Medical savings accounts: the newest medical cost reduction tool for employers.
In the private sector, the driving force behind the increased compensation in the form of benefits versus wages is the federal tax system. Because wages are taxed as income while employer-paid health insurance premiums are not, companies have a major incentive to shift compensation to this form. And as inflation in the 1960s and '70s pushed up the marginal tax rates for workers, the incentive to receive income in the form of fringes increased.
In 1960, for example, 19 percent of national health costs were paid by the government, 20 percent by private insurance, and 58 percent by consumers (out of pocket), while 3 percent came from charities. By 1991 those percentages had shifted to 42, 32, and 22 percent, respectively [FIGURE 1 HAS BEEN OMITTED], with charities constituting only 4 percent. Thus, the proportion of health care costs covered through insurance or federal government programs has risen dramatically. The implication of this trend is that as consumers pay a smaller part of their health care while government and the insurance industry increase their roles, market forces are less able to hold down health costs in proportion to demand for services. Figure 2 [FIGURE 2 HAS BEEN OMITTED]shows the rise in consumer prices since 1947 compared to the rise in the cost of medical care. Health care inflation has been considerably higher than general inflation, particularly since the introduction of Medicare and Medicaid in 1966. Any attempted "reform" of the health care system logically needs to deal with the problem of reintroducing market forces to slow the rise in relative medical costs.
Health Care Demand
Economists constantly refer to the law of demand. This states that, ceteris paribus, a fall in the price of a good or service will induce buyers to purchase more of that product. Although there is little debate that this is an accurate description of consumer behavior in the markets for automobiles or clothes, there appears to be less consensus that this behavior actually occurs in the market for health care. An individual suffering a heart attack probably doesn't ask about prices while being admitted to the ER. And in nonlife-threatening situations, such as a child's earache or the flu, the decision to use health services is assumed by some to be quite unrelated to the cost of that service.
Empirical research by economists seriously questions this line of thinking. For example, the Rand Corporation found that the "elasticity" (defined as the percentage change in the amount purchased divided by the percentage change in price) for health care to be approximately -0.2, implying that a 10 percent increase in the real cost of health care would reduce its use by 2 percent. Further, although employees with high deductible insurance sought care less often, they did not differ significantly from the low deductible group in terms of the quality of health care sought. Both groups were just as likely to seek both effective and rarely effective health care. The study by Lyke (1994) reported "negligible effects on general measures of health for adults and children as a whole." However, cost-sharing through higher deductibles had a greater effect on the incidence of service use for the poor, reducing the use of services particularly for poor children. This suggests that special attention should be given to designing MSA plans for use with Medicaid.
Additional research by Feldstein and Gruber (1994) indicates that the elasticity of health care demand may be even higher than that reported in the Rand Corporation study. The implication of this is striking. Feldstein and Gruber simulated the effect on national (non-Medicare) health care spending of a switch to a health insurance policy with a 50 percent co-payment and a 10 percent income limit. That is, an individual with health costs of $3,000 and an income of $30,000 would pay $1,500 out of pocket. If he had health costs of $10,000, he would pay $3,000, which is 10 percent of his income for that year. The effect on aggregate health expenses is significant, according to their simulation, with potential savings of more than $100 billion annually!
One problem, of course, is that most people would balk at a health insurance plan that leaves them responsible for such potentially high out-of-pocket costs. What is needed is a market-based reform that encourages health consumers to shop carefully for health care but does not expose them to significant deductibles and co-payments.
Medical Savings Accounts
An innovative new approach to reintroducing market forces to health care is the Medical Savings Account (MSA). This proposed plan, which one of the authors helped design, is essentially a combination of a catastrophic plan with a high deductible and an employer-provided account for first dollar health expenses. A simple example suggests the logic of such an approach. First, assume a traditional health plan with an employee contribution of $300, a deductible of $400, and a 20 percent co-pay on the next $2,500 for a total potential out-of-pocket cost (OPC) of $1,200. The cost of such a plan is $5,300 (these are actual numbers for a school district with an MSA and a traditional plan). Thus, a family with $1,000 in health expenses would have an OPC of $820, which is $300 contribution plus $400 deductible plus 20 percent of $600.
An alternative MSA plan would involve the purchase of a catastrophic plan with a deductible of $3,000 and an MSA with $1,800. An individual with $1,000 in health expenses would pay for those health services out of the MSA, leaving $800 in pre-tax funds available to the individual for that year. This is not "use it or lose it" money. The funds could be taken in cash or rolled over for use in future years. If the individual had $2,000 in expenses, the first $1,800 would be paid for out of the MSA and the next $200 would be OPC. If he had health expenses greater than $3,000, the catastrophic plan would kick in and the OPC would be limited to $1,200--the same as with the traditional plan.
The cost to the employer of the traditional plan is $5,000 ($5,300 less the $300 employee contribution). The cost of the high deductible plan is $4,400. This includes a $2,600 premium for the catastrophic plan with a $3,000 deductible plus the $1,800 deposit to the family MSA. The savings are 12 percent! Obviously, this is good news for the employer. But both sides benefit from the MSA plan. As Table 1 shows, the employee is in a better cash position with the MSA under all potential outcomes. The worst the MSA employee will do is $1,200 out of pocket, which is the same maximum OPC under the traditional plan. If the family had no health expenses, then $1,800 in pre-tax funds would be available for the family to use. This concept is currently being used by more than 2,000 firms nationwide.
Why might MSAs reduce employee health expenses? The first reason entails a shift in employees' perceptions of ownership. Employees with MSAs are spending their own money and so are very careful about how and where their health dollars are expended. And when employees spend their own money carefully, they are also spending their employer's money carefully. This means fewer instances of employee claims above the high deductible amount, resulting in a lower employer catastrophic premium.
This assumes that the MSA is treated as cash by employees. However, if the employees treat the MSA as insurance with the idea that it should be used for health care expenditures during the year, there might be little or no effect on health care use. In this instance, the MSA is little more than a "flex" account that encourages pre-tax health care spending. Thus, different assumptions regarding how MSAs might influence use will show remarkably different outcomes in simulations. For example, Blue Cross Blue Shield of Ohio conducted a simulation on the use of MSAs that assumed no change in use with an MSA plan. For the MSA plan to cover its costs, the MSA premium would have to be about 33 percent higher under that assumption.
The American Academy of Actuaries (1995) conducted a simulation on changes in health care spending under an MSA plan. The analysis was conducted using three scenarios regarding the cash nature of the MSA. These included a 90 percent cash assumption, a 50 percent cash assumption, and a 10 percent cash assumption. Under the 90 percent assumption the MSA plan reduced costs by around 11 percent. Under the 50 percent assumption, in which half of the funds were seen as "insurance," the savings dropped to around 7 percent. Finally, the savings fell to around 2 percent when the cash assumption was lowered to 10 percent. Unlike the high co-pay simulation referred to earlier, the uncertain nature of how MSAs are viewed implies that a decrease in use does not automatically follow when switching from a traditional insurance plan to an MSA.
Another way in which MSAs might reduce health expenses is that the cost of administering an MSA may be considerably lower than with a traditional plan. Suppose, for example, that the amount of health spending under a traditional plan averaged $5,000 per employee. These numbers are adjusted for deductibles and co-pays so that they represent the average cost of the employee plan to the provider. Assuming 15 percent in administrative costs for this plan, the total cost would be $5,882 per employee, including $882 for administrative costs and profit. Now assume an employee MSA plan with a $3,500 deductible, $2,500 in the MSA, and potential out-of-pocket costs of $1,000. Because employees first spend from the MSA account and then have $1,000 in OPC expenses, it is reasonable to expect that overall health spending would decline. If we anticipate a reasonable drop of 5 percent, the average amount of employee health costs to the provider fall from $5,000 to $4,750. This is covered by the $2,500 in the MSA plus the $1,000 OPC plus $1,250 from the catastrophic plan. Assuming a 2 percent administrative cost for the MSA of $51 ($2,500/.98 less 2500) and $397 for the catastrophic plan ($2,250/.85 less 2250), the total MSA plan administration costs are $448. This gives an administrative savings of $434 ($882 less $448), which represents a reduction of 49 percent!
The biggest potential problem with MSAs is the current tax treatment. Although the total employer expense (in a for-profit entity) is tax deductible, the amount of funds in the MSA for an employee is considered wages and thus subject to income and FICA tax. So if an employee has $2,000 in an MSA and is in a 28 percent tax bracket, the tax due on those funds would be $866 ($560 in income tax plus 15.3 percent or $306 in FICA tax).
The current tax treatment is a strong disincentive to establish an MSA plan, since the full cost of a traditional plan would be excluded from employee income under current law. Some firms have tried to "get around" this adverse tax treatment by setting up the MSA as a Section 125 "flex" account. Under this scheme, the employee would elect to make a $2,000 deposit to a pre-tax flexible spending account. The employer would then agree to raise the employee's salary by that amount so that the account is effectively financed by the employer. In the above example, assume the $2,000 "flex" deposit and $1,000 in medical expenses for the year. Because the flex is a "use it or lose it" proposition, the employer would take back the remaining $1,000 at the end of the year and would then issue the employee taxable wages for $1,000. Obviously, this strategy might be frowned on by the IRS.
Fortunately, this problem may be eliminated in the near future. As of May 1996, Congress is considering MSA legislation as an amendment to the Kennedy-Kassebaum portability bill. Basically, this legislation would make MSAs a trust that could be used for Section 213 qualified medical expenses. Employer contributions to the MSA would be tax deductible for the employer and tax-free for the employee. Disbursements from the trust would also be tax-free if used for qualified medical expenses. Other withdrawals would be subject to income tax plus a 10 percent penalty if they occurred before age 59-1/2. Unlike an IRA or 401(k) plan, however, the investment income earned in the MSA is taxable annually.
Another potential concern with MSAs is that such plans will cause employees to shun preventive treatment, such as physicals or mammograms, in an attempt to conserve MSA cash. Although this is certainly possible, the evidence so far indicates that employees are receiving more preventive care than before. On reflection, this seems logical. Suppose the cost of a physical runs $300. With the traditional plan outlined above, the employee would be out $300 because of the $400 deductible. With an MSA, however, the employee would have the funds available and still have MSA cash left over. One of the groups happiest about MSAs is single parents. With the MSA, they can now afford to take children with minor illnesses to the doctor, whereas before they either had to come up with the deductible money or not see the physician at all.
Some people may argue that MSAs move in the opposite direction of managed care--which, says the American Academy of Actuaries, has been somewhat successful in holding down health costs. But there is no reason why MSAs, which are effective in promoting wise use of low dollar health spending, cannot be combined with the proven ability of HMOs to control high dollar expenses. Employees would have the benefit of potential MSA cash as well as some of the pressure on gatekeeper primary physicians to reduce unnecessary expenditures. Such reductions would be realized as consumers responded to the incentive to make better health care decisions.
A common criticism is that MSAs might encourage adverse selection. That is, healthy employees would sign up for MSAs in order to receive the MSA cash, whereas sicker employees would move to the traditional plans or HMOs. But a properly designed MSA should not have this effect. From Table 1 it can be seen that no matter what the health usage, employees are in a better financial position with an MSA than with a traditional plan. The relatively low deductible for the MSA should not have the effect of driving a huge number of sicker employees into HMOs.
Empirical Evidence on MSAs
Because MSAs are a relatively new concept, there is limited data on their effectiveness. What is available, however, is very encouraging. Barchet, Anderson, and Chapman (1995) surveyed 17 firms that used MSAs. Each firm had a $1,500 deductible for single employees and a $2,000 deductible for family coverage. The average MSA deposit for single coverage was $1,073 with an average maximum OPC of $427. For family coverage the MSA deposit averaged $1,380 with an average maximum OPC of $620. The traditional plans used by these employers had average maximum OPCs of $941 for single coverage and $1,979 for family coverage. Thus, under the worst case scenario both single and family coverage under MSAs had much lower OPC exposure than under traditional plans.
These data would seem to counter the concern that sicker employees would migrate to traditional plans and HMOs and healthier employees would select MSAs. In addition, the savings to employers were dramatic. Single coverage cost for MSAs averaged $2,133, as opposed to $2,319 for traditional plans. And the cost of a family MSA plan averaged $3,882, compared to $6,536 for traditional indemnity coverage. These numbers represent savings of 8 percent and 41 percent, respectively.
Bond, Heshizer, and Hrivnak (forthcoming 1996) analyzed data from 27 Ohio firms using MSAs. The average MSA contribution was $857 and $1,167 for single and family plans, respectively. This gave average maximum OPCs of $643 and $833 for those plans (since the deductibles were $1,500 and $2,000, respectively). The traditional plans had a maximum OPC of $960 and $2,188 for single and family coverage. Again, the lower maximum OPC under MSAs would argue against the possibility of MSA implementation causing adverse selection. In fact, it would seem to follow that few employees would want a traditional plan if offered an MSA. The current evidence supports this position. The average employer savings from the MSA plans was around 12 percent in this sample. But these employer savings are generated by "sharing" some of the plan cost reductions with employees through significantly lower OPCs.
For example, one firm in the sample had only family coverage for its employees. The cost of the traditional plan was $2,598. This plan had a deductible and maximum co-pay of $750 and $1,200, respectively, for a potential OPC of $1,950. the firm offered an MSA plan with a $2,000 deductible and a $1,000 MSA deposit. The cost of the high-deductible policy was $1,747. the total MSA plan cost was therefore $2,747 ($1,747 plus $1,000)--about 6 percent higher than the traditional coverage. If the employer had set the maximum OPC for the plans equal at $1,950, the MSA plan would have required only a $50 deposit to the MSA account ($2,000 less $1,950). This would have lowered the MSA plan cost to $1,797, which is 31 percent less than traditional coverage. When the MSA costs were reestimated using this assumption, the 27 firms had an average savings of around 34 percent! These savings occurred despite the fact that no employee would have a greater potential OPC under the MSA plan than under a traditional one.
It should also be recognized that most employees do not use all of the funds in the MSA. Barchet and colleagues found an average unspent balance of around $600 in single MSA plans and around $900 in family MSA plans. Well over 50 percent of the employees had funds remaining in their MSAs. Golden Rule Insurance refunded (or rolled over) an average of $602 in 1993, $1,002 in 1994, and $997 in 1995 per employee. Similar findings have been reported for Morris County Hospital, Dominion Resources, and Forbes Magazine.
The evidence thus far, then, indicates that MSAs lower employer health costs, reduce employee OPCs relative to traditional plans, and give most employees cash back at the end of the year. It should also be recognized that these very positive findings are occurring despite the current unfavorable tax treatment of MSAs. Giving MSAs the same tax exempt status as traditional indemnity insurance and HMOs should significantly increase the number of firms adopting them.
Although it is a complicated one, the problem of escalating health care costs is partly related to the rise in third party payments, which removes incentives for health consumers to economize on medical usage. Recent research suggests that switching to 50 percent co-pays (with a limit related to income) would reduce national health spending by more than $100 billion annually for non-Medicare patients. But despite its cost effectiveness, this type of plan would no doubt be unpopular with most individuals who have become used to generous health plans.
A more realistic method to make consumers more cost conscious in purchasing health care is the Medical Savings Account. This plan combines a high deductible catastrophic plan with an employer financed MSA for low dollar heath expenses. The plan costs less than traditional policies for two reasons. First, the MSA is perceived as employee money and, as such, employees are likely to spend it more carefully than under traditional plans, thus reducing their plan costs to the employer. Second, the MSA plan has significantly lower administrative costs. Data available on existing MSA plans indicate employer savings averaging about 10 percent and remaining MSA funds for employees of around $700 per account.
Possible criticisms of MSAs include the taxation of remaining funds in the account, reductions in the use of preventive care, thwarting the growth of cost-effective managed care, and causing adverse selection by healthy employees. Existing evidence, however, suggests that the only major problem with MSAs is their current tax status.
Table 1 Cash Flow For Employee Health Expenses Under MSAs vs. Traditional Health Plans
Health Expense Traditional Plan MSA $ 0 $ -300 $ +1,800 400 -700 +1,400 1,000 -820 +800 1,800 -980 0 2,000 -1,020 -200 2,900 -1,200 -1,100 3,000 -1,200 -1,200 4,000 -1,200 -1,200
American Academy of Actuaries, Medical Savings Accounts: Cost Implications and Design Issues, Washington, DC, May 1995.
S. Barchet, Medical Savings Accounts: A Building Block for Sound Health Care (Olympia, WA: Evergreen Freedom Foundation, March 1995).
S. Barchet, J. Anderson, and L. Chapman, "Medical Savings Accounts: An Option to Reduce Health Care Costs and Increase Health Care Satisfaction," ACA Journal, Autumn 1995, pp, 34-37.
M. Bond, B. Heshizer, and M. Hrivnak, "Reducing Employee Health Expenses with Medical Savings Accounts," Compensation and Benefits Review, forthcoming (September-October 1996).
J. Burry, Medical Savings Accounts: Bad Medicine for the U.S. Health Care System (Cleveland, OH: Blue Cross/Blue Shield of Ohio, 1994).
C. Carlstrom, "The Government's Role in the Health Care Industry: Past, Present, and Future," Economic Commentary, Federal Reserve Bank of Cleveland, June 1994.
Council for Affordable Health Insurance, various communications.
M. Feldstein and J. Gruber, "A Major Risk Approach to Health Insurance Reform," Working Paper No. 4852, National Bureau of Economic Research, Inc., September 1994.
M. Forbes, "Health Care Reform: An Alternative Proposal," CFO, December 1993, pp. 34-36.
M. Litow, "Examining Medical Savings Accounts with Guaranteed Issue, Community Rating, and Risk Pools," Milliman & Robertson, Inc., December 1994.
Luntz Research Company, Survey of Golden Rule Employees, 1995.
B. Lyke, "Medical Savings Accounts," Congressional Research Service Report for Congress, 94-597 EPW, 1994.
J. Newhouse, Free for All: Lessons from the RAND Health Insurance Experiment (Santa Monica, CA: RAND Corporation, 1993).
R. Thompson, Employee Benefits: A Tale of Success (Council Grove, KA: Ron Thompson and Associates, 1993).
"Virginia Utility Company Holds Health Premiums to 1 Percent," Employee Benefits News, July 1993, p. 12.
Michael T. Bond is an associate professor of finance at Cleveland State University, Cleveland, Ohio, where Brian P. Heshizer is an associate professor and Mary W. Hrivnak an assistant professor of management and labor relations.
|Printer friendly Cite/link Email Feedback|
|Author:||Bond, Michael T.; Heshizer, Brian P.; Hrivnak, Mary W.|
|Date:||Jul 1, 1996|
|Previous Article:||JIT II: a purchasing concept for reducing lead times in time-based competition.|
|Next Article:||Employment Nondiscrimination Act: on the cutting edge of public policy.|