September 27, 2000

HCFA’s Plan to Regulate The Medicaid Upper Payment Limit
by Leighton Ku

View PDF version

If you cannot access the file through the link, right-click on the underlined text, click "Save Link As," download to your directory, and open the document in Adobe Acrobat Reader.

The Health Care Financing Administration (HCFA) of the Department of Health and Human Services (HHS) plans to issue a proposed regulation soon to restrict a rapidly spreading Medicaid financing scheme that is costing the federal government significant sums and about which the General Accounting Office (GAO) and HHS' Office of Inspector General (OIG) have raised strong warnings.(1) Under this financing mechanism, a state pays selected nursing homes, hospitals or other institutions more than the actual costs the facilities incur for medical services they provide. The state then requires these health care providers to transfer most of the extra payments back to the state. The state draws down federal matching funds based on the inflated payments it has made to the providers. As a result, the state collects additional federal money without contributing any state funds. The federal Medicaid funds gathered through these schemes can be used by states for any purpose they choose, including for activities that are neither related to health care nor authorized by Congress.

This practice, although apparently legally permissible (the GAO has referred to it as a loophole in the current rules), runs contrary to the basic principle that the federal government and states share the costs of the Medicaid program. The practice effectively enables states to increase the federal government's share of Medicaid costs (and decrease the state share), without Congressional approval.

In many cases, these financing arrangements do not improve the quality of health care provided or benefit health care providers. The financing mechanisms frequently operate in a manner that siphons extra federal money to state coffers without affecting the provision of health care. To date, this has been particularly true in financing arrangements that involve nursing homes. On the other hand, in some cases these financing arrangements have been used to provide important additional resources to safety net hospitals that provide care for the uninsured and HCFA's regulation ought to be sensitive to this distinction.

States using these arrangements generally have a variety of alternative ways to secure fiscal resources, including making different policy choices about the use of state budget surpluses and tapping tobacco lawsuit settlements. Most states that are employing this financing scheme to secure added federal dollars are not in fiscal difficulty, as is evidenced by the fact that most of them have cut state taxes in the past few years.

Some states claim the additional federal funds they have secured through the use of these financing arrangements have been used for Medicaid expansions or improvements. It is not clear, however, that this has occurred to any significant degree. The validity of this claim is difficult to determine, but if the claim were true, one might expect to find that the states using these practices have somewhat broader Medicaid eligibility criteria than states not employing them. In fact, the opposite is the case — the states using these financing arrangements have narrower Medicaid eligibility criteria, on average, than states not using them.

These financing mechanisms are now proliferating. If no action is taken, these practices will cause federal Medicaid expenditures to spiral upward by billions of dollars in future years. The resulting cost increases might eventually be used to justify new efforts to cut Medicaid or alter its basic character. In the 1990s, widespread state use of a variant of this loophole, along with other factors, caused federal Medicaid costs to rise at alarming rates; these cost increases became a significant factor in an effort that culminated in Congressional approval of a proposal to replace Medicaid with a block grant. (The proposal was not enacted because of a presidential veto.) At a minimum, the additional federal costs that will result from the increasing spread of these financing practices are likely to make it harder to secure support in coming years for the provision of new resources for further expansions in Medicaid or the State Children's Health Insurance Program (SCHIP) that are aimed at reducing the number of uninsured.

HCFA plans to publish a proposed regulation in the next few weeks to prevent these financing arrangements from spreading further and triggering billions of dollars of unnecessary federal expenditures. Although the precise contents of the regulation will not be known until the regulation is published, HCFA has suggested it will seek to limit the scope of this loophole while providing a multi-year "transition period" to let states and providers restructure their financing arrangements gradually.(2)

Some in Congress are reportedly considering an effort to attach a "rider" to an appropriation or other bill to block HCFA from proceeding with this rule. This analysis finds such an action would be unwise. HCFA should complete action this year. The Congressional Budget Office estimates that blocking the regulation would increase federal costs by $1.5 billion in fiscal year 2001 alone. The added costs would be higher in subsequent years and, if the regulation is blocked, state use of these arrangements is likely to escalate. It should be noted that if Congress refrains from blocking the regulation now, it will not lose the ability to act at a later time to modify the regulation. Congress always can act at a later date if it concludes, after reviewing the final regulation and examining these issues, that the rule needs to be changed. For example, if subsequent analyses support the belief that the final rule would significantly harm selected safety net hospitals, Congress could establish a more straightforward and accountable method of increasing funding for those hospitals, rather than continuing the current abuse-prone financing arrangements.



Since its creation in 1965, the fundamental principle in Medicaid financing has been that the federal government and the states share the program's costs. For each state dollar spent, the federal government contributes one to four dollars in matching payments. In 2001, the Medicaid program will cost $219 billion, of which $124 billion — or 57 percent — will be borne by the federal government.(3) The Medicaid statute gives states substantial authority to design and administer the program. The requirement that states share in the cost helps to ensure they act prudently in stewarding federal resources.

In the late 1980s and early 1990s, state abuse of a similar Medicaid mechanism, called disproportionate share hospital (DSH) payments, placed this relationship in jeopardy.(4) Many states began using complex accounting maneuvers to increase the federal matching payments without the states having to expend any additional state funds. By the early 1990s, states were using this accounting loophole to draw down billions of dollars in additional federal funds.

These financing mechanisms involving DSH payments contributed to an explosion in federal Medicaid expenditures in the late 1980s and early 1990s, which in turn provided some of the impetus for efforts in the mid-1990s to block-grant Medicaid or place caps on it. Rancorous disputes ensued between the federal government and the states about DSH funding arrangements, which culminated in a series of laws enacted in 1991, 1993 and 1997 that tightened the DSH rules and limited the maximum DSH payments that states may receive.(5) Even with these limitations, the federal government spent an estimated $9 billion for DSH payments in fiscal year 2000.

The new financing arrangements that now are spreading — and that are the subject of this analysis — are generally known as "upper payment limit" (UPL) arrangements. They bear strong similarities to the DSH financing mechanisms and essentially are a variant of those practices. Both types of arrangements use complex accounting gimmicks to secure additional federal funds for states without actual state matching contributions. Also like the DSH schemes, the UPL arrangements have been used for various purposes; some UPL arrangements have helped support safety net hospitals that care for Medicaid patients and the uninsured, while other UPL arrangements do not aid health care providers and are designed primarily to provide a windfall for state governments.

One key difference between the older DSH and the newer UPL financing arrangements is that the DSH program has been subject to close scrutiny. Congress acted in 1991, 1993, and 1997 to curb the worst abuses in DSH financing schemes. In contrast, the federal government currently has almost no regulatory authority today to limit UPL abuses. Under current regulations, HCFA has little option but to approve state proposals to exploit the UPL financing mechanism.

Research from the Urban Institute indicates that in recent years, the federal cost of UPL financing arrangements has burgeoned, rising from $313 million in 1995 to $1.4 billion in 1998.(6) Preliminary data from HCFA suggest the federal cost may be at least twice as high by 2001, with a potential federal cost of more than $3 billion.(7)


How Does the UPL Loophole Work?

Before describing the Rube Goldberg-like accounting arrangements inherent in UPL practices, it may be useful to discuss the key concept underlying these financial arrangements. A state makes inflated payments to a select group of nursing homes, hospitals or other health care facilities that a county or other local government owns, with the payments being in excess of the actual cost of the medical services these institutions provide to Medicaid beneficiaries.(8) The state then requires these providers to give back much or all of this extra money to the state in the form of "intergovernmental transfers." The state uses the large payments it has made to the providers to claim a large federal matching payment, which will equal at least 50 percent of the payment the state has made to the providers. The state thus receives these federal matching dollars without having put up a commensurate amount of state funds.

Three steps are involved in a UPL financing arrangement.(9)

Figure 1 presents data concerning a recent example of the use of this mechanism by Pennsylvania, as reported by HHS' Office of the Inspector General.(11) On June 14, 2000, the state paid $697.1 million in supplemental payments to 23 county nursing homes. Since Pennsylvania has a 54 percent federal matching rate, it received $393 million in federal matching funds (which is 54 percent of the $697.1 million payment the state made to the nursing homes). The nursing homes, in turn, returned $695.6 million of the $697 million to the state, doing so on the same day they received these payments from the state. The result was a small net gain to the nursing homes of $1.5 million — the amount of intergovernmental transfers is typically set so that no provider incurs a net loss — and a windfall for the state government of $392 million. (The state paid a net amount of $1.5 million to the nursing homes while receiving $393 million from the federal government.) Although, the federal government paid a large amount to the state, apparently no additional health services were secured for this money.

9-27-00health.jpg (37473 bytes)

Essentially, the only "real" money in such a transaction is the federal matching money. Both the state and the providers secure net financial gains without any contribution of state matching dollars. In this example from Pennsylvania, the state made most of the money, and the nursing homes kept little. UPL arrangements also can be structured to let the providers keep much or most of the money.

OIG and GAO have found that other states, including Alabama, Nebraska and Michigan, have arrangements similar to Pennsylvania's that are designed primarily to divert federal Medicaid funds to the state.(12) The OIG concluded that: "States did not base the enhanced payments on the actual costs of providing services or increasing the quality of care to Medicaid residents of the targeted nursing facilities. The counties involved in the enhanced payment scheme provided little or none of the sham enhanced payments to the participating nursing facilities to provide services to Medicaid residents."


What is the Upper Payment Limit and How Would the Forthcoming HCFA Regulation Change It?

The size of these financing schemes is governed by what is known as the "upper payment limit." Federal law gives states considerable flexibility regarding payments to health care providers, but it stipulates that, in general, Medicaid payments can be no higher than the amount that Medicare would pay for the same service.(13) Medicare's equivalent payments form the "upper payment limit" for Medicaid. The payment rates that states use in Medicaid are usually lower than the Medicare rates, with the exact gap varying by state and type of medical service.

The test of whether Medicaid payments exceed this "upper payment limit" is not based on the Medicare payment level for a single procedure or even on the payment level for all services that a single provider delivers. Instead, the upper payment limit is the aggregate amount of all payments that could be made to an entire "class" of providers if every provider were paid the Medicare rate for all services. Medicaid regulations currently establish two classes of health care providers: state-owned facilities and non-state providers, with the class of non-state providers including both local-government-owned facilities and private providers. To illustrate how the upper payment limit works, we use a hypothetical example.

Let's say that the gap between the Medicaid payments a state makes to all county-owned nursing homes in the state and the equivalent amount that Medicare would pay is $200 million. Let's also assume that the gap between the Medicaid payments the state makes to private nursing homes and the Medicare payment levels is $800 million. The upper payment limit for this class of providers, which encompasses both local government-owned providers and private providers, would consequently be $1 billion more than the amount the state actually pays. To exploit the upper-payment-level loophole, this state could make an extra, or supplemental, payment of $1 billion to the county-owned nursing homes, secure virtually the entire $1 billion back from these nursing homes as an intergovernmental transfer, and receive at least $500 million in federal matching funds for engaging in this maneuver. The state is allowed to use the maneuver — and to direct the entire $1 billion in supplemental payments to county-owned nursing homes despite the fact that the gap between the actual payments these facilities receive and the Medicare payment rate is $200 million — because, as noted, the upper payment limit applies to an entire "class" of providers and private facilities are in the same class as the county-owned facilities.

HCFA has intimated that the proposed regulation it plans to publish would tighten the UPL limits by making county or local government-owned facilities a separate class from private facilities.(14) That would not eliminate the potential for states to make supplemental payments but would greatly reduce the possible size of these payments and narrow the scope of these financing maneuvers. Depending on how the regulation is drafted, this might mean that under the above example, the maximum amount of supplemental payments the state could make to county nursing homes would be one-fifth of the amount the state now can make (i.e., $200 million rather than $1 billion).


UPL Arrangements Distort Medicaid Financing

As noted, one effect of these practices is that states can increase the federal government's share of Medicaid expenses without Congressional approval. While this appears legal, it is contrary to the spirit of the Medicaid statute.

OIG has estimated that Pennsylvania has increased the federal matching rate for its total Medicaid program from 54 percent to 65 percent in fiscal year 2000 by using these financing arrangements. The GAO has noted that New Jersey's pending UPL proposal could lift the federal share of Medicaid expenses that state receives from 50 percent to 60 percent. The GAO also estimates that Michigan increased the federal share of Medicaid costs it received from 56 percent to 68 percent by using similar practices in the past.(15)

UPL transactions also have another negative side-effect: they can distort apparent Medicaid spending trends and thereby inject confusion into policy debates. Some states have begun to raise alarms that their Medicaid budgets are on the rise again, pointing as evidence to growing total Medicaid spending (i.e., state plus federal spending) in their states. As shown above, however, UPL systems can increase apparent total Medicaid spending while decreasing the actual expenditure of state funds. Some of the complaints about rising Medicaid costs and their effects on state budgets rely on figures that are inflated because they reflect the use of these financing mechanisms and thus make total Medicaid expenditures in a state — and the drain on the state budget — appear larger than they actually are (because the total expenditure figures include the extra federal matching payments and fail to net out the intergovernmental transfer revenues from providers that help finance the transactions).(16) The appropriate measure of Medicaid's actual cost to a state is the amount of Medicaid expenditures financed from the state's general fund revenues, a measure that excludes federal matching payments and nets out the revenues contributed through intergovernmental transfers.

It is worth recalling that in the early 1990s, Medicaid spending rose very sharply in substantial part because of the explosion in Medicaid DSH payments, which shot up almost twenty-fold from $403 million in 1990 to $8.0 billion in 1992. This was interpreted as a sign that Medicaid was out of control and threatening to wreak havoc on state budgets, even though states were actually using DSH payments to reduce their share of program expenditures. The so-called Medicaid "cost crisis" was a major contributing factor in the push of the early and mid-1990s for proposals to restrict Medicaid funding by eliminating or limiting the program's entitlement status, such as by converting the program to a block grant or capping it.(17) Both houses of Congress approved such changes in 1995; the changes were not enacted only because of a Presidential veto. Concerns about rapid Medicaid spending growth in this period also brought federal Medicaid eligibility expansions to a halt until the creation of SCHIP in the 1997 Balanced Budget Act. Congress expanded Medicaid eligibility in each year from 1984 to 1990, but then cost concerns brought this legislative trend to a standstill.


What is Known about Current and Proposed UPL Arrangements?

Information about the extent to which states are using UPL schemes is fragmentary: HCFA, OIG and GAO are still collecting data on this matter. It appears that 19 states have at least one approved UPL financing arrangement (some of these states have proposals pending for additional UPL financing mechanisms), while nine states have proposals pending for UPL systems, and three states have initiated discussions with HCFA about submitting a UPL proposal. As these figures indicate, UPL financing schemes show signs of spreading rapidly. If left unchecked, they are likely to increase federal expenditures by billions of dollars.

Some earlier information about these financing arrangements is available from an Urban Institute study. In a survey the Institute conducted in 1998, the Urban Institute found that 12 of the 40 responding states were using UPL mechanisms at that time.(18) The study reported these UPL systems primarily involved hospitals and that the financial gains under these arrangements were being reaped principally by the hospitals, rather than the states. Of $1.4 billion in additional federal funds being secured through these arrangements, $1.3 billion were going to benefit county facilities (mostly hospitals) while relatively little, about $100 million, was being retained by the states. Although it thus appears that these UPL funds did reach hospitals in these states — particularly public hospitals in California and Illinois — the UPL mechanisms in question were designed so the states contributed virtually none of the additional money and the federal government provided virtually all of it.

The nature of UPL systems appears to have changed substantially since 1998, however, with the changes adding urgency to HCFA's current efforts to prevent these financing mechanisms from proliferating. The more recent UPL systems seem to be based primarily on county nursing homes rather than hospitals and apparently are being used to benefit state governments, with few of the added dollars going to the health care providers. Although there is potential for misuse of UPL financial arrangements involving either hospitals or nursing homes, there is more evidence of this type of abuse in the nursing home-based arrangements.


Do States Need Additional Federal Funds?

Some state officials defend the use of UPL financing arrangements, arguing that their states need the additional federal funds and that the funds help to pay for Medicaid and other health care programs, including program expansions. It is difficult to evaluate such statements, since a state's "need" for additional revenue is not absolute but is relative to other competing budget and political priorities. It should be noted, however, that most states are in the midst of a period of economic prosperity and have substantial budget surpluses.

Table 1 presents data about several measures of the fiscal status of states that currently have or are proposing UPL arrangements. Collectively, these states had state budget balances of $21 billion in state fiscal year 2000.(19) Most of these states had good, positive balances although a few states, such as Alabama, Arkansas, New Hampshire, and Tennessee, faced tight fiscal circumstances. Together, the group of states using or proposing to use UPL mechanisms cut taxes a total of $4.6 billion for the year 2000, although a few states with fiscal problems had to raise taxes. Overall, the strong trend was to cut state taxes. All except four of these states reduced taxes at least once in the past four years.

In addition, these states have state tobacco settlements worth a total of $5.6 billion in 2001. Preliminary data indicate that only a portion of those funds, which were based on the value of total (state plus federal) Medicaid expenditures for treatment of smoking-related illnesses, have been used for health-related purposes.

A final potential alternative resource for these states is money they have made from their use of similar financing mechanisms in their Medicaid DSH programs. In state fiscal year 1997, the latest year for which data are available, the states using or proposing to use UPL schemes garnered an additional $2.1 billion in federal funds from DSH, kept in state coffers. Federal DSH allocations have been reduced since then, and it is reasonable to think that states' DSH profits have declined somewhat, although recent data are not yet available.

Table 1
Fiscal Status of States with Approved or Proposed Medicaid UPL Arrangements
  FY 2000 state balance(20)
(mil. $)
FY 2000 balance as% of budget (20) FY 2000 tax changes enacted in 99(21)
(mil. $)
# of past 4 years with state tax cut(22) FY 2001 tobacco settlement
(mil. $)
FY 1997 state DSH profits(23)
(mil. $)
Alabama* 41 0.8% 147 1 112 (25.0)
Alaska 867 37.9% 0 1 24 6.0
Arkansas 0 0.0% 11 0 57 (0.5)
California* 3,012 4.6% (295) 4 884 376.0
Georgia 545 3.8% 0 3 170 74.0
Illinois* 1,350 5.9% 82 2 322 168.0
Indiana* 1,617 17.8% (233) 3 141 109.0
Iowa* 574 12.0% (8) 4 60 8.0
Kansas 318 7.2% 28 3 58 32.0
Louisiana 58 1.0% (10) 4 156 462.0
Massachusetts* 1,706 8.7% (68) 4 280 227.0
Michigan* 1,285 13.9% (376) 3 301 not avail.
Minnesota* 2,370 20.5% (2,084) 3 462 (17.0)
Missouri 435 6.1% (478) 3 158 288.0
Montana 165 15.1% 7 1 29 (0.0)
Nebraska* 271 11.6% 100 2 41 not avail.
New Hampshire* 0 0.0% 617 0 46 not avail.
New Jersey* 1,174 6.0% (70) 3 268 3.0
New Mexico* 143 4.2% (2) 2 41 not avail.
New York 1,170 3.2% (1,092) 4 884 18.0
North Carolina* 38 0.3% 6 3 162 158.0
North Dakota* 41 5.3% (2) 2 25 0.7
Oregon* 526 10.8% (93) 1 80 19.0
Pennsylvania* 1,511 7.8% (328) 2 398 not avail.
South Carolina* 464 8.7% (6) 3 82 32.0
South Dakota 37 4.8% 20 0 24 0.7
Tennessee* 212 3.1% not avail. 0 169 0.0
Washington 1,175 11.6% (478) 1 142 154.0
Total 21,105 6.4% (4,605)   5,574 2,093
    (natl. avg.)        
* State has at least one approved UPL arrangement in September 2000. The other states have pending proposals. Three additional states, Florida, Texas and Wisconsin have initiated discussions with HCFA about potential UPL arrangements.

It certainly is true that states must make difficult budget decisions and work hard to balance their budgets. But the data indicate these states generally could have made fiscal choices other than to use UPL mechanisms. For example, Pennsylvania, which has one of the most visible UPL arrangements, had a substantial state budget surplus in 2000 and recently reduced taxes. These states understandably believe it is to their advantage to use these financing arrangements to divert federal resources to state coffers, using lawful means. Taxpayers in other states, however, who ultimately pay for federal expenditures, might wonder whether it is fair for their federal taxes to be used to enlarge budget surpluses and effectively help to fund tax cuts or other program expenditures in states with UPL systems.

Some states defend the fact that they have siphoned off so much of the windfall funds they have captured through UPL arrangements (and have left providers with so little) by arguing that the extra money is rebudgeted to support Medicaid or other health care expenditures. It is not possible to determine the validity of this argument. Money is fungible; the additional funds go in general state coffers and can be mixed with other money. There is no way to ascertain the exact source of the money going to Medicaid. If $100 million retained by a state from UPL transactions is used to support Medicaid, this could mean that $100 million in other state money that otherwise would be used for Medicaid becomes available for another budget function, such as road construction or sports arenas. It is impossible to know whether states' Medicaid or health care budgets would be lower than they are today in the absence of these additional funds.

Another way to try to assess the claim that the additional funds help support state Medicaid programs is to examine whether states with UPL systems have broader Medicaid eligibility criteria than other states. We compared the Medicaid eligibility criteria for families in the states with approved UPL financing schemes to the criteria for states with no approved or pending UPL arrangements. Medicaid eligibility for families was actually a little higher in the states with no UPL systems than in the states with UPL systems. In states without UPL systems, the average income threshold for a family of three was 85 percent of the poverty line in the year 2000. In the states with UPL systems, the average threshold was 77 percent.(24)


How Might Safety Net Providers Be Affected?

The current, incomplete evidence suggests that UPL systems involving nursing homes have been used primarily to divert funds to state governments, while UPL systems that involve hospitals have tended to provide hospitals with additional resources. This suggests that efforts to limit UPL systems might harm some hospitals unless alternative sources of funding can be developed. Some discussions concerning the forthcoming HCFA regulations have focused on the reliance on UPL funds of California public hospitals and Cook County Hospital in Chicago.

HCFA will need to be cautious in regulating UPL systems that involve hospitals, as the current evidence suggests the hospital-based mechanisms have been less abused. Even so, the hospital-based UPL systems merit scrutiny for three reasons. First, even if UPL systems involving hospitals historically have helped hospitals, such systems could be structured in the future to divert more money to state governments, like the nursing home-based schemes. New UPL systems for hospitals need careful review.

Second, states have other methods to help hospitals, most notably through their Medicaid DSH programs. As shown in Table 1, the Urban Institute study indicated that in 1997 the state of California had a windfall of $376 million and Illinois of $168 million, secured through the manipulations of their DSH programs.(25) States could restructure their DSH programs so that more of the gains are directed to safety net hospitals, rather than being diverted to state coffers.

Third, it is not clear that additional funds provided to public hospitals are used to provide more health care; they might simply supplant other local funds. For example, a recent University of Chicago study analyzed hospital financial data from California for the years 1990 to 1995. It found that every additional dollar in DSH payments that public hospitals in California received was associated with a one dollar reduction in local government subsidies, so that "virtually none of the billions of dollars received by these facilities results in improved medical care quality for the poor."(26)


Taking Reasonable and Prudent Regulatory Action

HCFA is expected to issue a proposed regulation in the next few weeks and to complete the rulemaking by the end of this year. The proposed regulation should serve three important public policy purposes.

Given the history of the Medicaid DSH program, it seems reasonable to assume there eventually will be federal legislation in this area, even after HCFA issues its regulation. HCFA's regulatory solution is not the only possible mechanism to check the growth of these financing arrangements. In addition, both OIG and GAO have suggested there may be a need for Congressional action to help curtail questionable financing schemes.(27) OIG has recommended, for example, that states be required to demonstrate that additional payments actually are available to the facilities and that these funds are used to help patients. GAO has suggested that states should not be able to pay government-owned facilities more than the actual costs of care.

If Congress wishes to modify these rules in the future, it will have that legislative option. It can do so after it reviews the HCFA regulation. Since the regulation has not yet been issued and data about state UPL arrangements are so fragmentary, there are no sound estimates of the effects the regulation would have on specific hospitals. However, after the rule has been issued and during the transition period that HCFA has said it would provide, Congress could more carefully analyze the effects of the new rules and decide - before the rules are fully in effect - whether to modify the rules or to take some action to cushion the effects on certain providers. For example, if analyses indicated that specific safety net hospitals would be harmed by the rule, Congress could enact legislation that would provide subsidies to such providers in a more straightforward and accountable fashion than through the current UPL arrangements.

If the proposed rule is blocked now, however, it is likely that abuses will continue to spread, and it will become even harder to reel in the abusive financing practices in the future. We might therefore view the forthcoming HCFA regulation as the first step in a longer process of determining appropriate federal policy in this area. Letting HCFA act quickly to put regulations in place should stop the abuses from proliferating and give Congress time to act later if it so chooses.

End Notes:

1. Testimony of Kathyrn Allen, U.S. General Accounting Office before the Senate Finance Committee, Sept. 6, 2000. Testimony of Michael Mangano, Office of the Inspector General, Dept. of Health and Human Services, before the Senate Finance Committee, Sept. 6, 2000.

2. Testimony of Timothy Westmoreland, Director, Center for Medicaid and State Operations, HCFA, to the Senate Finance Committee, Sept. 6, 2000.

3. Based on the March 2000 Congressional Budget Office baseline. The extent to which the federal government matches state costs depends on the per capita income in each state. In wealthier states, the federal government pays 50 percent of the total cost. In poorer states, the federal share can rise as high as 83 percent.

4. Disproportionate share hospitals are those that serve a high proportion of Medicaid and low-income uninsured patients, as designated by the state Medicaid agencies, and therefore become eligible for special payments (DSH payments). Although the original legislative intent was to help safety net hospitals, many states designed their DSH policies to divert a large share of the funds to state coffers instead. As noted later, these abuses led to a series of legislative changes.

5. Jocelyn Guyer, Andy Schneider and Michael Spivey, Untangling DSH: A Guide for Community Groups to Using the Medicaid DSH program to Promote Access to Care, Boston MA: Access Project, 2000. Andy Schneider, Stephen Cha and Sam Elkin, "Overview of Medicaid DSH Provisions in the Balanced Budget Act of 1997," Center on Budget and Policy Priorities, Sept. 3, 1997. The 1997 Balanced Budget Act ratchets down the level of federal DSH funds that any state can receive from fiscal year 1998 through 2002. In this session of Congress, there are proposals to freeze DSH allotments at the 2000 levels rather than further reduce them.

6. These are conservative estimates based on data from 40 states. See Teresa Coughlin, Leighton Ku and Johnny Kim, "Reforming the Medicaid Disproportionate Share Hospital Program in the 1990s," Urban Institute, Jan. 2000, forthcoming in Health Care Financing Review.

7. Westmoreland, op cit. At this point, HCFA has not been able to determine a more rigorous estimate of the federal budget impact.

8. In addition to nursing homes and hospitals, these rules can be applied to residential institutions for people who are mentally retarded or who have developmental disabilities, but there are no known examples of such financing arrangements with regard to residential institutions.

9. UPL arrangements go by different names in different states. Some states call them "supplemental payment programs" because of the mechanism for making supplement payments to providers, while other states call these arrangements "intergovernmental transfer" programs because of the mechanism by which providers return funds to the state.

10. Privately-owned facilities are barred by federal law from making equivalent donations to the state Medicaid agencies.

11. Mangano, op cit.

12. Mangano, op cit. and Allen, op cit.

13. The noteworthy exception to this rule is that Medicaid DSH payments can be made above the upper payment limit for hospitals. Thus, hospitals may receive supplemental UPL payments as well as DSH payments.

14. Westmoreland, op cit.

15. Mangano and Allen, op cit.

16. Many states also look at state budgets excluding federal matching revenue, but might still have distorted apparent state Medicaid expenditures if they do not subtract the amount of intergovernmental transfer funds that are paid by health care providers.

17. Teresa Coughlin, Leighton Ku and John Holahan, Medicaid Since 1980: Costs, Coverage and the Shifting Alliance Between the Federal Government and the States, Washington, DC: Urban Institute, 1994, pages 91-97.

18. Coughlin, et al.,2000, op cit. One state responded to the survey, but did not provide data about its UPL system.

19. The state balance is its cumulative surplus, which may include Rainy Day Fund reserves.

20. Source: National Association of State Budget Officers, Fiscal Survey of States: August 2000.

21. Source: Tax Analysts. "State Tax Actions 1999," State Tax Notes, March 20, 2000. Positive numbers are tax increases, while negative numbers are tax cuts.

22. Source: National Conference of State Legislatures. State Policy Reports, 18(11), 2000.

23. Source: Coughlin, et al. 2000, op cit. The sum of gains by state hospitals and state "residual" gains.

24. In these comparisons we assumed that all the income was earned income.

25. In DSH, states can profit by either taking in more revenue from providers and the federal government than they spend in DSH payments or by making excess payments to state-owned hospitals. See Coughlin, et al. op cit.

26. Mark Duggan, "Hospital Ownership and Public Medical Spending," National Bureau of Economic Research Paper 7789, July 2000, and forthcoming, Quarterly Journal of Economics, Nov. 2000.

27. Mangano and Allen, op cit.