Health Care Reform And The 47 Percent

Republican presidential candidate Mitt Romney’s secretly videoed remarks regarding the number of Americans “dependent upon government” have sparked a great deal of debate over who in the public receives what from the state.

The Census Bureau’s latest figures confirm that 49 percent of Americans received benefits from at least one federal program.[1] This total, however, mixes apples and oranges insofar as it includes 35 percent of individuals who rely on means-tested benefits and those who might be viewed as having “earned” their benefits, including nearly 50 million recipients of Social Security and much smaller numbers of individuals who qualify for unemployment benefits, workers’ compensation, veterans benefits, etc.

Let’s examine one segment from the former category: the 80.5 million low-income Americans on Medicaid—26.4 percent of the entire population.

As discussed in previous posts, Medicaid is a seriously flawed program. From a beneficiary standpoint, it generally offers quite comprehensive benefits—rivaling or exceeding those available to most workers through employer-sponsored coverage. However, Medicaid also severely underpays providers.[2] Consequently, only 53 percent of doctors accept all or most new Medicaid patients (compared to 87 percent who will accept all or most new patients with private insurance). This is why Medicaid ranks as “America’s worst health care program”—in some cases, worse than no coverage at all.

Medicaid is equally bad from the standpoint of good governance. As many Americans learned for the first time from the Supreme Court decision last June, the federal government is not permitted to order states around. For this reason, the federal government has always had to rely on carrots rather than sticks to induce states to take part in Medicaid. These carrots historically have been quite generous, offering even the richest states one federal dollar for every state dollar put into the program.[3] The poorest states were able to get more than three federal dollars for every state one. Even so, it took states such as Arizona 16 years before they were willing to take the federal bait and set up a Medicaid program.

Decades of experience have shown that the program is a bit like the Roach Motel—far easier to get into than to get out. Consider the incentives posed by this matching formula. First, it greatly distorts state priorities since every state dollar put into Medicaid buys anywhere from $2 to $4 worth of care. When lawmakers weigh $4 of medical care against, for example, the $1 of education spending that the same state dollar could instead be used to purchase, it should not surprise anyone that Medicaid has ended up being the fastest growing component of most state budgets. Today, in most states, Medicaid spending exceeds spending on elementary and secondary education—accounting for nearly one quarter of the average state’s budget.

Of equal importance, when states are faced with cutting spending, every dollar of Medicaid savings results in $2 to $4 fewer dollars spent in that state, so there is a disproportionate adverse economic impact that accompanies efforts to tighten spending. The incentive to spare Medicaid from spending cuts whenever possible is amplified by the recognition that the state’s citizens have been taxed by Uncle Sam for Medicaid. So if they don’t use those dollars, some other state will. Put a different way, an average state that saves Uncle Sam $1 in Medicaid costs only will save its own citizens 2 cents in federal taxes. All of these perverse incentives create a “ratchet” effect that generally propels Medicaid spending upward and inhibits spending reductions even when the states face fiscal pressures.

One of the Affordable Care Act’s most serious flaws is that it did nothing to address these fundamental flaws and, instead, was designed to increase Medicaid eligibility by 38 percent in 2014.[4] It would be far more humanitarian and efficient to give the program’s recipients a fixed amount of dollars and let them purchase private health insurance just like everyone else. After all, when people cannot afford food, we do not give them a card that allows them to buy food only at grocery stores willing to sell to them at deeply discounted prices. We instead provide them with a fixed amount of dollars that they can use to purchase food in the same locations as everyone else.

This approach might not work very well for Medicaid patients who already are sick enough to require long term care services. Thus states might need to continue to be responsible for figuring out the best way to address the needs of such chronically ill patients. Still, rather than retain the “more-you-spend-more-you-get” matching formula, a more sensible solution would be to replace it by using the same block grant approach successfully used for the Clinton administration’s welfare reform in 1996. Giving each state a fixed dollar amount would create strong incentives to carefully weigh spending priorities both across various state programs and within Medicaid, along with far stronger incentives to control costs (since states will pocket every dollar of savings).


[1] In raw numbers, 149.8 million out of a population of 305.6 million receive some form of cash or non-cash benefit. Of these, 107.2 million receive means-tested benefits, meaning that income is used to determine eligibility for benefits; the three largest such programs are Medicaid (80.5 million), Food Stamps (48.3 million), and nutrition assistance provided through the WIC (Women, Infants and Children) program (23.1 million).

[2] Specifically, Medicaid paid hospitals 34 percent less than private health insurers in 2009 (Figure 1), while paying doctors 42 percent less (Figure 2).

[3] Mississippi qualified for a federal matching rate of 75.67%. But note that under the stimulus bill, all states received an enhanced federal match that increased Mississippi’s federal matching rate to 84.86%. See this table for matching rates for every state.

[4] According to the Medicare actuary, the number enrolled in Medicaid would have risen from 60.6 million in 2014 to 83.6 million that year (Table 2). As it turned out, the Supreme Court in June decided that the law unconstitutionally coerced states into expanding; the remedy was to make expansion optional without the threat of losing all federal Medicaid funding; consequently, Medicaid will end up enrolling about 6 million fewer people by 2022 than had originally been expected, but that obviously was not the intent of the designers.