Jul 2005

HSAs: More Than a (Tax) Shelter, Not Quite a House

Carla Y. Willis, PhD
Virtual Mentor. 2005;7(7):526-531. doi: 10.1001/virtualmentor.2005.7.7.oped3-0507.


Health care economists and policy experts have advanced a variety of ideas for helping the 45 million people who lack health insurance and the millions more who are underinsured. Their proposals range from market-based health savings accounts and income-based tax subsidies to a government single-payer system. This is one of 3 perspectives VM has solicited on health care financing in the US that explore some of these initiatives for expanding coverage.

Health savings accounts (HSAs) are an encouraging development in the market for health insurance, creating a new opportunity for affordable coverage, and engaging patients as allies in managing costs and achieving high value for health care spending. While HSAs have generated a flurry of market activity and media interest since their establishment in 2004, they should not be seen as the centerpiece of a strategy to expand coverage to the uninsured, but rather as complementary to more fundamental reform. This is because HSAs fail to redress the central distortion plaguing the US health care system—the way the federal government currently subsidizes health insurance. Specifically, because people covered through an employer (the majority of insured Americans) effectively pay for coverage with untaxed dollars, those in higher tax brackets receive the biggest tax breaks for their coverage. This mammoth tax subsidy for health insurance—hidden and indefensibly regressive—should be restructured to provide assistance to those who most need help obtaining coverage.

What Are HSAs?

Like their precursors, medical savings accounts, and their close cousins, health reimbursement arrangements, HSAs combine a tax-advantaged savings account earmarked for medical expenses with a high-deductible health insurance plan. Lower insurance premiums offset, in-part or in-full, the amount used to fund the account, which, in turn, can be used to pay expenses before the deductible has been met. The money belongs to the account holder, and unspent balances accumulate and accrue interest, tax-free, from year to year. Should the individual or family meet the high deductible, health plan coverage kicks in, typically in the form of a preferred provider organization (PPO), with little to no cost sharing and limits on total out-of-pocket costs. Because patients are literally spending their own money (most of the time), they have greater control over their health care decisions, while assuming greater financial responsibility for those decisions.1

Overblown claims have been made about the power of HSAs to either rescue the health care system or further erode coverage. On the one hand, HSAs can counter the forces that price health insurance out of reach for growing numbers of people, but only up to a point. On the other hand, preliminary evidence generally does not substantiate fears that HSAs attract only the "healthy and wealthy," or that they lead to widespread skimping on health care. However, the jury is still out, and it is important to take such concerns seriously and continue monitoring the situation.

The Upside

An affordable coverage option. HSAs provide a coverage option to those who prefer to purchase true insurance—protection against the financial consequences of low-probability, high-cost events—rather than coverage that, to a large extent, amounts to prepayment of anticipated and routine health care. High deductibles keep premiums down, bringing coverage within reach for many low-income individuals and families. Indeed, of the more than 1 million people now covered by HSAs, roughly a third were previously uninsured, and 40 percent have annual incomes below $50 000.2,3

Locus of decision-making with patients and physicians. Because people are spending their own money before the deductible has been met, high deductibles shift the locus of cost-consciousness, and thus decision making, from third-party payers to patients and those who care for them. Such cost-consciousness is a welcome antidote to the widening divide between public expectations and rising health care costs and could lessen the need for heavy-handed managed care or government controls. HSAs encourage patients to comparison shop among treatment options, as well as among physicians and hospitals. Note that only a critical mass of savvy shoppers is needed to make health care markets more responsive for everyone.

Safeguards against underutilization. Several features of HSAs protect against inappropriate underutilization of care. First, the savings account itself allows enrollees to set aside money to pay for medical expenses, particularly expenses incurred before the deductible has been met. Second, federal law requires HSA high-deductible health plans to provide a stop-loss limit on patient out-of-pocket expenses, $5100 for individuals and $10 200 for families in 2005. Someone with a chronic or expensive medical condition could find an HSA more affordable than conventional coverage because of the premium savings, out-of-pocket spending limits, and tax advantages (not to mention that they would have greater control over health care decisions).

Preventive coverage. Finally, there is an important exception to the federal requirement that patients with HSAs meet the deductible before insurance coverage is triggered: preventive care. Health plans are permitted, but not required, to cover preventive services, quite broadly defined, even if the deductible has not been met. Such "first-dollar" preventive coverage appears to be the norm for HSA plans. Even without it, having HSA coverage might encourage patients to seek health information that, in turn, prompts greater use of preventive services or adoption of healthy lifestyle habits. HSAs could also increase aversion to future costs of expensive or chronic conditions. In any case, more evidence is needed on the actual impact of HSAs on patient behavior, and on mechanisms through which such impact occurs.

Second-Order Effects

State regulatory response. Perhaps the greatest impact of HSAs will be their indirect and broad effects on health care markets, particularly on the supply side. In order to conform to HSA guidelines, a dozen or so states have recently repealed certain benefit mandates, coverage requirements that contribute to high premiums [4]. The impact of such regulatory change goes far beyond the market for HSAs, creating greater scope for affordable coverage regardless of plan type.

Price transparency. HSAs can also foster systemic change by creating pressure for greater price transparency. The current insurance system, private and public, insulates and, at the same time, bars health care professionals from competing for patients on the basis of price. In the meantime, advocates for the uninsured have launched high-profile assaults on arcane hospital billing practices that charge markedly higher rates to the uninsured than to the insured (whose health plans negotiate rates). A number of states now require hospitals to post price schedules for common services, and several large hospital systems have voluntarily agreed to do so. Thus, the interests of the uninsured and the consumer-driven health care movement are converging to make price information more publicly available.

Other types of information. Of course, informed decision making requires more than just price information. Patients need to be able to compare health care providers and treatment options (again, not everyone need do so for markets to work). HSAs contribute momentum to the development of systems to evaluate and report quality, both clinical and non-clinical aspects. Similarly, HSAs generate demand for tools to help patients make treatment decisions and manage their health.

Incentives for cost-reducing technology. Looking into the future, HSAs create incentives for the innovation of cost-reducing technology. It is no coincidence that the historical growth of third-party payment of health care expenses has been accompanied by many impressive but expensive advances in medicine. In contrast, cost-sensitive patients seek less expensive, better value approaches to diagnosis and treatment. For example, the lack of insurance coverage for laser eye surgery has fueled vigorous competition and falling prices in that market. New technologies and strategies that contain costs and provide value will benefit everyone, not just those with HSAs.

Potential Downsides

Selection and skimping. The potential downsides of HSAs should be neither glossed over nor allowed to overshadow the upside. A special edition of the journal Health Services Research examined early empirical evidence on the characteristics and behavior of enrollees in consumer-driven health plans.5 That body of research, along with more recent HSA-specific data, provides mixed or inconclusive evidence about possible skew in enrollment toward the young, the healthy, and the well-to-do, or about inappropriate skimping on care. Any such problems are certainly not glaring.

Limited impact on aggregate costs. The majority of medical expenses are generated by a small percentage of high-cost patients, and HSAs have little to no direct influence on individual spending once the deductible has been met. These realities limit the ability of HSAs to rein in aggregate health care spending. Nonetheless, as noted earlier, HSAs can serve as a catalyst for cultural change that includes greater awareness of costs. Administrative savings are often invoked as an advantage of HSAs, but this line of reasoning rests on the questionable assumption that patients effectively keep their health care receipts stashed in a shoe box, bypassing costly claims processing unless the deductible is reached. HSAs have also been viewed as a vehicle for personal retirement savings, easing fiscal pressure on Medicare and Social Security. However, even under optimistic assumptions, projected account balances fall miles short of average medical expenses in retirement.6

Limited expansion of coverage. Although HSAs are a step in the right direction toward expanding coverage, it is a small step. The impressive portion of HSA enrollees who previously lacked health insurance represents a drop in the bucket of the 45 million or so uninsured. Rapid expansion of HSAs would have to accelerate even more and draw heavily from the ranks of the uninsured in order to appreciably increase the rate of coverage.

A regressive tax break. For all their promise, it is an inescapable fact that HSAs (like the tax treatment of employment-based coverage) provide bigger tax breaks to those in higher tax brackets. Low-income people may find ample reason to choose HSAs despite receiving little or no tax advantage. But as a matter of public policy on health care, HSA tax breaks work in the wrong direction, lack inherent virtue, and exist merely as an enticement to prudent behavior. On a related matter, some have proposed allowing health insurance premiums for HSA high-deductible plans (or all types of health plans, depending on the proposal) to be paid for with pre-tax dollars, regardless of whether coverage is job-related. This would remove the tax bias favoring employment-based coverage, but further entrench existing inequities across income groups, thwarting efforts at more fundamental reform.

Fundamental Reform to Expand Coverage

Because workers pay no taxes on compensation that takes the form of health benefits, the government subsidizes health insurance to the tune of nearly $200 billion in foregone tax revenues.7 The lion's share of this subsidy goes to those with the highest incomes because, again, the higher the worker's tax bracket, the greater the tax break he or she realizes. Most perversely, those who earn too little to owe income taxes get no tax benefit, and neither do people who are not offered or who decline health benefits.

It would make far more sense for the government to instead give people money to buy health insurance and to base the amount they receive on their income. This is a straightforward way of saying that the tax exclusion for employment-based coverage should be replaced with income-related tax credits or vouchers for the purchase of health insurance. Allowing sliding-scale credits or vouchers to apply regardless of source of coverage would simultaneously level the tax bias favoring job-based coverage and that favoring higher-income households. Credits or vouchers could be used within, or even instead of, the Medicaid system. Of course, they would have to be generous enough to make coverage affordable for people of all income levels and be implemented in a user-friendly way, such as being available when premium payments are made.8

In addition to being a more efficient, fair, and transparent way for the government to support health coverage, tax credits or vouchers would allow individuals and families to choose coverage that suits their needs. Because dissatisfied enrollees could switch health plans (during open enrollment periods), insurers would have to respond to demand for ready access, high quality, and lower costs. People could choose HSAs or plans with varying degrees of managed care, and insurers would innovate new, more affordable coverage options. (The tax treatment of HSAs would have to be modified to prevent "double dipping" from both a tax-free account and a tax credit.)

A critical element of market-based reform is rationalization of the current maze of market regulations, which has inadvertently contributed to the number of uninsured. Fair ground rules would include modified community rating, guaranteed renewability, and subsidization of high-risk individuals from general tax revenues (eg, through high-risk pools or risk-adjustment), and the regulatory environment should enable rather than impede private market innovation. Both free-market mechanisms and government regulations are needed to meet societal goals of collectively financing the medical expenses of people with predictably high costs, while not unduly driving up premiums for the general population.

The swift response by individuals, employers, insurers, and states to the authorization of HSAs is testament to the power of enabling legislation to achieve market-based reform. As with HSAs, the spillover effects of tax credits would be as important as the direct ones, expanding coverage and choice through affordable, high-value insurance options. To be sure, implementing such fundamental change requires overcoming daunting political and budgetary obstacles. A reduction in the number of uninsured, in and of itself, would reduce the burden of uncompensated care borne through taxes and insurance premiums, making the situation somewhat less grim.9,10

More to the point, revenues generated by no longer shielding employee health benefits from income taxes would go a long way toward funding tax credits and vouchers. If revoking the tax exclusion seems untenable, so too do rising levels of uninsured and widespread anxiety about precarious coverage. The worsening situation could actually embolden our political leaders to at least take first steps in the right direction, such as phasing in a cap on the amount of health benefits excluded from taxable income or starting with narrowly targeted tax credits. So, while we keep an eye on how the HSA market evolves, we must also press for more fundamental reform that brings direct relief to those most at peril of being uninsured.


  1. More detailed descriptions of HSAs can be found at and Accessed June 29, 2005.

  2. America's Health Insurance Plans (AHIP) Center for Policy and Research. Number of HSA plans exceeded one million in March 2005. Washington, DC; May 2005. Available at: Accessed June 29, 2005.

  3. eHealthInsurance. Health Savings Accounts: the first year in review, January – December 2004. Available at: Accessed June 29, 2005.

  4. Jenkins, HW, Jr. O Health-Care Leader, where art thou? The Wall Street Journal, June 1, 2005; A-21.

  5. Gauthier AK, Clancy CM, eds. Health Services Research. 2004;39:part II: 1049-1233. Available at: Accessed June 29, 2005.

  6. Fronstin P, Salisbury DL. Health care expenses in retirement and the use of Health Savings Accounts. Employee Benefits Research Institute Issue Brief, No. 271, Washington, DC, EBRI; July 2004. Available at: Accessed on June 29, 2005.

  7. Sheils J, Haught R. The cost of tax-exempt health benefits in 2004. Health Aff. Web Exclusive, February 25, 2004. Available at: Accessed on June 29, 2005.

  8. For detailed descriptions of a tax credit proposal, see Expanding Health Insurance: The AMA Proposal for Reform. Available at: Accessed June 29, 2005; Palmisano DJ, Emmons DW, Wozniak GD. Expanding insurance coverage through tax credits, consumer choice, and market enhancements. JAMA. 2004;291:2237-2242; and Pauly MV. Keeping health insurance tax credits on the table. JAMA. 2004; 291: 2255-2256.

  9. Hadley J, Holahan J. The Kaiser Commission on Medicaid and the Uninsured. The Cost of Care for the Uninsured: What Do We Spend, Who Pays, and What Would Full Coverage Add to Medical Spending?[Issue Update]. May 10, 2004. Available at: Accessed June 29, 2005.

  10. Stoll K, Families USA. Paying a Premium: The Added Cost of Care for the Uninsured. Washington, DC. June 2005. Available at: Accessed June 29, 2005.


Virtual Mentor. 2005;7(7):526-531.



The viewpoints expressed in this article are those of the author(s) and do not necessarily reflect the views and policies of the AMA.