From the beginning of the debate over health-care reform, President Obama has advocated a major expansion of Medicaid — already a rapidly growing entitlement. Congress took heed: The Senate bill, which is currently before the House, would put millions of additional Americans onto Medicaid. In an analysis of this bill, the chief actuary at the Centers for Medicare and Medicaid Services claimed Medicaid expansion would be responsible for about 50 percent of the reduction in the uninsured population, at a projected 10-year cost of $395 billion.

Fortunately, we already have ample evidence of how expanding Medicaid affects taxpayers and the quality of health care. TennCare, enacted in 1994, was Tennessee’s attempt to reduce the number of uninsured individuals through a major expansion in Medicaid. In addition to traditional Medicaid beneficiaries, TennCare enrolled individuals who were uninsurable because of pre-existing conditions and individuals not eligible for insurance through either their place of employment or the government.

The idea was that the state would squeeze savings out of its Medicaid program through the use of managed care, thus offsetting the cost of this vast expansion in coverage. In essence, Tennessee officials promised a free lunch — an expansion of insurance coverage without increasing costs to the state.

Within months of the implementation of TennCare, enrollment swelled by half a million individuals, so that more than a quarter of the state’s 5,000,000-plus population was enrolled. Although private insurance was somewhat crowded out by government insurance, the number of uninsured individuals in Tennessee dropped substantially.

At the same time, costs exploded. While inflation-adjusted per-capita Medicaid spending in other states increased by an average of 71 percent between 1994 and 2004, the corresponding increase in spending on TennCare was 146 percent. Democratic governor Phil Bredesen was forced to restructure TennCare dramatically beginning in 2004, calling the program “a disaster” and stating he wouldn’t “let TennCare bankrupt our state.”

Yet even though TennCare failed to control costs, if covering more people led to improved health outcomes for the targeted population, then the program could be called a success. The best way to find out if health quality improved is to compare the health outcomes for Tennesseans with those of individuals in similar states that didn’t vastly expand their Medicaid programs.

In this analysis, I contrasted trends in Tennessee’s mortality rates before and after TennCare with the same data for the eight states bordering Tennessee: Alabama, Arkansas, Georgia, Kentucky, Mississippi, Missouri, North Carolina, and Virginia.

While the mortality rate doesn’t say everything about health status, it’s likely that healthy individuals have lower mortality rates. The essence of my findings is that Tennessee compared much less favorably to the surrounding states after the enactment of TennCare than before its enactment.

The change in Tennessee’s mortality rate between 1990 and 1994 basically mirrored that of the surrounding region. After the enactment of TennCare, however, when surrounding states were experiencing robust declines in their mortality rates, the decline in Tennessee was much more modest. In the four years following TennCare, the average decline in mortality rates in the surrounding states was 5.2 percent, compared with a 2.1 percent decline in Tennessee. Tennessee had the slowest decline in the region using both standard and age-adjusted mortality rates.

The mortality data suggest that instead of improving Tennessee’s health-care quality, TennCare may have caused it to decline. In any event, the state’s health outcomes clearly didn’t improve after TennCare.

Tennessee’s experience with TennCare demonstrates that the free lunch promised by President Obama of increased coverage with reduced expenditures will carry a cost — and won’t be very nourishing. But more important, as Democrats attempt to change our health-care system, they should provide some evidence that their policies will work — not just in expanding coverage, but in promoting health among the impacted population. The evidence from TennCare points in the opposite direction.

– Brian Blase is the Health Policy Graduate Fellow at the Heritage Foundation (www.heritage.org). He is finishing work on a Ph.D. in economics at George Mason University.