Financing Medicare: Short and Long Term Implications


Presented to

Subcommittee on Health Care
Committee on Finance
U.S. Senate


Gail R. Wilensky, Ph.D.
John M. Olin Senior Fellow
Project HOPE
February 12, 1997

Mr. Chairman and members of the sub-committee, thank you for inviting me to appear before you. My name is Gail Wilensky. I am a John M. Olin Senior Fellow at Project HOPE, an international health education foundation, and Chair of the Physician Payment Review Commission. I am also a former Administrator of the Health Care Financing Administration. However, I am here today to present only my own views on Medicare, and my testimony should not be regarded as representing the position of Project HOPE or PPRC.

In my comments, I will discuss some of the short and long term implications of the financial crisis facing Medicare. My concern is that most of the reforms considered in the last session of Congress and those already being raised in this session do not resolve the long term problems of Medicare and in many cases, not even the intermediate financing needs of Medicare. The public needs to be more aware of the magnitude of the changes needed to keep the Hospital Insurance Trust Fund afloat until the baby boomers begin to retire, as well as the changes that will be needed to accommodate the baby boomers.

The Need for Reform

Although Medicare has been one of our most popular social programs, it has some major weaknesses and is in serious need of reform. The most significant problems concern the financing of the program and the need to restructure Medicare so that the benefits that have been promised to our seniors can be delivered at a sustainable rate of spending.

Medicare’s current financing problems pose short term, intermediate term and long term difficulties for the program. In the short term, Medicare Part B represents a major drain on the budget since three-quarters of its spending is financed from general revenue. This spending exacerbates the deficit and makes it more difficult to reach a balanced budget. In the intermediate term, the Hospital Insurance (HI) Trust Fund will become bankrupt in the next four years and under current projections will accumulate enormous deficits over the next ten years. In the longer term, with the retirement of the baby boomers, Medicare is not financially viable and its future insolvency raises serious questions about the design of a program that will be sustainable in the 21st Century.

Spending Rates and Solvency Issues

At a time when spending in the private sector has slowed significantly, spending on Medicare continues at unsustainable rates. Although private sector growth rates exceeded Medicare rates in the 1980’s, this trend has been reversed since 1991. In 1996, private sector spending increased at a rate of 3.2 percent, Medicare at a rate of 8.5 percent.

Using the recently released Congressional Budget Office January 1997 baseline estimates, Medicare is still projected to grow at a rate of almost 8 1/2 percent per year over the next five year budget period. Comparatively, during this same period, total Federal Budget Outlays are only projected to grow at an average annual rate of 5.2 percent and the Gross Domestic Product is projected to only grow at an average annual rate of 4.8 percent.

The projections for the solvency of the HI Trust Fund remain alarming. Most of the lower spending growth (.5 percent per year) projected for Medicare from the 1997 baseline came from Part B reductions. The Trust Fund is still projected to be bankrupt in 2001, with accumulated deficits of more than half a trillion dollars by 2007.

In order for the Trust Fund not to be completely exhausted before the end of 2007, there needs to be $450 billion dollars of accumulated policy changes. As a CBO memo dated Jan. 29, 1997 indicates, there are a variety of ways this could be accomplished but all of them require a dramatic departure from present spending levels or a substantial infusion of new funds.

For example, if the growth rate in spending for the Trust Fund were reduced from the expected level of 7.7 percent to 3.4 percent for the entire period, 1998 to 2007, solvency would continue until 2007. Reductions in the growth rate could be postponed until 1999 or 2000 but the subsequent rates of growth would have to be reduced even further in order to maintain solvency through 2007. Alternatively, the combined employer-employee HI payroll tax could be increased by one-third, starting in 1998. All of these proposals involve a more radical change than any of the proposals of the last session had contemplated.

Yet another alternative is to transfer a portion of the current obligations of the Trust Fund to another source of funding, as has been proposed by the Clinton Administration. The main appeal of the transfer is that it “buys time” by extending the life of the Trust Fund without having either to reduce spending or raising taxes to the degree otherwise needed. The transfer of a portion of the home health care benefit (sort of) into Part B has been justified at a policy level on the grounds that approximately half of home care is no longer associated with a hospital stay and is therefore no longer logically associated with Part A.

However, the terms of the transfer of $80 billion of home care should be considered carefully because of the precedent it sets in transferring an obligation into what effectively is the general revenue of the Treasury. Normally, when an expense is brought into Part B, a portion of the total spending becomes part of the premium paid by the elderly and the expense itself is subjected to a 20 percent coinsurance charge. This is not being done for the home health care transfer. While an argument can be made that the separation of Medicare into Parts A and B, with two separate streams of funding is an archaic holdover from Medicare’s inception, removing the limited cost constraints that now exist without reforming the entire program is very risky.

The problems which have been receiving the most attention involve financing Medicare until 2002 and the implications of keeping the Trust Fund solvent for the next decade. Although the problems are less immediate, the implications of the impending retirement of the baby boomers are profound. In 1995, Medicare enrollees represented 13.6 percent of the population and Medicare spending as a percentage of GDP was 2.6 percent. In 2010, when the first of the boomers start to retire, Medicare enrollees will be 15.1 percent of the population and spending on Medicare is expected to be 4.5 percent of GDP. By 2030, when the last of the boomers will be retiring, Medicare enrollees are projected to represent 22 percent of the population and Medicare spending as a percentage of GDP is projected to be at 7.5 percent or almost three times what it was in 1995.

Present Structure of Medicare

There has been an enormous change in the organization and delivery of health services in the private sector. While not all of the changes have been regarded as desirable, there has also been a noticeable decline in spending growth for the private sector, at least to date.

Despite all of the changes occurring in the private sector, Medicare continues to remain primarily a fee-for-service program, with limited availability of and participation in any form of managed care. The projections for 1997 indicate an expected enrollment of 4.4 million seniors in risk-based HMO’s, representing 11.5% of all enrollees. While the enrollment in HMO’s has grown rapidly over the last several years, and is expected to continue growing rapidly for the next decade, even by 2007, it is expected that two thirds of the Medicare population will still remain in the traditional program.

There are several reasons that explain the relatively small numbers of seniors in managed care, but one of the most important reasons is the limited types of non-HMO managed care options available to the Medicare population, the very population that most needs and probably most desires flexibility. Medicare Select, a PPO offering for Medigap, is finally available across the country and a heavily regulated type of point-of-service plan was made available in 1996 but is not yet available everywhere. A Medicare Choices demonstration is setting up a number of provider service network and partial capitation models of managed care, but it will be years before an evaluation of this limited set of options is likely to be available. Even promising demonstrations may not result in changed legislation.

In addition to the limited options that have been available and the lack of incentives for the elderly to be cost conscious, there are also some significant problems with the way payments are made to HMO’s. These problems relate both to the geographic variations that occur across the country and the lack of adequate risk selection adjustments.
Payments to HMO’s reflect the Medicare spending per capita that occurs within the geographic area. These payments, called the Adjusted Average Per Capita Cost (AAPCC), vary enormously from a high of more than $750 per person per month to a low of $220 per person per month.

Differences in the AAPCC primarily reflect very different practice styles and volumes of services used, although to a small extent, they also reflect differences in costs of living. Not surprisingly, HMO growth has been greatest in the areas where the capitation rate is very high and HMO’s are able to offer many extra benefits at no additional cost to the senior.

By setting the capitation payment rate at 95% of the rate of spending that occurs under the traditional program and having the traditional program operate as an open-ended entitlement, the government guarantees it cannot save money by having seniors choose an HMO, other than the 5% it would save assuming there was no favorable risk selection. The issue of risk selection, however, has raised the possibility that the capitation payment may actually cost the government money. This would happen if the elderly choosing HMO’s are healthier than the elderly in their same age/sex categories and if they would spend less than 95% of the average were they to stay in traditional Medicare. While it appears that the elderly choosing an HMO use less services and are healthier the year before they enter an HMO, it is unclear whether favorable selection persists over time for the very large numbers of seniors who remain in HMO’s.

In sum, the present structure of Medicare hardly makes it surprising that it is facing financing problems. The elderly have limited options in the health care plans available to them. Medicare pays most of the costs for the services it covers and almost all of the elderly have coverage that is supplemental to Medicare, either privately purchased Medigap or Medicaid. That means there is little reason for an elderly person to seek out cost-effective physicians or hospitals, or to use lower cost durable medical equipment, laboratories or outpatient hospitals. Hospitals and physicians also have little reason to provide cost-effective care if there is any medical gain to be had from providing more services and some reason to fear legal repercussions if they do less than they might have done and the patient has an adverse outcome. Payments to capitation plans follow payments in fee for service and to the extent risk selection occurs, may even cost the government money. Ultimately, we need to reward the elderly for choosing more cost-effective health care, to provide incentives for physicians and hospitals to order and prescribe cost-effective medicine, and be willing to share the savings which an aggressive reorganization of health care can produce.

Restructuring Medicare

The use of a better designed AAPCC, the payment currently used for HMO’s, could become the basis for a voucher type payment which would encourage more cost-effective choices by seniors. In order to make this transformation, it would be necessary to redesign the determinants of the AAPCC to make it more stable than it is now and to take better account of risk selection than appears to occur. It also needs to be decoupled from spending in an open-ended entitlement, either being set by competitive bid or by administrative fiat. Government spending in the traditional Medicare program also needs to be limited to the same rate of increase as occurs in the capitated plans if seniors are to be encouraged to choose between traditional Medicare and capitated plans and among capitated plans on the basis of their cost-effectiveness and the seniors’ own preferences

The closest approximation of the structure I believe would be a better model for a reformed Medicare program is the Federal Employees Health Benefits Plan (FEHB) in which traditional Medicare would become one of the plan offerings and the premium payment by the government would be the same irrespective of the choice made. This model assumes many more choices available to seniors than is currently available, an annual enrollment process, more information available to seniors about the choices available, monitoring or control of the enrollment process and oversight of plan performance. The level of payment could be set by a weighted average of plans available, by a competitive bid of plans in an area or as a percentage increase over existing rates.

Some Specific Medicare Problems

Aside from the more general issues of restructuring Medicare, there are some specific problems currently facing the Medicare program which need to be resolved.

The traditional Medicare program is expected to be the dominant program for at least the next decade. There are management strategies which are routinely used by the private sector to improve efficiency which could be used in Medicare. These include physician profiling, case management, practice guidelines and the bundling of payments. Several of these have or are currently being tried by HCFA as part of a demonstration but HCFA needs to be able to take the early results of demonstrations and move on them more rapidly than has seemed possible in the past.

There are several areas in Part A of Medicare that need to be addressed in the near term. These include strategies for slowing the growth in home health, hospice and other non-hospital services through prospective payment, co-payments or other policies, and reforming payment methodologies for hospital outpatient departments.

There are also several areas concerning physician payment under Part B that need to be addressed in the near term. These include the implementation of resource-based practice expense values and the move to a single conversion factor under the relative value system.

Finally, there are several areas concerning the AAPCC that need to be addressed, regardless of whether it becomes the basis for restructuring the entire Medicare program. These include limiting the extreme variations in the AAPCC and introducing better risk adjusters. Also both as part of the AAPCC and as a more general issue in Medicare, reforms need to be made in the payments that are being made for graduate medical education.

Much of the attention in the months ahead will be on ways to produce Medicare savings needed for a balanced budget bill. It will be very important that the Congress be selective about the types of short term savings that are pursued to be sure that they are consistent with a reformed Medicare structure.

It is possible to accommodate the need for short-term revenue increases and also set the stage for the more fundamental changes in the incentives, information and options that are needed to reform the Medicare program. Since it will take some time to restructure Medicare and to realize the gains from reforming Medicare, it is important that these reforms be started as soon as possible. This session of Congress is none too soon to start.