In their wondrous 1957 interpretation of a Gershwin classic, Ella Fitzgerald and Louis Armstrong sing, “You like potato and I like potahto/You like tomato and I like tomahto.” Their back-and-forth has echoes in the current debate over what to call the TennCare III demonstration, approved on January 8 by the former CMS Administrator, Seema Verma. (TennCare is the name of Tennessee’s Medicaid program). The state calls it a “block grant;”  Administrator Verma calls it an “aggregate cap.” We think it’s more like a windfall. Whatever you call it, TennCare III is a radical change in financing of the 55-year old Medicaid program—a change that the Administrator does not have the authority to approve for one day, much less for 10 years, as her approval attempts to do.

Let’s start at the beginning. Since its enactment in 1965, Medicaid has guaranteed federal funding to participating states for the costs of health and long-term care services to low-income children, adults, individuals with disabilities, and the elderly. State spending on covered services for eligible individuals is matched by the federal government on an open-ended basis at an average rate of 65 percent.  The more that states have to spend in order to meet the needs of their low-income residents, the more the federal government will contribute to match those state expenditures. This open-ended matching is particularly important during times of natural disasters or public health emergencies (think pandemic).  Of course, if there is no state spending, then by definition there is no federal matching.

TennCare III jettisons this arrangement. The details are complex and opaque, in part because critical portions of the 221 pages of Waiver and Expenditure Authorities and Special Terms and Conditions are incomplete, internally inconsistent, and generally not good enough for government work—almost as though someone told the staff to put their pencils down in order to get the document out the door before the January 20 midnight hour. But the broad outlines are clear enough.

The federal government will continue to match Tennessee’s Medicaid spending at its regular matching rate, but only up to a fixed amount each year, for each of the next 10 years. That amount—let’s call it a cap—can be adjusted up (or down) if enrollment grows (or declines) more than 1 percent.  The per enrollee spending amounts used to make this adjustment grow at a generous rate of over 5 percent annually for the first five years; after that, both the caps and the adjustment amounts will be renegotiated, creating significant uncertainty for state policymakers, providers, and beneficiaries.

Capping federal Medicaid spending had been former Administrator Verma’s North Star. Early in her tenure, she set forth her vision in a speech to state Medicaid directors. She strongly supported the Trump Administration’s 2017 effort to enact a Medicaid cap, which fell short of Senate passage by one vote.  Having failed to change the Medicaid statute through the legislative process, former Administrator Verma had been trying since early 2018 to rewrite it by Executive Branch fiat, her calls for humility in government to the contrary notwithstanding. In Tennessee she finally found a willing partner (even though they can’t agree on what to call their arrangement).

The former Administrator’s ideological fixation on capping federal Medicaid spending is crystal clear.  But why would a state with among the highest rates of uninsurance and rural hospital closures in the nation decide to let the federal government dump the risk of increases in health and long-term care costs onto it in the middle of a pandemic—much less lock itself into a 10-year limit on federal funds knowing that medical care inflation is completely unconstrained?

In a nutshell: free federal funds. Under the demonstration, if Tennessee spends less during a year than the cap for that year, it gets to keep up to 55 percent of the federal share of the “savings”—without spending a dime of its own money.

For example, assume the spending cap for a demonstration year, after adjustments for enrollment growth, is $10 billion (roughly the total amount of TennCare’s spending on Medicaid benefits in FY 2019), and that total spending on Medicaid benefits for that demonstration year is, say, $9.5 billion. Under TennCare III, the state would be able to keep up to 55 percent of the federal share (66.1 percent in FY 2021) of the $500 million in “savings” (0.55 * 0.66 * $500 = $181.5 million)—without spending any of its own funds.

How high a percentage the state could keep will depend on its performance as measured by quality metrics which are somehow not specified anywhere in the hundreds of pages of Special Terms and Conditions.  The state will be able to use its “share” of whatever federal funds it “saved” to refinance existing state-funded programs, such as “Community and Faith-based Clinics” for the uninsured. But since federal dollars are fungible once they hit a state’s treasury, this is simply optical. Practically speaking, the state will be free to use the “savings” in federal Medicaid funds as it pleases, on items that have nothing to do with Medicaid.

So where would the “savings” come from? Good question, particularly since Tennessee’s Medicaid program relies almost exclusively on managed care organizations (MCOs) to control costs and has among the lowest per enrollee spending of any state Medicaid program ($6,486 per enrollee in FY 2019, lower than any states other than AL, FL, NC, NM, and SC). There is no evidence that Tennessee needs additional incentives to reduce its Medicaid spending.

One clear target is prescription drugs. Under TennCare III, the state is allowed to limit its coverage of high-cost drugs by closing its formulary—a radical departure from the beneficiary protections under current Medicaid rules for drug coverage that require coverage of most FDA-approved drugs.  Tennessee says it will provide only the same number of drugs in a class as is required under a commercial benchmark plan, which of course is not designed to cover the high needs beneficiaries that Medicaid serves. The state will allow beneficiaries to obtain “clinically appropriate” non-formulary drugs through an appeals process.  Tennessee, however, is free to define what constitutes a clinically appropriate drug and to set up the appeals system as it sees fit.

There are other potential sources of “savings,” such as lowering provider reimbursement rates.  Notably, the state could also reduce enrollment within specific eligibility groups—children, adults under 65, individuals with disabilities, etc.—by using administrative red tape, such as frequent redeterminations.  So long as the enrollment levels do not fall by 1 percent or more from the base enrollment for that eligibility group in SFY 2019, the state’s aggregate cap would not be adjusted downward, leaving the state to “share” in all of the resulting “savings.” (Tennessee is subject to the pandemic disenrollment freeze if it accepts the 6.2 percentage point increase in its matching rate).

According to the Governor’s press release, TennCare III creates “an unprecedented opportunity for Tennessee to be rewarded for its successful administration of TennCare”—in the form of  federal Medicaid funds that he says will be used to” further improve the health of TennCare members and Tennessee communities.” The “opportunity” is in fact unprecedented.  And that is precisely the problem.

Medicaid is a federal-state matching program. If a state spends its own funds on behalf of an eligible individual for a covered service, the federal government matches that expenditure.  But there has to be a state expenditure. If a state doesn’t spend money purchasing covered services on behalf of Medicaid beneficiaries, the federal government doesn’t pay any matching funds. Both the state and the federal government save, in proportion to their regular federal-state shares. Under TennCare III, the state will be able to draw down federal matching dollars even though it has not made an expenditure of its own funds.

This is not the way Medicaid works, and the Administrator does not have the authority to rewrite the program’s financing. Section 1115 of the Social Security Act authorizes the Secretary to waive state Medicaid plan requirements found in section 1902 of the Social Security Act., but only section 1902. The matching requirements—fundamental to the Medicaid program—are in section 1903.

The Administrator might argue that the federal government is matching state spending on Medicaid and that all that the  “shared savings” provision does is raise the state’s effective federal matching rate from its nominal 66 percent to something higher. That would in fact be the effect of this provision, and that is precisely what section 1115 does not authorize the Secretary to do. The formula for the federal matching rate for each state is set forth in section 1905(b) of the Social Security Act. Again, section 1115 only authorizes the Secretary to waive requirements of section 1902. To allow the Secretary to change the matching rates of individual states—whether directly or indirectly through the fiction of “shared savings”—would invite partisan political corruption of the nation’s largest federal funding program for states.

There is, of course, a legal way to improve the health of low-income Tennesseans and their communities.  It’s called Medicaid expansion. And it comes with a statutory 90 percent federal matching rate, considerably more generous than Tennessee’s regular 66 percent matching rate. If Tennessee took up Medicaid expansion, there would be less need to refinance Community and Faith-based clinics for the uninsured, because there would be 260,000 fewer uninsured Tennesseans.

You like “Block Grant.” I like “Aggregate Cap.”

Let’s Call the Whole Thing Off.