Medicaid Post-SCOTUS: Nevermind Whether States Choose to Expand; It Appears States Have the Choice to Shrink

Print Friendly, PDF & Email

Richard Evans / Scott Hinds / Ryan Baum


203.901.1631 /.1632 / .1627

richard@ / hinds@ /

twitter.jpg @SecSovHealth

July 9, 2012

Medicaid Post-SCOTUS: Nevermind Whether States Choose to Expand; It Appears States Have the Choice to Shrink

  • SCOTUS ruled that Congress cannot withhold all of a state’s Medicaid funding if the state refuses to participate in the Affordable Care Act’s (ACA) Medicaid expansion
  • Many states’ Medicaid eligibility levels are well above the federal minimums, in large part because of a separate ACA provision (maintenance of effort, or MOE) that requires states to maintain eligibility at levels of generosity established at the peak of the last economic cycle. The penalty for violating this ACA provision – losing all Medicaid funding – is the same as for not participating in the expansion, i.e. the penalty for reducing eligibility is the same penalty the Court just ruled unconstitutional
  • If the MOE provision is in fact invalid, states would have the ability to reduce eligibility to the federal minimums – disenrolling as many as 13 million current beneficiaries and reducing current Medicaid spend by $40B
  • Thus the question is not whether states participate in the expansion or stay at current eligibility levels; rather the question is whether the states fall back to (or very nearly to) the federal minimums (lose 13M beneficiaries and $40B in spending) or participate in the expansion (gain 16M beneficiaries and $70B in spending). Unless the MOE provision is valid or CMS can negotiate a compromise, the outcome inevitably varies by state
  • Of the 13M beneficiaries who exceed the federal minimum eligibility requirements, 4.6M are in Republican states. Of the 16M beneficiaries that could become eligible in an expansion, 8.8M are in Republican states. If Republican states reduce eligibility to federal minimums and forego the expansion, 2014 Medicaid enrollment would be roughly 13.4M less than if all states had participated in the expansion
  • Maine has announced plans to officially challenge the MOE requirement, and we expect other states to soon follow
  • The risk of Medicaid disenrollment is an obvious negative for the Medicaid HMOs (e.g. AGP, CNC, MGLN, MOH, WCG). Less obvious are effects on Hospitals; Medicaid disenrollment would reduce patient volumes and increase uncompensated care; however because most disenrolled Medicaid beneficiaries can be covered under the exchanges, any such patients switched to commercial carriers would pay hospitals at the considerably higher commercial rate. Drug and Generic Manufacturers potentially would benefit for similar reasons, namely a shift of demand from lower to higher priced payors

The relevance of the Court’s ruling on the Medicaid provisions of the Affordable Care Act (ACA) may be considerably larger than the market believes, as the ruling appears to give states the opportunity to reduce their current Medicaid eligibility standards back to the federal minimums

Recall that states’ Medicaid eligibility levels have been ‘frozen’ at peak economic cycle levels by the maintenance of effort (MOE) requirements, firstly of the American Recovery and Reinvestment Act (ARRA) and now by the MOE requirements of the ACA. Recall also that the Court’s ruling on ACA’s Medicaid provisions found that Congress cannot threaten to withdraw all of a state’s Medicaid funding if the state chooses not to participate in the expansion of Medicaid under ACA[1]

The MOE provisions of ACA call for precisely the same penalty – loss of all Medicaid funding – to be levied against states that reduce Medicaid eligibility to levels below those in place as of the date of passage of ACA[2]. It follows that ACA’s MOE provisions appear to violate the same test as the expansion provisions, and that the MOE provisions may therefore be invalid

When we first wrote about this on the day of the ruling, we argued that for simple Keynesian reasons all or nearly all of the states would participate in the expansion[3]. Separately, we argued that some states would take advantage of the MOE provisions having (apparently) become invalid by disenrolling Medicaid beneficiaries who were not eventually going to be picked up under the beneficial state:federal cost sharing provisions of the expansion. At that point in time we felt the validity of the MOE provisions was a nuance of the ruling with little investment relevance, since the ‘MOE-eligible’ but ‘not expansion eligible’ enrollees are in fact quite a small group[4] – consisting largely of children in households with incomes above the 133% FPL expansion threshold. In essence we saw the MOE provision as little more than an adjustment to the terms of states’ participation in the economically beneficial Medicaid expansion

On further reflection, this framing is far too narrow – we now recognize that if both the expansion and the MOE are optional, the door may be open for states to reduce Medicaid eligibility all the way back to the federal minimums

The states’ share of Medicaid spending for current enrollees is on average 43%, as compared to the (eventual) 10% share of spending the states will shoulder for Medicaid beneficiaries picked up in the expansion. We estimate the states presently spend $16B annually (excluding federal matching dollars) on enrollees whose incomes exceed the federal minimum Medicaid standards. In contrast, Urban Institute estimates the states would spend roughly $21B on the expansion (if all participate, and again excluding federal dollars) over the six-year period from 2014 to 2019, i.e. less than $4B per year[5]. In state budget terms, reducing Medicaid eligibility to the federal minimums saves four times more than participating in the expansion would cost. Thus the budgetary[6] question to states becomes: “Should we continue paying $0.43 on the dollar for beneficiaries above the federal minimum AND find the $0.10 on the dollar to participate in the expansion; OR should we save the $0.43 on the dollar by reducing eligibility to the federal minimums?”

Crucially, the ACA provides premium support for households with incomes at or above 100%FPL; thus if states roll back Medicaid eligibility to the federal minimums, many of those affected would be eligible for subsidized coverage (at no cost to the states) on the exchanges. Ironically, the fact that many of these disenrolled beneficiaries can go to the exchanges makes the states’ option of reducing Medicaid eligibility to the federal minimums all the more viable to the states, and all the more threatening to the federal government

Exhibit 1 summarizes the federal minimums. Pregnant women and children aged 5 and below are eligible up to no less than 133% FPL; children aged 6 to 18 are eligible up to no less than 100% FPL; parents of dependent children are eligible up to each state’s 1996 welfare eligibility standard (usually less than 50% FPL); coverage of childless adults is not required; and, the aged and disabled are eligible up to at least 75% FPL. Almost every state offers at least some coverage above these minimums, and the extent to which eligibility exceeds the minimums varies greatly by state[7]. Of the $16B currently being spent by states above the federal minimums, Republican[8] states are spending $5.5B of this, and stand to reduce their Medicaid spending by 9% (from current levels) if eligibility is rolled back to the minimums. For Democratic states, the potential savings are $10.4B, or on average 12.6% of these states’ current Medicaid spending. 4.6M current beneficiaries may be at risk of disenrollment in Republican states; in Democratic states 8.1M appear to be at risk (Exhibit 2). Medicaid (exclusive of federal matching) consumes on average 8.2% of expenditures in Republican states, and 9.2% in Democratic states (Exhibit 3). Assuming for the moment that all states were to participate in the expansion, Republican states would account for 55% of new enrollees (8.8M total), an increase of 35% from current enrollment; Democratic states would account for 39% of the expansion (6.3M enrollees), an increase of 29% from current enrollment (Exhibit 4). Politically ‘split’ states account for the remaining 5% of new enrollees

What’s happening now / what happens next

Appendix 4 summarizes states’ stated intentions thus far on whether or not to participate in the expansion. The signals mirror state politics: the 13 states supporting the expansion (either committed or ‘leaning’ in favor) are all Democratic, while the 11 states opposing the expansion (either declining to participate or ‘leaning’ in opposition) all are Republican

Very little has yet been said on the record by states about whether they are considering rolling Medicaid eligibility back to the federal minimums. Despite this, we would argue that a middle-of-the-road approach (maintain current eligibility but don’t participate in the expansion) makes almost no sense – if not participating in the expansion saves $1 and falling back to the federal minimums saves an additional $4, why refuse the expansion but maintain coverage in excess of the minimums? Accordingly, we think this whole issue ultimately sorts itself into states that either participate in the expansion at one extreme, or reduce eligibility to (or very nearly to) federal minimums at the other extreme

The next step is for the states to test whether CMS can hold them to the MOE provisions. Maine’s Attorney General has publicly stated his intention to challenge the MOE provision by filing a petition to disenroll over 20,000 beneficiaries this October, and the petition is to be formally made within the next three weeks[9]. We expect other challenges in addition to Maine’s. At some point – presumably quite soon given states’ interest in the matter – we expect one or more states will move to reduce eligibility in such a way that forces CMS to either exercise its related statutory authority or acknowledge that it has none

The states appear to have some considerable advantages. CMS is unlikely to admit it cannot enforce the MOE provisions without a fight; beyond slow-walking states’ petitions its only options are to either sue the states that violate the provisions or enact the prescribed penalty by eliminating the offending states’ Medicaid funding. Eliminating a state’s Medicaid funding simply isn’t going to happen, which reduces CMS’ ‘real’ options to one: sue the offending state. This sets the stage for a ruling (or more likely, rulings) on the validity of the MOE provision in the judiciary, where it appears likely the provision would fall. Rather than create such a test case by suing the states, CMS might tend to allow smaller petitions (like Maine’s 20,000 person petition) in the hope that a state won’t muster the political will to attempt disenrolling a large number of beneficiaries. Ultimately CMS has a weak hand; a precedent-setting challenge is practically inevitable, and to our reading of the law CMS appears unlikely to win

Where is this going, and why does this matter

The consequences to the federal government of losing this fight are considerable, which raises the possibility of a negotiated solution. Average total payments for a currently enrolled non-elderly / non-disabled Medicaid beneficiary are around $3,000, of which the federal government pays about 57%, or roughly $1,710. Persons[10] disenrolled by the states and with incomes at or above 100% FPL are eligible for subsidized insurance on the exchanges, where the federal share of the subsidy is 100%. In the income range under consideration (generally 100% to 133% FPL), federal subsidies are geared such that the net cost to the beneficiary of the second least expensive silver plan (70% actuarial value) offered on the local exchange is no more than 2 percent of income. CBO estimates the average individual premium for a bronze plan (60% actuarial value) will be about $4,750[11]; adjusting for actuarial value this implies a silver premium (i.e. a potential subsidy) of roughly $5,542. Two percent of income for persons at the applicable FPL levels is between $223 and $297, meaning the federal subsidy is between $5,318 and $5,244[12]about triple the current federal cost of covering that beneficiary under Medicaid

Beyond the obvious fiscal strains implied by tripling federal spending per beneficiary, there’s the practical matter that the ACA simply doesn’t authorize that type of spending. The Act contains a failsafe provision that caps federal subsidies to persons buying coverage on the exchange at 0.504% of GDP in years after 2018. CMS’ own actuary estimated[13] that subsidies would be above this level (approximately 0.518% GDP) by 2018, despite having made no provision for state disenrollment of Medicaid beneficiaries, and very limited[14] provisions for migration of employer-sponsored beneficiaries into the exchanges. If states drop back to (or toward) the federal Medicaid minimums (and/or if employers shift employees onto exchanges) then the Act is seriously underfunded, a potentially existential threat

Thus the relevance of the states being able to reduce Medicaid eligibility to federal minimums extends beyond state budget woes and election-cycle partisanship. If the Republican states can make reduction of Medicaid eligibility a credible threat, this creates a negotiating lever to be used either narrowly for purposes of further modifying the Act, or more broadly as a negotiating chip in the looming debate over the 2013 budget. I.e. even if the (Republican) states had no real intention of reducing eligibility (though for the record we believe they do), the threat of reducing eligibility – if made credible – is of considerable practical use. Accordingly we’re convinced the ‘post-SCOTUS’ ACA news flow will soon shift away from whether states participate in the expansion, to the more fundamental issue of whether states reduce Medicaid eligibility to the federal minimums

Investment relevance

Whether the states ultimately reduce eligibility, we believe the threat of reductions becomes increasingly apparent. This implies substantially higher risk premiums for companies with large exposure to Medicaid, particularly the Medicaid HMOs (AGP, CNC, MGLN, MOH, WCG). At first glance we might be concerned over volume-sensitive names, particularly Hospitals, since patients shifted out of Medicaid may not purchase coverage on the exchanges even if they’re eligible. And, even if these patients purchase[15] coverage, their unit consumption of healthcare may be lower than on Medicaid, as co-payments under Medicaid are dramatically less than co-payments would be under plans purchased on the exchanges. The bad news: this implies less demand from the 13 million beneficiaries who might be lost if states reduce eligibility to federal minimums, and from the 16 million beneficiaries who would be newly eligible (but potentially not picked up) under the expansion. The good news: the demand that does come from these beneficiaries, to the extent they gain coverage on the exchanges, would come at a much higher price, since Medicaid reimbursement schedules (Exhibit 5) would no longer apply. The net effect on Hospitals is not entirely clear, and almost certainly depends on how many states and thus how many patients go one way or the other. The effect on volume-sensitive sub-sectors (e.g. Non-Rx Consumables) other than Hospitals unfortunately would be negative, since the units sold by these names go at the same price regardless of whether the patient is a Medicaid or commercial beneficiary. Drug and Generic manufacturers presumably would benefit for the same reason Hospitals may benefit – shifting of demand from very low Medicaid pricing to much higher commercial pricing

  1. See pages 45 – 58 of the ruling
  2. See ACA Sec. 2001(b)(2)
  3. At most the states spend $0.10 of each Medicaid dollar for expansion enrollees; an average of roughly $0.82 of the $0.90 spent by the Federal government on the expansion enrollees appears to go directly into the states’ economies (more if we consider multiplier effects)
  4. Roughly 1 million children and $2B of annual (federal plus state) spending
  5. States’ share of expansion spending changes from 0% to 10% over this period, i.e. for the record the sustained incremental annual commitment is somewhat larger than $4B – but it’s still nowhere near as large as the savings a state could realize by reducing eligibility to the federal minimums
  6. We emphasize ‘budgetary’ rather than ‘economic’ savings; while it’s very clear the states save budget dollars by reducing Medicaid enrollment, it’s less than clear whether this ultimately is a wise economic move
  7. Appendix 1 provides current enrollment, average payments per enrollee, and total payments by beneficiary type and state; Appendix 2 provides enrollment and spending in excess of federal minimums by state. Appendix 3 provides a summary of state legislature control, governorship, and presidential voting by state
  8. If all 3 of legislative control, the governorship, and the 2008 presidential vote are under a single party we classify the state as Republican ‘R’ or Democratic ‘D’; if only 2 of the 3 are under a single party we classify the state as ‘Lean’ R or D. In states with a split legislature having voted in the 2008 presidential race for a candidate in a party other than the party of the current governor, we classify that state as ‘Split’
  9. C. Weaver and L Radnofsky, “States Interpret Ruling to Cut Medicaid Now”, WSJ, July 6, 2012
  10. Non-alien
  11. Letter from CBO Director Douglas Elmendorf to the Hon. Olympia Snowe, January 11, 2010
  12. I.e. $5,542 less either $223 or $297
  13. Richard Foster, “Estimated Financial Effects of the “Patient Protection and Affordable Care Act,” as Amended”, April 22, 2010
  14. In fact we believe far too limited; please see: “Why Employers are Likely to Drop Health Insurance – A Simplified View,” SSR, July 11, 2011
  15. For many of these patients the coverage would or could be free. Subsidies geared to keep silver plan premiums at or below 2 percent of income are sufficient to more than pay for bronze coverage, so bronze coverage would in many cases be free
Print Friendly, PDF & Email