The (Senate) Politics of Medicaid Reform; Why Medicaid Shrinks Under Block Grants; Who Wins or Loses

Print Friendly
Share on LinkedIn0Tweet about this on Twitter0Share on Facebook0


Richard Evans / Scott Hinds / Ryan Baum

203.901.1631 /.1632 / .1627

revans@ / shinds@ /


March 2, 2017

The (Senate) Politics of Medicaid Reform; Why Medicaid Shrinks Under Block Grants; Who Wins or Loses

  • Per-beneficiary federal Medicaid spending varies greatly by state, in large part because states are required to spend their own money in order to get federal matching. Block grants seek to do away with the state / federal co-funding arrangement; if states no longer have to spend their money to get federal money, they’ll all want as much federal money as they can get
  • At one extreme, Texas has 9.5% of the nation’s poor but receives just 6.3% of federal Medicaid dollars; at the other extreme New York has 6.2% of the nation’s poor but receives 9.8% of federal Medicaid dollars. Once the requirement to spend state funds to get federal funds is gone, there’s no way Texas Republicans will agree that New York Democrats should receive more than twice as many federal dollars per person of need
  • The path of (political) least resistance is to give all states an equal amount of federal Medicaid dollars per person of need. This implies an enormous re-shuffling of where federal Medicaid dollars go – and a reduction in overall spending. States that lose federal Medicaid funding are likely to see their programs shrink; however states that gain federal Medicaid funding are more likely to simply reduce state funding, keeping their Medicaid programs at or near their current levels of generosity. On net, this means a drop in total national Medicaid spending of at least 8 percent, with risks of far greater declines
  • If the best-case scenario is that Medicaid shrinks, this plainly isn’t good news for Hospitals or Medicaid-predominant HMOs. Impacts vary depending on which states hold a majority of each firm’s beds or enrollees; if the national program shrinks 8% as we’ve estimated, HCA and THC see minor Medicaid funding gains (2.6% and 1.1% because of their concentration in states with large gains in federal Medicaid funds). UHS and LPNT see 3.5% Medicaid funding declines, the balance of their exposures to states with gains and states with losses. QHC and CYH see almost no net impact. Among the Medicaid-predominant HMOs WCG sees a modest (2.3%) gain in Medicaid funding; MOH sees a more significant (5.5%) decline, and CNC a more modest (2.3%) decline. In all cases risks are to the downside

As part of the effort to repeal and replace the Affordable Care Act (ACA) Republicans have proposed eliminating the Medicaid expansion, and restructuring federal financing of Medicaid to block grants and/or per-capita caps. This note considers the political economics of both changes, with a focus on what appears to be a likely rate limit: Republican votes in the US Senate

Rolling back the Medicaid expansion negatively impacts 17 Republican senators

In 2015, the states whose Medicaid programs expanded under the ACA received approximately $61B in federal funds for persons who enrolled in these states’ expansions. Republican position papers, and the Republicans’ draft legislation, call for eliminating the ACA Medicaid expansion outright. For states whose Medicaid programs expanded under the ACA, Exhibit 1 shows the annual federal funds flowing to states for expansion enrollees (black line, left axis), and the number of Senate seats held by Republicans in each of these expansion states (red columns, right axis). The simple message is that eliminating the Medicaid expansion would reduce federal funds received by states with 17 Republican senators[1]. Even under the reconciliation rules that Republicans are using for the repeal side of the repeal / replace effort, Republicans need 50 senate votes to succeed. With only 52 Republican senators in total, and 17 in states that would lose federal funds if the ACA’s Medicaid expansion were eliminated, rolling back the expansion seems unlikely


Shifting Medicaid to block grants and/or per-capita caps is similarly complex, regardless of whether Republicans succeed in rolling back the Medicaid expansion. Block grants and per-capita caps both fix the federal contribution to states’ Medicaid spending at some level, then inflate future federal contributions at CPI without regard to the number of persons enrolled (block grants), or at CPI adjusted for the number of persons enrolled (per-capita caps). Both approaches are being offered in draft legislation, and which approach sticks isn’t terribly relevant to the key themes of this note. So even though the two approaches differ in important ways, for simplicity and ease of reading we’ll use the term ‘block grants’ to cover both approaches

The ‘Year One’ problem – deciding who gets what share of federal block grants

Deciding year one block grant levels by state is THE major sticking point. States’ current Medicaid spending reflects state by state differences in both economics and ideology – with the former variable undoubtedly being the more impactful. Under current law federal Medicaid contributions are in proportion to states’ spending; this means that to get more federal money states must spend more of their own money. This requirement naturally limits how much states are willing to spend, especially states with weaker economies, and/or ideological tilts that would tend to limit Medicaid benefits. Block grants eliminate the link between state and federal spending, and in so doing dramatically upset the basis on which federal Medicaid spending is allocated to states. To put it simply, under block grants federal funds flow to states without regard to states’ Medicaid spending –once this constraint is removed, states with low federal funding are unlikely to support a system in which other states permanently receive much higher funding. Exhibit 2 shows the difficulty of the block grant starting point. Each state’s share of total US persons ‘in need’[2] is shown as a grey column, and each state’s share of total federal Medicaid spending is shown as a green column. States are sorted along the x-axis according to the difference in these two percentages, with states having a higher share of the nation’s poor than of federal Medicaid spending on the left. At the extremes, Texas has 9.5% of the nation’s poor but receives just 6.3% of total federal Medicaid dollars; New York has 6.2% of the nation’s poor but receives fully 9.8% of federal Medicaid dollars. This clearly reflects the current-law reality that New York was willing to spend more state dollars on Medicaid – thus attracting more federal dollars – however if block-grants are set on trailing spending under current law (as is the direction of the current draft legislation), Texas would be agreeing to a program under which New York receives more than twice the federal funding per resident in need that Texas receives – permanently. This seems incredibly unlikely, especially when we consider that there are 40 Republican senators in the 26 states that would be accepting federal Medicaid amounts, per person-in-need, that are permanently lower than funding levels received by the remaining states


The logical and seemingly fair alternative would be to allocate federal Medicaid dollars according to where persons in need live – simply divide total federal Medicaid dollars by the number of US persons meeting eligibility requirements, then send those federal Medicaid dollars to the states where those persons live. Expanding federal Medicaid spending under the Republican-led Congress is a non-starter, so such an allocation is by necessity a zero-sum game. Exhibit 3 shows how the game would play out – the black line (left axis) shows the percentage change in federal Medicaid spending by state if federal Medicaid funds were allocated solely on the basis of the number of persons in need living in each state, and the columns show the number of Democratic (blue) or Republican (red) senators by state. At the extremes, Vermont (with one Democratic senator; Sanders is technically an Independent) would see a 45% decline in federal Medicaid dollars received, and Georgia (two Republican senators) would see a 98% rise in federal Medicaid dollars received. Bear in mind that the percent changes are dramatic, and dramatically different by state; however the outcome is one in which federal Medicaid dollars per person in need are equal across all states. As fair as this outcome might be, the politics are still extremely difficult. Note that this method results in a net shift of resources from blue to red states – something that’s feasible with a Republican president and bicameral Republican majorities, but also something that would invite fierce Democratic opposition. More to the point, note that despite the net blue-state to red-state shift this method still results in a drop in federal Medicaid dollars to states with 10 Republican senators[3], at least 8 of whom must vote in favor of this allocation method in order for it to go forward. This too seems unlikely, at least in the absence of offsetting concessions being granted to these states from other federal programs. For the record, we calculated how much additional federal Medicaid spending would be required to ensure that all persons of need receive the same federal Medicaid support regardless of where they live, and also that no more than two Republican senators (sorry, Alaska) see declines in their federal Medicaid funds. That added federal Medicaid spend level is roughly $95B to $100B – also a non-starter


Untying the knot between state and federal Medicaid spending is both the genesis of the Year One problem, and an essential feature of block grants

To summarize briefly, rolling back the Medicaid expansion requires at least 15 (of 17 negatively affected) Republican senators to vote against their states’ economic interests; shifting to a block grant mechanism in which grants are set on trailing year spending requires at least 38 (of 40 negatively affected) Republican senators to agree to permanently take less free federal money than is received by other states; and, setting year one block grants on the basis of identical federal funding per person in-need regardless of state of residence means either 8 Republican senators (of 10 negatively affected) voting against their states’ interests, or the expansion of federal Medicaid spending by $95B – $100B

The only obvious path out of this trap is to maintain the link between state and federal Medicaid spending. If states have to put up their own money to get federal Medicaid money we can maintain the disparity in federal Medicaid flows by state (and by person in-need); conversely if states do not have to put up their own money to get federal Medicaid money we believe it is unrealistic to expect any state will accept fewer federal Medicaid dollars per person in need than any other state. Unfortunately, this potential solution also leads to a dead-end. In Republican circles the shared funding link between state and federal spending attracts a lot of the blame for the current size of the Medicaid program. To be more specific, the notion is that shared funding is inherently inflationary. The fact that state spending on Medicaid attracts federal inflows at higher rates than state spending on other programs raises the likelihood of states spending available funds on Medicaid. And, the fact that states only receive a fraction of savings from Medicaid cost-savings initiatives means these are pursued less aggressively than if states realized all the savings. For this reason, to Republicans’ minds delinking state and federal spending appears to be a sine qua non of Medicaid reform. If state and federal Medicaid spending remain linked, much (or even very nearly all) of the motive for block grants is lost

The path of least resistance = the most likely option

Having considered the available options, the easy conclusion is that there exists no simple path through the US senate to either rolling back the Medicaid expansion, or restructuring Medicaid to a block grant program. There is however a feasible path, which involves a modification of the least complicated option, namely allocating current federal Medicaid spending equally across persons in need regardless of their state of residence (producing the change in federal Medicaid flows by state shown in Exhibit 3), and then promising the Republican senators from Arkansas, Pennsylvania, West Virginia, Maine, Ohio, Kentucky, and Alaska gains in federal spending from some program(s) other than Medicaid. The total drop in annual federal funding to these seven states under this form of block grant is roughly $5B at 2015 Medicaid spending levels – which in the context of overall federal flows of funds to states is not an enormous figure; and, the great majority of this drop falls on just two states, Kentucky and Ohio[4]

The path of least resistance results in a drop in overall Medicaid spending, of at least 8 percent, with further downside in the absence of rigid minimum coverage requirements

In states that received fewer federal Medicaid dollars under block grants, we assume these states continue to commit the same amounts of their own state funds to their Medicaid programs. However in states that received larger federal Medicaid payments under block grants, there is every likelihood that these states drop their state funding in proportion to the increased federal dollars received – i.e. these states keep their Medicaid programs the same size. The general result is that states with decreased federal Medicaid dollars see their programs shrink, and states with increased federal Medicaid dollars see their programs remain the same size, with the net result being an 8% decline in total national (federal + state) Medicaid spending. Exhibit 4 shows change in total Medicaid spending by state under this scenario; note that six states see increases in federal Medicaid dollars that exceed their current state spending, such that their programs grow even if these states completely eliminate state funding of their Medicaid programs. These states and the percentage changes that would occur in their Medicaid programs are South Dakota (+2.3%), Idaho (+4.3%), Utah (+14%), Alabama (+17%), Georgia (+24%), and Florida (+8.1%)


This estimated 8% drop in total Medicaid spending under block grants arguably is a best-case scenario. Without a requirement for states to spend in proportion to federal dollars received, there’s every reason to believe that state Medicaid funding could fall, even in excess of any gain in federal funds to a given state, unless reform legislation requires states to extend at least a minimum level of services to persons with incomes below a given eligibility limit. Where such limits might be set by reform legislation is simply unclear; however our guess is that these limits won’t be sufficiently rigid to hold the drop in overall program spending at just 8%

Investment relevance

Hospital chains and the Medicaid-predominant HMOs are the publicly traded firms most at risk from the re-shuffling of, and decline in, Medicaid spending under the most feasible block grant scenario. To get a sense of which of these names are more or less affected under the base case scenario of an 8% decline in total program spending, we simply weighted the likely change in total Medicaid spending by state by the share of each firm’s beds or enrollees located in each state. Under these assumptions UHS and LPNT are the hospital chains most negatively affected; however we would expect these firms to see Medicaid spending declines of only about 3.5% despite the full 8% drop in the broader program. THC and HCA actually see small rises in Medicaid funding in the states they serve (Exhibit 5). HMO effects are similarly modest, with the potential exception of MOH, who would see an estimated 5.5% drop in premiums under block grants.[5] WCG would see an estimated 2.3% gain, with changes in premiums for the remaining names falling in between these extremes (Exhibit 6). Note that these estimates reflect the best case scenario of an 8% drop in total Medicaid spending


  1. Portman(OH), McConnell(KY), Paul(KY), Toomey(PA), McCain(AZ), Flake(AZ), Young(IN), Gardner(CO), Moore Capito(WV), Heller(NV), Grassley(IA), Ernst(IA), Boozman(AR), Cotton(AR), Hoeven(ND), Murkowski(AK), Sullivan(AK)
  2. We defined ‘in-need’ as living in a household with an income at or below 137% of the federal poverty level (FPL), regardless of age, disability status, or presence of children. As it turns out, where we set the FPL limit and whether we account for age, disability, and children makes almost no difference to our calculation of what share of the nation’s persons in need live in which state
  3. Boozman(AR), Cotton(AR), Toomey(PA), Moore Capito(WV), Collins(ME), Portman(OH), McConnell(KY), Paul(KY), Sullivan(AK), Murkowski(AK)
  4. Lost annual federal Medicaid funds by state, in 2015 Medicaid dollars: Arkansas $71M, Pennsylvania $542M, West Virginia $344M, Maine $202M, Ohio $2B, Kentucky $1.6B, Alaska $235M
  5. This assumes that changes in spending on Medicaid HMO premiums are the same as changes in total Medicaid spending for each state


©2017, SSR, LLC, 225 High Ridge Rd, 2nd Floor, Stamford, CT 06905. All rights reserved. The information contained in this report has been obtained from sources believed to be reliable, and its accuracy and completeness is not guaranteed. No representation or warranty, express or implied, is made as to the fairness, accuracy, completeness or correctness of the information and opinions contained herein. The views and other information provided are subject to change without notice. This report is issued without regard to the specific investment objectives, financial situation or particular needs of any specific recipient and is not construed as a solicitation or an offer to buy or sell any securities or related financial instruments. Past performance is not necessarily a guide to future results. In the past 12 months, through a wholly-owned subsidiary SSR Health LLC has provided paid advisory services to Pfizer Inc (PFE), BioPharmX (BPMX), Gilead Sciences (GILD), Bristol-Myers Squibb (BMY) and Sanofi (SNY) on both securities-related and non-securities-related topics

Print Friendly