The Federal Employees Health Benefits Program (FEHBP) works well in providing decent, affordable health coverage for the 8.5 million people it serves, including the president, members of Congress, federal employees, retirees, and their families. The most tangible real-world example of "managed competition," FEHBP is a system of competing private health plans in which the government contributes a relatively fixed amount toward the employees coverage and employees pay a premium based on the cost of the individual plan they choose.
Largely because of its ability to constrain cost growth reasonably well with limited government intervention, the program has been proposed by some political leaders and analysts as a model to replace the current Medicare program, to cover small businesses and the uninsured, or, in some cases, to cover the entire nation. This analysis finds that FEHBP would represent a substantial improvement over the high premiums and limited benefits currently faced by small businesses and uninsured adults. The approach would not work well, however, for the older, sicker populations served by Medicare. Not only is the FEHBP model likely to lead to discrimination against ill or disabled beneficiaries, but Medicares large-group purchasing clout would be diminished, program administrative costs would rise, and, as a result, costs to government and beneficiaries alike would grow. While FEHBP insures retirees, this coverage is largely supplemental to Medicare and thus does not bear the full risk of health services for an elderly, less healthy population.
This report describes FEHBP and how it has worked over the years, examines how it might work if applied to Medicare or small businesses and the uninsured, and assesses whether the model would be an improvement over current systems of health coverage.
Overview of FEHBP
Enrollment and plan participation. The largest employer health insurance program in the United States, FEHBP insures about 3 percent of all Americans. As of 2003, 133 plans, offering 188 coverage options, are participating in FEHBP. One dozen of these options are offered by nationwide preferred provider organization (PPO)/fee-for-service plans; six of these nationwide plans are available to specific groups of federal employees (e.g., Foreign Service); and the remaining plans are local HMOs. Enrollment, however, is concentrated in just a few plans: over half the enrollees are enrolled in Blue Cross Blue Shield and another quarter are in one of the other national PPO/fee-for-service plans.
Government contribution and benefits. The government's contribution toward the cost of the beneficiary's premium is the lesser of 72 percent of the average FEHBP plan premium, weighted by enrollment, or 75 percent of the premium for the plan chosen. The enrollee pays the difference. There is no standard prescribed minimum benefit package, and benefits vary from plan to plan. In general, the benefits offered by large employer plans are richer than those in FEHBP; benefits in Medicare, meanwhile, are not as generous, principally due to that program's lack of prescription drug benefits and catastrophic coverage.
Premium-setting. Most PPO/fee-for-service FEHBP plans are experience-rated; premiums are based on expected costs, plus a small service charge. Plans are not strictly at risk and can recover losses either through tapping into reserve funds or increasing the premium in a subsequent year. HMOs are community-rated, charging rates comparable to those charged to nonfederal groups.
Cost history. Since 1969, FEHBP has experienced slightly lower premium growth per enrollee than private health insurance overall (10.6% vs. 11.0%) but higher growth than Medicare (8.9%) (Figure ES-1). In the last four years, Medicare has outperformed FEHBP by a far greater margin, with premiums growing at only about one-third the FEHBP rate. In addition, Medicare's administrative costs as a percentage of total claims cost have been far lower than FEHBPs (2% vs. 7%15%).
Risk selection and plan participation. Throughout its history, FEHBP has suffered from adverse risk selection, which occurs when sicker beneficiaries gravitate toward certain plans. Indeed, the program nearly lost Blue Cross Blue Shield in the early 1980s as a result of adverse selection. Although more than 84 percent of retirees choose PPO/fee-for-service plans, FEHBP does not employ risk adjustment of any kind to account for differences in enrollees' health; consequently, HMOs benefit from a younger and therefore healthier enrollment. Some HMOs nevertheless have experienced enrollment so low that they have had to withdraw from the program. Unstable plan participation has been a problem for FEHBP, as it has for the Medicare+Choice program. Insufficient enrollment or unpredictable health care utilization led to more than 100 FEHBP plan withdrawals between 2000 and 2002.
Converting Medicare to the FEHBP Model
Proposals to convert Medicare to the FEHBP model would require either that basic, fee-for-service Medicare compete on the same basis as private health plans or that the program move entirely to a system of competing private plans. Most proposals would have the government provide a fixed dollar amount based on a percentage of the average plan premium (e.g., 85%). Under this "defined contribution" approach, beneficiaries would pay the difference between the premium of the plan selected and the government's contribution. Proponents say that government costs would be reduced, government involvement minimized, and plan choice enhanced. Our analysis finds, however, that the FEHBP model poses serious risks to Medicare beneficiaries as well as taxpayers.
Costs. As a single, large governmental purchaser of care, Medicare achieves lower administrative costs and lower provider payment rates than private plans do. To date, private plans have not demonstrated a value-added advantage over Medicare's inherent cost-savings advantages. It may be possible for the federal government to reduce its financial exposure under a FEHBP approach by reducing its share of the premium, changing the "benchmark" plan to one whose price is lower than average, or reducing the benefit package. Unless these changes are draconian, however, these changes may not compensate for the more favorable payment rates Medicare now enjoys and the much lower growth rates that result. The government may also save money if competition results in enrollees choosing lower-priced plans. But, again, the savings FEHBP achieves through competition have not been greater than savings from Medicare's system. Moreover, if the choice of lower-priced plans is not accompanied by an adequate risk-adjustment mechanism and plans do not fully assume risk (neither of which has occurred), any resulting savings may be illusory.
Unless competition succeeds in lowering overall cost growth substantially, all the other measures to increase savings to the government will simply increase costs to the beneficiary. A lower government share of the premium or reduced benefit package will have to be made up by higher beneficiary spending. Medicare currently pays only 57 percent of the total health expenses of beneficiaries. Any further reductions in that share will be felt by most beneficiaries, but especially by the sicker and more disabled individuals who are heavy users of health services.
Choices and complexity. Increasing the number of plan choices carries with it a greater risk of confusion for Medicare beneficiaries. Four million elderly beneficiaries have Alzheimer's disease; 2 million are in nursing homes, many with cognitive impairments, and 12 million have less than a high school education.
Risk selection. The average FEHBP enrollee is 46 years old and in good health. On the other hand, one-third of Medicare enrollees have serious physical or cognitive impairments, accounting for two-thirds of Medicare outlays. These skewed expenditures create a strong incentive for private plans to market to younger, healthier enrollees, causing premiums in traditional Medicare to spiral upward. Given that HMOs in FEHBP and in Medicare have experienced favorable selection to date, an expansion of private plan participation in Medicare is likely to have a similar result.
Plan stability. Both Medicare and FEHBP have suffered from instability in private plan participation. For older, sicker Medicare beneficiaries, the potential disruption in care from plan withdrawals or instability in plan provider networks is particularly worrisome.
Government intervention. Although FEHBP has operated with little government intervention, its spending is only about one-tenth that of the Medicare program. It seems unlikely that government policymakers will remain immune from plan and provider pressure if the stakes are raised to Medicares level.
Using FEHBP to Cover Small Businesses and the Uninsured
Proposals to extend FEHBP to small businesses and the uninsured are based on the following assumptions: that large group pooling would result in lower premiums for small businesses and individuals than they can obtain on their own; that benefits would be improved; and that coverage would become available for persons currently ineligible due to prior health problems. The likely effects of extending FEHBP to these groups are summarized below.
Costs and access. FEHBP has much lower administrative costs than the small group and individual insurance markets, where they can range from 20 to 50 percent. The large group pooling that FEHBP offers could also lower premiums, unless only sicker individuals and higher-risk businesses opt for such coverage. Premium assistance or tax credits may be needed to ensure that healthy people also enroll. No one would be prohibited from enrolling on the basis of poor health.
Plan choices. FEHBP's wide array of plan choices and provider networks is significantly greater than that offered to individuals and small businesses. Some plans, however, may be unwilling to participate due to the broadening of government involvement in health coverage. Providing federally subsidized reinsurance and risk-adjustment may help reduce the risk to plans and make participation more palatable.
Administrative complexity. Administering a program for 44 million uninsured people across the country is far more complex than doing so for 8.5 million federal employees and retirees who are geographically more concentrated.
Conclusion
If extended to Medicare, the FEHBP model poses serious risks to Medicare beneficiaries and likely risks to the taxpayer as well. It would provide no significant, demonstrable improvement in the current system while opening the door to increased financial risk for enrollees. If applied to small businesses and the uninsured, however, the FEHBP model has certain tangible advantages: the promise of lower premiums, better benefits, greater choice, and more stable coverage than what is currently available.