Gail R. Wilensky, Ph.D.
John M. Olin Senior Fellow
June 29, 1995
7500 Old Georgetown Road þ Suite 600
Bethesda, MD 20814-6133
(301) 656-7401 þ (301) 654-0629
Mr. Chairman and Members of the Committee. Thank you for inviting me to appear before the Committee to testify about the recent growth of the Medicaid program, the program's administrative structure, the need for greater flexibility, and the use of the waiver process. I am currently the John M. Olin Senior Fellow at Project HOPE, an international health education foundation, and the chair of the Physician Payment Review Commission. I am here today, however, not to represent either of these organizations, but to speak from my experiences as Health Care Financing Administration Administrator during the Bush Administration.
The Need For Reform
Medicaid has played a vital role in helping states finance the health care needs of our nation s most vulnerable citizens. Medicaid is a state-administered program, funded jointly by the Federal and state governments, that currently funds services for about 37 million low income people. States receive Federal matching funds to finance Medicaid based on a formula that varies with the state s per capita income. The Federal share varies from 50% to 83% of the total funding. The Medicaid program allows states substantial flexibility, in terms of the services that can be provided, the populations that can be covered, and the way that the program is administered.
Although it is frequently maligned for providing expensive, fragmented care to people in high cost and inappropriate settings, by most measurements Medicaid has met its basic objective of providing health care to selected categories of low income populations. We should applaud the successes that Medicaid has achieved, but we should also acknowledge the need for changes -- changes that will reduce the growth in spending and lessen the burdens that have been imposed on the states responsible for administering the program.
A variety of reasons have been given to explain the growth in Medicaid spending, which in most states has made Medicaid the fastest growing component of the state's budget and the second largest category of state expenditures after education. These reasons include increasing caseloads, additional requirements that have been placed on the states by the Federal government, and the discovery of what were effectively Federal-only dollars during the early 1990s.
At least part of the increased spending can be attributed to the increased use of mandates during the last decade. These mandates included new populations that had to be covered -- such as all pregnant women and children up to age 6 with family income up to 133% of the poverty line, all children born after 1983 to families in poverty, and increased numbers of elderly through the Qualified Medicare Beneficiary program. -- and new services that had to be provided -- such as all needed services discovered during a screening visit under the EPSDT program, whether or not otherwise covered by the state's Medicaid program. It also included new requirements about how and by whom services were to be provided, particularly for people in nursing homes.
In addition to the mandates, new options were made available regarding the populations that could be covered under Medicaid, particularly women and children. This, along with the flexibility to bring in previously uncovered populations provided through the 1115 waiver process, allowed states to shift what previously had been state-only dollars to Medicaid, which made them Federal/state dollars.
A third factor explaining the rapid growth of Medicaid spending was the development by several states of creative financing strategies to fund their share of the match. This was a critical moment in the program's history, because it began to undermine the basic premise of the financial structure of Medicaid-- that funding be shared through a Federal match of State monies. Matching grants presume that those responsible for spending decisions have a reasonable stake in the program's costs. In fact, the only real cost containment mechanism that exists in the Medicaid program is the State s share of the costs. By its structure, Medicaid is an open ended matching program, with no limit on Federal payment.
The discovery of strategies by states to enhance their Federal matching share, at little or no cost to themselves, along with the other pressures to increase spending, had the not very surprising result of producing explosive growth rates during the early 1990 s. Medicaid, which had been growing at rates that varied between 8% and 12% earlier in the 1980 s, grew at rates of almost 19%, 32% and 28% for the years 1989-90, 1990-91 and 1991-92, respectively. These rates of growth have since slowed to rates closer to those of the 1980 s, with growth in the Medicaid program projected at rates between 9% and 11% for the rest of the decade. Although obviously less than the rates of the early 1990 s, these growth rates are far in excess of most of the rest of the Federal budget, which, excluding health and interest on the debt, is projected to grow at rates of 3.8% per year for the rest of the decade.
Lessons Learned From Donations and Provider Taxes
States showed remarkable creativity in the strategies they devised to enhance their share of Federal dollars. West Virginia started the process in 1986, but at least 30 states were involved by July of 1991. The specific strategies varied substantially, but basically each worked in the following way. A state borrowed money from providers through donation or tax programs. The money was used as the state's share of Medicaid and was matched at least dollar for dollar by Federal funds. The state would then increase Medicaid payments to reimburse providers for the donations or taxes they had paid. In many states, providers were guaranteed to get back at least as much as they donated or paid in provider-specific taxes through hold-harmless mechanisms. The funds were most frequently distributed via "disproportionate share" payment strategies (payments to hospitals providing a disproportionate share of services to low income populations) that allowed states to reimburse institutions in excess of the amounts spent providing care to low income people.
Legislation was passed in the fall of 1991 that limited the amount of revenue that could be used for purposes of Federal match from taxes that were limited to medical providers, eliminated the use of donation strategies, and limited the amount of funds that could be received as disproportionate share payments. States had a minimum of one year to comply with the new requirements (some states whose legislatures only met on a biannual basis had two years to comply). As frequently happens, some states had been much more aggressive than others in increasing their effective Federal match rate. Those states were allowed to maintain their high rates of provider taxes for a period of time and were allowed to keep very large levels of disproportionate share spending while other states were limited in what they could introduce or claim.
Legislation was also passed in 1993 which limited disproportionate spending allocations to the amounts institutions had spent furnishing hospital care to Medicaid-eligible and uninsured patients (less the amount they had received directly from Medicaid for providing services to Medicaid eligible individuals). This meant that the total money received from Medicaid couldn't exceed the cost of providing hospital services to low income populations, something which had happened with some frequency before the legislation was passed.
The experience with provider taxes, donations and disproportionate share spending was a rude awakening regarding the fungibility of money. In general, the 1991 legislation, combined with the 1993 legislation, shut down the abusive provider tax and donation funding arrangements which the states had adopted. Tennessee may be an exception. Tennessee used a new services tax which included hospital service and combined it with the use of a sales tax that exempted certain activities, such as those included in the new services tax. The state borrowed revenue from the new services tax and used it, in part, for its Federal match. There is an ongoing court case as to whether this practice violated the 1991 law, but, since there wasn't a clear baseline to start from, it will be very difficult to prove that adding a hospital tax to a pre-existing tax violated the law. It is also no longer currently relevant, as Tennessee has capitated its entire program into TennCare. To the extent that Tennessee had expanded its baseline to include questionable or even inappropriate expenditures, the state will carry more money forward into the future than it would have been able to do otherwise, with effectively no chance for the Federal government to ever recoup that money.
The bigger concern today has to do with intergovernmental transfers. Intergovernmental transfers was an area of concern in 1991, but those of us working on the issue at HCFA were unable to devise a rule which would distinguish between an intergovernmental transfer that represented a legitimate transfer between levels of government and the movement of funds which results in only new Federal money coming into the program. It is particularly problematic when the county or state is paying itself because it owns the hospital and is putting up its share of the match with an intergovernmental transfer. The absence of an ability to distinguish appropriate uses and abusive uses of intergovernmental transfers is a good reminder that money is fungible and that reliance on the use of state matching as a cost containment strategy is a genie that can never be put back into the bottle.
The Waiver Process
Although the amount of legislation and regulation associated with the Medicaid program is substantial, the flexibility provided by the waiver process allows the Federal government to waive any individual component of Medicaid as long as the resulting change is in keeping with the spirit of Title XIX. Thus provisions of state-wideness, duration and scope, freedom of choice, the Boren Amendment, etc., have all been waived as part of requests by states for more flexibility in designing their Medicaid programs.
The flexibility provided by the waiver process has allowed states to experiment with changes that have enabled states to improve the efficiency with which they provide care and have given states the ability to more effectively provide care for various difficult subgroups in the population, such as high risk teens, pregnant women, frail elderly and so forth. The 1115 waiver, which removes certain restrictions on the use of managed care delivery systems and also allows coverage of individuals not normally covered by Medicaid, has been especially popular. Since 1993, HCFA has granted 10 statewide 1115 demonstration waivers. Thirteen other states have 1115 waivers under consideration or discussion.
While much good has been associated with the waivers, there have also been areas of concern. Some of the concern has to do with the administrative complexity, burden and costs associated with the application process and subsequent monitoring by HCFA. Despite the efforts that I undertook to streamline the waiver process while I was at HCFA and the efforts that the current Administration has undertaken since then, states continue to feel that the reporting and filing requirements associated with the waiver process are costly and burdensome. State representatives are only too eager to show the piles of paper required for waivers or as part of the state plan amendment process (the fundamental statement of how a state's program will function or change during a future period of time). States continue to be frustrated at the time and expense required to include changes that have been tried in other states or are a regular feature of the private market.
States and the Secretary of Health and Human Services are forced to use HCFA's research and demonstration authority to continue implementing successful changes in state programs, since that is the only statutory authority available for their continued implementation. The most notable example is Arizona's AHCCS program. This program was introduced in 1983 and provides the state with authority to run its entire Medicaid program under a waiver. The State has used a combination of managed care programs in urban areas combined with primary case management in rural programs to produce a highly successful Medicaid program that has provided its recipients with greater access to physicians and other primary care personnel than exists under most Medicaid programs while experiencing lower rates of spending growth than has occurred nationally. Despite this fact, twelve years later Arizona must continue to operate under a research and development waiver, with the sham that suggests for process, because it does not conform to many of the requirements of Medicaid.
A second area of concern has to do with the determination of budget neutrality that appears to have been used in some of the waiver requests, when, in fact, common sense would suggest that provisions were included that make them "costers" to the Federal government. In particular, there are questions as to what the Federal government would have spent in the absence of the waiver as well as questions about the states' use of the Federal share of savings from managed care plans to fund coverage of people not otherwise included in Medicaid.
Let me give a few examples. When Hawaii requested a waiver during the early part of the Clinton Administration, it was allowed to calculate Federal contributions which would have occurred if Hawaii had taken advantage of all of the options, in terms of populations and services covered, available to it, even though Hawaii had not, in fact, included many of these options in their program. In my opinion, this use of 'hypothetical spending' in no way meets a common sense definition of the term "budget neutral," that is, equal to the amount which would have been spent in the absence of the waiver.
A second example involves the TennCare program. This is a program that has raised a lot of controversy because of the pressure that the Blues placed on providers to participate in the program. Much less attention has been given to the fact that Tennessee has claimed it was able to cover 362,000 previously uninsured individuals because of the savings associated with managed care, and has therefore been able to lay claim to all the additional Federal dollars generated from this in a transaction that has been designated as budget neutral. The Committee knows that I am a supporter of managed or coordinated care for vulnerable populations, but these claims of savings seem beyond belief, even to an ardent supporter. The General Accounting Office has raised similar questions concerning increased Federal spending for the demonstrations in Florida, Hawaii, and Oregon.
I am not necessarily against the individual expansions or the programs that have been produced by the creative use of "budget neutrality," but I am very concerned with what this does to the budgeting process and the ability of government to gain control over its own spending levels. Just as I didn't necessarily object to some of the programs that states funded with their provider and donation money, I believed then, as I believe now, that we must run government at levels that the public is willing to fund and not at levels that other well-meaning people believe is appropriate even in the face of public unwillingness to provide such funding levels.
The Choices for a Reformed Program
Tinkering with the existing program will not solve the fundamental problems of Medicaid -- an unsustainable growth in spending and burdensome, costly requirements on the states. The major alternatives facing the Congress are the use of a capped payment per person covered under Medicaid eligibility rules or the use of a block grant. The most significant difference between these strategies is that the capped payment per person option maintains an entitlement between the Federal government and the individual, whereas the block grant does not. This is an important philosophical issue that should be debated on its own merits, but there is also concern that any strategy other than a block grant strategy exposes the Federal government to uncertainty with regard to future spending and the potential to be "gamed" by the states in terms of claims for Federal funding.
Both of these strategies, but particularly the block grant, will relieve the states from much of the regulatory burden associated with the current system. The prescriptive nature of existing law and regulation would be replaced with a limited monitoring function for the Federal government. Rather than focussing on the specifics of process and inputs, as is currently the case, the Federal government's role would be one of monitoring and auditing. HCFA's role would be to audit the states to assure that Federal monies were being spent on health care and on the low income population, however they were defined in statute, and that the states were meeting whatever "maintenance of effort" requirements were put into law.
In addition, consideration is being given to putting a limited number of performance or outcome measures into law. The irony is that, if this occurs, Medicaid could require information from the states on access to care, use, or health outcomes for their low income populations which it does not now require, despite all the mountains of paperwork which HCFA demands from the states. In fact, because of a lack of uniform reporting requirements on use, claims, and expenditures, it is very difficult to assess the effects of the Medicaid program from HCFA data.
The result is that if block grants (or capped payments per capita) were to be adopted, states would be relieved of the burdensome reporting and filing requirements associated with the current regime, but would be expected to provide a minimum set of information on outcomes and performance. It seems to me that this is a trade worth making. Whether the increased flexibility will be worth the lower rates of growth being suggested to the governors will be an important determinant of the successful adoption of these changes.
A significant philosophical issue underlies the choice between the current
administrative structure with its division of responsibility between the Federal
government and the states and that implied by block grants. That issue is whether
the states, given some additional financial resources and with the Federal government
assuming a reasonable monitoring role, can be relied on to take care of their
most vulnerable populations, or whether only the Federal government can be presumed
to care about these vulnerable populations and that without its active involvement,
those populations will not receive adequate or appropriate care. This is an
issue that's worth debating.