Revised June 13, 2001

Amendment Would Create New Tax Shelter for Healthy, Affluent Individuals

by Iris J. Lav and Edwin Park


View this report in PDF

To access the PDF files, right-click on the underlined text, click "Save Link As," download to your directory, and open document in Adobe Acrobat Reader.

When the Senate considers the Patients' Bill of Rights, perhaps as early as this week, a floor amendment may well be offered that would make Medical Savings accounts (MSAs) universally available and substantially alter MSA policy in a number of other ways, with the goal of dramatically expanding use of MSAs. The amendment is based on a proposal that conservative activists have promoted for several years and that is included in the Administration's budget.

Medical Savings Accounts are tax-advantaged personal savings accounts that may be used by persons covered by high-deductible health insurance policies.(1) Funds in MSAs may be used to pay for a wide range of health care expenditures. The funds also may be retained in MSA accounts and invested in stocks and bonds (or other investment vehicles), with the investment earnings accumulating free of tax. Eventually, the funds in the accounts may be withdrawn not only for medical purposes but also for non-medical purposes such as retirement.(2)

Supporters of this proposal are expected to argue, as they have in the past, that expanding MSAs will expand health care coverage and reduce the ranks of the uninsured. Most health analysts disagree. Substantial expansion of MSAs would be much more likely to degrade health insurance coverage — and to increase the number of uninsured — than to make progress on these fronts.

These effects differ sharply from the effects that would be expected under legislation introduced by Senators McCain, Kennedy, Edwards and others (S. 284) to extend and modestly enlarge the current MSA demonstration project. S. 284 would improve and expand the demonstration project — and require GAO evaluation of the revised demonstration — to secure better information on the effects of MSAs. It would do without risking destabilization of health insurance markets and sharp increases in the cost of conventional health insurance and also without eliminating the current safeguards in the MSA demonstration project that discourage the use of MSAs as tax shelters. (See box below for more information on S. 284.)

The anticipated Senate floor amendment is likely to have a relatively modest cost in terms of the federal budget. (The Joint Tax Committee has estimated the cost of the Administration's MSA proposal at $5 billion over ten years.) This does not mean, however, that the amendment's consequences would be minor. Instead of the federal government bearing a large cost, much of the "cost" of the policies that the amendment would put in place would effectively be borne by individuals who would pay more in future years for conventional insurance or become uninsured or underinsured. Furthermore, if the amendment leads to more widespread use of MSAs than the current cost estimates anticipate — a distinct possibility if MSAs are aggressively marketed by financial institutions as tax shelters — the cost to the federal budget could substantially exceed the official cost estimates.

Amendment Likely to be Based on Administration Proposal

As noted, the anticipated amendment is likely to be patterned on an MSA expansion proposal that is one of the tax cuts in the Administration's budget that was not included in the recently enacted tax legislation. The Administration's proposal would effectively replace the MSA demonstration project that Congress established in 1996 with a policy making MSAs available to anyone who wants them. Currently, MSA use is limited to people who work for small employers, are self-employed, or are uninsured.(3)

The Administration proposal also would make MSAs much more attractive as a tax shelter to healthy, affluent individuals by removing or weakening many of the safeguards that Congress wrote into the MSA demonstration to prevent that from occurring.

The Administration's proposal also would circumvent rules now in effect under the MSA demonstration project that prevent employers from setting up MSAs in a manner that primarily benefits highly paid executives and effectively discriminates against lower-paid employees.

The Patients' Bill of Rights is supposed to be legislation that makes health care more accessible and responsive to consumers' needs. The anticipated MSA floor amendment moves in the opposite direction. It represents a sharp departure from the current design of MSAs and would be likely to have adverse consequences for health-care consumers. It carries the strong potential to drive up the cost of conventional, comprehensive insurance to such an extent — possibly more than doubling — that many Americans, including those most in need of health services because they are in poorer health, may no longer be able to afford health coverage. Instead of increasing coverage, such an MSA expansion is likely to enlarge the ranks of the uninsured.

Of particular concern are provisions that would significantly increase the appeal of MSAs as tax shelters for higher-income individuals (who tend to be healthier), thereby facilitating their participation in MSAs and further compounding the risk of triggering adverse selection in the health-insurance marketplace. Because of its potential to lead to sharp increases in health insurance costs, the anticipated MSA amendment could, if enacted, end up injuring consumers more than the other provisions of the Patients' Bill of Rights would assist them.


The MSA Demonstration

The bipartisan Health Insurance Portability and Accountability Act of 1996 established a demonstration to test and evaluate Medical Savings Accounts. The demonstration was designed to provide information about the effects of MSAs on workers, employers, and insurers and to do so without creating widespread, irreparable harm to the participants or the insurance market as a whole. Participation in the demonstration was limited to no more than 750,000 participants who are either employees of small businesses (businesses with 50 or fewer employees), self-employed individuals, or the uninsured. Participants could deduct contributions to MSAs in amounts up to certain specified levels. Other rules governing use of MSAs during the demonstration were designed to assure that these tax-advantaged savings accounts were used largely for the purpose of obtaining medical care and would not become a general-purpose tax shelter. The demonstration was originally scheduled to run through 2000 but was extended last year to December 31, 2002.


Senators McCain, Kennedy, Edwards and other sponsors of the Patients' Bill of Rights legislation coming to the Senate floor (S. 283, S. 872) also support legislation (S. 284) that includes modest MSA expansions. Because these provisions do not make MSAs available to anyone who wants one and do not eliminate safeguards that discourage the use of MSAs as tax shelters, S. 284 does not engender the same risks of higher health insurance premiums and extensive tax shelter benefits for higher-income taxpayers as the Administration's proposal does. S. 284 would:

  • Extend the MSA demonstration from 2002 to 2004.
  • Increase the limit on the number of MSA policies from 750,000 to 1 million.
  • Expand the definition of small businesses whose employees can use MSAs from firms employing 50 or fewer workers to firms employing 100 or fewer workers.
  • Require a new GAO study to determine the impact of MSAs on the cost of conventional insurance, on adverse selection, and on health care costs generally.

It should be noted, however, that S. 284 also contains various other tax and spending provisions unrelated to MSAs, some of which may have significant costs and budgetary impacts.

Extending the life of the demonstration in this manner may permit the collection of the data the GAO needs to evaluate MSAs and their impact on conventional insurance and health care costs. To date, lack of greater MSA utilization has prevented GAO from conducting the study mandated in the original demonstration project. The absence of such a study makes moving to universal availability of MSAs with weakened safeguards a particularly dangerous policy that poses serious risks to the availability of affordable health insurance coverage for those most in need of it.

The 1996 legislation required an evaluation by the GAO to determine the effects of MSAs on the insurance market and consumers. Among other issues, the evaluation was to study the extent to which MSAs fostered "adverse selection" — a situation in which younger and healthier individuals find MSAs financially advantageous and choose MSAs while older and less healthy individuals remain in conventional insurance. Such adverse selection would be highly problematic; if younger, healthier individuals (who have below-average medical costs) shift from conventional insurance to MSAs while older, less healthy individuals (who have above-average medical costs) remain in conventional insurance, the cost of conventional insurance necessarily rises, making it harder for employers and employees to afford. The GAO also was charged with studying the effect of MSAs on health care costs, including the cost of health insurance premiums. The intention was that Congress would be able to examine the results of the evaluation and, on the basis of those results, determine future policy regarding MSAs.

Relatively few individuals, however, have chosen to use MSA during the demonstration period. The IRS estimates that in 1999, some 54,000 tax returns reflected MSA contributions.(5) As a result of this low utilization, the GAO has been unable to conduct a full evaluation of the effects of MSAs. Nevertheless, one portion of the GAO evaluation was completed — a survey of insurers, conducted by Westat under contract to the GAO.

MSA proponents attribute the lack of popularity of MSAs during the demonstration period in part to various statutory safeguards included in the demonstration legislation to prevent abuse of MSAs. Another possible interpretation of the sparse usage of MSAs under the demonstration project is that MSAs are not attractive as a health insurance product per se and can gain acceptance only if MSA policies allow substantial abuse of the accounts as tax shelters. The Administration's proposal would make MSAs considerably more attractive as tax shelters.


MSAs and Adverse Selection

As noted, a major concern is that universal access to MSAs would trigger widespread adverse selection in the insurance market. If healthier people choose high-deductible insurance with MSAs, the pool of people covered by conventional, more comprehensive health insurance will tend to be sicker on average than it otherwise would be. If the pool of people who are conventionally insured incurs higher-than-average health-care costs because some of the healthier people are no longer in the pool (having switched to MSAs instead), the premiums for conventional insurance will necessarily increase.

MSAs pose a strong risk of engendering this type of effect. Young, healthy people who anticipate having low health care costs in the near future would likely choose to participate in MSA plans. They would do so because the MSA legislation allows participants to retain unspent health care dollars in their own accounts. Thus, people with low health care costs can accumulate tax-free earnings on those funds and use them as retirement savings or for other purposes.

On the other hand, older and less healthy people who judge they are likely to incur significant health care costs would be better off financially if they remained covered by conventional health insurance, which generally has lower deductible amounts and relatively low caps on out-of-pocket expenditures. As a result, the pool of workers that would retain conventional insurance if MSA use becomes widespread would incur higher average health care costs than the larger pool of workers covered by conventional insurance today. To accommodate those higher average health care costs, the premiums charged for conventional insurance policies would have to increase, perhaps dramatically.

At the higher premium rates, it is likely that significant numbers of employers would be unwilling to offer their employees conventional insurance. In addition, the resulting decline in the market for conventional insurance could lead some insurers to cease selling it. Instead of addressing the needs of the uninsured, an MSA expansion proposal of this nature thus threatens to increase the ranks of the uninsured, especially among the older and sicker individuals who most need health-care coverage.


The Administration's MSA Proposal

The Administration's proposal is likely to serve as the basis for a floor amendment to the Patients' Bill of Rights. The proposal would open up MSAs to use by all individuals and employees working in any size business, allow MSA use on a permanent basis, remove the limit on the number of people who may participate in MSAs, and make other changes in MSAs that would increase their attractiveness as tax shelters.

The Administration's proposal also would increase the maximum amount that can be deposited in an MSA each year. The current demonstration project places strict limitations on such deposits to prevent use of MSAs as general-purpose tax shelters.(7)

If deposits are held until retirement age, for example, there is no penalty for withdrawal for non-medical purposes. Even if funds are withdrawn for non-medical purposes before retirement age, there are a number of circumstances under which the value of the tax-free compounding of the deposits for some years outweighs the penalty on a non-medical withdrawal.

Finally, the Administration's proposal includes changes that would circumvent the rules under the current MSA demonstration that prevent employers from setting up MSAs in a manner that primarily benefits highly paid executives and effectively discriminates against lower-paid employees.



The Administration proposal likely to serve as the basis for a floor amendment to the Patients' Bill of Rights would greatly increase the potential for abuse of MSAs as tax shelters. The proposal would make MSAs more attractive as tax shelters and likely lead to more widespread use of MSAs by healthy, affluent individuals. As a result, the proposal would compound the risk that MSAs will result in health insurance becoming unaffordable or unavailable for many Americans, particularly those most in need of affordable, comprehensive health care coverage. The proposal consequently is likely to increase rather than reduce the number of Americans without health insurance.

End Notes:

1. In 2001, high deductible plans must have deductibles not less than $1,550 and not more than $2,350 for individual coverage and not less than $3,100 and not more than $4,650 for family coverage.

2. A penalty applies to the withdrawal of funds from MSAs for some, but not all, non-medical purposes. (There is no penalty for withdrawal for retirement.) MSA accounts can serve as a tax shelter, however, even in many cases in which a penalty would apply.

3. Technically, the Administration's proposal would make the demonstration project permanent, but this nomenclature is essentially meaningless. MSAs would no longer be a demonstration project in any meaningful sense of the word.

4. Note: If deposits are held until retirement age, they may be used without penalty for any purpose, including non-medical purposes.

5. IRS Announcement 99-95 (October 1, 1999).

6. U.S. General Accounting Office, Medical Savings Accounts: Results From Surveys of Insurers, December 31, 1998, GAO/HEHS-99-34, Appendix, p.14.

7. For individuals, the maximum amount that can be contributed annually under current law is 65 percent of the health insurance policy's deductible amount; for family coverage, it is 75 percent of the deductible amount.

8. Associated Press release by Vivian Marino, August 15, 1997.

9. Margaret O. Kirk, "Medical Accounts: Mixed Reviews," The New York Times, July 5, 1998.

10. U.S. General Accounting Office, Medical Savings Accounts: Results From Surveys of Insurers, December 31, 1998, GAO/HEHS-99-34, Appendix, pp. 15-16.