By Daniel Sutter

Since it was founded alongside Medicare in 1965, Medicaid's costs have exploded from $5.3 billion to $449 billion (adjusted for inflation). It now comprises about 3 percent of our gross domestic product. The uncertainties associated with Medicaid, along with Medicare and Social Security, pose a serious threat to America's long-term fiscal health.

How can we control Medicaid spending growth? We must first understand what's driving it.

If medical need - the number of low-income Americans needing coverage and the cost of providing their medical care - explains all of the growth, then reform might involve a divisive and painful approach, like slashing the health care of children and pregnant women.

But if the structure of Medicaid produces unintended growth and wasteful spending, a design flaw so to speak, real reform might be accomplished without harming the people who rely on it.

Medicaid's growth has not escaped the attention of academics, who have investigated the causes. They've turned up evidence that this design flaw - the program's matching-grant structure - is largely to blame. We should view this as good news: It means that we have an opportunity to better direct funding toward medical care for the poor.

As a joint federal-state program, Medicaid gets about 60 percent of its funding from Washington under matching grants and rules set by the Department of Health and Human Services. Each state, however, administers and operates its own Medicaid program, so each dollar they choose to spend is matched by the federal government.

The exact federal matching rates vary based on states' average incomes relative to the nation as a whole. Known as the Federal Medical Assistance Percentages (FMAP), they range from $1 to $3 for each dollar a state spends.

It's no secret that matching grants can be wasteful because states don't bear the full cost of their spending decisions. When a state spends an extra dollar, between 50 and 75 cents will come from Washington. Similarly, half or more of any savings from eliminating waste returns to Washington, reducing the motivation for states to be careful stewards of tax dollars.

But the tale is a little more complicated. Washington mandates that all states cover certain groups, like poor children and the disabled, and certain types of medical care. States have discretion to expand their programs beyond these minimums to new groups or types of medical care. This discretionary spending is where we would expect to see the influence of Medicaid's matching grants.

Studies consistently find evidence that the matching-grant structure increases discretionary spending. The higher a state's FMAP, the more it spends in order to leverage federal dollars.

Politics and interest groups also play a role. States controlled by Democrats tend to spend more, and despite Medicaid being commonly considered a health-insurance program for the poor and children, nearly two-thirds of spending actually goes to the elderly and disabled - two groups with especially powerful lobbies.

The effect of matching grants can also be seen across states. Federal participation - in particular, higher FMAPs for states without adequate resources to cover their low-income populations - was intended to equalize spending among affluent and poorer states. Yet overall, high-income states receive more federal dollars. They have used the matching-grant formula to push half the cost of very generous Medicaid systems to the federal government, helping explain why such a low proportion of Medicaid funds are spent on health care for poor adults and children.

This misallocation of resources has important consequences. Medicaid reimbursement rates for doctors and hospitals remain low, and patients have difficulty finding doctors who will accept them. One study found that Medicaid patients had worse surgical outcomes - controlling for age, illness, and other factors - than Medicare patients and even the uninsured. Billions are spent on discretionary coverage in high-income states, while reimbursement rates for basic coverage remain low.

Block grants, which hold federal aid at a fixed level regardless of how much states choose to spend, are not a new idea. Welfare reform in the Clinton era switched cash assistance from matching grants to block grants with great success, and Rhode Island has been experimenting with block grants for Medicaid.

As the facts about Medicaid's flaws come to light, this approach becomes more appealing. If matching grants, more so than medical needs, are fueling discretionary spending, then would switching to block grants actually harm patients? States could still offer optional coverage with their own money, but in a more equitable way. More importantly, they could help redirect federal assistance to the neediest among us.

Daniel Sutter is an affiliated senior scholar with the Mercatus Center (www.mercatus.org) at George Mason University, a professor of economics at Troy University, and the author of the new study "The Political Economy of Medicaid Expansion: Federalism, Interest Groups, and the ACA."