ACA Post-SCOTUS – What Matters Now

Print Friendly, PDF & Email

Richard Evans / Scott Hinds / Ryan Baum


203.901.1631 /.1632 / .1627

richard@ / hinds@ /

twitter.jpg @SecSovHealth

June 29, 2012

ACA Post-SCOTUS – What Matters Now

  • The Court’s rule modifying the Medicaid expansion (you can make the additional federal funds conditional on state participation in the expansion, but not all Medicaid funds) appears to be violated by the ACA’s Medicaid maintenance of eligibility (MOE) provisions
  • MOE provisions require the states to retain Medicaid eligibility standards from the date of ACA passage until roughly 2019. In effect these provisions require the states to carry boom-level Medicaid generosity through bust-level (i.e. lower) state revenues and (i.e. higher) Medicaid enrollment. The ACA makes all federal Medicaid support conditional on complying with the MOE provisions; this appears to directly violate the rule created to address the Medicaid expansion
  • States are likely to agitate for eliminating the MOE provisions. Being close to the (2014) start date, losing MOE probably has less to do with reducing spending (+/- $1B in 2013) and more to do with who does the spending. If MOE falls, states will push what Medicaid enrollees they can onto the exchanges, where these enrollees will be subsidized entirely by federal dollars. Such changes would reduce Medicaid enrollment projections, and may be material to Medicaid HMOs
  • Repeal is a long shot at best. The odds of a Romney win AND a 60 seat Senate super-majority are infinitesimal; repeal can thus only proceed under reconciliation. Even these odds are incredibly long
  • Setting aside the MOE question, states will participate in the Medicaid expansion for simple Keynesian reasons – for every $0.10 they spend they appear to get around $0.82 in benefit
  • Consensus seems to frame the mandate in terms of post-2014 volume gains, but we think these will disappoint. Employers are likely to shift employees to exchanges, where fewer buy insurance and those who do purchase buy less. Today’s uninsured will benefit from the exchanges, but because of the high marginal costs of buying coverage – even on the exchanges – fewer will buy than is generally expected
  • Hospitals now have a clear source of pricing power. Health plans should see 2014 as an enrollment land-grab; however the marginal enrollee has little familiarity with health plan names and no primary care relationship, but recognizes the name(s) of the preferred local hospital(s). Hospitals are the recognizable brands that draw the marginal enrollee; this means pricing power pre-2014
  • IPAB is here to stay; the closer its start date (2015) looms, the tougher it is to muster the pay-fors to eliminate it. IPAB can only work by reducing the price of innovative (device, drug, biotech) products used in Medicare beneficiaries’ care, and the necessary reductions are quite large
  • The AWP replacements likewise are here to stay – with ACA clear, the states should soon move away from AWP, followed eventually by larger commercial buyers. The information asymmetries embedded in AWP enable large generic dispensing margins; when AWP goes these margins fall
  1. Medicaid maintenance of eligibility (MOE) requirements may be challenged

Paraphrasing the Court, the federal government may not make all Medicaid funding conditional on states’ participation in the Affordable Care Act (ACA) expansion of Medicaid (emphasis added):

“We have upheld Congress’s authority to condition the receipt of funds on the states’ complying with restrictions on the use of those funds, because that is the means by which Congress ensures that the funds are spent according to its view of the “general Welfare.” Conditions that do not here govern the use of the funds, however, cannot be justified on that basis. When, for example, such conditions take the form of threats to terminate other significant independent grants, the conditions are properly viewed as a means of pressuring the states to accept policy changes.”

This simple test appears to be violated elsewhere in the Act, specifically in language requiring states to maintain Medicaid eligibility standards in place as of the date of the Act’s passage. ACA Sec. 2001(b)(2), (emphasis added):


Subject to the succeeding paragraphs of this subsection, during the period that begins on the date of enactment of the Patient Protection and Affordable Care Act and ends on the date on which the Secretary determines that an Exchange established by the State under section 1311 of the Patient Protection and Affordable Care Act is fully operational, as a condition for receiving any federal payments under section 1903(a) for calendar quarters occurring during such period, a State shall not have in effect eligibility standards, methodologies, or procedures under the State plan under this title or under any waiver of such plan that is in effect during that period, that are more restrictive than the eligibility standards, methodologies, or procedures, respectively, under the plan or waiver that are in effect on the date of enactment of the Patient Protection and Affordable Care Act.”

The preceding passage clearly makes states’ receipt of any federal Medicaid funds[1] between the passage of ACA and a point after full effect of the ACA (about 2019) conditional on states maintaining current eligibility standards for Medicaid. Recall that states with Medicaid eligibility standards more generous than the federal minimums were ‘frozen’ at these levels of generosity, first by the American Recovery and Reinvestment Act (ARRA), and then by the ACA. The states have on many occasions agitated for these maintenance of eligibility (MOE) standards to be eased, as the standards they’ve had to carry through the recession were set during the preceding period of economic strength. If our reading of the Court’s ruling and the Act are correct, we would expect some number of states to challenge whether CMS has the authority to enforce the ACA’s MOE provisions. Exhibit 1 provides a summary of states’ levels of coverage generosity relative to the federal minimum, as well as the states’ share of total Medicaid spending, and their current politics. Importantly, CBO issued a memorandum assessing the impact of repealing MOE provisions when

assessing HR 1683[2]. From these and other sources it seems as though the MOE provisions may be less relevant to total health spending (relying heavily on CBO’s federal estimate, we think combined federal and State Medicaid outlays would drop by $1B or less in 2013 if MOE were repealed, though it’s not immediately clear what CBO is assuming with respect to how many states lower eligibility and by how far) than to the ‘forms’ of spending. E.g. without MOE provisions states may be able to more quickly put current Medicaid enrollees who are above pre-ACA minimums and not subject to the ACA expansion into the exchanges; this would make these enrollees eligible for federal subsidies at no cost to the states, and alleviate the states’ Medicaid cost shares. Our immediate thought is that these ‘MOE beneficiaries’ represent potential enrollment losses that may or may not be material to any number of Medicaid HMOs

  1. Full repeal of the ACA is a VERY long shot …

The current Republican leadership has promised (yet another) repeal vote; there is no chance this succeeds

More important, if Romney wins in November, he has promised to repeal the ACA and may be compelled to at least make the gesture. As a practical matter (for the moment barring use of reconciliation) Romney can repeal only with a simple House majority or greater and a Senate super-majority of 60 or more Republicans. At current polling, the odds of the latter are incredibly steep (Exhibit 2). The odds in the presidential race favor Obama slightly, and the odds of the Republicans keeping the House are quite high. However the odds of the Republicans having even a simple Senate majority are only about 50:50. The odds of a 60 seat Senate majority are absurdly long – Republicans would have to keep all of their current seats, win all of the toss-ups, have King (I) win in Maine and join the caucus, AND win 7 of the races where odds favor Democrats. Based on current polling, the odds of this last step alone – taking at least 7 of the seats in which Democrats are favored – are 0.01%. The odds of everything necessary for a 60 seat majority are longer still

Thus repeal outside of reconciliation procedures[3] is very nearly impossible. Mathematically, repeal using reconciliation isn’t impossible but the odds are still long. Repeal under reconciliation requires three events, each harder than the one before: 1) Romney has to win AND the Republicans have to take a Senate majority (these things tend to go together, so crudely the odds of both happening are ‘a little’ less than the current independent odds of either – say 40% or less); 2) the Senate parliamentarian has to allow the reconciliation measure (this process is more mechanical than political – it can be gamed a little but not a lot – so the repeal measure has to be ‘shaped’ to reconciliation rules); and 3) the reconciliation measure has to survive points of order (reconciliation measures are relatively unshielded in floor debate, and accordingly are more likely to be gutted than measures passed under other mechanisms). Maybe the cumulative odds of repeal under reconciliation are 20%

These odds ignore the political costs, which are enormous, especially at the start of a new President’s term. We can’t rule out an ACA repeal, but the purely mathematical / mechanical odds are quite long, and the politically adjusted odds longer still

  1. Why the states all participate in the Medicaid expansion

Ignoring the separate question of whether some states try to roll back eligibility requirements ahead of 2014, we think all states participate in the ACA’s Medicaid expansions after 2014, for simple Keynesian reasons

For the first three years of the expansion, states pay nothing for the additional federal dollars; states’ full funding obligation of 10% of the expansion isn’t reached until 2020 (Exhibit 3). At 100% federal funding the expansion is plainly beneficial to states’ economies – it’s pure gain. At 90% funding, it’s still better to be in than out. We estimate that roughly 82% of Medicaid spending stays in the state (Exhibit 4). I.e. stripping out supply costs which may benefit out-of-state (or country) suppliers, the balance of spending tends to benefit local labor and facilities – and non-supply spending is about 82% of Medicaid spending. Thus in exchange for spending $0.10, states receive $0.82 in direct benefit. This ignores multiplier effects, which are considerable

  1. Why post-2014 volume gains may be disappointing

With the mandate having been confirmed, we can now expect the states to produce a reasonably consistent ‘set’ of health insurance exchanges. If the mandate had been over-turned we would have expected the opposite condition

This national backdrop of 50 comparable health insurance exchanges creates an opportunity for employers to back away from health coverage, and the employer-level economics of ACA provide the driver

As we’ve previously shown[4], if we make the simple assumption that an employer wants to deliver a given level of wage and health coverage as efficiently as possible, for most subsidy-eligible households this can be done more efficiently by grossing up the employee’s wage, and having the employee buy their coverage on the exchange. In Exhibit 5, Employer A offers employer-sponsored insurance (ESI); Employer B offers no insurance but grosses up wage so that the employee can buy the same coverage offered by ‘A’ on an exchange. In either case, employees receive the same level of health coverage and same take home after-tax (and after health premiums) wage. Looking at this same scenario from the perspective of the employer (Exhibit 6), this compensation package costs Employer B (who grossed up wage and sent the employee to the exchange for insurance) $2,666 less than is paid by Employer A (who provides employer-sponsored insurance). The $2,666 can go to the employer, the employee, or be shared by both – who gets the $2,666 is irrelevant – the fact that it’s there is the key. For most subsidy-eligible households, the most efficient way to provide a package of wage and health benefits is to gross up wages and have the employees buy coverage on the exchange. The economics are simple, and they argue that employers put subsidy-eligible employees on the exchanges

The politics of this shift are complex at a national level, but still simple at the level of the employee. The two packages offer the same coverage and the same wage, i.e. ignoring optionality they have equal value. Including optionality, i.e. thinking in real-world terms, Employer B’s offer is more attractive. B’s employees can keep all of the wage gain and buy no insurance, some of the wage gain and buy less insurance, or none of the wage gain and buy the same insurance they would have gotten from A. Thus B’s offer is not only cheaper to create, it’s more attractive. Finally, because B’s offer is particularly attractive to younger and healthier workers, B has an even greater reason to go the route of grossing wages higher and having employees buy coverage on the exchange

Thus an employer-sponsored insurance (ESI) to health insurance exchange (HIE) shift reduces per-capita unit demand relative to what it would have been if ESI had remained the norm. Persons shifted to HIEs are less likely to buy coverage[5], and the coverage they buy is less generous[6]

This ultimately is bearish for both Hospitals and Commercial HMOs. Hospitals not only see less post-2014 volume than expected, they face growing collection headwinds on the commercial volumes that do emerge. In the case of an ESI beneficiary covered by a plan with an actuarial value (AV) of 0.82, the hospital collects $0.18 on the dollar from the patient at a +/- 30% rate, and the $0.82 balance from the insurer at a 98+% rate. On the HIEs with a beneficiary covered by a plan with an AV of 0.65, the hospital has to collect $0.35 from the patient at the +/- 30% rate

Under a shift to HIEs, Commercial HMOs lose beneficiaries (not everyone shifted buys coverage; and, those shifted may buy elsewhere), average contract values fall, and operating costs per beneficiary rise (contracts shift from large multi-state employers to individual contracts on exchanges). Remarkably, current valuations allow for these fundamental negatives, but they are important negatives nonetheless

  1. Hospitals have more pre-2014 pricing power than may be obvious[7]

Notwithstanding our conservative views on 2014 enrollment gains, health plans should view 2014 as a rare land-grab of enrollment. The marginal enrollee has little familiarity with health plan brands, and is relatively unlikely to have a primary care relationship, but is highly likely to recognize local hospitals

It follows that hospitals are the brands that steer marginal enrollees in 2014, and it further follows that plans need brand name hospitals in 2014 more than ever – i.e. hospitals have solid commercial pricing power as ’14 approaches

This appears to have happened in the roll-out of MA’s health reform, where teaching hospitals had margin gains in excess of both their MA and national peers (Exhibit 7)

  1. It looks like IPAB is here to stay – which means pressure for innovators’ margins

Congress – both chambers, both parties – hate IPAB, but as IPAB’s implementation date draws closer (2015), the pay-fors required to eliminate IPAB grow larger. We think the political economics favor IPAB’s survival

IPAB is charged with keeping per-beneficiary Medicare spending growth within a close range of GDP or CPI, depending on the year, but cannot do this by reducing eligibility or raising premiums, and generally cannot do this by lowering payment rates to hospitals or physicians

This leaves only inputs to care as a source of savings. Commodity inputs are an unlikely source of savings – prices are set (efficiently) by competition. Innovators lose by default – higher margin inputs to care are the only obviously available source of savings available to IPAB

It follows that IPAB likely operates by reducing the costs of higher margin (i.e. innovative) inputs into Medicare beneficiaries’ care (Exhibit 8)

  1. The AWP replacements also are here to stay – which means pressure on generic dispensing margins

The ACA includes the statutory authority for CMS to create (and/or make public) two alternatives to Average Wholesale Price (AWP), specifically Average Manufacturer Price (AMP) and National Average Drug Acquisition Cost (NADAC)

The Secretary of Health has been proceeding on both the AMP and NADAC fronts since August of last year[8]. Alabama shifted to AAC – essentially an Alabama-specific version of NADAC – and by our estimates saved up to 37% per generic Medicaid Rx filled versus their previous AWP reimbursement formula. Now that the fate of ACA is clear, we expect the other states to quickly follow suit

Once the states ‘leave’ AWP, the only users would be commercial buyers. Because AMP or NADAC are better for commercial buyers, we expect them to also leave AWP, beginning with the larger / sophisticated buyers

Because information asymmetries associated with AWP appear to enable very large generic dispensing margins in the drug trades (PBMs especially, also Drug Retail and Drug Wholesale), the replacement of AWP by either alternative index (neither of which has the information asymmetries associated with AWP) can be expected to reduce generic dispensing margins[9]

  1. Section 1903(a) refers to the Social Security Act language that authorizes federal Medicaid payments to the states
  2. “HR 1683 State Flexibility Act”, May 11, 2011, CBO (
  3. Which provide a pathway for passage of Senate measures on a simple majority
  4. Please see “Why Employers are Likely to Drop Health Insurance – A Simplified View July 11, 2011, SSR; and “Post-2014 Reform Related Volume Gains are Modest” March 2, 2011, SSR
  5. If no subsidy-eligible workers are shifted to HIE, we estimate the Act would increase the number of insured workers by just over 5%. If at the other extreme all workers were shifted to HIE, we estimate the number of insured workers would fall by just more than 5%. The relationship between the percent of workers shifted to HIEs and the change in percent coverage of workers is linear, but the demand effects are not: as workers shift to HIEs they buy less expensive coverage, so the demand loss is exponential (see next footnote)
  6. The average actuarial value (AV, the proportion of claims paid by the plan) of ESI is about 0.82; we estimate the average AV of plans bought on HIEs will be around 0.65
  7. Please see “Hospitals Stable to Improving Net Pricing Power” January 24, 2012, SSR
  8. Please see “CMS Takes Concrete Steps Towards Replacing AWP” August 18, 2011, SSR; “CMS Starts to (Kind of) Publish AMP September 26, 2011, SSR; and A Detailed Comparison of the AWP Replacements: AMP v. NADAC October 3, 2011, SSR
  9. Please see “PBM Pricing Post-AWP – An Estimate of Sustainable Earnings Power November 14, 2011, SSR
Print Friendly, PDF & Email