Vous êtes sur la page 1sur 28

No.

569 May 30, 2006


�������

Health Savings Accounts


Do the Critics Have a Point?
by Michael F. Cannon

Executive Summary

Health savings accounts, or HSAs, are a new To address some of those shortcomings,
health insurance option that became available in President Bush proposes to reduce the price dis-
2004. HSAs couple a tax-preferred savings account tortions further, through higher HSA contribu-
(the HSA) with high-deductible health insurance. tion limits and tax credits for individuals who
Enrollees or their employers, or both, make tax-free contribute to their HSAs or who purchase their
contributions to the HSA. Enrollees use the funds own HSA-compatible insurance. Although those
in their HSAs to purchase medical care until they steps would be helpful, HSAs should be expand-
reach their deductibles. At that point, health insur- ed further still to give individuals full ownership
ance begins paying part or all of enrollees’ medical of and control over all their health care dollars.
expenses. Unfortunately, HSAs (and proposals to expand
HSAs reduce government’s influence over them) have become politicized. Critics contend
consumers’ medical decisions by reducing the that HSAs benefit only the healthy and the
price distortions created by the federal tax code. wealthy and that HSAs are ineffective or even
However, HSAs as they exist today do not elimi- harmful. In most cases, criticisms of HSAs fall flat.
nate those distortions. Current HSA law restricts In some cases, the critics do have a point. However,
consumers’ health insurance choices, makes it the failures they identify stem not from HSAs or
difficult for the chronically ill to save for their proposals to expand them but from the problems
future medical needs, and discourages cost shar- that HSAs are meant to correct. Expanding HSAs
ing above the health insurance deductible. would help to correct those problems faster.

_____________________________________________________________________________________________________
Michael F. Cannon is director of health policy studies at the Cato Institute and coauthor of Healthy Competition:
What’s Holding Back Health Care and How to Free It (2005).
HSAs Introduction saving for one’s medical expenses. Thus, HSA
emerged from a deposits largely enjoy the same tax-preferred
Health savings accounts, or HSAs, are a new status as employer-provided health insurance.
recognition that type of health insurance option that first
federal tax laws became available in 2004. As a concept, howev-
distort the er, HSAs have been discussed by policymakers How HSAs Work
and economists for decades.1 Originally called
private-sector “medical savings accounts,” HSAs emerged HSAs combine a high-deductible health
decisions that from a recognition that federal tax laws distort insurance policy with a tax-free savings
the private-sector decisions that people make account (the HSA).3 Enrollees or their employ-
people make about how to pay for health care and how ers, or both, make tax-free contributions to the
about how to pay much health care to consume. HSA. Enrollees can withdraw funds from their
for health care Federal tax law treats employer-provided HSAs tax-free for out-of-pocket medical
health insurance differently than cash wages. expenses. Once an enrollee reaches her health
and how much Whereas cash wages are subject to federal pay- insurance deductible, the insurance coverage
health care to roll taxes (15.3 percent), federal individual begins to pay for part or all of her medical care.
income taxes (which range from 0 percent to People with HSA coverage must have a “qual-
consume. 35 percent), and state individual income taxes ified high-deductible health plan,” or HDHP.
(which range from 0 percent to 10 percent), Essentially, a health plan covering an individual
the amount that employers pay for employee must have a deductible no lower than $1,050
health benefits is not subject to those taxes. (In and a limit on out-of-pocket spending no high-
some cases, neither are employee contribu- er than $5,100. For families, the HDHP must
tions.) That encourages workers have a deductible no lower than $2,100 and a
limit on out-of-pocket spending no higher than
1. to obtain health insurance through an $10,200.4 HDHPs may cover some preventive
employer, because the tax savings lower services below the deductible, but they are not
the relative price of employer-sponsored required to do so.
coverage; HSA holders or their employers, or both,
2. to demand comprehensive third-party can make tax-free contributions to the HSA.
health coverage, because any medical The maximum contribution that individuals
care purchased through their employer can make in 2006 is the amount of their
plan is effectively tax-free; and health insurance deductible or $2,700,
3. to consume more medical care than whichever is lower. Those with family coverage
they otherwise would, because compre- may contribute the lesser of their insurance
hensive third-party coverage insulates deductible or $5,450. HSA holders between
patients from the cost of care at the the ages of 55 and 64 may make additional
point of service. “catch-up” contributions of $700 in 2006.5
Funds from the HSA may be used to cover any
Those responses contribute to the rising cost of medical expense; withdrawals for this purpose
medical care and health insurance.2 Conversely, are not taxed. Unused HSA funds remain in
the tax code penalizes people without employ- the account, earn tax-free interest, roll over
er-sponsored health insurance. from year to year, and follow the account hold-
HSAs were proposed by economists to er wherever she goes.
reduce the distortions of consumer behavior In theory, HSAs will make patients more
and to encourage more sensible decisionmak- prudent consumers of medical care, because
ing about how much medical care to consume consumers are more cost conscious when
and how to pay for it. HSAs partially level the spending their own money. Patients will ask
playing field between employer-sponsored their providers more questions about the costs
health insurance and self-insurance—that is, and benefits of different options, such as

2
brand-name versus generic drugs. One survey ly to exceed their insurance deductibles year
found that people with HSAs or similar cover- after year.
age were 50 percent more likely to ask President Bush has proposed small steps
providers about costs, 33 percent more likely that would address many of those shortcom-
to seek out treatment alternatives, and three ings. He proposes (1) to increase the annual
times more likely to choose a less-expensive limit on HSA contributions from the amount
treatment alternative.6 Greater cost conscious- of an enrollee’s health plan deductible to the
ness on the part of patients should induce health plan’s limit on out-of-pocket costs, (2)
providers to pay greater attention to price and to create tax credits that would eliminate the
quality as well. Just as HSAs are making price distortions between individual and
patients more cost conscious, entrepreneurs employer contributions to an HSA, (3) to cre-
are responding with services that make access ate tax credits that would eliminate the price
to care more convenient and affordable. distortions between individually purchased
Since HSAs became available in January and employer-purchased HDHPs, and (4) to
2004, the number of Americans enrolled in allow people to use their HSA funds to pay the
HSA-compatible health insurance has grown premiums of HSA-compatible health insur-
to three million, though the number who ance. The higher proposed contribution limits
have opened HSAs is smaller and less cer- (up to $5,250 for individuals and $10,500 for
HSAs should be
tain.7 HSA asset growth has outpaced the ini- families) would help more people save for expanded to give
tial growth of IRA assets. Observers have var- their future medical needs. Those proposals individuals full
iously projected that there will be six million would provide tax parity to millions who are
HSAs with $5 billion in assets by 20088 and unfairly punished by the tax code. ownership of and
as many as 25 million HSAs holding $75 bil- Although those steps would be helpful, control over all
lion by 2010. One report projects that HSAs HSAs should be expanded further still. First,
will cover 1 of every 10 individuals by 2010.9 HSA contribution limits should be raised so
their health care
HSAs do not eliminate the price distortions that nearly all individuals could take 100 per- dollars.
that federal tax laws create between health and cent of their health benefits as a tax-free cash
nonhealth expenditures. Nor do they com- deposit into their HSA. For example, the con-
pletely eliminate the price distortions that fed- tribution limits could be raised to $8,000 per
eral tax laws create between different types of individual and $16,000 per family. Second,
health expenditures (i.e., employer-paid insur- consumers with such “large HSAs” should be
ance premiums, individually purchased insur- permitted to use those dollars to purchase
ance, self-insurance, and out-of-pocket medical health insurance tax-free from any source.
spending). People without access to employer- Large HSA holders could purchase coverage
sponsored health insurance still face large tax from their employer, the individual market, or
penalties when purchasing health insurance or elsewhere. Third, Congress should eliminate
contributing to an HSA. the HDHP requirement and allow large HSA
More concretely, HSAs unnecessarily restrict holders to purchase whatever health plan they
consumer choice. Current law requires enrollees wish. Those changes would give individuals
to accept a government-designed health insur- full ownership of and control over all their
ance policy in order to save tax-free for their health care dollars.10 President Bush’s propos-
medical needs. Chronically ill patients, for exam- als take measured steps toward that larger goal.
ple, cannot combine an HSA with a health plan
that covers prescription drugs below the cata-
strophic deductible. Moreover, chronically ill Criticisms That Fall Flat
patients are effectively prevented from accumu-
lating savings in their HSAs because they cannot The president’s HSA expansion proposals
contribute even one penny more than the have put HSAs back on the national agenda.
amount of their deductible, even if they are like- Unfortunately, those proposals have become

3
unnecessarily politicized, drawing fire from for individuals ($1,901) and families ($4,070).15
the president’s critics.11 Though many criti- After two years, the premium savings alone
cisms of HSAs come from the left side of the could more than cover the average annual
political spectrum, it is worth noting that early deductible. Therefore, HSAs would allow
supporters included former senate majority healthy people to save money that they are cur-
leader Tom Daschle (D-SD) and former sena- rently “throwing down the premium hole,”
tor John Breaux (D-LA).12 while still maintaining protection against large
Detractors charge that HSAs benefit only medical bills.
people who are healthy or wealthy, or both. Savings of that magnitude could leave
They also argue that HSAs will fall short of millions of Americans better off. In 2001, 60
proponents’ claims, are unfair to low-income percent of nonelderly individuals with health
workers, and may even be harmful. This sec- insurance spent less than $1,000 on medical
tion explores the least defensible criticisms of care, and 90 percent spent less than $5,000.16
HSAs. Merely by being a good deal for the healthy,
HSAs would benefit the vast majority of the
Health Savings Accounts Don’t Work for population.
Those Who Are Not Healthy Moreover, most individuals could accumu-
Critics most often allege that HSAs cou- late substantial HSA assets over time, because
pled with high-deductible health insurance most of those with high health expenses in a
are a good deal if you are healthy but a bad given year have lower expenses in subsequent
deal if you are sick.13 Supporters typically years. Economists Matthew Eichner of Colum-
respond by citing data showing that HSAs are bia University and David Wise of Harvard
attracting enrollees of all ages; this is meant to University simulated the balances that workers
suggest that because HSAs attract older indi- could accumulate in HSAs. The researchers
viduals, they must be attractive to less-healthy assumed that workers saved $2,000 per year in
individuals. However, many older people use HSAs beginning at age 25 and invested those
very little medical care. One study found that funds in stocks. Even after accounting for life-
the demographic profile of workers who chose time health expenditures, Eichner and Wise
HSA-like plans was similar to that of workers found that by age 60, 90 percent of men would
who chose traditional plans, yet those who accumulate more than $150,000 in their HSAs,
chose the traditional plans nonetheless had and 90 percent of women would accumulate
significantly higher past medical expenses.14 more than $100,000. The authors concluded
that, by funding an HSA, “the typical employee
HSAs and the Healthy would see a substantial increase in assets at
HSAs do enable generally healthy people to retirement.”17
save, tax-free, the money that they would other-
wise spend on unnecessary health coverage. HSAs and the Sick
Data from the Kaiser Family Foundation sug- What about people who are sick? The best
gest that the premium savings that come from evidence available shows that, on average, those
switching to high-deductible coverage can be in ill health fare no worse with high-deductible
considerable. In 2005 the average employer-pro- insurance than they do with other types of cov-
Merely by being a vided health plan cost $4,024 for self-only cov- erage, including “free” health care. Over a 12-
good deal for the erage and $10,880 for family coverage. The aver- year period, the RAND Health Insurance
age annual premium for an HSA-compatible Experiment studied 2,000 families that were
healthy, HSAs HDHP was $2,700 for self-only coverage and randomly assigned to different types of health
would benefit the $7,909 for family coverage. The difference in insurance. Some of those families were given
vast majority of premiums was $1,324 for individuals and “free” health care. Others were assigned cover-
$2,791 for families, which is enough to cover age with cost sharing, including some plans
the population. about 70 percent of the average HSA deductible that resembled HSAs.

4
Health outcomes for people with high partisan Congressional Research Service HSA features
deductibles were overall no worse than for observes: most likely to be
those with any other type of coverage. That is
despite the fact that those with high- Some less healthy people may find HSA unappealing to
deductible insurance consumed far less med- plans attractive because they enable them less-healthy
ical care.18 Those findings were consistent to circumvent the restrictions of man-
regardless of whether an individual was high- aged care plans. Conversely, some healthy
consumers are
or low-income and regardless of whether an people may find them unattractive the current limits
individual was initially in good or ill health.19 because they are very risk-averse; they on tax-free
The investigators wrote, “Our results show would prefer to pay more for comprehen-
that the 40 percent increase in services on the sive insurance with low deductibles. contributions
free-care plan had little or no measurable Older people may find HSA plans attrac- and the lack of
effect on health status for the average adult.” tive because of the tax advantages: being health insurance
In fact, those with high deductibles had fewer in higher tax brackets (since average earn-
restricted activity days.20 As discussed further ings increase with age until people are in choices.
below, there is little evidence to suggest that their 50s), their tax savings from contri-
high-deductible insurance results in worse butions would be greater. People who are
outcomes for those in ill health. The RAND 55 but not yet 65 years of age would also
experiment provides significant evidence to be attracted by the additional catch-up
suggest that it does not. contributions they may make. By the
Nonetheless, some critics allege that HSAs same token, younger people with low
will be unattractive to less-healthy individuals. incomes may consider the HSA tax
Yet many scholars have argued the opposite: advantages inconsequential.24
that HSAs can be attractive to those with high
expected medical expenses. Moreover, expand- Thus the appeal of HSAs cannot be reduced
ing HSAs—by permitting larger contributions to blanket statements that they are good for
and more health insurance choices—would the healthy and bad for those with high
make them even more attractive to those indi- health care costs, or vice versa.
viduals. Finally, people who frequently inter- The HSA features most likely to be unap-
act with the health care system stand to bene- pealing to less-healthy consumers are the
fit the most if HSAs encourage patients to be current limits on tax-free contributions and
more prudent consumers and force providers the lack of health insurance choices. Under
to become more competitive. the current contribution limits, for example,
Some researchers, including critics of the chronically ill likely would use up all their
HSAs, note that HSAs would attract un- HSA deposits in a given year and have little
healthy as well as healthy individuals. Schol- opportunity to save for future medical needs.
ars at the RAND Corporation have projected President Bush’s proposal to increase the
that, “depending on the size of the cata- contribution limits would help some chroni-
strophic [deductible], waste from the exces- cally ill individuals. The president also pro-
sive use of generously insured care could be poses allowing employers to make larger con-
reduced, and [HSAs] would be attractive to tributions to the HSAs of chronically ill
both sick and healthy people.”21 Others have workers than to the HSAs of other workers.
reached similar conclusions.22 Some evidence However, those proposals still would not
even suggests that the chronically ill prefer allow HSA contributions in excess of an
the more flexible access to providers that enrollee’s out-of-pocket exposure. Thus
comes with HSA-like plans.23 some chronically ill enrollees still would have
This suggests that predicting who will no opportunity to build up savings in their
prefer HSAs and who will prefer more tradi- HSAs. As discussed further below, the prohi-
tional health insurance is difficult. The non- bition of any below-the-deductible coverage

5
for treatment of chronic illnesses also makes HSAs Will Lead to Adverse Selection
HSAs unappealing to those patients. Some observers predict that HSAs will
cause premiums to rise for “traditional”
Health Savings Accounts Shift Costs to health plans that offer more comprehensive
Workers third-party coverage. If healthy workers leave
Some critics maintain that HSAs allow traditional insurance products for HSAs,
employers to shift the cost of health benefits then traditional plans will be left with a less-
to workers.25 Rather than provide compre- healthy pool of customers. More formally,
hensive third-party health coverage, employ- HSAs would experience advantageous selection
ers will switch to HSA plans, making workers by healthier workers, while traditional plans
shoulder a greater share of their health care would experience adverse selection by less-
costs. That criticism rests on a misunder- healthy workers. That would cause tradition-
standing of the economics of employee al insurance premiums to rise, which in turn
health benefits. Rather than shift costs to could encourage additional healthy people to
workers, HSAs give workers more control leave traditional plans, causing those premi-
over their compensation. ums to increase further. Thus the rational
Economists generally agree that the cost choices of individual workers could make tra-
Rather than shift of health benefits is already borne by work- ditional coverage increasingly unaffordable.
costs to workers, ers.26 Health benefits, cash wages, and other Some scholars have predicted that premiums
HSAs shift to remuneration are all components of an for traditional plans could rise by as much as
employee’s total compensation. Employees two-thirds.28
workers greater bear the burden of an increase in the cost of Though the evidence thus far is mixed,29 it
control over the health benefits because employers react to is probable that HSAs will lead to some
such increases by reducing other forms of degree of risk selection. The following hypo-
health benefits compensation. That is one reason why real thetical illustrates the effects that HSAs may
portion of their nonwage benefits increased 26.2 percent have on traditional coverage and why con-
compensation. between 1989 and 2004, but real wages rose cerns about risk selection should not stand
just 7.5 percent.27 Workers understand this in the way of encouraging and expanding
intuitively: an employer that does not offer HSAs.
health benefits must offer higher cash wages
to compete for workers. Workers who choose Selection and Subsidies: A Hypothetical
the job with health coverage bear the cost of Suppose an aging professor proposes to a
that benefit in the form of the higher cash student an insurance policy, which we will
wages forgone. HSAs cannot shift to workers call Plan A. The policy would cover 100 per-
costs that workers already bear. cent of their medical expenses, and each
Rather than shift costs to workers, HSAs would pay a premium equal to one-half of
shift to workers greater control over the health their combined medical expenses. From the
benefits portion of their compensation. professor’s point of view, Plan A is a great
Before 2004 Congress generally exempted deal. The student’s premiums would heavily
workers’ health care dollars from taxation subsidize his larger medical expenses. From
only if workers surrendered control over the vantage point of the student, Plan A
those dollars to an employer. In contrast, would be less desirable. Her expected medical
HSA deposits, which the employee owns and costs would be well below the proposed pre-
controls, enjoy the same tax treatment as mium, meaning that she could purchase
employer-provided health benefits. HSAs equivalent coverage at a much lower price.
thereby allow workers to reclaim control over She likely would turn the offer down, but we
a part of their compensation that the govern- will assume she does not.
ment previously encouraged them to surren- When we expand Plan A to 1,000 students
der to their employers. and 1,000 aging professors, all of whom pay

6
the same premium, we see that it would be Initially, the professors would not find
socially undesirable as well. First, because it Plan B very attractive because it would great-
would heavily subsidize the professors’ medical ly reduce the subsidies they received from
consumption, it would encourage the profes- students. (For that reason, they might even
sors to waste resources on low-value health resist Plan B’s introduction.) However, as
care. In fact, the professors could be expected their premiums under Plan A became less
to consume care that provides little or no ben- and less affordable, professors would also
efit, and even care that is harmful.30 Plan A begin to switch to Plan B. This response is
might even induce the students to do the same, socially desirable, because professors (and
because the cost of care would be zero at the students) would pay for medical care below
point of service. Second, increased consump- the deductible directly, rather than through
tion by either group would make Plan A more the financial instrument of insurance.
expensive for the entire pool, even those who Because participants would be spending
did not increase their consumption. Third, their own money, they would be less wasteful
Plan A would discourage socially desirable consumers.
behaviors. Because neither students nor profes- Eventually, premiums for Plan A could
sors could reduce their premiums by quitting rise so much that all professors would move
smoking or monitoring their diabetes, fewer to Plan B. If that happened, Plan B would
would do so. Again, if this were the students’ exhibit the same dynamic as Plan A did: stu-
only health insurance option, many would dents would again subsidize professors,
likely go uninsured rather than pay such exces- because the premiums would be based on the
sive premiums. But we will assume that all stu- average risk of the entire pool. The subsidy
dents participate. would be smaller than in Plan A, because
Now suppose the group alters the contract. Plan B would pay only those expenses above
The group agrees that students and professors the deductible. Still, each student would be
should have the choice of another health plan, paying more than it would cost to insure her-
Plan B. Plan B’s premiums would be deter- self, because she would be less likely than the
mined the same way as Plan A’s: by averaging “average” enrollee to have medical expenses
the medical expenses of all the people in that that exceed the deductible.
pool. However, Plan B would offer no coverage Now consider one final change to the con-
until one’s medical expenses exceeded a high tract. Suppose the group also allows mem-
deductible. The high deductible would ensure bers to choose from Plans C through Z, each
that Plan B’s premiums would be lower than of which has a different deductible and sets
Plan A’s, because fewer enrollees would file premiums according to an individual’s age,
claims and those claims that were filed would gender, behavior, and known medical condi-
be smaller. Enrollees who switched to Plan B tions. Again, we would expect students to
could deposit the premium savings in an migrate to plans that gave them their pre-
account. ferred deductible at the lowest price. That
Plan B would experience advantageous would further reduce the subsidies that pro-
selection by the students, because it would fessors received from students.
narrow the wide gap between the premiums Even though the professors would be likely
they would pay and the benefits they would to resist, these changes would also be socially Concerns about
receive under Plan A. Conversely, Plan A would desirable. All participants would have more risk selection
fall prey to adverse selection by the professors. health insurance choices. Lower premiums
As more students opted for Plan B, Plan A’s would mean that fewer students would go should not stand
pool of enrollees would become professor- without health insurance. Students would have in the way of
heavy, causing that plan’s premiums to rise. As more disposable income to spend on other
a result, additional students would leave Plan items (e.g., tuition and books). Professors
encouraging and
A, and the cycle would repeat itself. would seek to reduce their insurance premiums expanding HSAs.

7
Though HSAs by choosing higher deductibles, which would rather than the (higher) expected costs of
may reduce encourage them to be even more careful con- their coworkers. If so, that too would reduce
sumers. Allowing premiums to vary according subsidies to higher-cost workers.
hidden subsidies to one’s expected medical expenses would cre- The above hypothetical illustrates that the
to sicker workers, ate financial incentives for participants to take risk selection process produces desirable
better care of themselves. If participants can results. First, a level playing field for all types
they do not reduce their insurance premiums by quitting of health insurance, including self-insurance,
preclude smoking, monitoring their blood sugar, lower- respects the right of consumers to choose how
subsidizing ing their blood pressure or cholesterol, and the to plan for their medical needs. Second, it
like, then more participants will do so. makes health insurance more affordable for
those workers Though the professors would lose the many who now face unnecessarily high health
in other ways. cross-subsidies they received under Plan A, insurance premiums. Third, all workers,
those losses would essentially be temporary healthy and unhealthy alike, likely will gravi-
transition costs. The higher health insurance tate toward insurance with higher deductibles.
premiums for today’s professors would convey That will encourage all workers to be more
to today’s students the importance of saving cost conscious medical consumers and should
for their future medical needs. Thus tomor- force health care providers to become more
row’s professors would face greater incentives efficient. Fourth, consumers will take better
to save for their future medical needs. Because care of their health when doing so can save
their current premiums would be lower, they them money on their premiums.
would be better equipped to do so. HSA critics fear that risk selection will
also produce undesirable outcomes. First,
Risk Selection in Job-Based Insurance many less-healthy individuals may be left
Plan A in this hypothetical roughly paying more for their health insurance or
describes how employment-based insurance with less-comprehensive coverage, or both.
pools operate. Some healthy workers opt out However, if adverse selection causes premi-
of the pool. The healthy workers who do ums for traditional health plans to rise dra-
enroll in an employer-based plan often pay matically, that, by definition, indicates that
more (in premiums and reduced wages) than those plans were economically inefficient
they cost to insure. Those extra payments and covered medical expenses that were not
subsidize the medical spending of workers insurable. Second, and of greater concern,
with higher medical expenses. All involved sicker individuals may not be able to afford
face incentives to overconsume medical care the high out-of-pocket expenses for which
and diminished incentives to take care of they will be responsible, and they may not be
themselves. able to afford health insurance at all. Yet
Critics fear that HSAs (and proposals to veiled redistribution under the rubric of
expand them) will reduce the subsidies that health “insurance” is not the way to address
workers with high expected medical expenses that concern. Such redistribution produces
(“professors”) receive from healthier workers the overconsumption and other harmful
(“students”). Within firms, healthier workers incentives that we currently see in the private
may gravitate toward an HSA option (akin to health care sector.
Plan B), reducing subsidies for those who Though HSAs may reduce hidden subsi-
remain in traditional plans. Critics also fear dies to sicker workers, they do not preclude
that HSAs may induce healthier workers to subsidizing those workers in other ways.
leave employer plans and purchase coverage Other options include government subsidies
on the individual market (Plans C–Z). That or private charity, including assistance from
option may be more attractive to some work- family and friends, churches, civic associa-
ers if premiums in the individual market cor- tions, and uncompensated care from hospi-
respond to their expected medical costs, tals and doctors. Employers are also an

8
option. To the extent that cross-subsidiza- pay older workers less than firms that do not
tion currently exists within firms, it may be provide coverage; the lower wages partly offset
understood as an implicit agreement under the added costs those workers impose on the
which employers pay workers with high med- company health plan. That means that many
ical expenses more than those with low med- high-cost workers are already paying more for
ical expenses. There is nothing about HSAs their health coverage—they are just paying in
that prevents people from perpetuating such the form of reduced wages rather than higher
agreements. Indeed, HSAs could be used to premiums. Firms are responding to the rising
facilitate them: President Bush has proposed cost of health insurance by increasing
allowing employers to contribute more to the deductibles, coinsurance, and copayments. As
HSAs of chronically ill workers than to those in the above hypothetical, greater cost sharing
of other workers. However, those agreements reduces the subsidies that “professors” receive
would have to be explicit, and the subsidies from “students.” That strategy reduced the
visible. average annual premium increase for employ-
Explicit subsidies are in fact preferable to ment-based coverage by 2 to 4 percentage
hidden subsidies. Explicit subsidies give firms, points each year from 2002 through 2004, and
workers, and voters an opportunity to weigh somewhat less than that in 2005.36 (At present,
costs and benefits. Hidden subsidies deny the that process is less kind to the “professors”
The fact that
public that opportunity. The fact that subsi- than when HSAs reduce cross-subsidies, subsidies might
dies might be rejected if explicit is no argument because the “professors” must meet the added be rejected if
for preserving hidden subsidies. As Harvard out-of-pocket expenses with after-tax dollars.)
University economist Martin Feldstein argues, Second, research indicates that significant explicit is no
“It is inappropriate in a democracy to use a pooling occurs in the individual health insur- argument for
deliberately opaque system to achieve a redis- ance market. That is because insurers in that
tribution of income that would be rejected if market have long offered products that keep
preserving
proposed in a more transparent way.”31 premiums relatively stable even when the pur- hidden
chaser’s health status deteriorates. Some subsidies.
Much Ado about Nothing? scholars hypothesize that those subsidies will
Nonetheless, the risk selection resulting increase as more workers move from employ-
from HSAs may not be as dramatic as critics er pools to the individual market.37
fear. First, the extent of “pooling” (i.e., cross- Finally, insofar as HSAs enable health care
subsidization) that takes place within firms and health insurance markets to operate
may be less than is commonly believed. A more efficiently, they will reduce the need for
number of factors act to reduce risk pooling such subsidies, and their cost.
within firms. Workers sort themselves into
firms, and into health plans within firms, in HSAs Will Not Expand Health Insurance
ways that reduce risk pooling. Younger, Coverage
healthier workers disproportionately choose Some observers contend that a combination
jobs without health insurance, which limits of risk selection and worker migration toward
the ability of employer plans to pool risk. the individual health insurance market would
Within firms, younger and healthier workers encourage many employers to stop offering
tend to choose less-comprehensive health health insurance. Economist Jonathan Gruber
plans (or decline coverage altogether),32 has estimated that, although President Bush’s
which further reduces pooling.33 HSA proposals would enable some currently
Firms also work to defeat risk pooling. uninsured people to obtain coverage, even
Employers often screen job applicants for more people would lose the coverage they have.
health risks34 and tend to adjust cash wages to On balance, Gruber predicts that President
compensate for easily observable differences in Bush’s proposals would increase the number of
risk.35 That is, firms that offer health insurance Americans without health insurance by

9
600,000, primarily as a result of employers ceas- individual markets. Given that Gruber was
ing to offer health benefits.38 looking at a population of more than 200 mil-
HSA supporters tend to dismiss such lion individuals and found a change of 600,000
claims by pointing to data that seem to show covered lives (or 0.3 percent), it is likely that
that HSAs are extending coverage to the that finding is sensitive to the underlying
uninsured. However, the evidence that sup- assumptions.
porters cite is less than definitive. A survey of It is also important to put those projections
insurers conducted by the insurance lobby, in context. The Census Bureau reports that
America’s Health Insurance Plans, found from 2000 to 2004 the number of Americans
that approximately one-third of consumers with employment-based health insurance
purchasing HSA-compatible insurance in the dropped by 3.7 million from 64 percent to 60
individual and small group markets were pre- percent of all Americans.44 In other words, more
viously uninsured.39 A report by the online than 900,000 Americans lost employment-
insurance broker eHealthInsurance found based coverage on average in each of those four
similar results in the individual market. That years. That is in part because premiums for
report also found that one-half of those with employer-provided insurance increased a total
incomes below $15,000 who purchased HSA- of 73 percent from 2000 to 2005.45 HSAs, and
compatible coverage in the first half of 2005 proposals to expand them, are efforts to arrest
were previously uninsured.40 those trends by encouraging more cost con-
However, as Columbia University’s Sherry scious behavior.
Glied notes, a significant share of new pur- However, Gruber acknowledges that his
chasers of any type of health insurance on the model did not include “any effect [that HSAs
individual market will be previously unin- might have] on behavior in either the non-
sured.41 Thus those data do not reveal whether group or employer markets.”46 That is, his
HSAs are any more attractive to the uninsured model accounts for the effects of the presi-
than other types of coverage. A survey by the dent’s proposals on workers’ decisions to buy
Blue Cross Blue Shield Association did com- health insurance, but not for the effects of
pare HSAs with other types of coverage. It those proposals on workers’ decisions to pur-
found that 12 percent of those purchasing chase health care. The latter are crucial to
HSA-eligible insurance were previously unin- understanding the overall effects of expand-
sured, compared to 6 percent for traditional ed HSAs. The president’s proposals would
coverage.42 While those results are promising, change the relative prices of different types of
further study is in order. insurance. However, they would have an even
Gruber’s projection that the president’s more dramatic effect on the prices that
HSAs will have proposals would reduce the number of patients pay for medical care, by encouraging
Americans with health insurance by 600,000 is patients to pay directly for a larger share of
complicated plausible, but so are many other outcomes. their medical care. Medical consumption in
effects on Gruber himself notes that, like all such econo- turn has a large impact on the cost of health
health insurance metric projections, his are “fraught with uncer- insurance. Yet Gruber assumes no secondary
tainty.” His model required him “to make or tertiary effects of consumers facing higher
coverage that important assumptions as to the attractiveness out-of-pocket prices, such as more cost con-
cannot be known of HSAs relative to traditional non-group sciousness or downward pressure on prices.
in advance, insurance options, and as to the ultimate pre- HSA supporters expect that HSAs will
miums that will result in a much-broadened make medical care and coverage more afford-
particularly if high-deductible plan marketplace.”43 As noted able primarily as a result of the behavioral
one assumes away earlier, projections about the appeal of HSAs to changes they will induce on the part of both
particular individuals are not simple. Nor is patients and providers. If supporters are cor-
changes they are projecting the relative prices of HDHPs and rect, those unmeasured effects could easily
likely to induce. traditional plans in both the job-based and overwhelm the effects that Gruber does mea-

10
sure. Moreover, any benefits of downward pres- War II, workers have been able to prepay their The HSA concept
sure on prices would be enjoyed by those with medical expenses tax-free through compre- was vetted and
and without HSAs. In Gruber’s defense, the hensive third-party insurance. In contrast,
effects that HSAs will have on medical con- saving for one’s medical needs could be done endorsed by
sumption and provider behavior are even more only with after-tax dollars, which diverted prominent
difficult to estimate than the effects on cover- resources from banks to insurers and
age levels. HSAs will have complicated effects arguably contributed to America’s low level
economists,
on health insurance coverage that cannot be of national saving. HSAs stand to benefit the including Nobel
known in advance, particularly if one assumes financial services industry—and harm the laureates.
away changes they are likely to induce. insurance industry—by granting self-insur-
ance the same tax advantages as third-party
HSAs Are Special Interest Legislation insurance. Yet in the decades-long debate
Sen. Richard Durbin (D-IL) has likened over HSAs, the financial services industry was
HSA expansion proposals to special interest inexplicably absent.
legislation. On a television news program, he
referred to HSAs as
Criticisms That Raise
the brainchild of the Golden Rule Serious Issues
Insurance Company back in the Ging-
rich era. . . . That’s why they are being pro- Some questions that critics pose about
posed. This is not good policy. We can the effects of HSAs do raise serious issues.
have private sector health insurance, but However, those criticisms are less an indict-
we shouldn’t be buying on to these ment of HSAs than of the problems that
schemes that haven’t been well vetted HSAs are designed to correct. Those prob-
and well thought out.47 lems include a tax code that unfairly discrim-
inates against workers and reduces their
In fact, the HSA concept was developed in the health insurance choices and a health care
1970s by Jesse Hixson, an economist with the sector that is unresponsive to patients. The
American Medical Association, and Paul way to address those underlying problems,
Worthington, a fellow at the Social Security however, is not to repeal or limit HSAs but to
Administration.48 Over the subsequent 30 expand them.
years, that concept was vetted and endorsed by
prominent economists, including Nobel lau- Will HSAs Encourage People to Skimp
reates.49 The concept drew significant support on Necessary Care?
from the results of the RAND Health Insur- Some observers fear that, because HSAs
ance Experiment, the most rigorous scientific are coupled with high-deductible insurance,
experiment ever conducted on the effects of they will discourage people from obtaining
different health insurance designs. That aca- needed medical care. Avoiding or delaying
demic pedigree fairly well establishes that needed care today could result in more costly
HSAs and proposals to expand them are not episodes of care down the road, making
designed to benefit the health insurance HDHPs a blunt cost containment tool.50 The
industry, or any particular industry. preponderance of evidence suggests that
Moreover, if any industry stands to gain high-deductible health insurance does not
from making HSA contributions tax-free, it induce people to skimp on care in ways that
is not the insurance industry but the finan- harm their health. However, it is nonetheless
cial services industry. Until recently, the tax possible that high-deductible insurance may
treatment of employer-provided health bene- do so in some circumstances. Moreover, the
fits put banks at a competitive disadvantage existing HDHP requirement restricts con-
relative to insurance carriers. Since World sumers’ ability to choose their own health

11
plans. That requirement should be removed deductible insurance masked harmful reduc-
or relaxed to give consumers and insurers the tions in use among certain subgroups. Some
right to experiment with coverage designs people argue that the RAND experiment pro-
that reduce overall costs. vides evidence of such harmful skimping.
One survey reports that people with HSAs The RAND investigators did find one area
or similar coverage “were significantly more where the cost-sharing plans were associated
likely to avoid, skip, or delay health care with a worse health outcome: the cost-shar-
because of costs than were those with more ing plans appeared to do a worse job of con-
comprehensive health insurance, with prob- trolling blood pressure than “free” care. That
lems particularly pronounced among those resulted in a slightly higher mortality risk for
with health problems or incomes under low-income subjects with hypertension who
$50,000.”51 However, surveys tell us nothing were enrolled in the cost-sharing plans.53
about the health effects of reducing medical However, that finding fails as a criticism
expenditures. of HSAs. First, it is possible that that effect
The RAND experiment found that high- was the result of chance.54 Second, even if it
deductible health insurance and other forms was not the result of chance, the comparison
of cost sharing reduced health care expendi- is inapt. Only patients with access to “free”
Surveys tell us tures, but those reductions in use resulted in care experienced a reduced mortality risk.
nothing about the no adverse health effects overall. However, Most privately insured individuals today do
health effects of that is not to say that high-deductible insur- face some degree of cost sharing. Thus, the
ance never induced harmful reductions in relevant comparison would be between high-
reducing medical use. Rather, it suggests that any harmful deductible plans and plans with lower levels
expenditures. reductions were offset by beneficial reductions of cost sharing. Using that comparison, the
in use. The RAND investigators found that RAND investigators found no health effects:
compared to “free” coverage, greater cost “In comparing results within the group of
sharing induced people to reduce their use of cost-sharing plans, we could detect no differ-
appropriate and inappropriate care in equal ences among the plans.”55 Third, the
proportions: improved mortality risk was only among the
poor, who already have access to “free” care
We interpret the proportionate changes through Medicaid. Fourth, the RAND inves-
in appropriate and inappropriate care tigators note that “the effect of having a
to mean that the benefits of the addi- screening exam . . . is almost as large as the
tional appropriate care were offset by effect of free care; this result suggests that a
the additional inappropriate care re- screening exam might be an effective alterna-
ceived by those with full coverage. In tive.”56 Current HSA rules allow insurers to
other words, the increased inappropri- provide free blood pressure screening. The
ate care was not just zero-benefit care; it RAND Health Insurance Experiment pro-
actually had negative effects.52 vides no evidence that HSAs will lead to
harmful skimping on medical care.
In other words, “free” coverage encour- Since the RAND experiment, at least one
aged some patients to obtain medical care study has suggested that greater cost sharing
that was on balance harmful. Examples for prescription drugs could result in greater
include adverse side effects from unnecessary emergency room use by and hospitalization
prescriptions, harmful drug interactions, and of the chronically ill.57 Although these find-
other treatment-induced injuries. ings are far from definitive, it is entirely pos-
Therefore, even though there were no sible that insurers could reduce the incidence
adverse health effects from high-deductible of expensive episodes of care by offering full
insurance overall, it is possible that the bene- or partial coverage of certain expenses, begin-
ficial reductions in use induced by high- ning at a low (or zero) deductible. Drugs that

12
treat chronic illnesses would be a prime can- philosophy underlying HSAs to use them to
didate. restrict consumers’ health insurance choices.
To reduce health care costs overall, public Second, allowing HSA-compatible insur-
policy should allow insurers and consumers to ance to create financial incentives for cost-
experiment with coverage designs that encour- effective care below the deductible would
age cost-effective care while discouraging make HSAs an even more valuable tool for
inappropriate care. In general, that requires promoting appropriate care and discourag-
generating information on effective medical ing inappropriate care. Just as Congress has
care, making that information available to no business dictating to consumers what
patients, and giving patients incentives to use type of health insurance they should pur-
it. HSAs are an indispensable part of those chase, it has no competence to design cost-
efforts. HSAs encourage patients to ask more effective health coverage.60
questions—and demand answers—about the Finally, the HDHP requirement is the
cost-effectiveness of treatment options. They largest single factor inhibiting HSA growth;
provide strong incentives for patients to forgo removing it would encourage HSA growth.
unneeded medical care. At the same time, they Most Americans are not accustomed to high-
provide strong incentives to obtain appropri- deductible health insurance. In 2005 more
ate medical care, because HSA holders will than 75 percent of workers with self-only PPO
want to protect their HSA deposits from even coverage had deductibles below $500, while 10
bigger medical expenses later on. percent of all covered workers faced separate
Yet the rigid HDHP requirement prevents prescription drug deductibles averaging just
insurance companies from providing addition- $122.61 Deductibles upward of $1,050 are
al incentives for patients to obtain cost-effec- intimidating to consumers accustomed to low
tive care. For example, that rule prevents any out-of-pocket exposure. Removing or relaxing
coverage below the deductible for drugs that this requirement would allow more individu-
treat an existing injury, illness, or condition— als to open HSAs with their existing coverage,
even if covering those drugs would save the which would create more HSA holders.
plan and the patient money in the long run. Allowing coverage below current HSA
HSA-compatible insurance can cover preven- deductibles would reduce patients’ incentives
tive care below the deductible for conditions to be cost conscious. That is why removing
that are asymptomatic or from which the the insurance requirement should be cou-
patient has recovered. Indeed, HSA plans avail- pled with reforms that give consumers own-
able to federal workers often provide better ership of the dollars used to purchase their
coverage of preventive care than traditional health coverage and that limit the overall Public policy
plans.58 But HSA-compatible insurance cannot amount of medical expenditures that indi- should allow
provide any drug coverage below the de- viduals may deduct from taxes. Large HSAs
ductible for diabetics, asthmatics, or patients are one option.62 If workers own the dollars insurers and
with multiple sclerosis. used to purchase their health coverage, there consumers to
Congress should remove or relax the is no need for Congress to specify what type
HDHP requirement currently tied to HSA eli- of insurance consumers purchase.
experiment with
gibility. First, doing so would give consumers coverage designs
a greater choice of health insurance designs. Why Do HSAs Allow Wealthy Individuals that encourage
Current rules require some HSA holders to to Shelter Funds?
carry higher health insurance deductibles A number of researchers have debunked the cost-effective
than they would prefer and even require other common assumption that HSA enrollment care while
HSA holders to purchase lower deductibles would skew heavily toward the wealthy.63 Dana discouraging
than they would prefer.59 HSAs are intended Goldman and Jesse Malkin of the RAND
to give consumers more control over their Corporation posit that “the prototypical HSA inappropriate
health care dollars; it is inconsistent with the user will only be slightly wealthier than those in care.

13
Furman’s conventional insurance, and HSAs would be accumulate tax free, and withdrawals
interpretation attractive to the seriously ill in high tax brack- also are tax free as long as they were [sic]
ets.” Just what are these high tax brackets? “This used for medical expenses. No other sav-
falters because he group includes employees earning under the ings vehicle in the U.S. tax code offers the
compares HSAs $94,200 Social Security [payroll tax] ceiling— potential for both tax-free contributions
certainly a well-compensated group but hardly and tax-free withdrawals. Second, under
to the wrong tax the CEOs of America.”64 An earlier study of the President’s proposals, HSA contribu-
deductions. HSAs also estimated, “The median [HSA] user tions would qualify for a 15.3 percent tax
would be only slightly richer than people in credit, a substantial advantage as com-
conventional FFS and HMO plans, which pared to contributions to 401(k)s or
already contain substantial tax breaks.”65 IRAs, for which no such tax credit is pro-
Still, many observers object that the tax ben- vided. Third, unlike IRAs, HSAs have no
efits of HSAs do skew heavily to the wealthy. income limits. The exceedingly generous
Some argue that this is because HSAs receive HSA tax breaks thus are available to very
more advantageous tax treatment than other high-income families.66
tax-preferred savings accounts. Jason Furman
of New York University and the Center on Furman’s description of HSAs is accurate.
Budget and Policy Priorities writes: However, his interpretation of those facts fal-
ters because he compares HSAs to the wrong
First, HSA contributions are tax de- tax deductions. The relevant comparison is
ductible, amounts earned on the ac- not other tax-preferred savings vehicles but
counts (which can be invested in stocks, the tax exclusion for employer-provided
bonds, real estate, or other investments) health insurance premiums.

Figure 1
Average Federal Tax Savings from Health-Related Deductions by Family Income Level, 2004

$3,000 $2,780
$2,640
$2,500
$2,134
$2,000
Savings

$1,448
$1,500
$1,231

$1,000
$725

$500 $292
$102
$0
k k k k k k k k+
< $1 0 $ 10 k– $2 0 $ 20 k– $3 0 $ 30 k– $4 0 $ 40 k– $5 0 $ 50 k– $7 5 75 k– $1 00 $ 10 0
$
Income

Source: John Sheils and Randall Haught, “The Cost of Tax-Exempt Health Benefits in 2004,” Health Affairs Web
Exclusive, February 25, 2004, pp. W4-106–W4-112.

14
Figure 2
Distribution of Federal Tax Savings from Health-Related Deductions: Percent of Total Tax
Savings by Family Income Level, 2004
$100k+
26%

Below $40k
19%

$75k–$100k
22%

$40k–$50k
10%

$50k–$75k
23%
Source: John Sheils and Randall Haught, “The Cost of Tax-Exempt Health Benefits in 2004,” Health Affairs Web
Exclusive, February 25, 2004, pp. W4-106–W4-112.

The tax treatment of HSAs was designed to limited by income because the tax breaks for
provide tax parity between employer-spon- employer-sponsored insurance are not limited
sored health benefits and financing medical by income.
care through personal savings. The reason Though Furman criticizes the president’s
why (1) contributions to an HSA and (2) dis- HSAs proposals for heavily skewing the tax
bursements from an HSA for medical expens- savings toward high-income households,
es are untaxed is so that they will mirror the Figures 1 and 2 show that the distribution of
tax treatment of (1) employer-sponsored tax savings from the exclusion of employer-
health insurance premiums and (2) claims provided health premiums and other health-
paid by employer-sponsored health insur- related deductions already skews heavily
If HSAs dispro-
ance—both of which are untaxed. The effect is toward wealthier families. This is true in portionately
to level the playing field between employer- terms of both the amount that wealthy fami- benefit the
sponsored third-party insurance and personal lies save and their share of the total tax bene-
savings. The president proposes to make non- fits.68 If HSAs disproportionately benefit the wealthy, it is
employer HSA contributions “qualify for a wealthy, it is because they are intended to off- because they are
15.3 percent tax credit” because employer con- set the distortions of an even larger tax loop- intended to offset
tributions to an HSA are exempt from the 15.3 hole that already does so.
percent federal payroll tax.67 In other words, All of these skewed tax benefits are the prod- the distortions
the proposed credit would level the playing uct of deductions that reduce one’s taxable of an even larger
field between HSA contributions made by income under a system in which the wealthy
employers and HSA contributions made by pay higher marginal tax rates. If your marginal
tax loophole that
individuals. Finally, HSA enrollment is not tax rate is 40 percent and your neighbor’s is 30 already does so.

15
Figure 3
Percentage Increase in Insurance Premiums over Previous Year, by Year and Type of Plan

POS HMO PPO Traditional/Indemnity HSA/HRA

10
9 8.5
8
8 7.2 7.3 7.4 7.5
7 6.4 6.6

6
Percent

5
4
2.8 2.6
3
2
1
0
2005 2006

Source: Deloitte Consulting, “Reducing Corporate Health Care Costs 2006 Survey,” January 2006, p. 3, http://www.
deloitte.com/dtt/cda/doc/content/us_chs_red_cor_hea_costs_0106.pdf.

percent, and if you each reduce your taxable income workers (like everyone else) have vary-
income by $100, then your tax savings will be ing preferences for risk and must evaluate HSA-
$40 and your neighbor’s will be $30. That compatible coverage relative to the alternatives.
inequity stems, not from the size of the deduc- Nonetheless, HSAs are less attractive to low-
tion, but from the combination of deductions income workers than they could be, because the
and rising marginal tax rates. High marginal tax code continues to discriminate against
income tax rates and health-related deductions those workers. Expanding HSAs would make
both result in economic inefficiency. Ideally, them a better deal for low-income workers.
Congress should discard all health-related A number of features make HSA coverage
deductions and reduce marginal income tax potentially attractive to low-income workers.
rates. HSAs are merely a second-best option.69 Premiums for high-deductible insurance are
typically much lower than premiums for tradi-
Are HSAs a Bad Deal for People with tional coverage. That alone makes HSA-com-
Ideally, Congress Low Incomes? patible coverage more affordable. In addition,
should discard all HSAs offer the biggest tax breaks to people those premiums have been increasing more
with the highest incomes and require enrollees slowly than premiums for traditional plans. In
health-related to carry insurance with significant out-of-pock- 2005 and 2006 premiums for HSA-compatible
deductions and et exposure. Thus some critics contend that and similar insurance grew at roughly one-third
reduce marginal HSAs are a bad option for those with low the rate of premiums for traditional plans
incomes. In fact, HSAs can be attractive to (Figure 3). In fact, eHealthInsurance reports
income tax rates. many low-income households. Whether high- that in 2005 premiums for HSA-eligible poli-
HSAs are merely deductible insurance is the best option for a cies decreased by 15 percent from the previous
low-income household, however, is a decision year.70 Those results could stem from healthier-
a second-best that should be made only by those individuals, than-average people enrolling in such plans.
option. not by their employers or Congress. Low- Nonetheless, premiums for HDHPs rise less

16
rapidly for all enrollees, if only because any rate were exempt from payroll taxes.72 The only Expanding HSAs
of increase is applied to a smaller base. way low-income workers can purchase health would make them
As noted earlier, healthy low-income insurance with payroll-tax-exempt earnings is
workers might be drawn to HSAs because if an employer purchases it for them. Yet low- a better deal for
they would no longer be required to purchase income workers are the least likely to be cov- low-income
more coverage than they need or to pay more ered by job-based insurance. That means that
than necessary to cover themselves. One those workers must purchase health insur-
workers.
study by HSA critics concluded, “Wealthier ance and contribute to their HSAs with after-
workers may be more willing to bear the tax dollars.
financial risks of increased out-of-pocket lia- President Bush has proposed remedying
bilities, but our results show that on average, that problem through tax credits that reim-
lower wage workers would benefit from burse workers the payroll taxes they pay on
switching to [HSA]/catastrophic plans.”71 earnings that are contributed to an HSA or
Though many health policy analysts would that go toward the purchase of an HDHP on
prefer to see such workers obtain comprehen- the individual market. That would be a posi-
sive third-party insurance, low-income work- tive step, but the president’s proposal would
ers pay for those benefits through either high- still require those workers to purchase govern-
er premiums or reduced wages. Thus, to claim ment-prescribed health insurance. Allowing
that high deductibles are bad for those work- workers to take all their health benefits as a
ers is to claim that low-income workers would cash contribution to a large HSA would give
be better off with less take-home pay (and low-income workers full tax parity and full
comprehensive insurance) than with more control over their health insurance decisions.
take-home pay (and catastrophic insurance).
HSA critics should neither presume to know Are Patients Unable to Shop Around for
the risk preferences of low-income workers Health Care?
nor dictate their own preferences to those Critics claim that HSAs unreasonably
workers. Ultimately, only low-income workers expect patients to shop around for care, even
can decide for themselves what type of cover- though there are many instances in which
age best meets their needs. patients cannot. Many episodes of care are
However, the critics are correct that HSAs emergencies, and useful price and quality
fail low-income workers. First, current HSA information is scarce even for nonemergency
rules do not let workers choose the health care.73 Yet patients can shop around for many
insurance they prefer. Just as HSA opponents health care needs. In those situations, HSAs
should not dictate to low-income households correct the perverse incentives that have sup-
their preference for comprehensive third-party pressed useful price and quality information.
coverage, HSA supporters should not dictate Indeed, markets are already responding to
their preference for high-deductible insurance. consumer demand for such information.
Low-income workers should have the right to HSAs also allow patients to benefit from the
control all of their health care dollars and not negotiating clout of insurers in areas where
be penalized if they purchase more (or less) patients cannot negotiate themselves. Though
health insurance than HSA supporters (or critics are correct that price and quality infor-
critics) think they should. mation is scarce, that is not a shortcoming of
Second, HSAs discriminate against low- HSAs. Indeed, it is a problem that HSAs are
income workers because current law fails to intended to correct.
provide tax parity between employment- Most health care spending occurs in cir-
based coverage and health insurance pur- cumstances under which the patient can com-
chased on the individual market. One study parison shop. For example, emergency room
found that low-income workers would be very care accounts for only 3.3 percent of health
attracted to HSAs, but only if contributions expenditures.74 Hospital and nursing home

17
care combined account for 45 percent of per- that prohibit their disclosure.80 Meanwhile,
sonal health care expenditures,75 yet many various legal barriers, including federal anti-
hospital expenditures are discretionary.76 fraud laws and Medicare’s administrative
Spending on physicians, prescription drugs, pricing system, make it nearly impossible for
home health care, and other services accounts cash-paying patients to negotiate their own
for 55 percent of personal health care expendi- discounts.81 As a result, hospitals often
tures.77 Those data suggest that a large share charge HSA holders and the uninsured more
of health care spending does allow time for than other payers.
considering one’s options. HSAs and other forms of consumer-direct-
However, a lack of useful information on ed care are already correcting this situation.
prices and quality frustrates patients’ efforts to First, many carriers have begun to provide use-
shop around. For example, the Government ful price information to their patient-cus-
Accountability Office reports that insurers tomers. Knowing that their HSA enrollees
offering HSA-compatible health plans to feder- need more price information, insurers such as
al workers provided spotty price and quality UnitedHealthcare, Aetna, and Humana have
information to their enrollees. The plans pro- begun disclosing their negotiated rates to
vided some information on average prices enrollees.82 They and other carriers are devel-
Far from being charged by physicians and pharmacies but oping tools to help customers make informed
a shortcoming of never the actual prices that enrollees would pay decisions.
HSAs, the lack (with the exception of prices at mail-order phar- Many providers have begun posting their
macies). The plans provided scant information prices as well, some of them combining trans-
of information on the quality of hospitals and even less on the parent price information with new modes of
on health care quality of physicians, in large part because such delivery. Firms such as MinuteClinic and
quality information does not exist.78 RediClinic are opening clinics in retail stores.
prices and quality Far from being a shortcoming of HSAs, the Those clinics offer convenient access to basic
is among the lack of information on health care prices and care, with posted prices that are usually less
problems that quality is among the problems that HSAs are than the cost of a doctor’s visit. A service called
designed to correct. American patients are so TelaDoc offers members quick telephone
HSAs are heavily insured that providers—especially hos- access to a doctor (usually within 40 minutes)
designed to pitals—have little incentive to provide patients for $35 per consultation and a $4 monthly
correct. with meaningful price and quality informa- membership fee. A company called Doctokr
tion. Some 83 cents of every dollar spent on provides a similar service. There is also a small
health care comes from someone other than but growing movement of “cash-friendly”
the patient, usually insurers, employers, or physicians who openly advertise their prices.
government.79 Insulating patients from the Services such as CashDoctor.com post those
cost of their medical decisions reduces their prices online for patients. The online service
incentive to demand price and quality infor- HealthGrades provides average prices for
mation, which reduces the likelihood that numerous hospital services for a nominal fee.
providers will supply it. Online services such as Subimo.com and
In fact, overreliance on third-party pay- WebMD.com are competing to provide cus-
ment actually penalizes cash-paying patients, tomers with user-friendly information on hos-
such as the uninsured and HSA holders. pital quality.
Rather than engage in open price competi- Second, HSAs provide layers of protection
tion, private third-party purchasers negotiate from excessive prices even when patients can-
exclusive discounts from largely meaningless not shop or negotiate directly with providers.
“list prices” set by providers. The discounts Carriers of HSA-compatible insurance typi-
are available only to the particular employer cally extend the discounts they negotiate
or health insurer and are typically hidden with providers to their HSA enrollees even
from public view by contractual provisions when those enrollees are paying out of pock-

18
et. One survey found that more than 90 per- price information, there will be no need to
cent of consumers enrolled in HSA-compati- force providers to supply it: providers will pub-
ble plans have a preferred provider network lish it, or perish. But as long as patients have no
and that those plans generally give patients interest in price information, forcing providers
access to the discounted rates that the insur- to supply it would be a useless exercise and a
er negotiates with providers.83 Thus HSAs costly distraction.
lower prices for cash-paying patients even
before they come through the emergency Will HSAs Affect Only a Portion of
room doors. Health Care Spending?
Finally, by making more patients cost Critics claim that HSAs will do little to
conscious, HSAs will encourage other efforts reduce overall health spending. They rightly
to hold down prices. Civic groups have argue that, although consumers would be
proven capable of negotiating discounts on much more cost conscious when consuming
hospital charges for cash-paying patients.84 medical care below the deductible, once they
Princeton economist Uwe Reinhardt writes, reached the deductible consumers would
“Partly under pressure from consumers and have little incentive to be cost conscious.
lawmakers and partly on their own volition, Because the overwhelming majority of med-
many hospitals now have means-tested dis- ical spending comes from a small share of
counts off their chargemasters for uninsured patients who would exceed the deductibles,
patients, which bring the prices charged the detractors question whether HSAs would
uninsured closer to those paid by commer- have much of an impact on health spending
cial insurers or even below.”85 Such efforts at all. One study estimated that allowing
become more likely as the number of cash- workers to choose HSAs would reduce overall
paying patients increases. health spending only modestly.87 Similar
If HSAs are permitted to grow and make results have been found in the spending pat-
more patients cost conscious, a critical mass terns of people with HSA-like coverage.88
of price sensitive patients will push more Despite that legitimate concern, HSAs as
insurers and providers to furnish transpar- they exist today have the potential to inject
ent, competitive prices. However, that may cost consciousness into a significant share of
not be sufficient. Congress and state legisla- medical spending. If every nonelderly, fully
tures should also eliminate regulatory and insured individual in 2001 had had health
legal barriers to price transparency and com- insurance with a $1,000 deductible, 28 percent
petition. In particular, Medicare’s adminis- of that group’s total medical spending would
trative pricing systems, which effectively set have fallen below that deductible.89 Thus, if
price floors for private payers, will have to be adopted widely, HSAs could bring cost con-
abandoned. sciousness to about one-third of the medical
Unfortunately, the Bush administration spending of that group. There is also evidence
has proposed forcing doctors and hospitals to to suggest that patients with HSAs and similar
post their prices.86 That would be unwise, plans do remain cost conscious even when HSAs as they
because doing so would guarantee that the way cost sharing disappears. One survey found
prices are presented will serve the purposes of that such patients were more likely to seek
exist today have
regulators, rather than the needs of patients. alternative treatments even after they had the potential to
The Bush administration has announced it reached their out-of-pocket maximums.90 inject cost
will publish, in a patient-friendly manner, the Moreover, some HDHPs also have cost
prices that federal programs such as Medicare sharing above the deductible, which injects consciousness
pay doctors and hospitals. That step probably cost consciousness into an even larger share of into a significant
will not hurt, and it could even help. However, overall medical expenditures. The current
the administration’s efforts to force price dis- limit on total cost sharing in a self-only HDHP
share of medical
closure should stop there. If patients demand is $5,100. Suppose that in 2001 all nonelderly, spending.

19
Current HSA fully insured individuals had had an HSA plan dent’s proposed HSA expansion, is small. Over
contribution with a $1,050 deductible and 20 percent coin- the next five years, HSAs are projected to gen-
surance above the deductible. Even though an erate a revenue loss equal to 2 percent of the
limits effectively individual’s total cost sharing would have revenue loss created by the tax exclusion for
penalize those been limited to $5,100, the total amount of employer-provided health insurance. Even
medical spending subject to cost sharing with President Bush’s proposals in effect, the
who do not obtain would have been $21,300. That would have revenue loss over that period would be just 7
100 percent introduced price sensitivity into more than 60 percent of that larger figure (Figure 4).
coverage above percent of medical expenditures by all The financial condition of the federal
nonelderly fully insured individuals.91 Medicare program is also relevant. Current
the deductible. However, the current HSA contribution revenue sources cannot finance all the benefits
limits discourage such coinsurance. Annual that Medicare has promised to current and
HSA contribution limits are tied to the future retirees. As of 2006 Medicare carries an
amount of the HDHP deductible. Therefore, unfunded liability of $70.8 trillion. In other
consumers cannot cover such coinsurance words, if the federal government were to
with HSA contributions in a given year. As a deposit funds in an interest-bearing account
result, current HSA contribution limits effec- to cover all of Medicare’s future deficits, it
tively penalize those who do not obtain 100 would have to deposit $70.8 trillion.93 As a
percent coverage above the deductible. point of reference, that is about five times the
Raising HSA contribution limits would size of the U.S. gross domestic product. Filling
help correct that flaw. The president has pro- this shortfall would require an immediate and
posed allowing employers and individuals to permanent tax increase equal to 25 percent of
make tax-free HSA contributions up to their wages, which is so large as to be politically
HDHP’s limit on out-of-pocket costs, or as infeasible.94
high as $5,100 for those with self-only cover- In short, cuts in Medicare benefits are
age and $10,200 for those with family cover- inevitable. Given that inevitability, Congress
age. Doing so would encourage cost sharing should be doing everything it can to encour-
above the deductible and bring cost con- age current workers to save for their health
sciousness to even more medical decisions. care needs in retirement. The quickest way to
An even better solution would be to raise do that is to make HSAs more appealing by
the contribution limits high enough so that allowing them to be combined with any type
nearly all workers could take the full value of of insurance and to increase HSA contribu-
their health benefits as a cash contribution to tion limits. The revenue loss to the govern-
their HSAs. Then workers would be cost con- ment would be a fraction of the amount that
scious about every penny they spent on med- Americans would have available in their
ical care and health insurance premiums. HSAs to pay for their present and future
medical needs.
Would Expanding HSAs Increase the
Deficit? Are HSAs Unpopular?
Critics correctly note that expanding Critics have pointed to survey research
HSAs would lead to a revenue loss for the fed- suggesting that HSAs are thus far unpopular
eral government. The president’s proposals with consumers. Supporters counter that the
would lead to an estimated revenue loss of number of people covered by HSA-compati-
$47 billion from 2007 through 2011.92 While ble insurance tripled from one million to
important, that revenue loss must be under- more than three million between March
stood in context. 2005 and January 2006. That is a less than
Compared to revenue losses under other perfect measure of HSA growth or populari-
provisions in the tax code, the revenue loss ty. First, that survey counted only people
from current HSA law, or even from the presi- with HSA-compatible insurance, not those

20
Figure 4
Projected Federal Revenue Losses from Bush HSA Proposals and Other Items, 2007–2011
$1,000
$889
$900

$800
$700
Billions of Dollars

$600
$471
$500
$400

$300 $243 $233


$200
$200

$100 $64

$0
Employer Mortgage Charitable 401(k)s Child Tax HSAs with
Health Interest Contributions Credit Proposed
Benefits Deductions

Source: U.S. Office of Management and Budget, Analytical Perspectives, Budget of the United States Government,
Fiscal Year 2007, February 6, 2006, Table 17-3, “Effect of Proposals on Receipts,” pp. 265, 296, http://www.white
house.gov/omb/budget/fy2007/pdf/spec.pdf.

who have actually coupled that insurance stemmed from a lack of price information
with an HSA. Second, it is unclear how many with which to comparison shop.97 Other sur-
of those three million “covered lives” chose veys report different findings. A survey by the
HDHPs themselves and how many had an Blue Cross Blue Shield Association found
HDHP chosen for them. Those who pur- that individuals with HSA-compatible insur-
chased HSA-compatible insurance for them- ance were consistently more satisfied with
selves in the individual market appear to their coverage than those in traditional
account for fewer than half of the three mil- plans.98
lion covered lives. It is certain that some por- There are good reasons not to draw any
tion of the remainder had HSAs thrust upon firm conclusions based on current survey
them by their employers.95 research. First, the available surveys do not
Overall, consumer satisfaction surveys directly measure satisfaction among HSA None of
have so far produced mixed results. A survey holders. So far, there are relatively few HSA the available
by the Employee Benefits Research Institute enrollees. Consumers with HSAs are there-
and the Commonwealth Fund found that fore typically lumped in with those who are surveys
individuals with comprehensive third-party qualified to open an HSA but have not done measures
coverage were significantly more satisfied so, or others with similar consumer-directed consumer
than were those with HSAs and similar high- plans. As a result, none of the surveys mea-
deductible plans.96 Likewise, a survey by sures consumer satisfaction with HSAs satisfaction with
McKinsey & Company found that fewer than alone, or at their full potential. Second, some HSAs alone,
half of consumers were as happy with HSAs dissatisfaction inevitably stems from unfa-
as they had been with their previous tradi- miliarity. Consumers with HSAs and similar
or at their full
tional plans. Part of their dissatisfaction coverage have had much less experience with potential.

21
HSAs will have those options, because those options have regulatory efforts. The quickest and surest way
to reach a critical been in existence for only a few years.99 This to build that critical mass and a political con-
source of dissatisfaction can be expected to stituency for HSAs would be to allow them to
mass in the dissipate over time. be coupled with any type of health insurance.
marketplace Finally, HSAs may be unpopular for rea- Some degree of complexity is inevitable so
sons that should not sway policymakers. long as HSAs enjoy special tax status. To
before they can be HSAs are intended to be disruptive, which obtain tax breaks, enrollees must contend with
expected to effect will inevitably cause dissatisfaction in some nettlesome contribution limits and account
a systemic change corners. We should expect that consumers for all their HSA withdrawals. Until all health-
who lose the cross-subsidies they used to related tax breaks can be eliminated (as sug-
like widespread receive (i.e., the “professors” in the above gested above), such complexity is inevitable. Yet
transparent price example) will be dissatisfied. To wit, the EBRI the complications added by the rigid HDHP
competition. survey found the difference in out-of-pocket requirement are not inevitable. To open an
expenses was the source of the greatest gap in HSA, millions of Americans would have to give
satisfaction between people in HSA-like up their current health insurance. HSA sup-
plans and traditional plans.100 We should porters can and should make HSAs simpler by
also expect that as HSAs give consumers removing that requirement.
greater control over their health care dollars, None of the consumer satisfaction sur-
uncompetitive providers will express their veys tells us what we need to know most: the
discontent. One analysis predicted that the types of insurance and medical care con-
growth of HSAs would cause a “major dis- sumers would choose if they controlled all
ruption in the health care industry.”101 their health care dollars and all their health
Yet these sources of dissatisfaction are not care decisions. To find those answers requires
an indictment of HSAs. HSAs are designed to expanding HSAs and removing all restric-
eliminate inefficiencies and hidden cross- tions on HSA holders’ insurance choices.
subsidies. If that causes some dissatisfaction,
it means that HSAs are achieving their pur-
pose, not that they should be abandoned. If Conclusion
we stop robbing Peter to pay Paul, Paul’s dis-
satisfaction should not persuade us to Health savings accounts are an important
change course. step toward restoring market forces and
Nonetheless, HSA supporters should be reducing government involvement in the
very concerned about the frustration HSA health care sector. Proponents expect that
holders feel with (1) the lack of information to HSAs will make medical markets more effi-
help them be cost conscious consumers and cient by encouraging patients to be more par-
(2) the complex rules and restrictions that simonious consumers. There is some evi-
come with HSAs.102 HSAs will have to reach a dence that this is happening.
critical mass in the marketplace before they Many of the criticisms that have been
can be expected to effect a systemic change like lodged against HSAs are unfounded.
widespread transparent price competition. However, not all criticisms can be easily dis-
Currently, only about 1 percent of privately missed. HSAs are not a panacea, and the rules
insured individuals has HSA-compatible cov- governing them make them less appealing
erage. Some observers project a large boost in than they could be, particularly to sick or
HSA enrollment in 2006.103 But as of February low-income individuals. Those rules should
2006, 45 percent of American adults had never be relaxed with an eye to letting individual
heard of HSAs, and only 29 percent even consumers control all of their health care
claimed to know what an HSA is.104 Moreover, dollars and decisions.
without a broad base of political support, Whatever the strengths and weaknesses of
HSAs could be undermined by legislative or HSAs, they are now a reality and appear to be

22
growing. Nonetheless, it is important to build and Consumer-Driven Health Plans: Findings
from the EBRI/Commonwealth Fund Consumer-
a constituency for preserving and expanding ism in Health Care Survey,” Employee Benefits
HSAs in order to forestall legislative or regula- Research Institute Issue Brief no. 288, December
tory efforts to undermine them. The most 2005, p. 1, http://www.ebri.org/pdf/briefspdf/EB
important thing Congress could do to encour- RI_IB_12-2005.pdf.
age HSA growth would be to allow HSAs to be 7. Jeff Lemieux, “January 2006 Census Shows 3.2
coupled with any type of health insurance. Million People Covered by HSA Plans,” America’s
That would allow millions of Americans to Health Insurance Plans Center for Policy and
open HSAs without giving up the coverage Research, March 9, 2006, p. 2, http://www.ahip.
org/content/default.aspx?docid=15302.
they now have. Further, Congress should
allow workers to take the full amount of their 8. Katy Henrickson, Bradford J. Holmes, and Eric
health benefits as a cash deposit in their HSAs, G. Brown, “Will Health Plans Profit from HSAs?
and then allow workers to use those funds to Forecast of Health Savings Account Assets and
Transactions,” Forrester Research, April 7, 2005,
purchase health insurance of their choosing http://www.forrester.com/Research/Document/
from any carrier they wish. Those steps would Excerpt/0,7211,36469,00.html.
give workers ownership of every one of their
health care dollars, solidify support for HSAs, 9. Aamer Baig, Jeff Nuckols, and Amy Dawson,
“Seizing the HSA Opportunity,” DiamondCluster
and truly put consumers in the driver’s seat. International Insight, August 2005, pp. 3–4, http:
//www.diamondcluster.com/Ideas/Viewpoint/P
DF/HSA%20DiamondCluster%20Single.pdf.
Notes 10. See Cannon and Tanner, pp. 65–73; and
1. See, for example, John C. Goodman and Gerald Michael F. Cannon, “Combining Tax Reform and
Musgrave, Patient Power: Solving America’s Health Health Care Reform with Large HSAs,” Cato
Care Crisis (Washington: Cato Institute, 1992); and Institute Tax & Budget Bulletin no. 23, May 2005,
Victoria Craig Bunce, “Medical Savings Accounts: http://www.cato.org/pubs/tbb/tbb-0505-23.pdf.
Progress and Problems under HIPAA,” Cato
Institute Policy Analysis no. 411, August 8, 2001, 11. See, e.g., Elana Schor, “Dems Try to Do to
http://www.cato.org/pubs/pas/pa411.pdf. HSAs What They Did to Social Security Plan,”
Hill, February 8, 2006, http://www.hillnews.com/
2. For a fuller discussion of the problems created thehill/export/TheHill/News/Frontpage/020806
by excessive health coverage, see Michael F. Cannon /hsa.html.
and Michael D. Tanner, Healthy Competition: What’s
Holding Back Health Care and How to Free It 12. See, e.g., John Breaux et al., dear colleague let-
(Washington: Cato Institute, 2005). ter re “Medical Cost Containment Act (S. 2873),”
September 8, 1992, author’s files.
3. This introduction to HSAs is not meant to be
comprehensive. The U.S. Department of the 13. See, e.g., Harry Reid et al., “Democrats Call on
Treasury has a highly useful guide to HSAs on its Bush to Offer Real Healthcare Reform in State of
website at http://www.treas.gov/offices/public- the Union,” news release, January 27, 2006, p. 3,
affairs/hsa/faq.shtml. http://help.senate.gov/Min_press/2006_01_27.pdf.
4. Figures are for 2006 and are indexed for infla- 14. “A number of CDHP [consumer-directed
tion annually. health plan] sponsors have used demographic data
to suggest that these plans have not dispropor-
5. The amount allowed for catch-up contribu- tionately attracted low-risk members. . . . We found
tions rises by $100 annually until it freezes at that although demographic data did not reveal
$1,000 in 2009. favorable risk selection in the [CDHP], better mea-
sures of risk based on prior use and prior cost
6. Vishal Agrawal et al., “Consumer-Directed unanimously indicated risk segmentation taking
Health Plan Report—Early Evidence Is Promising: place.” Laura Tollen, Murray N. Ross, and Stephen
Insights from Primary Consumer Research,” Poor, “Risk Segmentation Related to the Offering
McKinsey & Company, June 2005, p. 4, http://mck of a Consumer-Directed Health Plan: A Case Study
insey.com/clientservice/payorprovider/Health_Pla of Humana Inc.,” Health Services Research 39, no. 4,
n_Report.pdf. See also Paul Fronstin and Sara R. Part II (August 2004): 1167–87, http://www.
Collins, “Early Experience with High-Deductible cmwf.org/usr_doc/hsr_39_4p2_1167_Tollen.pdf.

23
15. Gary Claxton et al., Employer Health Benefits: Part II (August 2004): 1107, http://www.cmwf.org/
2005 Annual Survey (Washington: Kaiser Family usr_doc/hsr_39_4p2_1091_Parente.pdf.
Foundation/Health Research and Educational
Trust, 2005), pp. 96–97, http://www.kff.org/insur 24. Bob Lyke, Chris Peterson, and Neela Ranade,
ance/7315/upload/7315.pdf. The figures reported “Health Savings Accounts,” Congressional Re-
are average premiums and are only suggestive of search Service Report for Congress no. RL32467,
the actual premium savings that HDHPs offer. updated March 23, 2005, pp. 24–25, http://www.
belmont.edu/hr/CRSReportforCongress.pdf.
16. Phil Ellis, “Do Health Accounts Promote
Better Spending?” slide presentation to the 25. See, e.g., Reid et al., p. 3.
American Enterprise Institute, June 29, 2005,
http://www.aei.org/docLib/20050628_Ellis.pdf. 26. See Lawrence H. Summers, “Some Simple
Economics of Mandated Benefits,” American
17. Health expenditures were based on the actual Economic Review 79, no. 2, Papers and Proceedings
claims experience of 600,000 workers at a Fortune of the Hundred and First Annual Meeting of the
500 company. Matthew J. Eichner and David A. Wise, American Economic Association (May 1989):
“Little Saving and Too Much Medical Insurance: 177–83; Jonathan Gruber, “Health Insurance and
Medical Savings Accounts Could Help,” in Personal the Labor Market,” in Handbook of Health Economics,
Saving, Personal Choice, ed. David A. Wise (Stanford: vol. 1A, ed. Anthony J. Cuyler and Joseph P.
Hoover Institution Press, 1999), pp. 86–92. Newhouse (Amsterdam: Elsevier, 2000): 690–95;
Katherine Baicker and Amitabh Chandra, “The
18. Joseph P. Newhouse et al., Free for All? Lessons Labor Market Effects of Rising Health Insurance
from the RAND Health Insurance Experiment Premiums,” NBER Working Paper no. 11160,
(Cambridge, MA: Harvard University Press, 1996), February 2005, p. i, http://www.nber.org/papers/
p. 197. w11160; Dana Goldman, Neeraj Sood, and Arleen
Leibowitz, “Wage and Benefit Changes in Re-
19. Ibid., p. 210. sponse to Rising Health Insurance Costs,’’ NBER
Working Paper no. 11063, January 2005, p. 1,
20. Ibid., p. 243. http://www.nber.org/papers/w11063; and Mark
Pauly and Bradley Herring, Pooling Health Insurance
21. Emmett B. Keeler et al., “Can Medical Savings Risks (Washington: AEI Press, 1999), pp. 69–70.
Accounts for the Nonelderly Reduce Health Care
Costs?” Journal of the American Medical Association 27. Bureau of Labor Statistics, Office of
275, no. 21 (June 5, 1996): 1666. Compensation Levels and Trends, “Employment
Cost Index: Constant Dollar Historical Listing,”
22. See Sherry A. Glied and Dahlia K. Remler, “The April 29, 2005, tables 2a, 5a, 8a, ftp://ftp.bls.gov
Effect of Health Savings Accounts on Health /pub/suppl/eci.ecconst.txt.
Insurance Coverage,” Commonwealth Fund Issue
Brief no. 811, April 2005, pp. 5–6, http://www. 28. Marilyn Moon, Len M. Nichols, and Susan Wall,
cmwf.org/usr_doc/811_Glied_effect_hlt_sav “Medical Savings Accounts: A Policy Analysis,”
ings_accounts_coverage.pdf; and Dwight McNeill, Urban Institute, March 1996, p. 14, http://www.
“Do Consumer-Directed Health Benefits Favor the urban.org/uploadedPDF/msa.pdf; and Kenneth E.
Young and Healthy?” Health Affairs 23, no. 1 Thorpe, “Medical Savings Accounts: Design and
(January–February 2004): 186–93, http://content. Policy Issues,” Health Affairs 14, no. 3 (Fall 1995): 258,
healthaffairs.org/cgi/reprint/23/1/186.pdf. This http://content.healthaffairs.org/cgi/reprint/14/3/2
study simulated the distributional effects of health 54.pdf, citing American Academy of Actuaries,
reimbursement arrangements (HRAs), which are “Medical Savings Accounts: Cost Implications and
similar to HSAs. HRAs combine (usually high- Design Issues,” Public Policy Monograph 1, May
deductible) insurance with a line of credit for out- 1995, p. 9.
of-pocket medical expenses. Unused credit rolls
over from year to year but is forfeited when an 29. Stephen T. Parente et al., “Assessing the
employee changes jobs. HRAs do not give workers Impact of Health Savings Accounts on Insurance
ownership of those funds as HSAs do. Nonethe- and Coverage Costs,” unpublished manuscript,
less, scholars have studied HRAs for insights into December 2005, p. 10.
the likely effects of HSAs. This paper refers to
HRAs as “HSA-like” plans, or “similar coverage.” 30. This is not conjecture. See, e.g., Newhouse et
al., p. 356.
23. Stephen T. Parente, Roger Feldman, and Jon B.
Christianson, “Employee Choice of Consumer- 31. Martin Feldstein, “Rethinking Social Insurance,”
Driven Health Insurance in a Multiplan, Multi- Presidential Address for the American Economic
product Setting,” Health Services Research 39, no. 4, Association, January 2005, American Economic Review

24
95, no 1 (March 2005): 7, http://www.nber.org/feld President’s Health Insurance Budget Proposals,” p. 2.
stein/aeajan8.pdf.
44. Carmen DeNavas-Walt, Bernadette D. Proctor,
32. Bowen Garrett, “Employer-Sponsored Health and Cheryl Hill Lee, “Income, Poverty, and Health
Insurance Coverage: Sponsorship, Eligibility, and Insurance Coverage in the United States: 2004,”
Participation Patterns in 2001,” Kaiser Commission U.S. Census Bureau Current Population Reports
on Medicaid and the Uninsured, report no. 7116, no. P60-229, August 2005, p. 60, http://www.cen
July 26, 2004, pp. 14, 19, 22, http://www.kff.org sus.gov/prod/2005pubs/p60-229.pdf; and author’s
/uninsured/upload/Employer-Sponsored-Health- calculations.
Insurance-Coverage-Sponsorship-Eligibility-and-
Participation-Patterns-in-2001-Full-Report.pdf. 45. Claxton et al., p. 17.

33. See, e.g., Sherry Glied, Revising the Tax Treatment 46. Gruber, “The Cost and Coverage Impact of the
of Employer-Provided Health Insurance (Washington: President’s Health Insurance Budget Proposals,” p. 2.
AEI Press, 1994), pp. 6–8.
47. “Your World with Neil Cavuto,” transcript,
34. Ibid. FOXNews.com, January 31, 2006, http://www.fox
news.com/story/0,2933,183446,00.html (“This is
35. Pauly and Herring, pp. 69–70. a partial transcript . . . that was edited for clarity.”)

36. Bradley C. Strunk, Paul B. Ginsburg, and John P. 48. Bunce, pp. 8–9, 34.
Cookson, “Tracking Health Care Costs: Declining
Growth Trend Pauses in 2004,” Health Affairs Web 49. See, for example, Milton Friedman, “How to Cure
Exclusive, June 21, 2005, p. W5-292, http://con Health Care,” Public Interest 142 (Winter 2001): 8–9,
tent.healthaffairs.org/cgi/reprint/hlthaff.w5.286v1. http://www.thepublicinterest.com/archives/
pdf. 2001winter/article1.html; Feldstein; and Vernon L.
Smith, “Trust the Customer!” Wall Street Journal,
37. Mark Pauly et al., “Individual versus Job-Based March 8, 2006, p. A20.
Health Insurance: Weighing the Pros and Cons,”
Health Affairs 18, no. 6 (November–December 50. Fronstin and Collins, p. 1; and Beth Fuchs and
1999): 28–44, http://content.healthaffairs.org/cgi/ Julia A. James, “Health Savings Accounts: The
reprint/18/6/28.pdf. Fundamentals,” National Health Policy Forum Back-
ground Paper, April 11, 2005, p. 19, http://www
38. Jonathan Gruber, “The Cost and Coverage .nhpf.org/pdfs_bp/BP_HSAs_04-11-05.pdf.
Impact of the President’s Health Insurance
Budget Proposals,” Center for Budget and Policy 51. Fronstin and Collins, p. 1.
Priorities, February 15, 2006, p. 5, http://www.
cbpp.org/2-15-06health.pdf. 52. Newhouse et al., p. 356.

39. Lemieux, pp. 4–5. 53. Ibid., pp. 208–11.

40. “Health Savings Accounts: The First Six Months 54. It is important to note that Newhouse et al.
of 2005,” eHealthInsurance, July 27, 2005, p. 11, argue that this finding was not a result of chance
http://www.ehealthinsurance.com/content/Report (pp. 197–201).
New/072705HSA6mosReportFinal.pdf.
55. Newhouse et al., p. 197.
41. Personal correspondence with Sherry Glied,
March 14, 2006, author’s files. See generally 56. Ibid., p. 210.
Kathryn Klein, Sherry Glied, and Danielle Ferry,
“Entrances and Exits: Health Insurance Churning, 57. Dana P. Goldman et al., “Pharmacy Benefits
1998–2000,” Commonwealth Fund Issue Brief no. and the Use of Drugs by the Chronically Ill,”
855, September 2005, http://www.cmwf.org/usr_ Journal of the American Medical Association 291, no. 19
doc/klein_855_entrancesexits_ib.pdf. (May 19, 2004): 2349, http://jama.ama-assn.org
/cgi/content/abstract/291/19/2344. The authors
42. Maureen Sullivan, “Health Savings Accounts— note, “Other studies have found mixed evidence on
the Consumer Perspective,” Blue Cross Blue Shield this issue.”
Association, September 28, 2005, p. 6, http://bcb
shealthissues.com/events/consumer/sullivan_pre 58. “All HDHPs offered preventive care cost shar-
sentation.ppt. ing for in-network providers that was the same or
lower than the traditional plans. In addition, while
43. Gruber, “The Cost and Coverage Impact of the all HDHPs covered preventive services before the

25
deductible, those services were not always covered 2005, p. 4, http://www.ehealthinsurance.com/
before the deductible by the traditional PPO content/ReportNew/072705HSA6mosReportFin
plans.” U.S. Government Accountability Office, al.pdf.
“Federal Employees Health Benefits Program:
First-Year Experience with High-Deductible Health 71. Len M. Nichols, Marilyn Moon, and Susan
Plans and Health Savings Accounts,” GAO-06-271, Wall, “Tax-Preferred Medical Savings Accounts
January 31, 2006, p. 15, http://www.gao.gov/new. and Catastrophic Health Insurance Plans: A
items/d06271.pdf. Numerical Analysis of Winners and Losers,”
Urban Institute, April 1996, p. 28, http://www.
59. The limits on out-of-pocket exposure in HAS- urban.org/UploadedPDF/winlose.pdf.
qualified HDHPs effectively constitute a maxi-
mum deductible. Thus “high-deductible health 72. Keeler et al., p. 1670.
plans” might be more appropriately termed
“high-but-not-too-high-deductible health plans,” 73. See, e.g., Damon Darlin, “You Think 401(k)’s
or HBNTHDHPs. Are Hard to Manage? Try Health Accounts,” New
York Times, February 18, 2006.
60. See, for example, the benefits structure of the
federal Medicare program. 74. David Kashihara and Kelly Carper, “National
Health Care Expenses in the U.S. Civilian Non-
61. Claxton et al., pp. 82, 105. institutionalized Population, 2003,” Agency for
Healthcare Research and Quality Statistical Brief
62. See Cannon. no. 103, November 2005, p. 1, http://www.meps.
ahrq.gov/papers/st103/stat103.pdf.
63. See, e.g., Reid et al., p. 4.
75. U.S. Centers for Medicare & Medicaid
64. Dana Goldman and Jesse Malkin, “The HSA Services, “An Overview of the U.S. Healthcare
Mirage,” United Press International, February 20, System: Two Decades of Change, 1980–2000,”
2006, http://www.upi.com/HealthBusiness/view. CMS Chart Series, p. 6, http://www.cms.hhs.gov/
php?StoryID=20060220-110738-4830r. TheChartSeries/downloads/us_health_chap1_p.
pdf (accessed April 10, 2006).
65. Keeler et al., p. 1670.
76. See, e.g., John E. Wennberg, “Variation in Use
66. Jason Furman, “Expansion in HSA Tax Breaks Is of Medicare Services among Regions and Selected
Larger—And More Problematic—Than Previously Academic Medical Centers: Is More Better?”
Understood,” Center on Budget and Policy Priorities, Duncan W. Clark Lecture, New York Academy of
February 4, 2006, p. 5, http://www.cbpp.org/2-4- Medicine, January 24, 2005, p. 6, http://www.
06tax.pdf. dartmouthatlas.org/atlases/NYAM_Lecture.pdf.
67. To make this tax credit mirror precisely the tax 77. Some portion of physician expenditures
benefit extended to employer contributions to an relates to care delivered in hospitals. U.S. Centers
HSA, the president proposes to limit this credit “by for Medicare & Medicaid Services, p. 6.
the amount of wages in the [individual’s] payroll
tax base.” U.S. Office of Management and Budget, 78. U.S. Government Accountability Office, pp. 6,
Analytical Perspectives, Budget of the United States 8.
Government, Fiscal Year 2007, February 6, 2006,
Table 17-3, “Effect of Proposals on Receipts,” p. 79. U.S. Centers for Medicare & Medicaid
254, http://www.whitehouse.gov/omb/budget/ Services, p. 4.
fy2007/pdf/spec.pdf.
80. Uwe E. Reinhardt, “The Pricing of U.S. Hospital
68. John Sheils and Randall Haught, “The Cost of Services: Chaos behind a Veil of Secrecy,” Health
Tax-Exempt Health Benefits in 2004,” Health Affairs 25, no. 1 (January–February 2006): 62, http://
Affairs Web Exclusive, February 25, 2004, pp. W4- content.healthaffairs.org/cgi/reprint/25/1/57.pdf.
109, W4-110, http://content.healthaffairs.org/
cgi/reprint/hlthaff.w4.106v1.pdf. 81. See, e.g., Joel Millman, “How the Amish Drive
Down Medical Costs,” Wall Street Journal, February
69. Without any reduction in tax rates, eliminat- 21, 2006, p. B1, http://online.wsj.com/article/SB11
ing health-related deductions would effectively be 4048909124578710.html.
a tax increase.
82. Darlin.
70. “Health Savings Accounts: The First Six
Months of 2005,” eHealthInsurance, July 27, 83. Lemieux, p. 3.

26
84. See, e.g., Millman, p. B1. 93. Boards of Trustees, Federal Hospital Insurance
and Federal Supplementary Medical Insurance
85. Reinhardt, p. 62. “A hospital’s chargemaster is Trust Funds, 2006 Annual Report of the Boards of
a lengthy list of the hospital’s prices for every sin- Trustees of the Federal Hospital Insurance and Federal
gle procedure performed in the hospital and for Supplementary Medical Insurance Trust Funds (Wash-
every supply item used during those procedures.” ington: Government Printing Office, 2006) pp. 65,
105, 116, http://www.cms.hhs.gov/ReportsTrust
86. Kate Schuler, “White House Looks to Price Funds/downloads/tr2006.pdf.
Transparency to Lower National Health Care
Costs,” CQ Today, March 13, 2006. “‘It’s going to be 94. See Cannon and Tanner, p. 28.
imposed upon you,’ Hubbard said at a meeting last
week of the Federation of American Hospitals.” 95. At least one-third of the 500,000 or more indi-
viduals who get HSA-compatible insurance through
87. Keeler et al., p. 1669. a small employer appear to have no other options.
Lemieux, p. 5. Fronstin and Collins, p. 9, find that
88. Stephen T. Parente, Roger Feldman, and Jon “among individuals covered by an employment-
B.Christianson, “Evaluation of the Effect of a based health plan, those in CDHPs or HDHPs were
Consumer-Driven Health Plan on Medical Care more likely than those with comprehensive insur-
Expenditures and Utilization,” Health Services ance not to have a choice of health plan.”
Research 39, no. 4, Part II (August 2004):1189–1209,
http://www.cmwf.org/usr_doc/hsr_39_4p2_1189 96. Fronstin and Collins, p. 1.
_Parente.pdf.
97. Agrawal et al., pp. 10–13.
89. Ellis, slide 3; and personal correspondence
with Phil Ellis, June 29, 2005. 98. Sullivan, pp. 7–9.

90. Agrawal et al., p. 4. 99. Fronstin and Collins, pp. 24–25.

91. Expenditure data for 1997 from Alan C. 100. Ibid., p. 10.
Monheit, “Persistence in Health Expenditures in
the Short Run: Prevalence and Consequences,” 101. “The migration to lower premium, high-
Medical Care 41, no. 7, Supplement (July 2003): III- deductible health plans is certain to restrain payer
53–III-64; inflated to 2006 dollars using growth in revenue growth. . . . By our estimates, $200 billion
per capita national health expenditures from U.S. of revenue will be in play as healthcare payers face
Centers for Medicare & Medicaid Services, Table 1, a daunting challenge of holding on to their exist-
“National Health Expenditures Aggregate and per ing revenue streams while consumers, employers,
Capita Amounts, Percent Distribution, and Average and financial services firms stand to gain.” Aamer
Annual Percent Growth, by Source of Funds: Baig, Jeff Nuckols, and Amy Dawson, “Seizing the
Selected Calendar Years 1960–2004,” http://www. HSA Opportunity,” DiamondCluster International
cms.hhs.gov/NationalHealthExpendData/down Insight, August 2005, p. 4, http://www.diamond
loads/tables.pdf, and “National Health Care Ex- cluster.com/Ideas/Viewpoint/PDF/HSA%20Dia
penditures Projections: 2005–2015,” January 2006, mondCluster%20Single.pdf.
p. 4, http://www.cms.hhs.gov/NationalHealthEx
pendData/downloads/proj2005.pdf. 102. Ibid., pp. 7–8.

92. U.S. Office of Management and Budget, Table 103. Fronstin and. Collins, p. 4.
17-3, “Effect of Proposals on Receipts,” p. 265.
This revenue loss estimate is confined to the pres- 104. Kaiser Family Foundation, “January/February
ident’s proposals that would reduce the federal 2006 Health Poll Report Survey: Selected Findings
tax code’s distortion of taxpayers’ health care on 2006 State of the Union Address and Health
decisions. HSA-related proposals that would add Care (Toplines),” February 2006, p. 3, http://www.
new distortions are excluded. kff.org/kaiserpolls/upload/7460.pdf.

27
OTHER STUDIES IN THE POLICY ANALYSIS SERIES

568. A Seismic Shift: How Canada’s Supreme Court Sparked a Patients’


Rights Revolution by Jacques Chaoulli (May 8, 2006)

567. Amateur-to-Amateur: The Rise of a New Creative Culture by F. Gregory


Lastowka and Dan Hunter (April 26, 2006)

566. Two Normal Countries: Rethinking the U.S.-Japan Strategic


Relationship by Christopher Preble (April 18, 2006)

565. Individual Mandates for Health Insurance: Slippery Slope to National


Health Care by Michael Tanner (April 5, 2006)

564. Circumventing Competition: The Perverse Consequences of the Digital


Millennium Copyright Act by Timothy B. Lee (March 21, 2006)

563. Against the New Paternalism: Internalities and the Economics of Self-
Control by Glen Whitman (February 22, 2006)

562. KidSave: Real Problem, Wrong Solution by Jagadeesh Gokhale and Michael
Tanner (January 24, 2006)

561. Economic Amnesia: The Case against Oil Price Controls and Windfall
Profit Taxes by Jerry Taylor and Peter Van Doren (January 12, 2006)

560. Failed States and Flawed Logic: The Case against a Standing Nation-
Building Office by Justin Logan and Christopher Preble (January 11, 2006)