purposes upon retirement. As a result, MSAs can be used as a tax shelter.
Despite the findings of an array of analyses by respected research institutions that widespread use of MSAs could destabilize the health insurance market (findings the demonstration project has failed to dispel), the Administration is proposing a package of MSA changes that have long been pushed by insurance companies that sell MSA policies and conservative policy institutions. The Administration proposes to repeal most current protections and limitations on MSAs, to make MSAs more lucrative as tax shelters for affluent, healthy individuals (and hence more attractive to such individuals), and to allow unlimited expansion of MSAs across the country. (23) The risks of such a course are great.
Adverse Selection and Its Effect on Conventional Health Insurance
Research by the RAND Corporation, the Urban Institute, and the American Academy of Actuaries has found that premiums for conventional insurance could more than double if MSA use becomes widespread. (24) This is because of the extensive "adverse selection" in the health insurance market that would likely ensue. If MSAs become broadly available, substantial numbers of healthy, affluent individuals may opt for them in lieu of conventional, employer-based group insurance policies. As a result, those remaining in group insurance would be less healthy, on average, and premiums for conventional group insurance would have to increase.
High deductible policies and MSAs are most attractive to younger, healthy individuals, because such individuals do not expect to incur significant health costs and thus can anticipate accumulating significant amounts in their tax-advantaged MSA accounts. MSAs can be particularly attractive to higher-income individuals, a group that also tends to be in better health than people with lesser incomes, since they can benefit handsomely from the tax sheltering advantages of MSAs, which are worth the most to those in higher tax brackets. The attraction of MSAs to healthy, affluent individuals would be significantly enhanced under the Administration's proposal, which would alter the rules governing MSAs in ways that would make the accounts more lucrative as tax shelters. While evidence is very limited, in a preliminary survey of insurers in 1998, the General Accounting Office found evidence that MSAs may indeed be encouraging adverse selection in health insurance markets. (25)
The increase in premium costs that would be expected to result if use of MSAs becomes widespread and significant numbers of healthy individuals withdraw from employer-based plans could lead many employers either to cease offering coverage or to raise the percentage of premium costs that their employees must pay. Such steps would make insurance less affordable and likely cause more people to become uninsured.
Increased Use of MSAs as a Tax Shelter
The Administration's proposal would also make MSAs more attractive as a tax shelter to healthy, affluent individuals by removing or weakening many safeguards Congress enacted to prevent MSAs from turning into a significant tax shelter opportunity.
MSAs are similar to tax-deductible Individual Retirement Accounts in that the deposits an individual makes in these accounts are tax deductible and the earnings that accumulate in the accounts are tax-free. (The funds in the account are never taxed as long as they either remain in the account or are withdrawn for medical purposes; the funds are subject to taxation if withdrawn for non-medical purposes, just as funds in tax-deductible IRAs are subject to taxation when they are withdrawn.) MSAs differ, however, from IRAs in one key respect -- there are no income limits on MSAs that prevent wealthy people from making tax-deductible contributions to them. Indeed, the higher an individual's income, the greater the tax benefit an MSA provides. By opening MSAs for widespread use and eviscerating a number of the current limitations and safeguards on MSAs, the Administration's proposal would essentially enable high-income individuals to circumvent the IRA income limits by using MSAs for the same purpose -- as tax shelters to accrue substantial assets over time on a tax-advantaged basis. At retirement, funds can be withdrawn from MSAs penalty-free for non-medical purposes. In the same GAO survey of insurers, there were indications that MSAs were in fact being marketed primarily as tax-sheltered savings vehicles rather than as sources of tax-sheltered funds for reimbursement of medical expenses. The report stated that insurers were targeting certain segments of the insurance market such as highly-paid professionals, farmers, ranchers, partnership firms and association groups. (26)
1. U.S. Department of Treasury, General Explanations of the Administration's Fiscal Year 2003 Revenue Proposals (February 4, 2002), p. 18-21.
2. Jonathan Gruber, Tax Subsidies for Health Insurance: Evaluating the Cost and Benefits, National Bureau of Economic Research (February 2000); Judith Feder, Cori Uccello, and Ellen O'Brien, The Difference Different Approaches Make: Comparing Proposals to Expand Health Insurance, Kaiser Family Foundation (October 1999); Leonard E. Burman and Amelia Gruber, First Do No Harm: Designing Tax Incentives for Health Insurance, National Tax Journal (May 2001); Linda Blumberg, Health Insurance Tax Credits: Potential for Expanding Coverage, Urban Institute (August 2001).
3. Jonathan Gruber, Written Testimony before the House Ways & Means Committee (February 13, 2002).
4. CBPP analysis of Health Interview Survey, 1997.
5. U.S. Census Bureau, Health Insurance Coverage: 2000 (September 2001); Elisabeth Simantov, Cathy Schoen and Stephanie Bruegman, Market Failure? Individual Insurance Markets for Older Americans, Health Affairs (July/August 2001).
6. CBPP.
7. Lori Achman and Deborah Chollet, Insuring the Uninsurable: An Overview of State High-Risk Health Insurance Pools, Commonwealth Fund (August 2001).
8. Karen Pollitz, Richard Sorian and Kathy Thomas, How Accessible is Individual Health Insurance for Consumers in Less-than-Perfect Health?, Kaiser Family Foundation (June 2001).
9. Achman and Chollet. See also Sally Trude and Paul B. Ginsburg, Tax Credits and Purchasing Pools: Will This Marriage Work?, Center for Studying Health System Change (April 2001).
10. U.S. General Accounting Office, Private Health Insurance: Potential Tax Benefit of a Health Insurance Deduction Proposed in H.R. 2990, GAO/HEHS-00-104R (April 2000).
11. Simantov.
12. eHealthInsurance, The Cost and Benefits of Individual and Family Health Insurance Plans (June 2001). Based on the study's cost-per-member-per-month estimates.
13. Burman.
14. Blumberg.
15. Gruber, Written Testimony; Feder et al.; Memorandum from Joint Committee on Taxation to the Senate Finance Committee (September 13, 2001).
16. Gruber, Written Testimony.
17. Centers for Medicare and Medicaid Services, "The State Children's Health Insurance Program Annual Enrollment Report" (February 6, 2002).
18. Office of Management and Budget, Analytic Perspectives (February 4, 2002), p.297.
19. Lisa Dubay, Expansions in Public Health Insurance and Crowd-Out: What the Evidence Says, Kaiser Family Foundation (October 1999); Kathleen Call et al., "Who Is Still Uninsured in Minnesota? Lessons from State Reform Efforts," Journal of the American Medical Association (October 8, 1997), p.1191-95; Leighton Ku, Marilyn Ellwood et al., "The Evolution of Medicaid Managed Care Systems and Eligibility Expansions," Health Care Financing Review (Winter 2000); Jeremy Alberga, Wisconsin's BadgerCare Program Offers Innovative Approach to Family Coverage, Robert Wood Johnson Foundation (January 2001); Amy Lutzky and Ian Hill, Has the Jury Reached a Verdict? States' Early Experiences with Crowd-Out Under SCHIP, Urban Institute (June 2001); Richard Kronick and Todd Gilmer, Insuring Low-Income Adults: Does Public Coverage Crowd-Out Private?, Health Affairs (January/February 2002).
20. Matthew Broaddus and Leighton Ku, Nearly 95 Percent of Low-Income Uninsured Children Now Are Eligible for Medicaid or SCHIP, Center on Budget and Policy Priorities (December 2000).
21. Leighton Ku and Matthew Broaddus, The Importance of Family-Based Insurance Expansions: New Research Findings about State Health Reforms, Center on Budget and Policy Priorities (September 2000); Jeanne Lambrew, Health Insurance: A Family Affair, Commonwealth Fund (May 2001); Lisa Dubay and Genevieve Kenney, Covering Parents Through Medicaid and SCHIP: Potential Benefits to Low-Income Parents and Children; Kaiser Family Foundation (October 2001); Elizabeth Gifford, Robert Weech-Maldonado and Pamela Farley Short, "Encouraging Preventive Health Services for Young Children: The Effect of Expanding Coverage to Parents," Pennsylvania State University, presentation at the Academy for Health Services Research and Health Policy Conference, Atlanta, June 12, 2001.
22. These proposals are analyzed in the following papers. Edwin Park, Health Insurance Proposals in Administration's Budget Could Weaken the Employer-Based Health Insurance System, Center on Budget and Policy Priorities (February 2002); Edwin Park, Administration's Budget Includes Additional Health Tax Cuts That Primarily Benefit Higher-Income Individuals, Center on Budget and Policy Priorities (February 2002).
23. U.S. Department of Treasury, General Explanations of the Administration's Fiscal Year 2003 Revenue Proposals (February 4, 2002), p. 26-29.
24. Emmett B. Keeler, et al., "Can Medical Savings Accounts for the Nonelderly Reduce Health Care Costs?", Journal of the American Medical Association (June 5, 1996), p.1666-71; Len M. Nichols et al., Tax-Preferred Medical Savings Accounts and Catastrophic Health Insurance Plans: A Numerical Analysis of Winners and Losers, Urban Institute (April 1996); and American Academy of Actuaries, Medical Savings Accounts: Cost Implications and Design Issues (May 1995).
25. U.S. General Accounting Office, Medical Savings Accounts: Results from Surveys of Insurers (December 1998), GAO/HEHS-99-34.
26. GAO.