Manhattan Institute for Policy Research.
Subscribe   Subscribe   MI on Facebook Find us on Twitter Find us on Instagram      


Issues 2012

No.24 September 2012


Improving Health, Decreasing Costs

Paul Howard, Senior Fellow
Manhattan Institute.



Executive Summary

Medicaid, the joint state-federal health-care program for the poor, presents a large and rapidly growing fiscal burden for states, crowding out spending on other vital public programs. One significant Medicaid policy reform proposed by Republican presidential candidate Mitt Romney and his running mate, Paul Ryan, would block-grant federal support for Medicaid in return for broad state flexibility in designing and implementing Medicaid coverage.

We consider criticisms of Medicaid block grants and suggest a block-grant design (based on bipartisan block-grant welfare reforms enacted in 1996) that encourages states to experiment with Medicaid reforms to help control costs while improving patient outcomes. Federal block-grant goals should encourage states to:

  • Design coverage and care arrangements that improve health outcomes for Medicaid recipients through a variety of care options and benefit structures (including co-pays and preferred provider network designs).
  • Coordinate and deliver care in the most cost-effective fashion, with an emphasis on preventive care and wellness programs that encourage patients to take more responsibility for maintaining good health.
  • Target federal resources in relation to a state’s relative population of low-income and disabled, adjusted for cost of living (this would ensure that poorer states with relatively larger poor and disabled populations would receive more federal aid than wealthier states).
  • Encourage the uptake of high-quality private insurance.

Introduction: Making the Case for Block-Granting Medicaid

Medicaid spending is spiraling out of control in states and crowding out other critical investments. Largely, this is because Medicaid’s open-ended financing structure (on average, the federal government pays about 60 percent of Medicaid costs, with states paying the rest) rewards wealthier states for spending more but makes it very difficult for any state to restrain Medicaid spending, since they essentially have to cut $2 of Medicaid spending to reap $1 in state savings.

Overly prescriptive federal Medicaid rules also make it cumbersome and complicated for states to experiment with innovative and cost-effective ways of operating their Medicaid programs that can both reduce costs and improve on the historically poor health associated with the program. Before states can make any systemic changes to their Medicaid programs, they must secure permission from the federal Centers for Medicare & Medicaid Services (CMS) through an elaborate and time-consuming waiver process. States are also limited in their ability to manage Medicaid expenses by the Patient Protection and Affordable Care Act’s (PPACA) Maintenance of Effort (MOE) provisions, which have prevented them from making any change in Medicaid eligibility requirements since PPACA was passed in 2010.

Often, the only tool that states can use to restrain spending (particularly during economic downturns) is to slash reimbursements to providers—leaving Medicaid enrollees struggling to find doctors who accept Medicaid coverage. And delayed or denied health-care access leads to worse health-care outcomes for America’s most vulnerable citizens.

PPACA is poised to radically expand Medicaid coverage beginning in 2014—adding to an already crushing state burden of Medicaid costs. The “perfect storm” of Medicaid overspending, excessive regulation, and PPACA expansion is leading to serious discussions among governors and state Medicaid directors about block-granting federal Medicaid funding in return for significant state flexibility in Medicaid program design and administration. Block grants would shift Medicaid financing from an open-ended federal entitlement to a fixed level of federal funding, in return for broad state flexibility in how those funds are spent.

A notable proponent of a Medicaid block grant is House budget chairman and Republican vice-presidential nominee Paul Ryan. Chairman Ryan’s Medicaid proposal (broadly embraced by the Romney campaign) entails block-granting the federal share of Medicaid payments (which would grow more slowly than historical rates) in return for enhanced state flexibility. Critics of block grants—and, in particular, of the Ryan block-grant proposal—argue that it would inevitably harm the nation’s poorest citizens by forcing states to radically reduce future Medicaid spending. History—as well as current Medicaid experiments—suggests that these concerns are unwarranted, or at least could be addressed through a careful block-grant design.

Block Grants: A Short History

Block grants as a solution to the challenges posed by fragmented public programs like Medicaid are far from a new, or even radical, idea. In 1949, the Commission on the Organization of the Executive Branch suggested that “a system of grants should be established based upon broad categories—such as highways, education, public assistance, and public health—as contrasted with the present system of extensive fragmentation.” As part of the Omnibus Budget Reconciliation Act (OBRA) of 1981, Congress consolidated 50 federal aid programs into nine consolidated block grants, giving states more program flexibility and control over how federal funds were spent.

A 1995 Government Accountability Office (GAO) report found that the OBRA block grants were generally successful, especially in program areas where the states already had extensive experience managing local programs. GAO suggested three lessons learned from these early block-grant experiments: focus federal rules on broad accountability for results; adjust funding to reflect “need, ability to pay, and variations in the cost of providing services”; and carefully transition complex categorical programs to block grants.

Lessons from these earlier block grants were later applied to a much larger block-grant program: welfare reform. Here, governors played a major role in designing and implementing prototype state welfare-to-work programs and using the lessons learned to inform federal welfare-reform legislation: the Temporary Assistance for Needy Families (TANF) program in 1996, which turned the operation of the nation’s cash welfare program over to the states with limited (but clear) federal purposes and oversight. Welfare reform is certainly the nation’s most successful, and highest-profile, block-grant program, and advocates of Medicaid block grants have much to learn from that example.

Criticisms of Medicaid block grants today are the same criticisms that were leveled at TANF in 1996: that it would cause a “race to the bottom” in levels of cash assistance, greatly increase poverty, and lead to significant suffering among poor families. These concerns proved unfounded, as the TANF program is currently recognized as one of the nation’s top bipartisan success stories.

States are America’s incubators for policy innovation, and block-granting Medicaid should allow states to more effectively operate the program to better control costs and—just as important—tailor programs to meet the needs of specific state Medicaid caseloads and improve health outcomes for patients. Concerns about increased state control of the Medicaid program should also be weighed against the very real shortcomings of the current system.

The Current Medicaid System Is Failing the States as Well as the Poor

Today, Medicaid is the largest or second-largest line item on every state budget, crowding out spending on education, public safety, roads, and infrastructure. As a recent report from the bipartisan State Budget Crisis Task Force noted, Medicaid spending now accounts for 24 percent of general funds and 16 percent of state funds in state budgets—and because Medicaid spending is growing much faster than tax revenues, the structural budget gap “can no longer be absorbed without significant cuts to other essential programs like education, unpopular tax increases or both.”[1]

The current Medicaid financing structure is fundamentally regressive, in that wealthier states capture far more federal Medicaid funds than poor states because they can afford to spend more. Perversely, this means that Medicaid benefits (and health-care access) for poor recipients can vary widely, based on where they live.

For instance, while New York has only 7 percent of the national population, it accounts for about 14 percent of all national Medicaid spending (projected at $54.1 billion and growing in the 2012–13 fiscal year). New York spends 60 percent more per Medicaid enrollee than the national average and spends more across every category of Medicaid enrollee (see chart below). In 2009, New York spent 46 percent more per adult enrollee ($4,277 vs. $2,926) than the national average, 94 percent more per disabled enrollee ($30,000 vs. $15,453), and 70 percent more per elderly enrollee ($22,494 vs. $13,186), according to the Kaiser Family Foundation. Spending on children was comparatively modest: New York spent only about 8 percent more than the national average ($2,505 vs. $2,313).

Higher spending, however, does not automatically equate to better health outcomes. Despite lavish spending, New York’s health-care outcomes generally range from poor to average (and New York ranks last in avoidable hospital admissions). Indeed, New York policymakers have been quick to admit that the state’s program offers poor value for program recipients as well as for state taxpayers.

At the same time, Medicaid reimbursements are so low in many states (about 60 percent of what private insurance pays) that many primary-care doctors and specialists refuse to accept it, leaving recipients with insurance but no real health-care access. A recent Health Affairs study[2] found that over 30 percent of physicians (both primary care and specialists) would not accept new Medicaid patients. (PPACA does raise primary-care provider rates for Medicaid to Medicare levels—but just for two years. New York is requesting more Medicaid funding from the federal government for primary-care capacity through a recently submitted waiver, but specialist rates will not increase.) Other studies suggest that Medicaid patients have worse outcomes for some cancers, major surgeries, and longer lengths of hospital stays, compared with patients with private insurance or Medicare.[3]

Open-ended spending under the program makes Medicaid a magnet for waste, fraud, and abuse. Since 2003, GAO has designated Medicaid as a high-risk program because of its “size, growth, and inadequate fiscal oversight.”[4] GAO estimates that “improper payments” account for about 10 percent of annual expenditures—in a program that cost state and federal taxpayers over $400 billion in 2010. Some experts suggest that because federal controls for detecting waste, fraud, and abuse in the program are poor, the true cost could be 20-30 percent.[5]

Unfortunately, the Obama administration has elected to radically expand a deeply flawed program rather than fix it. PPACA extends Medicaid coverage up to 138 percent of the Federal Poverty Level (FPL) to all uninsured Americans beginning in 2014. The expansion accounts for close to half of PPACA’s insurance expansion—about 15.9 million new enrollees by 2019.[6] Washington will be footing the lion’s share of the bill for PPACA’s Medicaid expansion,[7] at least initially, but this has done little to reassure many governors whose states are already facing crushing Medicaid costs.

PPACA picks up 100 percent of the cost of the new Medicaid expansion, before gradually phasing federal support down to 90 percent in 2019. That sounds like a great deal, but in reality, states will take a serious financial hit. In particular, states with large uninsured populations that are eligible under current Medicaid rules (but not enrolled) will absorb significant administrative costs and potentially devastating coverage costs as millions of low-income Americans who are eligible under current federal Medicaid funding rules sign up for new coverage.[8] This is because states will pay a much larger share for enrollees who are eligible under the current program rules. The states are skeptical that Congress, which faces its own massive debt problems, can afford to sustain the 90 percent federal match for new populations in future years.[9]

States are right to be concerned about their future Medicaid exposure, according to Matt Salo, executive director of the National Association of Medicaid Directors: “Their future share may sound small, but it represents billions in new spending that could require cutbacks of other more popular programs, such as education or transportation, or else raising taxes.”[10]

Texas, which has the nation’s highest uninsured rate, is a prime example of the added expenses facing states. Under PPACA, Texas would receive $76 billion in new federal funds over the next decade to cover more than 2 million people through Medicaid. But state officials predict that they’ll be on the hook for $15–$16 billion in new costs over the same period—because an estimated 800,000 Texans will sign up under the old match rates. Also, only 31 percent of Texas doctors are accepting new Medicaid patients, down from 42 percent just two years ago—foreshadowing the worsening access challenges facing millions of new Medicaid recipients beginning in 2014.

Green Shoots: Early Examples of State Innovation

Rhode Island, Indiana, and New York are providing important examples of state-based innovations that show how state policymakers can utilize enhanced Medicaid flexibility to implement patient-focused Medicaid reforms that can help lower costs and improve health-care efficiency.

Rhode Island has shown how variation on a block-grant (capped allotment), along with broad federal flexibility, can allow states to improve their Medicaid programs. Beginning in 2009, Rhode Island accepted a five-year cap on combined state and federal Medicaid spending of $12.075 billion as part of a global waiver from CMS, based on historical spending trends and actuarial projections of the future rate of spending. The waiver gave the state badly needed flexibility in administering its program and reduced federal reporting requirements, allowing the state to report on broad quality measures on a quarterly basis. The state is fully responsible for any spending above its cap.

The primary goals of the state’s waiver were to: “rebalance” the state’s long-term care program (including preventive care, increased access to home- and community-based services, and improved care coordination across public programs); ensure patient access to a primary-care medical home; establish cost-effective strategies for procuring Medicaid services (including selective contracting); and utilize (where appropriate) federal funding for services that were not generally eligible for a Medicaid match but that might be cost-effective. The waiver aimed to improve the coordination of public services across state agencies through investments in IT. A December 2011 independent assessment of the Rhode Island global waiver program by the Lewin Group found that:

  • Rebalancing long-term care decreased the average number of nursing-home users over the three-year study period by 3 percent and increased the number of home- and community-based services by 9.5 percent. This strategy saved the state an estimated $35.7 million.
  • New rate setting and acuity adjustments for nursing-home care saved the state $15 million.
  • Children with special needs and adults with disabilities were automatically enrolled in care-management programs to ensure that they had access to a medical home. Care management reduced total expenditures compared with fee-for-service care. Care management also lowered emergency-room (ER) use and improved access to physician care, resulting in a conservative savings estimate of $5 million.
  • Children and adults with chronic illnesses (asthma, diabetes, cardiac conditions, and mental health disorders) in care management had lower ER use and improved access to physician services.

To date, the state projects that it has saved $100 million through changes related to the global waiver and that the waiver helped reduce the projected Medicaid spending rate from 8 percent to 3 percent annually.

Indiana has introduced consumer-directed health plans for low-income uninsured Hoosier residents under its Healthy Indiana Plan (HIP). HIP operates under another federal waiver and covers residents who make too much to qualify for the state’s traditional Medicaid plan—from 19 to 200 percent of FPL. Using funding reallocated from the state’s Medicaid Disproportionate Share Hospital (DSH) funding along with cigarette tax revenues, HIP pairs catastrophic coverage with a health savings account (HSA) called a Personal Wellness and Responsibility (Power) account. The state subsidizes the accounts based on income, but enrollees must (if financially able) pay monthly payments into Power accounts of up to 5 percent of gross family income.

Power accounts allow enrollees to pay for routine out-of-pocket health-care costs, but HIP includes first-dollar coverage for preventive services like mammograms and PSA tests—up to $500 annually. Members who can but don’t pay into their accounts are excluded from HIP for a year—giving enrollees a sense of responsibility for maintaining their own care.

As of 2010, HIP had a very low (1.86 percent) rate of nonpayment, and HIP members were likelier to maintain coverage than traditional Medicaid recipients were. To date, HIP has nearly 50,000 enrollees, and 99 percent of enrollees surveyed said that they would reenroll.

About 76 percent of HIP’s population use preventive services, and enrollees are less likely to use ERs than traditional Medicaid recipients are. For HIP members who made contributions to Power accounts, ER use declined by 15 percent during their first six months in the program. Even enrollees whose Power accounts were 100 percent state-subsidized showed a 5 percent decrease in ER use. HIP members are more likely to use generic drugs than the commercially insured, and 94 percent of HIP clients surveyed said that they were satisfied with the program.

New York has embraced a global cap on a substantial fraction of its Medicaid expenditures through the Department of Health (the cap excludes spending on disabled Medicaid recipients and spending on mental health and addiction-related services). Providers have been given increased flexibility in how they go about reducing costs. The cap will increase at 4 percent annually, based on a ten-year rolling average for medical inflation—but this is still significantly below historical growth rates for the program. (From 1991 to 2009, spending in New York’s Medicaid program grew at an average annual rate of 6.2 percent, according to a December 2011 report from the Centers for Medicare and Medicaid Services.) New York is seeking a waiver from CMS to move populations (such as the disabled) that are currently in fee-for-service Medicaid into managed-care arrangements, recognizing the need for even further-reaching reforms to control costs and improve outcomes.

Moving Beyond the Status Quo: How to Design a Workable Block Grant

Without comprehensive changes to Medicaid’s financing and administrative structure, along with broad program flexibility from Washington, Medicaid reform will continue to be episodic and halfhearted. This does not mean that federal policymakers should simply hand states the cash and then walk away. Flexibility should be implemented within a general benchmark of core benefits (such as physician and hospital care) combined with broad reporting requirements on health outcomes that will allow federal policymakers and voters to monitor state progress on improving the care delivered to Medicaid patients.

Once again, the example of TANF in 1996 is instructive. States received a block grant under TANF in return for meeting a handful of broad federal program goals, while funding was frozen at 1996 levels—where it has remained for the last 16 years. A Medicaid block-grant program modeled on TANF should attract broad bipartisan support and offer policymakers a way to avoid the Medicaid trap—high costs, poor outcomes—that has bedeviled policymakers and program beneficiaries alike for nearly 50 years.

Under TANF, any activity that can (reasonably) meet federal goals is permissible for the states. Policymakers should follow the TANF model and establish streamlined Medicaid reporting requirements (focused on health outcomes) and set broad goals that would standardize federal support, encourage improved patient health, and reduce access challenges associated with Medicaid coverage. For a Medicaid block grant, broad federal goals should include:

  • Designing coverage and care arrangements that improve health outcomes for Medicaid recipients through a variety of care options and benefit designs (including co-pays and preferred provider network designs).
  • Coordinating and delivering care in the most cost-effective fashion, with an emphasis on preventive care and wellness programs that encourage patients to take more responsibility for maintaining good health.
  • Targeting federal resources in relation to a state’s relative population of low-income and disabled, adjusted for cost of living (this would ensure that poorer states with relatively larger poor and disabled populations would receive more federal aid than wealthier states).
  • Encouraging the uptake of high-quality private insurance.

This approach—block-granting federal Medicaid funding based on state poverty rates with adjustments for care acuity and cost of living (and perhaps a modest adjustment for geographic variation in the cost of care)—would also have the advantage of ending the fiscal shell games that states play with the program today and would reduce Medicaid fraud, waste, and abuse. The initial block grant should be created to hold states financially “harmless” for current Medicaid funding levels, while ramping down the rate of spending over time to sustainable levels.

A block grant would allow states full latitude to wrap other social but nonmedical services—such as supportive housing for the mentally ill homeless—around Medicaid funding to encourage the best and most efficient delivery of services to these populations and to encourage rehabilitation where possible.

Medicaid should be explicitly countercyclical, with federal assistance growing in the event of a prolonged recession but subsequently falling back to previous levels. States could be required to use at least a portion of program savings from reduced Medicaid spending to create a rainy-day fund as a backstop against an economic downturn. And Congress should rescind (or at least greatly scale back) PPACA’s MOE restrictions.

The goal of Medicaid reform would be to move as many recipients as possible from public support to private, portable health insurance; but for practical and political reasons, states should be allowed to continue their current fee-for-service Medicaid programs, or adopt other, more centralized (i.e., single payer) health-care arrangements. They should be allowed to offer additional benefits or expand subsidized coverage up the income ladder with purely state funds.


States have only begun to scratch the surface of Medicaid innovation. Preferred provider networks, competitive bidding, and other selective contracting tools could be used to lower costs and improve the quality of care available to Medicaid patients.

Partnerships with retail health clinics staffed by nurse practitioners in stores like Wal-Mart could redirect Medicaid patients with nonemergency conditions away from expensive ERs and improve basic health access. Innovative health IT programs—such as predictive risk scoring and data mining of care-utilization patterns—offer states an opportunity to identify high-risk patients and focus on keeping them healthy, before hospitalization or other high-cost services are needed. Allowing more seniors to use Medicaid funds to purchase the services they need to live at home, as opposed to entering expensive nursing homes, is another promising approach.

In short, policymakers have many available options for improving care for vulnerable populations while reducing excessive Medicaid cost growth.

The single biggest difference between welfare reform and Medicaid reform is that there is no magic bullet—such as moving women on welfare into work—that will work for every Medicaid population. That is precisely why states need much more flexibility—and much better incentives—to coordinate care and experiment with a wide variety of tailored approaches for improving health outcomes while keeping spending on a sustainable trajectory. A carefully tailored Medicaid block grant would improve the lives of the millions of low-income Americans who rely on Medicaid coverage for safety-net care.


  1. Paul Volcker and Richard Ravitch, “State Budget Crisis Task Force Report”, July 2012., p. 15.
  2. Sandra L. Decker, “In 2011 Nearly One-Third Of Physicians Said They Would Not Accept New Medicaid Patients, But Rising Fees May Help”. Health Affairs, August 2012 vol. 31 no. 8 1673-1679.
  3. Scott Gottlieb, “Is Medicaid Worse than No Insurance at All?,” Wall Street Journal, March 10, 2011; and Avik Roy, “The Medicaid Mess: How Obamacare Makes It Worse,” Manhattan Institute Issues 2012, No. 8, March 2012.
  4. “States Reported Billions More in Supplemental Payments in Recent Years,” GAO, July 2012,
  5. Malcolm Sparrow, testimony before the Senate Committee on the Judiciary: Subcommittee on Crime and Drugs, May 20, 2009.
  6. John Holahan and Irene Headen, Urban Institute and Kaiser Commission on Medicaid and the Uninsured, “Medicaid Coverage and Spending in Health Reform: National and State-by-State Results for Adults at or Below 133 Percent FPL,” May 2010,, p. 2.
  7. The new Medicaid match, however, only covers benefits for newly enrolled recipients—not the significant administrative costs that states bear from expansion. The benefit package for these beneficiaries is also specified by PPACA, under “essential health benefits.” States cannot reduce these benefits, although they can enlarge them.
  8. A matching rate called the Federal Medical Assistance Percentage (FMAP). Poorer states get a larger FMAP match, with richer states getting less (down to a floor of 50 percent).
  9. Some critics believe that this fear is unfounded and argue that there is little history of the federal government reneging on its commitments to joint programs like Medicaid, pointing to the state Children’s Health Insurance Plan (CHIP) as an example of increasing federal commitment to expanding insurance coverage. This did, in fact, happen in 2007 and 2009. However, SCHIP is a relatively small component of Medicaid spending. While the U.S. spends approximately $400 billion annually on Medicaid, CHIP only spends $10.6 billion. Given that the federal government is currently in its fourth year of averaging annual deficits in excess of $1 trillion and U.S. debt is above 70 percent of GDP, there is simply no precedent for the budget crisis in past congressional decisions. President Obama has already proposed, as part his 2013 budget proposal, adopting a single “blended rate” for Medicaid-related programs. In short, the enhanced 90 percent match will be very difficult to maintain.
  10. “States Balk at Expanding Medicaid”, Kaiser Health News, July 2, 2012.


The Manhattan Institute, a 501(c)(3), is a think tank whose mission is to develop and disseminate new ideas
that foster greater economic choice and individual responsibility.

Copyright © 2015 Manhattan Institute for Policy Research, Inc. All rights reserved.

52 Vanderbilt Avenue, New York, N.Y. 10017
phone (212) 599-7000 / fax (212) 599-3494