Medicaid Spending: A Brief History.

Medicaid spending growth has varied greatly over time. This article uses financial and statistical data to trace the history of Medicaid spending in relation to some of the major factors that have influenced its growth over the years. Periods of varying growth are divided into eight "eras," ranging from program startup in 1966 through the post-welfare reform period. Average expenditure and enrollee growth for each era are presented and briefly discussed. Finally, some factors are mentioned that are likely to affect future growth in the Medicaid program.

Introduction

From less than $1 billion in 1966, Medicaid has grown to a program whose expenditures are expected to top $200 billion in fiscal year (FY) 2000 (Health Care Financing Administration, 2000). During the same period, enrollment has increased from 4 million to 33 million, and per-enrollee spending from less than $200 to more than $6,000. Medicaid spent about $4 per U.S. resident in 1966 and will spend nearly $750 per resident this year. This article reviews the history of Medicaid spending in relation to major events that have driven its growth in various “eras.” The approach used is adapted from that found in Muse et al. (1985). The need for brevity necessitates omitting mention of many important aspects of Medicaid's history, some of which are discussed elsewhere in this issue of the Review. Two excellent sources of pertinent information on factors affecting Medicaid spending over the years are Congressional Research Service (1993) and Coughlin et al. (1994).

Data Sources

Expenditures in this article have been derived from Medicaid Financial Management Reports (Form HCFA-64 and its predecessors). These forms have been in use since the inception of the Medicaid program and represent the most complete and accurate source of information on Medicaid spending. Expenditures are on a total computable cost basis, (i.e., both Federal and State shares are included) and include benefits and administrative costs.

Enrollment data presented here are taken from annual Medicaid Statistical Reports (Form HCFA-2082) for the period 1975-1998. Earlier data on Medicaid enrollment are derived from information found in Institute for Medicaid Management, (1978) and internal HCFA documents. Enrollee data have been adjusted to a full-year-equivalent (person-year) basis, which takes into account the number of months a person is enrolled during the year (e.g., one person enrolled for 6 months is counted as one-half a person-year.) Since many persons are enrolled for less than the full year, the person-year measure is smaller than measures based on unduplicated counts of individuals ever enrolled during the year (called “eligibles” in Form HCFA-2082).

All years cited refer to the Federal FY as currently defined (October 1 – September 30), and all data have been converted to this basis.

Eras of Medicaid Spending History

As Figure 1 demonstrates, Medicaid spending over the years has followed a typical “exponential” growth pattern, with periods of both faster and slower growth relative to the long-term trend. Following Muse et al. (1985) these periods of varying growth have been divided into “eras,” which are briefly discussed. Components of growth rates during these eras are summarized in Table 1.

Figure 1

Medicaid Expenditures and Enrollment 1966-1999

Table 1

Medicaid Expenditure Growth, by Era

Era	Description	Annual Compound Rate of Growth
		Total Expenditures	Enrollees	Price Inflation1	Growth in Expenditures per Enrollee in Excess of Price Inflation
		Percent
1966-1971	Program Startup	52.3	32.2	4.0	10.7
1972-1976	Early Amendments	17.9	4.9	6.5	5.5
1977-1981	Medical Inflation	14.8	-0.7	8.4	6.7
1982-1984	Retrenchment	7.8	-0.3	4.5	3.4
1985-1990	Program Expansion	11.8	2.5	3.8	5.2
1991-1992	Taxes and Donations, DSH	27.3	12.2	3.4	9.7
1993-1996	Experimentation	7.9	3.6	2.2	1.9
1997-19992	PRWORA, BBA	5.6	-0.4	1.6	4.4

Measured by the gross national product implicit price deflator.

Statistics for 1997-1999 do not include State Children's Health Insurance Program.

NOTES: DSH is disproportionate share hospital. PRWORA is Personal Responsibility and Work Opportunity Reconciliation Act of 1996. BBA is Balanced Budget Act of 1997.

SOURCES: Expenditures: Medicaid Financial Management Reports (HCFA-64 and predecessors). Enrollment: 1966-1974 (Institute for Medicaid Management, 1978); 1975-1998 Medicaid Statistical Reports (HCFA-2082); 1999 projections (Health Care Financing Administration, 2000).

Program Startup (1966-1971)

The growth of Medicaid during the first 6 years of its existence is typical of most State-based programs at their inception. A number of States implemented programs immediately while others needed several years to get underway. By 1971, annual spending had reached $6.5 billion, and enrollment had topped 16 million. Initial projections of Medicaid forecast less than one-half of this spending level, primarily because analysts greatly underestimated the extent to which States would offer coverage of optional eligibility groups—especially the medically needy—and optional services. Enrollment growth also greatly exceeded original expectations.

As shown in Table 1, expenditures increased by more than one-half, on average, each year during the startup period, while enrollment grew at an average annual rate of nearly one-third, reaching by 1971 almost one-half of what it would be at the end of the century. Moreover, the rapid growth in covered services resulted in per-enrollee growth that exceeded economywide inflation by nearly 11 percentage points.

Early Amendments (1972-1976)

The next 5 years of Medicaid's history were heavily influenced by major amendments to the Social Security Act (SSA) that were passed by Congress in late 1971 and 1972. The 1972 amendments created the Supplemental Security Income (SSI) program, which federalized existing State cash assistance programs for aged and disabled persons. Nearly all beneficiaries of SSI also receive Medicaid coverage, and the outreach efforts undertaken with the implementation of SSI resulted in significant increases in enrollment among the aged and disabled in Medicaid, averaging nearly 8 percent per year during the period.

The 1971-1972 amendments also added as optional Medicaid covered services intermediate care facilities for the mentally retarded (ICF/MR) and inpatient psychiatric services for beneficiaries under age 22. Residents of these facilities, and the disabled in general, are among the most expensive groups in Medicaid.

Taken together, the 1971-1972 amendments contributed to total expenditure growth averaging 18 percent per year during the 1972-1976 period. Driven by the growth in enrollment of persons with disabilities, total Medicaid enrollment grew at an average rate of almost 5 percent per year, and by 1976 it had reached 20.7 million, a level from which it would not vary by more than a few percent for the next decade.

Medical Inflation (1977-1981)

The period of the late 1970s was marked by sharp increases in economywide inflation and even higher increases in medical prices. General inflation rose at an annual average of 8.4 percent during the 1977-1981 period, peaking at nearly 11 percent in 1980. At the same time, there were no significant legislative expansions of Medicaid eligibility or services during this period, and welfare caseloads were stable or declining. Although Medicaid enrollment actually declined by an average of 0.7 percent per year between 1976 and 1981, annual Medicaid expenditure growth averaged nearly 15 percent.

Retrenchment (1982-1984)

The tremendous growth of the previous decade led Congress and the Reagan Administration to consider ways to reign in Medicaid spending. Administration attempts to place caps on the program failed to pass Congress. However, in the Omnibus Budget Reconciliation Act of 1981 (OBRA-81), Congress did institute a 3-year reduction in Federal financial participation, cutting Federal matching rates by 3.0, 4.0, and 4.5 percentage points in FYs 1982, 1983, and 1984, respectively, for States whose growth exceeded certain targets. OBRA-81 also reduced eligibility for welfare benefits, thus making it harder for poor families to qualify for Medicaid.

To help States cope with reductions in Federal support, Congress enacted a number of flexibility provisions, which broadened State options for providing and reimbursing Medicaid benefits, as well as State authority to limit coverage under medically needy programs. In response, many States began to experiment with alternative delivery and reimbursement systems, such as health maintenance organizations (HMOs) and other capitated programs, home-and-community-based waiver programs, and prospective hospital payment. The focus in Medicaid began to change from merely paying claims to managing services and the cost of care as well. As a result of these changes and a drop in inflation pressures (general price increases averaged about 4.5 percent annually, about one-half the rate of the previous era) Medicaid expenditures grew at an annual average rate of less than 8 percent between 1981 and 1984, while Medicaid enrollment remained stable with an annual average of just under 20 million.

Program Expansion (1985-1990)

With continuing improvements in the economy and concern among policymakers that OBRA-81 may have spawned program contractions that were too harsh, Congress embarked in 1984 on a series of Medicaid expansions that continued each year through the end of the decade. The expansions affected nearly the entire spectrum of Medicaid enrollees from infants, children, and pregnant women to low-income Medicare beneficiaries, and other aged and disabled enrollees. Initially, States were offered options to expand coverage of these groups, but ultimately most of the options were converted by subsequent legislation into mandates, most notably in the Medicare Catastrophic Coverage Act of 1988 (MCCA). It was hoped that the increase in Medicare coverage of elderly and disabled persons under MCCA would help to offset part of the increased cost of the Medicaid mandates included in the bill. However, the Medicare provisions of the MCCA were repealed within a year, before any Medicaid savings impact could be realized.

Historically, Medicaid eligibility for low-income families had been linked to receipt of cash assistance under Aid to Families with Dependent Children (AFDC). The legislation of this era began to weaken this link by specifying eligibility criteria based on income in relation to Federal poverty guidelines. For infants, children, and pregnant women, this legislation introduced income-eligibility levels that were significantly higher than most States' AFDC payment levels and that were, unlike AFDC levels, indexed to the cost of living. For the low-income aged and disabled, similar poverty-based income thresholds were put in place, with benefits ranging from the full Medicaid package (which has remained optional with States) to coverage of just Medicare premiums and/or cost sharing (mandatory).

Besides these basic eligibility expansions, the 1984-1990 period saw the enactment of many other pieces of legislation, too numerous to mention here, that affected Medicaid eligibility, coverage, and reimbursement. A comprehensive treatment of these can be found in Congressional Research Service (1993).

Many of the expansions introduced between 1984 and 1990 were subject to delayed effective dates or phase-in provisions. (Coverage of children below the poverty level, for example, is still phasing in and will not be complete until 2002.) Thus, the full effect of this era's expansions was not felt during the period. Average annual caseload growth, which turned positive again at 2.5 percent per year between 1984 and 1990, jumped to over 12 percent in the following 2 years and continued to increase steadily through the mid 1990s (Figure 1). There were similar delayed impacts on Medicaid expenditure growth, which increased from the previous 3-year period to an average of 11.8 percent per year during 1984-1990, but the stage had been set for even greater growth in the 2 years that followed.

Taxes and Donations and DSH (1991-1992)

Perhaps no era in Medicaid's history has presented more dilemmas for policymakers, budget officials, and estimators than the short period from 1991 to 1992. The mandates of the previous era, the recession, and other factors all combined to put pressure on already strained State budgets, most of which were running deficits by 1991 or 1992. Increasing Medicaid caseloads (average annual growth of 12 percent) and mounting expenditures prompted some States to turn to alternative financing mechanisms, which relied on disproportionate share hospital (DSH) payments, combined with the use of provider donations or provider-specific taxes as sources of the State share of Medicaid spending.

Medicaid DSH payments, which were designed to help hospitals with a high proportion of low-income and Medicaid patients defray the impact of low reimbursements and uncompensated care, were required by law and, more importantly, not subject to the Federal limits that applied to all other types of Medicaid reimbursement. Thus a State could, if it wished to do so, increase DSH payments to a provider to any level it might choose, recoup the increased payment through a donation from or tax on that provider, and thereby receive essentially unlimited Federal matching funds with little or no increase in net State spending. By 1992, DSH payments had grown to more than $17 billion, or more than 15 percent of total Medicaid spending, and provider tax and donation programs were accounting for about $8 billion in State revenues (Coughlin et al., 1994). More than 30 States had or were planning to put provider tax or donation programs in place.

Concern over State efforts to shift costs to the Federal Government, and a desire to resolve the disputes that had arisen over the Administration's attempts to impose regulatory restrictions on tax and donation programs, led Congress in November 1991 to enact Public Law 102-234, the Medicaid Voluntary Contribution and Provider-Specific Tax Amendments of 1991. This legislation outlawed the use of most provider donations and restricted provider tax programs to those that were “broad based” and did not hold providers “harmless” for their tax payments. Moreover, it placed a statutory aggregate cap on DSH payments at 12 percent of Medicaid spending.

Medicaid spending growth, which averaged over 27 percent per year between 1990 and 1992, slowed considerably in the years following the enactment of Public Law 102-234, although DSH payments remain a significant share of total Medicaid spending.

Experimentation and Reforms (1993-1996)

The years that followed the cost explosion of the early 1990s saw the growth of a number of Medicaid reform efforts and experiments on the part of States. These included increased use of managed care and statewide health reform demonstrations under Section 1115 of the Social Security Act. By the end of 1996, more than 24 States, accounting for over 60 percent of Medicaid spending, had demonstration projects that were either approved or pending. This period also saw an improving economy, along with moderating price inflation (just 2.2 percent per year) and decelerating Medicaid caseload growth (averaging 3.6 percent, or about 30 percent of the previous era). Overall, Medicaid expenditure growth averaged less than 8 percent per year.

The slowdown in spending growth, however, did not come soon enough to deter congressional proposals to convert Medicaid into a block grant program. In 1995, Congress considered establishing the “Medigrant” program, which would have ended the Federal Medicaid entitlement and capped Federal matching funds. Though this provision was not adopted, the prospect of a capped program led States to accelerate spending in FY 1995, which was to be the base year for calculating the block grants (U.S. General Accounting Office, 1997). The resulting increase in 1995 expenditures contributed to a growth rate of less than 2 percent in 1996, the lowest one-year growth rate in Medicaid's history.

Welfare Reform and the Balanced Budget Act (1997-1999)

In 1996 and 1997, Congress passed two pieces of legislation that had significant impact on Medicaid. The Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (more informally known as “welfare reform”) effectively decoupled Medicaid from cash assistance for low-income families by replacing AFDC with a block grant program known as Temporary Assistance for Needy Families. Families meeting the requirements for assistance under the old AFDC rules continued to be eligible for Medicaid, although there is evidence that many such families did not retain their Medicaid benefits (Garrett and Holahan, 2000).

In 1997, Congress passed the Balanced Budget Act (BBA). Along with other provisions, the BBA gave States the option of setting up Medicaid managed care programs without the waivers that were usually required for such programs. More than one-half of all Medicaid enrollees are currently in some form of managed care program. The BBA also placed further restrictions on DSH spending. However, the most significant provision of the BBA from Medicaid's perspective established the State Children's Health Insurance Program (SCHIP), which authorized nearly $40 billion in Federal funding over 10 years (1998-2007) to provide health coverage to low-income children who did not qualify for Medicaid. States can use SCHIP monies to fund coverage of children through expansions of their Medicaid programs or through separate State programs under a new Title XXI of the Social Security Act. At present, about 40 percent of SCHIP funds are being spent under Medicaid. During FY 1999, more than 2 million children were enrolled under the combined Medicaid and separate SCHIP programs. (Note: The statistics on Medicaid growth in this era do not include the SCHIP program.)

The effects of welfare reform and a thriving economy resulted in 3 straight years of caseload drops in Medicaid (1996-1998), averaging about 0.4 percent per year. At the same time, annual expenditure growth slowed to the lowest levels of any era in the program's history, averaging 5.6 percent in 1997-1999. However, when the decreasing caseloads and general price inflation are factored out, real per capita Medicaid spending growth shows an upsurge since 1996, averaging 4.4 percent compared with less than 2 percent in the previous era (Table 1).

Future Trends

As this article shows, the factors that have driven Medicaid spending over the years have varied greatly from one era to the next, resulting in extreme variation in spending growth over time. This variation can generally be expected to continue into the future as new factors come into play. Factors that are likely to figure prominently in Medicaid's future growth include the following:

The cost of long-term care. Long-term care expenditures in Medicaid (institutional and community-based services) have steadily decreased as a share of total spending over the last 10 years or so—from about 45 percent in the late 1980s to 35 percent today—but can be expected to increase again as the baby boom generation ages.

The cost of prescription drugs, which averaged 15 percent annual growth during the most recent era and is approaching 10 percent of total Medicaid spending. These costs, like those of long-term care, can be expected to continue to be a significant factor in Medicaid spending as a result of the aging of baby boomers.

Managed care. The option to provide Medicaid coverage through HMOs and other types of prepaid health plans without a waiver is likely to result in even greater use of managed care in the future. Premiums for these plans currently account for about 15 percent of Medicaid spending and could exceed 20 percent within a few more years if present trends continue.

Medicaid “maximization.” Federal matching programs have always been popular with States; other things being equal, States would rather invest one dollar where it will do two dollar's worth of good. The availability of Federal Medicaid matching has thus led States over the years to adopt innovative strategies designed to obtain the greatest possible Federal funds. This was most noticeable during the taxes and donations and DSH era. Opportunities for maximization are likely to present themselves in the future and could again result in a sudden and unpredictable escalation of Medicaid spending.

Accounting for these and other factors will present a challenge to policymakers and estimators of Medicaid as they attempt to chart the course of the program into the 21st century.