As Tim Jost notes in his post on the proposal, the newest version of Graham-Cassidy aims to create a system—at least in the short term—in which, via a new, temporary block grant, all states would receive some amount of federal resources to provide at least some level of coverage to certain populations who would be ineligible for Medicaid. But this short-term pooling arrangement appears to suffer from a number of significant limitations. These limitations begin with its short-term nature, of course (the pool ends in 2026 and what happens after that is anyone’s guess).

In addition, however, the funding formula on which the pool rests, which supposedly accounts for the population that was previously insured (or that could have been insured) through the ACA’s Medicaid expansion, appears not to account for a significant proportion of the expansion, suggesting that the approach is deeply underfunded. Thus, even though the FAQ accompanying the proposal says that it “would replace the federal money spent on [among other things] Medicaid expansion,” in fact there appear to be millions of people missing in light of the way in which the allocation formula is designed.

The loss of federal Medicaid expansion financing and its replacement with a time-limited partial, capped fund, represents only part of Graham-Cassidy’s Medicaid plan. Like the earlier Senate GOP Better Care Reconciliation Act, from which Graham-Cassidy borrows, the measure also would fundamentally alter the underlying Medicaid program itself. Based on earlier Congressional Budget Office (CBO) cost estimates in connection with BCRA, one can assume only that these fundamental alterations would themselves result in hundreds of billions of dollars in reduced federal Medicaid funding for what has come to be known as the “traditional” Medicaid program.

Following In BCRA’s Medicaid Path

As with the earlier version of Graham-Cassidy, the new proposal mirrors the fundamental thrust of BCRA’s Medicaid provisions. While flying under the now-familiar banner of state flexibility, these provisions also amount to a large-scale rollback in federal Medicaid financing, accompanied by new options for eliminating benefits and services in the wake of a massive federal funding loss. The principal federal Medicaid funding reduction provisions included in the new proposal are:

  • a one-year elimination of federal funding for all Medicaid-covered services furnished by Planned Parenthood;
  • an end to federal funding of retroactive Medicaid eligibility, a basic safety-net feature of the program since 1965 that enables states and health care providers to cover the cost of health care for catastrophically ill people who were eligible but not enrolled at the time they experienced a high-cost event, whether illness or injury; and,
  • a permanent one-third reduction (phased in between FY 2021 and FY 2026) in the amount of matching funds for the traditional program that states can generate through lawful provider taxes.

The funding reductions in traditional Medicaid also include the earlier BCRA cap on the federal contribution to the traditional Medicaid program that CBO already has estimated will result in enormous federal funding reductions by holding spending well below the projected growth rate.

Like BCRA, Graham-Cassidy gives states some additional tools for coping with a large-scale funding loss in addition to those that already exist (i.e., eliminating optional populations such as many of the children and adults in need of long-term services and supports or eliminating optional services such as prescription drugs). These new tools, again mirroring BCRA, include allowing states to impose work requirements on traditional adults (since the expansion adults would be gone) whose eligibility is tied to poverty. Research suggests that the tool would add little to any savings other than to create barriers to enrollment and retention of coverage, since the vast majority of poor adults are either working, looking for work, unable to work because of illness, in school, or caring for family.

The proposal also would allow states to secure savings by reverting to a more frequent (semi-annual) eligibility redetermination process, eliminating the less burdensome (for both states and beneficiaries) annual renewal process established under ACA implementing rules. Again, this type of change might save money by disrupting coverage, helping drive down Medicaid spending.

In addition, like BCRA, the proposal would allow states to elect to run a portion of their traditional Medicaid programs as a block grant, with more freedom to eliminate eligibility groups whose coverage is now mandatory, eliminate mandatory benefits, and make other changes that reduce Medicaid spending by withdrawing assistance. As with work requirements and shortened eligibility periods, these tools might help reduce Medicaid spending while at the same time shifting more indigent health care costs directly back onto local budgets or other portions of states’ own budgets.

The New Approach To The ACA Medicaid Expansion: Ending It As We Know It

Where the proposal differs significantly from what came before—both the earlier version of Graham-Cassidy and BCRA—is in its treatment of the Medicaid expansion population. This version would simply eliminate the expansion group at the end of December 2019. After this date, states would no longer have an option to extend Medicaid to all low-income working-age adults, even at a reduced federal funding rate, as was the case with BCRA, and even if they are willing to introduce additional coverage disruption elements such as work requirements, shortened coverage periods, and other restrictions.

In place of this option would be, as Jost notes, a program of “short-term assistance for states and market-based health care grant program” (§ 106). Using the formula summarized by Jost, the federal government would cash out the ACA adult Medicaid expansion and combine it with federal premium and tax credit funding, as well as federal payments to states that have opted for a Basic Health Program. This new short-term fund then would be allocated to states under the formula Jost describes and that can be found in a special formula description also posted at Senator Cassidy’s website along with other explanatory materials and the legislative text. What happens over the long term to the federal funding, which under current law is ongoing but that ends after calendar year 2026 under the new proposal, is not discussed.

The short-term program funds apparently cannot be used to re-establish Medicaid coverage for those who previously were eligible or, for that matter, for people who once might have been eligible for Medicaid had their states chosen to expand. This is important, since in fact, as CBO has reported in the past, Medicaid is a significantly less costly form of health insurance than private coverage purchased in the commercial market. Instead, states would be limited to commercial insurance solutions and even would be given an option to use the short-term fund to move Medicaid-eligible people into commercial coverage. The short-term fund options include the following:

  • assisting “high-risk individuals” to buy health benefits coverage if they lack access to employer coverage;
  • premium stabilization assistance paid directly to insurers;
  • cost-sharing reduction assistance for people enrolled in the individual market;
  • moving Medicaid eligible people into private insurance “by establishing or maintaining relationships with health insurance issuers to provide such coverage;” and,
  • creating “coverage programs” for individuals who are not eligible for Medicaid assistance or child health assistance under the state plan” (§ 106(b)(4) as proposed).

In other words, the short-term fund would be used to divert people included in the fund into commercial insurance coverage as an alternative to either Medicaid or the Children’s Health Insurance Program (CHIP). In creating these alternative programs, states can seek federal waivers to allow insurers to waive numerous ACA protections including non-discriminatory pricing, essential health benefits, and premium rebates for plans that fail to meet medical-loss requirements.

Underestimating The Size Of The Affected Population And CHIP’s Actuarial Value

As Jost notes, the allotment formula for this temporary fund is extremely complex. The text of the bill as well as the accompanying explainers appear to link the size and distribution of the fund to a population that is considerably more limited than the population that would have qualified for Medicaid had all states expanded and that would have qualified for premium reduction and cost-sharing assistance (including Basic Health Program subsidies) had take-up been greater.

The FAQ suggests that the size of the funds made available to the states rests on certain mis-assumptions about the magnitude of the ACA adult Medicaid expansion population. Basically the fund undercounts the population by counting only people with incomes between 50 percent and 138 percent of poverty. Specifically, the FAQ states:

Q: Why was 50 percent – 138 percent FPL selected to share money between states?

A: The 50-138 percent FPL represents the population currently on Medicaid expansion. . . This extends below 100 percent FPL because some states did not expand traditional Medicaid coverage to 100 percent FPL prior to accepting Medicaid expansion.

But under pre-ACA Medicaid rules, adults who did not fall into one of Medicaid’s traditional assistance groups (caretaker relatives of minor dependent children; pregnant women; adults with disabilities) were excluded from Medicaid even if they had no income. In other words, a significant proportion of the ACA expansion population (both those who do receive benefits and those who would qualify for assistance were their states to expand) consists of working age adults with incomes below 50 percent of the federal poverty level. Indeed, 50 percent of the 2017 federal poverty level for a single person amounts to about $7,500 annually— fulltime work at the minimum wage. Apparently the federal distribution formula does not count these people when translating Medicaid expansion funds into temporary state block grant allotments.

This omission is significant. According to research conducted by SHADAC, about 36 million nonelderly, working age adults have incomes at or below 50 percent of the federal poverty level. This is a pretty serious underestimate of people who will need financial assistance. To be sure, some of this group already may be receiving Medicaid, but millions do not. Yet this part of the population seems to have simply disappeared. The proposal notes that states would no longer have to provide matching funds to qualify for the block grant, and states could potentially supplement the fund with additional financing to cover the considerable hole left by the formula’s omission of the poorest people. Indeed, the FAQ assumes that states will supplement the fund with their own expenditures—in addition to the expenditures they would need to make to offset the loss of indigent care financing resulting from hundreds of billions of dollars in lost federal Medicaid funding.

Furthermore, in terms of the scope of coverage assumed by the formula, the FAQ states that its provisions would mirror CHIP’s actuarial value, which the FAQ states is 70 percent. However, as MACPAC has reported, the actuarial value of CHIP plans is far higher than that, exceeding 90 percent. As a result, the Graham-Cassidy formula has built into it an enormous funding shortfall not only in terms of the number of former (or potential) Medicaid beneficiaries covered but also in terms of the value of the plans that states would be able to afford. According to the FAQ, the goal of the formula is to achieve parity among states in the distribution of federal funding, but the size of the fund to be distributed shrinks significantly compared to current law.

Under the bill, the total amount of funding now spent on subsidized care for low- and moderate-income families would shrink substantially. Even this diminished fund would end after 2026. The funds would be adequate to purchase coverage far more limited than that established under CHIP. The equalizing formula used to allocate the federal funding playing field for states, a specific goal of the proposal, means that states with costly medical care systems will experience even steeper shortfalls. With funding shortfalls this grave, and with the short-term fund ending after Calendar Year 2026, it is anyone’s guess what states are supposed to do to cope with such losses.

Author’s Note

The author’s work on Medicaid is funded by the Commonwealth Fund.