Of all federal formula grants, Medicaid is the largest. Accounting for more than 40% of federal grants, Medicaid cost the federal treasury roughly $78 billion in FY94. When state dollars and the federal match are combined, Medicaid expenditures have risen from $25 billion in 1980 to more than $150 billion this year, or nearly 15% of current U.S. health care spending. In the state houses over the last 15 years, Medicaid has grown from 9% of state budgets to about 20%. Medicaid's rapid growth is due to above-inflation increases in health care costs and to growth in the populations eligible for Medicaid.2
Under the current scheme, Medicaid provides federal matching funds to pay for between 50% and 83% of a state's expenses for providing health care to public assistance recipients and other medically-needy individuals. The average state receives 57% in federal matching funds. Per-patient spending on Medicaid services varies significantly from state to state. California has worked hard to control spending, while some other states have not. While greatly helping the state's budgetary situation, California's efficiency may pose serious problems if Medicaid is converted as-is to a block grant.
The current percentage reimbursement provided to any one state by the federal government depends on that state's per capita income relative to the national average.3 Higher income states receive lower reimbursement rates. California, New York, and 11 other states, receive the minimum 50% match because their per capita income is substantially above the national average. While a match as high as 83% is allowed by statute, that ceiling is not constraining any state -- the maximum state match at present is about 79% for Mississippi.
Of the $78 billion distributed amount in FY94, California received $7.1 billion or 9.1%, making it the second largest recipient state. The largest share went to New York, which received $10.3 billion or 13.1% of federal Medicaid dollars. However, the fact that New York's share was larger than California's was not due to a skewed matching formula -- both states are at the 50% match level. Rather it was due to the fact that New York's health care spending is much higher than California's.
In 1992, California served 4.5 million persons with $8.7 billion. That same year, New York served 2.6 million persons with $15.3 billion. Thus, California's per patient expenditures were barely one third that of New York's -- nearly $2,000 in California compared to nearly $6,000 in New York. The national average per-patient expenditure was slightly over $3,000. Because California's spending is lower, it receives fewer Medicaid dollars.
The General Accounting Office has criticized the current Medicaid distribution formula as inequitable4 and has recommended alternatives. It recommends shifting away from per capita income and instead using poverty figures. Such approaches -- most of which would significantly raise California's receipts -- will be discussed further below.
State-by-state differences in eligibility affects federal Medicaid receipts. While federal guidelines exist, eligibility for Medicaid coverage is currently determined by each state.5 Beneficiaries are divided into two groups -- those it must cover and those it may cover. For example, all states must cover individuals who already receive specific federal cash assistance.6 States' individual choices regarding coverage of those that may or may not be covered -- such as persons who, except for income and resources, would otherwise be eligible for cash assistance7 -- create much state-to-state variation in federal receipts. There are mandates and state options with respect to benefit levels as well -- states must provide a wide range of services to those they are required to serve, whereas benefit levels are generally lower for those states elect to but are not required to serve.
Long-term care (primarily nursing home8) expenditures account for one-third of Medicaid expenditures nationwide, but only 22% in California.9 Long-term care is very dependent on Medicaid. In 1993, Medicaid paid for more than half of all U.S. nursing home expenditures, and 61% of nursing home residents now rely at least in part on Medicaid.10 Persons may qualify for Medicaid after their medical expenses are deducted from their income or resources, -- only once much or all income is spent does the individual become Medicaid eligible. This process, know as 'spenddown', is commonly used to make nursing home residents eligible for Medicaid. California currently spends the least per capita of any state on Medicaid long-term care.11
One Medicaid cost control effort undertaken by California has been the shift to managed care. Nearly 1 million of California's 5.4 million individuals receiving Medi-Cal services now receive them through managed care providers; by the end of 1996, state officials hope that the managed care figure will rise to 3.4 million.12 Nationwide, 25% of Medicaid enrollees are in some form of managed care. In 1982, California was the first state to free its Medicaid program (Medi-Cal) and private insurance firms to negotiate with providers for the price of health care services.13
Proposals for Altering Medicaid
As Medicaid is currently an entitlement, the federal government allocates as many federal dollars for Medicaid as are required, and the program is not reauthorized periodically, though state programs are audited regularly for compliance with federal guidelines.
A number of possible Medicaid changes have been discussed recently, including a shift to a block grant approach accompanied by imposition of spending growth caps.
The House and Senate budgets currently under consideration each assume a significant reduction in the growth of Medicaid spending below levels currently anticipated. The Senate Budget Committee chairman's mark assumes lower federal Medicaid spending growth from 1996 to 2000 -- $780 billion over 7 years instead of the $955 billion predicted without a cap.14 The mark assumes average growth of 5% per year -- with faster growth in the first three years and slower growth thereafter. The mark suggests two alternative approaches for achieving the savings: (1.) a declining growth cap which would level out at 4% growth per year15 and (2.) an across-the-board matching rate reduction, cutting each state's matching rate by 18.7%.16 The House Budget Committee's mark assumes roughly the same change in anticipated spending ($770.6 billion) over the same 7-year period, urges converting Medicaid to a block grant, and assumes 4% growth by 2002.17
Several issues deserve particular attention by California as a shift of Medicaid to a block grant or another approach is discussed.
Nature of the Block Grant -- Congress could choose to create a single Medicaid block grant, or it could create several grants directed to specific populations (such as pregnant women or the elderly) or for specific services (such as long-term care or acute care).
Grant Basis and Base Year -- Of primary concern to California if a block grant approach to Medicaid is implemented would be the grant level. There has been discussion of freezing existing Medicaid expenditures at a certain base year or aggregate of several years. Thanks to aggressive control efforts, California has relatively low per-patient Medicaid spending. California and other low-expenditure states could be penalized for having squeezed out excess spending, and New York and other high-expenditure states could be rewarded for having not done so.
Should grants be frozen at a certain point in time, the pinpointing of that time or base year would be crucial. However, from California's standpoint, it would be greatly preferable to develop a different, more equitable allocation approach than to simply base the allocation on a prior year's level. For example, a more equitable allocation based on the number of persons below a certain income or poverty threshold would increase California's federal dollar allocation.
The Urban Institute found that basing block grants on current expenditures would freeze federal Medicaid payments per poor person (below 150% of poverty) at more than $2,000 for New York, Rhode Island, Massachusetts and Connecticut, compared to roughly $800 for California. The U.S. average was found to be about $1,000 per poor person.
On the other hand, the study suggested that "a more equitable allocation formula" such as basing grants on the number of poor persons in a state would yield vastly different results, and these results would be much better for California. California's federal allocation under such a scheme would increase 22%, from $6.8 billion currently to $8.2 billion. New York's allocation, on the other hand, would decrease 53%, from $9.8 billion to $4.6 billion. In general, the South, West, Midwest and Mountain states would gain under such a scheme, while states in the Northeast would lose.18
The best scenario for California would be to base the grant on numbers of patients actually served. While California's Medicaid expenditures in 1992 were 9.5% of the U.S. total, California served 14.8% of the total patients.19 Because of disparities with other states, however, building a broad base of support for this approach might be difficult.
Growth Caps -- Recent discussions have suggested the idea of a growth cap for Medicaid spending, perhaps at 4%, 5% or 8% per year. The Urban Institute predicts that, absent a change in policy, uncapped Medicaid expenditures would otherwise increase nationwide at an average annual rate of 9.8%, and in California at an average rate of 10.6%. If needs do grow at 10% and growth is capped at less than that, policy makers will need to consider if and how any resulting shortfall should be handled.
If growth is capped, care should be taken to ensure that California is not penalized for having kept spending low. If the current scheme is retained, states such as New York with more Medicaid "fat" to cut will have an easier time keeping to an expenditure-oriented growth cap than states which already run lean operations.20 An important question would be whether a cap is placed on growth in each state's allocation or on total U.S. Medicaid spending, with the latter being a better approach for growing states.
The Urban Institute study of state Medicaid spending variations concludes that block grants and expenditure growth caps would favor high income states. California, however, is an exception to that general rule. Despite California's relatively high income (ranking 12th of the 50 states), a block grant based on a FY95 level or a growth cap at a certain percentage would work against California's interests.
Eliminating the 50% FMAP Floor -- Under the current Medicaid formula, no state may have a matching percentage (FMAP) below 50%. If a state's FMAP would otherwise be below 50%, the formula raises it to 50%. One approach to reducing federal Medicaid spending discussed by the U.S. General Accounting Office and others, though less discussed during the current block grant debate, is elimination of the 50% floor for the FMAP.21 Eliminating the floor would reduce further the percentage reimbursement to states which theoretically are better able to pay for services themselves.
While California is at the 50% mark because its per capita income is relatively high, elimination of the 50% floor would not currently affect California as much as some other 50%-level states because California is very near that threshold. In fact, California would be above that threshold if the most current income data used were used; the 50% mark is retained for California only because the formula is based on a 3-year per capita income average. Several richer states (Connecticut, New Jersey, New York, Massachusetts, etc.) would be impacted significantly by a reduction in the 50% floor because their floor-less FMAP would be considerably lower. (If California's per capita income were to rise again compared to the rest of the nation,22 the removal of the 50% floor could have negative implications for the state at that stage.)
Acute Care vs. Long-Term Care -- Some have proposed creation of a separate block grant for acute care and another for long-term care. Because California now has the lowest long-term care spending per capita, a grant based solely on current spending levels would put the state at a severe disadvantage.23 Again, a per patient or per poor person approach would be preferable.
Disproportionate Share Hospital (DSH) Payments -- Medicaid provides for extra assistance to hospitals which have a high percentage of Medicaid-eligible patients. According to the California Association of Hospitals and Health Systems (CAHHS), 17% of California's hospitals provide 90% of Medi-Cal hospital services.24 California is currently under a Congressionally-imposed 12% cap on the share of its Medicaid dollars which may be distributed as DSH allotments.25 This places a $2.2 billion cap on its Medi-Cal DSH payments. The state restricts DSH payments to facilities serving roughly one-third or more Medi-Cal eligible patients. This differs from states which are more liberal with DSH payments, and sometimes permit DSH payments to hospitals with as few as 1% of patients eligible for Medicaid. An Urban Institute study found that an expenditure growth cap would have a somewhat smaller impact on states with high DSH payments because they have already been subject to the 12% DSH payment cap.26
Title XIX Requirements -- A number of states have complained about the burden of various federally-imposed requirements associated with the Medicaid program. If the program is altered, there will likely be a re-examination of these requirements. Under one such requirement, known as the Boren Amendment, a state must pay institutional providers rates which are "reasonable and adequate" for an efficiently and economically operated facility. Since a 1990 Supreme Court case established standing to sue, many hospitals and nursing homes have sued for higher, more "reasonable" reimbursement rates.27 (Medicaid reimbursements to providers are typically well below similar private payments.) If expenditures are capped and federal dollars restricted, there will be pressure to reduce the services mandated by the current Medicaid laws, as well as to further experiment with managed care and alternative reimbursement systems not currently permitted.
Treatment of Legal Immigrants -- As has been discussed with respect to various assistance programs, some proposals would exclude legal immigrants from receiving federal benefits from Medicaid, as well as other programs for low-income individuals. California is home to between 30% and 40% of the nation's legalized immigrants, so any restrictions of benefits could disproportionately impact the state. Moreover, more than 50% of immigrants who receive SSI or AFDC live in California.28
Treatment of Illegal Immigrants -- The Immigration and Naturalization Service estimates that 43% of the nation's undocumented immigrants reside in California, though precise figures are difficult to pinpoint. The Governor's Office estimates that in 1995-96, more that 300,000 illegal immigrants will be provided emergency- and pregnancy-related health care services in the state, at a cost of $395 million. The Urban Institute estimates that such costs in 1993 ranged from $133 million to $166 million.
Welfare Reform and AFDC Population Changes -- Eligibility requirements that change because of pending welfare reform legislation may impact numbers of persons served under Medicaid, regardless of whether Medicaid is shifted to a block grant. Thus, changes to the Medicaid program should be considered in concert with welfare reform.
Two Recent Formula Change Proposals-- The GAO studies referenced previously combined several of these factors and devised an alternative formula which would have increased California's Medicaid allocation considerably.29 During the 103rd Congress, Sen. Connie Mack (R-FL) and Rep. Karen Thurman (D-FL) proposed this formula in S. 1266 and H.R. 4047. The formula lowered the FMAP floor to 40%, switched to a poverty count from a per capita income count, and proposed a "total taxable resources" factor reflecting the ability of a state to pay for services itself. (It is important to note that the poverty factor and the floor drop would help California, while a total taxable resources factor would hurt California.) According to GAO, the resulting formula would have raised California's FMAP to 56.9%, increasing the state's 1991 allocation by $556 million or 12.8%. The GAO study also examined three other formula revisions which would have raised California's receipts by $344 million (7.9%) to $599 million (13.9%).
But better still for California would be basing allocations purely on numbers of poor persons. Again, the Urban Institute study referenced previously30 estimates 1993 federal allocations for California would have increased 21.8%, or nearly $1.5 billion -- from $6.8 billion to $8.2 billion -- if a poverty count (150% of poverty line) had been used.
Conclusions
Whether Medicaid is shifted to a block grant or not, there are several California-oriented issues which should be considered during the debate.
Some Medicaid revision proposals could penalize California for having been conscientious and efficient in managing its Medi-Cal program. Basing state block grant allotments at current levels could freeze in place inequitable disparities between states that would badly disadvantage California.
While formula rewrites are always difficult, it would be desirable to revise the Medicaid funding distribution mechanism. Better formulas for California would base federal funds on numbers of patients served or on the state's share of poor residents. A formula based on numbers of poor persons would shift funding primarily away from Northeastern states and toward California, and states in the South, West and Midwest.
With California expenditures estimated to grow at a rate above the national rate over the next decade, a low growth cap could further skew funding away from California.
Efforts to reduce or eliminate immigrants' benefits disproportionately impact California.


1 Information sources include The Urban Institute, The Kaiser Family Foundation, Federal Funds Information for States (a service of the National Governors Association and the National Conference of State Legislatures), National Association of State Budget Officers, U.S. General Accounting Office, Health Care Financing Administration, U.S. Bureau of Economic Analysis, Congressional Research Service, U.S. Census Bureau, U.S. Office of Management & Budget, California State Department of Finance, California Association of Hospitals and Health Systems, American Association of Retired Persons, various Congressional committees, California Institute.
2 Federal Funds Information for States reports that health costs rose at a 6.7 percent annual rate during the 1980s compared to a 4.6 percent average inflation rate. Families receiving AFDC increased from 3.6 million in 1982 to 5 million in 1994, while SSI recipients increased from 3.8 million to 5.8 million during the same period. Both groups are Medicaid eligible.
3 The Medicaid federal share for a state is currently determined by squaring the ratio of the national per capita income to the state's per capita income (PCI), with both PCI figures being based on a three-year average. The algebraic formula as delineated by the GAO (in Grant Formulas, A Catalog of Federal Aid to States and Localities, HRD-87-28, 1987, p. 119) reads:
Federal Share = 1 - [ .45 * (STATEPCI/USPCI)_ ]
4 See GAO reports: Medicaid Formula: Fairness Could Be Improved, GAO/T-HRD-91- 5, December 1990, and Medicaid: Alternatives for Improving the Distribution of Funds to States, GAO/HRD-93-112FS, August 1993, U.S. General Accounting Office.
5 According to OMB's Catalog of Federal Domestic Assistance, beneficiary eligibility generally includes "low-income individuals over age 65, blind or disabled individuals, members of families with dependent children, low-income children and pregnant women, certain Medicare beneficiaries and, in many States, medically needy individuals".
6 These include all persons receiving AFDC and most persons receiving SSI. In addition, states must by statute cover certain other individuals, including pregnant women, infants and children up to age six with family incomes up to 133% of the poverty level; certain persons within AFDC standards but who fail to qualify for AFDC for other reasons; families and disabled persons recently losing AFDC benefits due to increased employment; and persons losing SSI because of Social Security benefit increases. Medicaid, Issue Brief, Congressional Research Service, Education and Public Welfare Division, updated regularly.
7 Often these persons have incomes somewhat (perhaps one-third) above cash assistance eligibility thresholds but are also members of another targeted category, such as being of a certain age, having a disability, or being a member of a family with dependent children. Id.
8 In 1993, 62% of Medicaid long-term care expenditures were for nursing facilities, 22% were for intermediate care facilities for the mentally retarded (ICFs/MR), and the remaining 16% were for home and community-based, personal, and home health care. Ibid., quoting data from the Health Care Financing Administration.
9 In 1995, FFIS estimates California will spend $3.7 billion of its $16.9 billion Medi-Cal expenditures on long-term care, while the nation as a whole will spend $53.4 billion of its $155.1 billion Medicaid total. Issue Brief: Recent Trends in Medicaid Spending, Federal Funds Information for States, December 16, 1994.
10 Long-Term Care: Measuring the Impact of a Medicaid Cap, Public Policy Institute of the American Association of Retired Persons, April 1995. See also "National Health Care Expenditures, 1993," in Health Care Financing Review, Fall 1994, Vol. 16, No. 1, p. 264.
11 California spends $75 per capita on long-term care, while New York spends $496. State Variations in Medicaid: Implications for Block Grants and Expenditure Growth Caps, Policy Brief, The Kaiser Commission on the Future of Medicaid, March 1995, pp. 18, 22.
12 See Medicaid Managed Care: More Competition and Oversight Would Improve California's Expansion Plan, GAO/HEHS-95-87, U.S. General Accounting Office, April 1995. See also Statement of Leslie A. Margolin, President, CIGNA HealthCare of California, before the California State Senate Health & Human Services Committee, May 3, 1995.
13 See A Case History: California Health Care, by Margaret R. Campbell for the California Institute for Federal Policy Research, September 21, 1993, p. 3.
14 FY 1996 Balanced Budget Resolution Chairman's Mark, Republican Staff of the Senate Budget Committee, May 9, 1995.
15 Specifically, the mark suggests that the required savings could be achieved by restructuring Medicaid such that federal payments would be capped at 8% growth in FY96, 7% in FY97, 6% in FY98, 5%in FY99, and 4% in FY 2000 and thereafter. See page 550-4.
16 The mark notes that further programmatic changes would be required to slow the rate of growth in later years with the across-the-board reduction method.
17 The Budget Resolution for Fiscal Year 1996, House Committee on the Budget, May 10, 1996. While this document shows 4% per year growth, other committee documents reportedly assume even slower rates. One report indicates the committee actually assumes 1.9% per year growth from 2001 to 2002 and thereafter. See A Medicaid Block Grant is Likely to Lead to an Inequitable Distribution of Federal Funds, Center on Budget and Policy Priorities, May 17, 1995, p.1.
18 The only states not located in the Northeast which would lose under such a scheme would be South Carolina (-7.5%), West Virginia (-27.5%), Kentucky (-12%), Louisiana (-41%), Indiana (-3.9%), Minnesota (-4.7%), Alaska (-5.1%), and Washington (-13.7). The only state located in the Northeast which would gain would be Pennsylvania (+15.1%). The scheme shows 19 states plus D.C. losing funding and 30 states gaining funding (Arizona does not participate).
19 U.S. Health Care Financing Administration, unpublished data, as reported in Statistical Abstract of the United States, 1994, U.S. Department of Commerce, September 1994, p. 115.
20 The Urban Institute described the effect of locking in current levels as a block grant as follows: "States that have spent large amounts on a per capita or per poor person basis would continue to be able to do so, regardless of whether high-spending levels reflect a high level of need or simply inefficient program administration. It would also penalize those states that have successfully controlled expenditures -- they would face the same caps as states which have had more rapid historical growth. Urban Institute, Ibid.
21 See two GAO reports: Medicaid Formula: Fairness Could Be Improved, GAO/T- HRD-91-5, December 1990, and Medicaid: Alternatives for Improving the Distribution of Funds to States, GAO/HRD-93-112FS, August 1993, U.S. General Accounting Office.
22 California ranked 3rd in per capita income in 1980, compared to 12th in 1993 according to the Survey of Current Business, U.S. Bureau of Economic Analysis, 1994. Unpublished reports from the Tax Foundation, a nonprofit organization in Washington, D.C. which analyzes tax issues, predicts California's 1995 PCI rank may fall to as low as 18th.
23 The Urban Institute estimates that, under such a scenario, New York's payment from a long-term care block grant would be between 6 and 7 times that of California's. Ibid., p. 18.
24 Issue Brief: Medicaid Cuts Would Undermine Tattered Safety Net Program, California
Association of Hospitals and Health Systems, 1995, p.2.
25 California and 21 other states had DSH payments exceeding 12% of their total Medicaid payments in FY 1994. The highest share was New Hampshire's, at 42%. California's DSH share has been declining for several years since the 12% freeze was imposed. In FY 1994, California's DSH share was 14% and the U.S. average was 13.2%.
26 Policy Brief: The Impact of a Five Percent Medicaid Expenditure Growth Cap, A State Level Analysis, The Urban Institute for the Kaiser Commission on the Future of Medicaid, March 1995, p.3.
27 By mid-1992, Boren amendment lawsuits had been filed in 16 states. Medicaid Sourcebook, Congressional Research Service, January 1993, as quoted in Long Term Care, Public Policy Institute of the American Association of Retired Persons, infra.
28 Welfare Reform: Implications of Proposals on Legal Immigrants' Benefits, U.S. General Accounting Office, GAO/HEHS-95-58, February 1995.
29 Medicaid Formula: Fairness Could Be Improved, GAO/T-HRD-91-5, December 1990, and Medicaid: Alternatives for Improving the Distribution of Funds to States, GAO/HRD- 93-112FS, August 1993, U.S. General Accounting Office. See also a GAO letter to Senator Connie Mack (R-FL), GAO/HRD-93-17R, Medicaid Formula Alternative, March 2, 1993.
30 State Variations in Medicaid: Implications for Block Grants and Expenditure Growth Caps, Policy Brief, The Kaiser Commission on the Future of Medicaid, March 1995.