Policy Forum
Jul 2005

Medicaid's Untallied Costs

Michael F. Cannon, MD
Virtual Mentor. 2005;7(7):505-508. doi: 10.1001/virtualmentor.2005.7.7.pfor4-0507.


Medicaid occupies a special place among government programs for the poor. Public support for Medicaid is broader and deeper than for other safety net programs because the consequences of inadequate medical care can be much more immediate and severe than a lack of cash or even food. This may be one reason voters have heretofore accepted the rapidly growing tax burden Medicaid imposes.

Yet Medicaid's financial burden has become so severe that Tennessee has cut tens of thousands of beneficiaries from its program. Missouri is looking to replace its program by 2008. Medicaid has become states' largest expenditure, eclipsing elementary and secondary education. What was intended only to provide medical care for the truly needy has now grown larger than Medicare. Congress is struggling over how to cut less than 1 percent of Medicaid spending over the next 5 years. But cutting Medicaid simply adds to the number of uninsured…doesn't it?

Unfortunately, that's as far as today's debate goes. The untold story of Medicaid is that its greatest costs are not fiscal, or even primarily economic. But Medicaid's other costs rarely penetrate the current debate over the program's future.

Medicaid is welfare. Like other forms of welfare, it discourages work and charitable effort among the taxpayers who fund it. More troubling, though, is that it discourages self-sufficiency and encourages dependency among beneficiaries.

The phenomenon of "crowd-out"—where the presence of "free" public coverage pulls people away from private coverage—is the most-researched way that Medicaid encourages dependency. Researchers at the Robert Wood Johnson Foundation surveyed 22 leading studies and concluded that crowd-out "seems inevitable." Over half of these studies found evidence of crowd-out, and some even found that growth in public programs was completely offset by reductions in private coverage.1

For those on Medicaid, income and asset tests deter work effort and wealth accumulation—both crucial tools for escaping poverty. Recipients who earn too much risk losing an average $6000 in Medicaid benefits. Aaron Yelowitz of the University of Kentucky and MIT's Jonathan Gruber found that, rather than accumulate assets, non-elderly Medicaid-eligible households increase consumption in order to remain eligible. In 1993, Medicaid eligibility was associated with reduced wealth holdings equal to $1600 to $2000 in today's dollars.2

Even when recipients try to climb out of poverty, Medicaid frustrates their efforts long after they have left the rolls. Medicaid's annual tax burden is now over $1000 per capita. Half of Medicaid spending is financed by states, which rely heavily on sales taxes. Sales taxes are widely considered regressive in that they place a proportionally larger burden on lower-income earners.

In addition, Medicaid increases the cost of private medical care and health insurance. A program that offers services effectively free of charge to tens of millions of beneficiaries cannot help but fuel demand and medical inflation. Cost-shifting adds more fuel to the fire: Mark Duggan of the University of Maryland and Fiona Scott Morton of Yale University found that cost-shifting from Medicaid increases the price of non-Medicaid drug prescriptions by 13.3 percent.3 Thus if granny's pills cost $1000 per year, more than $117 of that represents Medicaid costs that are shifted to a private payer.

These costs are magnified by Medicaid's financing structure. States receive an average of $1.30 from Washington for every dollar they spend; spending $1 on Medicaid buys $2.30 of health care. This encourages states to expand their programs beyond what is necessary to assist the truly needy. According to the Urban Institute, about one-fifth of Medicaid-eligible adults and children have private coverage,4,5 which suggests that Medicaid currently provides coverage to many who could obtain it on their own.

A significant body of research suggests that, as it exists today, Medicaid encourages recipients to become dependent on government; encourages people to behave in ways that increase the cost of government and of health care, which makes self-reliance more difficult for their neighbors; and encourages state policymakers to get more people to behave that way. A worthwhile attempt to cut Medicaid costs would look beyond state and federal budgets and seek to minimize other costs as well.

Fortunately, the federal government has a roadmap for doing just that. In 1996, it ended the entitlement to federal cash assistance; block-granted to the states funds that were previously given out in proportion to what each state spent; and gave states greater flexibility in setting eligibility and benefits. The idea was first to stop encouraging states to foster dependency, and then to give states the flexibility they needed to discourage dependency instead.

Opponents predicted this would be disastrous for the poor. Yet caseloads plummeted and poverty decreased. By 2003, the poverty rate remained lower than at any point in the 17 years leading up to welfare reform.6 Although the robust economy played a part, many who opposed the 1996 law have since admitted that it accomplished a large measure of good.

The federal government could build on this success by applying these lessons to Medicaid. First, Congress could let states set their own rules regarding eligibility and benefits as it proposed to do in 1996. Next, it could stop encouraging Medicaid expansions by freezing payments to states at the 2005 amount, just as welfare reform froze payments to states at the 1995 amount. By itself, that would wipe out 96 percent of the cumulative 10-year federal deficit.7 Finally, Congress could give states maximum flexibility to use federal Medicaid funds to meet a few broad goals, such as:

  1. Targeting medical assistance to the truly needy;
  2. Reducing dependency;
  3. Reducing crowd-out of private effort, including charitable care; and
  4. Promoting competitive private markets for medical care and insurance.

At the same time, states and the federal government should refocus Medicaid on its original mission: aid for the truly needy. This means eliminating eligibility for those most likely to land on their feet. While some will call any such steps "draconian," Harvard economist George Borjas found that there may be a reverse crowd-out effect: when Congress cut immigrants from the Medicaid rolls in 1996, a surge in private coverage increased overall coverage levels for immigrants—a result that Borjas says cannot be explained by a robust economy.8 A good way to start refocusing Medicaid funds on the truly needy would be to prevent well-to-do seniors and their heirs from using Medicaid to pay for long-term care.

Opponents may argue that those who move from Medicaid to private insurance would end up with less coverage. Yet this is less than certain. The Urban Institute reports that "Medicaid-eligible adults with private health insurance coverage…are less likely to report unmet medical needs than their Medicaid-enrolled counterparts."9

But just as important as how much coverage people have is how they obtain it. When people work and become more productive, both they and society benefit. Offering people Medicaid in lieu of (allegedly) inferior private coverage, on the other hand, tells them that the way to get more is by doing less: work less, save less, cultivate less self-reliance. That is a recipe for dependency.

Providing medical assistance to the poor without fostering dependency is a delicate balancing act. And the costs incurred by getting it wrong don't get a line-item in the federal budget. Reforming Medicaid along the lines of the 1996 welfare law would allow the states to strike a better balance for all involved.


  1. Davidson G, Blewett LA, Call KT. Public Program Crowd-out of Private Coverage: What are the Issues?" Princeton, NJ: Robert Wood Johnson Foundation; 2004. Research Synthesis Report No. 5. This survey reports on 22 studies that found some evidence of crowd-out of public insurance, with results ranging from no evidence of crowd-out to crowd-out levels as high as 177 percent of increased enrollment in public programs.

  2. Gruber J, Yelowitz A. Public health insurance and private savings. J Polit Econ. 1999;107(6):1249-1274.
  3. Duggan M, Morton FS. The Distortionary Effects of Government Procurement: Evidence from Medicaid Prescription Drug Purchasing. NBER Working Paper No. w10930; November 2004.

  4. Davidoff AJ, Garrett B, Yemane A. Medicaid-eligible adults who are not enrolled: who are they and do they get the care they need? Urban Institute Policy Brief. 2001;A-48:2. Twenty-one percent of Medicaid-eligible adults and 27 percent of Medicaid-eligible children are reported as having private coverage.

  5. Davidoff AJ, Garrett B, Schirmer M. Children eligible for Medicaid but not enrolled: how great a policy concern? Urban Institute Policy Brief. 2000;A-41:1-2. The study reports, "Dual Medicaid and privately insured children were counted in the privately insured category," but does not state what portion of the privately insured category duals represent.

  6. US Census Bureau. Poverty Status of People by Family Relationship, Race, and Hispanic Origin: 1959 to 2003. Historical Poverty Tables; Table 2. Available at: http://www.census.gov/hhes/poverty/histpov/hstpov2.html. Accessed June 27, 2005.

  7. Congressional Budget Office. The Budget and Economic Outlook: Fiscal Years 2006 to 2015. Congressional Budget Office. 2005;56. See also: Congressional Budget Office. An Analysis of the President's Budgetary Proposals for Fiscal Year 2006. Congressional Budget Office. 2005;24 and author's calculations.

  8. Borjas BJ. Welfare reform, labor supply, and health insurance in the immigrant population. J Health Econ. 2003;22:956-957.

  9. Davidoff AJ, Garrett B, Yemane A, 1.


Virtual Mentor. 2005;7(7):505-508.



The viewpoints expressed in this article are those of the author(s) and do not necessarily reflect the views and policies of the AMA.