Cost Control in the Medicare Drug Bill Needs Premium Support, Not a "Trigger"

Report Health Care Reform

Cost Control in the Medicare Drug Bill Needs Premium Support, Not a "Trigger"

November 10, 2003 11 min read Download Report

Authors: Robert Moffit, Stuart Butler and Brian Riedl

The House-Senate conference committee on Medicare drug legislation is considering a "trigger" proposal to control program costs. This measure is intended to place a limit on the amount of general taxpayer revenue that would be devoted to cover any gap between the actual expenditures of Medicare and "dedicated sources" of revenue for the program--primarily payroll taxes and premiums.

The focus on devices to control costs is long overdue and a welcome development for current taxpayers and future generations. These Americans would be saddled with the huge, unfunded new obligations implicit in the drug legislation--estimated by Medicare Trustee Tom Saving as a two-thirds increase in the $3.8 trillion federal debt held by the public. Placing an effective dollar limit on the additional unfunded liabilities of the structurally insolvent Medicare program is one of two critical tests of an acceptable Medicare drug bill. The other is true structural reform of the program.1

For a cost-control device to satisfy the test, however, it must have the same characteristics that the Food and Drug Administration requires of new drugs. It must be "safe and effective." The trigger proposal, however, does not appear to meet this most basic requirement:

  • The trigger would not even come into play for at least 12 years--six Congresses from now and at a time when the politically powerful and huge baby-boom generation is reaching retirement. This means tough decisions on spending would be delayed until a far more politically difficult time.
  • The proposal has far weaker enforcement tools than budget controls already tried over the past 20 years with little success. Future Congresses could easily evade it--or simply ignore it.
  • If the trigger proposal somehow were to prove effective, it would not be safe for the objective of consumer choice and the availability of private plans. If a future Congress did feel obligated to ratchet down Medicare spending to meet the general revenue cap, it is hard to imagine lawmakers cutting specific services under the traditional Medicare program--other than imposing new and tighter price controls on doctors and hospitals. It is far more likely they would cut funding to private plans. The result would be a further exodus of private plans from the program, leaving traditional Medicare as the only "choice."

Protecting future taxpayers from the huge costs of an open-ended expansion of Medicare is no easy task. But a "supply-side" strategy of trying to clamp down on program spending is unlikely to succeed. That approach would either be ineffective or lead to ever-increasing price controls, micromanagement, and other vestiges of central planning.

Much more likely to succeed is a "demand-side" approach that limits the taxpayer subsidy for seniors enrolling in plans or purchasing services while providing modest-income seniors with enough assistance to guarantee them adequate services. This approach means cost control must be in the form of a "premium support" mechanism. Without such a mechanism, there will be no effective cost control in a Medicare drug bill.

What's Wrong With The "Trigger" Proposal

Currently, 70 percent of Medicare spending is funded by a combination of payroll taxes, premiums paid by Medicare beneficiaries, Social Security taxes, and net interest received from trust funds. The remaining 30 percent is funded through general tax revenues. As Medicare falls sharply into deficit over the next few decades, additional general tax revenues will be needed to close this funding gap--even without a drug entitlement. Those new general revenues would strain the federal budget and force a choice between substantial tax increases and deep spending cuts.2

Under pressure to limit taxpayer costs, House-Senate conferees are considering a proposal to limit general tax revenues to funding no more than 45 percent of Medicare spending in a given year. If the Medicare trustees project that this threshold will be exceeded in two consecutive years within the following seven years, the President would be required to include in the next budget proposal a plan to reduce Medicare's reliance on general tax revenues. Congress would then be required by law to take up this proposal quickly.

Regrettably, this proposal would not lead to meaningful Medicare reforms for two major reasons:

  • It would delay reform. Chart 1 shows that the percentage of Medicare spending funded by general tax revenues has usually fluctuated between 28 percent and 33 percent over the past 15 years and is not projected to reach 45 percent until 2023. The enactment of a Medicare drug benefit would move up this day of reckoning to perhaps 2017. But by that point, millions of baby boomers will be enrolling each year, costs will be soaring, and the program will be moving sharply into deficit. The political window for reforming Medicare before its costs reach crisis levels will have closed.

In the meantime, the benchmark for the trigger would create a false sense of security as well as political cover for lawmakers during the next decade. There would be little incentive to tackle the underlying spending pressures until the trigger is imminent. But pretending that everything is fine until 2017 or 2023 will only delay needed reforms and make future steps more painful to seniors or, perhaps more likely, more expensive to taxpayers.

  • It is not enforceable. Throughout the past 20 years, Congress and the President have enacted numerous laws intended to limit future spending. When it came time to write those budgets, those same lawmakers typically evaded or ignored their own rules.

For example, the Gramm-Rudman-Hollings Act in the 1980s and the discretionary spending caps and PAYGO laws of the 1990s set specific targets limiting spending and/or the budget deficit. Under both of these pieces of legislation, failure by Congress to adhere to those limits led automatically to immediate, across-the-board spending cuts (called sequestrations) until spending or the budget deficit reached the year's target.

While these laws were supposed to impose discipline automatically, spenders routinely voted to exempt certain programs or simply ignore the rules whenever they would have required anything beyond minimal spending reforms. Lawmakers usually passed new laws to write in new, weaker targets for the current year. On other occasions, they simply voted to ignore their own law altogether. Of course, immediately after bypassing the current year's restraints, lawmakers would turn around and enact seemingly tough, new restraints scheduled to begin in the following year. And the game would begin again.

The Likely Medicare "Trigger" Scenario

The proposal to limit the general revenue funding of Medicare is far weaker than even these past, ineffective spending process reforms. In the past, exceeding spending and deficit targets would trigger automatic spending reductions, but the Medicare proposal does not require any automatic controls or process reforms at all. The proposal merely requires the President to submit a plan to curb general revenue funding, and Congress must debate it.

Even those frail requirements are unenforceable. What if the President refuses to submit a reform plan or proposes steps few lawmakers could stomach? No matter how the law is written, nothing would force the President to submit a realistic plan. For political reasons, the White House could purposely submit a vague, short, weak proposal that it has no intention of seriously supporting.

What If a Congress Did Decide to Live by the Law?

In the very unlikely event that a future Congress decided to face down the baby boomers and actually put spending controls into place, how might those controls be distributed? History suggests two things, perhaps in combination.

One would be tighter price controls on doctors and hospitals throughout the program, moving Medicare and the entire health system further toward the centrally planned health systems of countries like Britain and Canada. But America's recent experience of ratcheting down costs in this way indicates the consequences of price controls--reductions in service reaching such a level that Congress is under pressure to raise spending again.

The other likely approach would be to reduce spending on private health plans serving seniors in order to protect the politically sensitive traditional fee-for-service Medicare program. But if private plans were forced to shoulder most of the cost-cutting burden, history shows that more and more of them would withdraw from the program. The result would be fewer private plans available, increasing the probability that in state after state the "fallback" provisions for government-sponsored drug benefits would come into force. The likely effect of supply-side spending controls with real teeth, therefore, would be to undermine the goal of assuring choice and private plans for seniors.

Cost Containment That Works
The FEHBP Model

A structural change in the Medicare program, based on a "demand-side" premium support model, would provide the best mechanism to control costs while improving the quality of health care to Medicare beneficiaries.

Beyond the capacity of the Federal Employees Health Benefits Program (FEHBP) to control costs better than Medicare,3 the FEHBP financing model can also effectively limit the costs to the taxpayer. In a 1999 report on the Breaux-Thomas proposal (a reform of Medicare based on the FEHBP), the staff of the National Bipartisan Commission on the Future of Medicare found that the proposal would have saved approximately $100 billion over a 10 year period (2000-2009) and would have progressively slowed the growth of Medicare spending as a percentage of the gross domestic product.4

How the FEHBP Model Works

Under the FEHBP model, the federal government pays a premium for each FEHBP policyholder. The premium amount is determined each year according to a formula set by law. In determining the government contribution, the government share is 72 percent of the average premium of all plans, based on plans' bids and weighted by the number of enrollees in each plan in the previous year. Each year, using this formula, there is a dollar cap on the amount the government will pay. It varies each year. For 2003, that capped amount is $2,840 for individuals and $6,490 for families.

The FEHBP formula includes one other key component. While the current formula formally provides that enrollees pay 28 percent of the premium, it also provides that in no case will the government pay any more than 75 percent of the premium for any health plan. When the government's 75 percent share exceeds the 72 percent of the enrollment-weighted average premium, the enrollee starts to pay all of the additional premium cost.

Applying the FEHBP
Model to Medicare

In the Medicare context, the government share of the plan payment would necessarily be larger and reflect the status quo. In fact, the government might pay about 90 percent of the overall premium cost of Parts A and B, plus the value of the drug benefit under the new Part D.5 But an FEHBP-style premium support mechanism would directly limit the taxpayer cost of Medicare automatically and more effectively than the proposed trigger.

The premium support system would have many advantages:

  • Medicare's initial annual payment to private plans could be based on the cost of the Medicare benefits. The government's premium contribution would initially equal the average cost to the government of traditional Medicare. At present, that cost is about $7,000 and rising by about 10 percent per year. With a drug benefit, the cost might be $700 to $1,000 higher, depending on the final details of the legislation.
  • Medicare's subsequent annual payments to private plans could be based on the weighted average of competing health plans, combined with an FEHBP-style dollar cap on the amount paid to those plans . In subsequent years, just like the FEHBP, the annual calculation of the government contribution to competing health plans in a reformed Medicare program could be based on the previous year's experience for enrollment weighting and the projected costs for the following year, as estimated by the health plans themselves, just as plans do in the FEHBP today. Congress could set the dollar cap on the basis of the existing FEHBP-style formula, the per capita spending for traditional Medicare (Parts A, B, and D), or a weighted average of the two.
  • With an annual cap, the government payments would be clear, predictable, stable, and limited . Government costs could be based primarily on--and be equal to--the benefits provided under the Medicare fee-for-service system. Each health plan would charge whatever premiums are required by its benefit package, either above or below the premium cost of the Medicare program.
  • With a fixed, but generous, government contribution, Medicare beneficiaries rather than taxpayers would bear the cost of the premium above the government contribution, just as federal workers and retirees do in the FEHBP today. As with federal employees and retirees, Medicare beneficiaries would be able to pay anything they wish to pay above that amount, as long as the benefits and rates are in reasonable relationship to one another. With a fixed government contribution and under extreme competition, high-cost plans would be pressured to provide better services for the money or cut their costs, or they would lose market share.
  • With a fixed government contribution, beneficiaries would reap any savings between the actual plan costs and the government contribution . Plans costing less than the traditional Medicare program (including Part B and Part D premiums) would pass on the entire premium savings to enrollees. This would be a substantial improvement over the current FEHBP policy, under which the government, as noted, collects 75 percent of all savings from choosing low-cost plans.
  • A premium support mechanism can control costs while protecting modest-income seniors. Beyond allowing beneficiaries to keep all savings from health plan choices, a premium support approach can allow a variation in the government contribution on the basis of income, geography, or other relevant factors. Given the enormous costs facing future taxpayers, it makes a great deal of sense to means-test government contributions to health plans. Such additional adjustments would actually be a substantial improvement over the current FEHBP policy.6


The Medicare drug debate has been long and tortuous, in large part because the fundamental issues were largely ignored in the rush to enact a politically popular benefit. Fortunately, though belatedly, these critical issues are emerging toward the end of the House-Senate conference deliberations.

One of the most important of these issues is achieving a Medicare drug benefit without a staggering increase in the program's already unsustainable unfunded liabilities. It would be unconscionable of Congress to enact a bill that increases the liabilities that are passed on to the next generation--and which would mean huge tax increases. Rather, the bill must reduce those liabilities.

The trigger proposal, however, will do little if anything to hold down the mushrooming taxpayer cost of Medicare. It could easily be evaded by politicians who are adept at circumventing or simply ignoring spending controls. Moreover, even if it did work, it would do so by increasing government controls on doctors and hospitals to the detriment of patients.

Needed instead is a firm commitment by Congress to an effective premium support mechanism. That mechanism would directly limit the subsidy to seniors while making the health industry compete to satisfy patients, not the regulations of government officials.

Stuart M. Butler, Ph.D., is Vice President for Domestic and Economic Policy Studies, Robert E. Moffit, Ph.D., is Director of the Center for Health Policy Studies, and Brian M. Riedl is Grover M. Hermann Fellow in Federal Budgetary Affairs in the Thomas A. Roe Institute for Economic Policy Studies at The Heritage Foundation.

1. Stuart M. Butler, "The Crucial Elements of an Acceptable Medicare Bill," Heritage Foundation Backgrounder No. 1667, July 16, 2003.

2. See Brian M. Riedl and William W. Beach, "New Medicare Drug Entitlement's Huge New Tax on Working Americans," Heritage Foundation Backgrounder No. 1673, July 30, 2003, and Daniel J. Mitchell, Ph.D., "Why Medicare Expansion Threatens the Bush Tax Cuts and Undermines Fundamental Tax Reform," Heritage Foundation Backgrounder No. 1672, July 25, 2003.

3. This comparison controls for differences in benefits between the FEHBP and Medicare. See Michael O'Grady, Health Insurance Spending Growth: How Does Medicare Compare? Joint Economic Committee, U.S. Congress, June 10, 2003. According to the report, Medicare spending grew an average of 6.7 percent per year over the past 20 years, while FEHBP grew at 6.5 percent. If FEHBP had not offered prescription drugs (like Medicare), the JEC found, FEHBP spending grew at only 5.8 percent. For a broader historical comparison of the FEHBP and Medicare, including cost control, see Walton Francis, "The FEHBP As a Model for Medicare Reform: Separating Fact from Fiction," Heritage Foundation Backgrounder No. 1674, August 7, 2003.

4. For the commission staff's analysis, see National Bipartisan Commission on the Future of Medicare, "Cost Estimate of the Breaux-Thomas Proposal," memo to the Medicare Commission, March 14, 1999, at

5. The authors are indebted to Walton Francis, an independent economist and consultant and a prominent expert on the functioning of the FEHBP, for specific suggestions for applying the FEHBP financing arrangement to a reformed Medicare. For a detailed description of this approach, see Walton J. Francis, "Using the Federal Employees' Model: Nine Tests for Rational Medicare Reform," Heritage Foundation Backgrounder No. 1675, August 7, 2003.

6. Recent research on the issue of adverse selection in the FEHBP shows that there is "little observable risk selection" in the program. The reason is that the generosity of the government contribution to private plans mitigates the impact of risk segmentation, and there is consequently little variation in the distribution of younger and healthier persons among plans. See Curtis S. Florence and Kenneth E. Thorpe, "How Does the Employer Contribution for the Federal Employees Health Benefits Program Influence Plan Selection," Health Affairs, Vol. 22, No. 2 (Spring 2003), pp. 211-218. Walton Francis has suggested yet another remedy to adverse selection: The legislative language could simply provide that the Secretary of HHS shall adjust the government premium contributions further to compensate for measured risk selection between private plans and traditional Medicare.


Robert Moffit

Senior Fellow

Stuart Butler


Brian Riedl
Brian Riedl

Senior Fellow, Manhattan Institute