Why Adverse Selection Pressures are Building Rapidly on the HIEs

Print Friendly, PDF & Email

Richard Evans / Scott Hinds / Ryan Baum / Hardy Evans


203.901.1631 /.1632 / .1627

revans@ / shinds@ / rbaum / hevans@ssrllc.com


December 7, 2015

Why Adverse Selection Pressures are Building Rapidly on the HIEs

  • Persons choosing to enroll in a health insurance exchange (HIE) plan arguably do so for one of three reasons: coverage is essentially free (for persons with lower incomes and correspondingly high subsidies, ‘Group 1’); anticipated health costs are higher than the net costs of acquiring coverage (‘Group 2’); or, net costs of acquiring coverage are higher than anticipated health costs (‘Group 3’)
  • Persons in Groups 1 and 2 are very likely to enroll, since in both cases enrolling obviously is economically beneficial. Unfortunately, combined health costs for Groups 1 and 2 are roughly 1.9x the combined premiums paid by (or on behalf of) Groups 1 and 2
  • As such the market needs significant enrollment from Group 3 in order for premiums and claims to balance. Persons in Group 3 can only be expected to enroll if the value they place on avoiding health cost risks is greater than their net costs of enrolling. Older patients (who are more risk averse) are more likely than younger patients to view this as a fair trade – so actual enrollees from Group 3 tend to be relatively old. Because the gap between premiums paid and costs incurred is smaller for older than for younger enrollees, the excess of premiums over health costs from Group 3 is too small to cover the excess of health costs over premiums from Groups 1 and 2
  • Premiums will grow faster than health costs in 2016, as underwriters attempt to cover the higher-than-expected health costs experienced in 2015. And, in 2016 and beyond, premiums paid by federally subsidized beneficiaries are likely to grow even faster than the premiums charged by health plans, because federal premium subsidies are geared to (slower) wage growth, rather than to (faster) health cost growth
  • As a result, from the Group 3 perspective the price of offloading health cost risks is growing faster than the underlying health costs – thus all else equal fewer and fewer Group 3 persons are likely to enroll. The likely result is accelerating premiums, as underwriters seek to cover claims for a shrinking pool of beneficiaries with higher average medical costs. Adverse selection, in other words
  • Policy options include allowing larger premium differences across age groups, reducing the scope of minimum essential coverage, allowing higher out-of-pocket maximums, merging HIE risks into healthier risk pools, increasing subsidies, and/or increasing penalties. None of these are remotely likely in 2016
  • Because adverse selection pressures build at an accelerating rate, we believe Congress will soon be faced with the binary choice of either making expansive HIE reforms, or allowing the HIEs to collapse. Our expectation is that Congress eventually will enact the requisite reforms, primarily because of the large numbers of persons currently receiving and/or eligible for subsidies
  • The publicly traded HMOs all appear to have relatively modest (<10% of members) exposure to the HIEs. We continue to prefer the Medicaid-predominant HMOs (e.g. CNC, MOH, WCG) to the commercial-predominant HMOs (e.g. AET, ANTM, CI, UNH). The former enjoy rising shares of a growing market where average contract values will increase; the latter face declining (employer-sponsored insurance) market shares and falling average contract values

Where we’re BULLISH: Biopharma companies with undervalued pipelines (e.g. AMGN, BMY, GILD, SHPG, VRTX); Biopharma companies with pending major product approvals (e.g. ABBV, ACAD, ADMA, ALIOF, AZN, BDSI, BIIB, BMY, CHMA, CLVS, CPRX, ENDP, GNMSF, ICPT, JAZZ, LLY, LPCN, MACK, MRK, NVS, PTCT, RLYP, RPRX, SHPG, SRPT, TEVA, UCBJY, ZSPH); SNY on undervalued basal insulin franchise and sales potential for Praluent (alirocumab), in addition to its undervalued pipeline; CFN, BCR, CNMD and TFX on rising hospital patient volumes; XRAY and PDCO on rising dental patient volumes and rising average dollar values of dental products and services consumed per visit; CNC, MOH and WCG on bullish prospects for Medicaid HMOs; and, DVA and FMS for the likely gross margin effects of generic forms of Epogen

Where we’re BEARISH: PBMs facing loss of generic dispensing margin as the AWP pricing benchmark is replaced (e.g. ESRX); Drug Retail as dispensing margins are pressured by narrowing retail networks and replacement of AWP (e.g. WBA, CVS); Research Tools & Services companies as growth expectations and valuations are too high in an environment of falling biopharma R&D spend (e.g. CRL, Q, ICLR); and, suppliers of capital equipment to hospitals on the likelihood hospitals over-invested in capital equipment before the roll-out of the Affordable Care Act (e.g. ISRG, EKTAY, HAE)


The HIEs charge too much to lay off risks, especially to those least interested (the young)

By sub-dividing potential beneficiaries into three distinct groups, we can better demonstrate why the health insurance exchanges (HIEs) appear to be suffering adverse selection losses, and why these losses are likely to continue:

  • Group 1: persons for whom coverage is essentially free[1];
  • Group 2: persons for whom likely[2] health claims paid exceed their net cost of obtaining coverage; and,
  • Group 3: persons for whom net costs of obtaining coverage are likely to exceed health claims paid

For simplicity we assume all persons in Groups 1 (free coverage) and 2 (health claims > coverage costs) enroll

Persons in Group 1 are very nearly a random draw of health risks, so assuming premiums have been set high enough to cover average health risks, total premiums paid by persons in Group 1 will more than cover the claims of the group. Specifically, we estimate paid health claims[3] in Group 1 are just 55% of premiums[4]

By definition, Group 2 consists of persons whose health claims exceed their net costs of acquiring coverage. Specifically, we estimate that total health costs in Group 2 exceed total premiums[5] by roughly 2.4x. There are a little more than twice the number of persons in Group 2 as in Group 1; on a combined basis we estimate that total health claims across Groups 1 and 2 are roughly 1.9x total premiums (i.e., a medical loss ratio or ‘MLR’ of 190)

It logically follows that the only way for the HIE market to balance (i.e. for premiums to match or exceed claims paid) is for a sufficient number of persons in Group 3 (coverage costs > health costs) to enroll. Since net premiums exceed anticipated health costs for persons in Group 3, the only practical reason for any of these persons to enroll is risk aversion – i.e. by enrolling, a person in Group 3 is paying premiums to avoid the risk of his or her out-of-pocket (‘OOP’) health costs exceeding the annual OOP limit ($6,850) for HIE-based plans

The odds of a Group 3 person enrolling change fairly dramatically with age, for two reasons. First and subjectively, older persons are more risk averse than younger persons; second and objectively, the gap between premiums and average health costs is (at least on the HIEs) much greater for the young (Exhibit 1). In a very real sense the current premium structure on the HIEs asks those who place the lowest value on risk avoidance (the young) to pay the highest price for risk avoidance – a fundamental but addressable flaw

Exhibit 2 compares self-assessed relative risk aversion scores[6] (green columns) by age group with the actual percentage of Group 3 persons who purchase coverage by age group (black line). Actual enrollment is even more age-skewed than self-assessed relative risk aversion, which is likely due to the fact that the young are being asked to pay larger risk premiums than the old


We estimate that 15% of current HIE enrollees fall into Group 1, 36% fall into Group 2, and 50% fall into Group 3. This implies that Group 3’s total claim costs would be roughly 15% of total premiums paid by Group 3. In aggregate, total costs across the three groups appear to be roughly 106% of premiums (Exhibit 3). We note that this 106 MLR estimate for the HIEs is very close to the apparent 104 MLR experienced by UNH on the HIEs (Exhibit 4)

Adverse selection pressures are likely to build at an accelerating rate

If our model is roughly indicative, as UNH’s HIE experience suggests it may be, then in the absence of substantial policy changes adverse selection pressures are likely to worsen. Let’s assume that Group 1 (free coverage) and Group 2 (health claims > coverage costs) continue to enroll as before – which is reasonable given that both groups have strong and continuing economic incentives to enroll. This implies that ‘balancing’ the HIEs – i.e. getting premiums to cover claims cost – is entirely dependent on getting a higher percentage of Group 3 persons enrolled (currently 22 percent across all age groups; see Exhibit 2, again)

Unfortunately the Group 3 enrollment rate is likely to fall, because beneficiaries’ premiums are inflating faster than health costs – meaning the cost of insuring against unexpected health costs is growing faster than the health costs themselves. We see this gap expanding for two main reasons. First, in 2016 the average underwriter must grow premiums faster than health costs simply to make up for the fact that premiums currently are lower than health costs. Second, in 2016 and beyond, even in the case of a constant sub-100 MLR the premiums paid by beneficiaries will tend to grow more rapidly than underlying health costs. This is because premium subsidies are indexed in such a way that if health costs grow more rapidly than wages (as is typically the case), federal subsidies will grow more slowly than health costs, which in turn means beneficiaries’ share of premiums tends to grow faster than health costs[7]

There are available solutions …

We see 6 options for addressing adverse selection pressures on the HIEs:

  1. Allow greater differences in premiums across age groups. The current maximum is 3x (premiums for oldest / premiums for youngest); however this is more narrow than the actual difference in health costs (average health costs for those aged 60-64 are 4.3x the average health costs for those aged 18-22). The 3x limit is simply too restrictive, and results in a greater difference between premiums paid and benefits received – i.e. a greater risk premium – for younger, healthier beneficiaries
  2. Reduce the minimum essential scope of coverage[8]. Narrowing the scope of coverage to exclude services that lack broad appeal can narrow the gap between premiums paid and benefits received for potential beneficiaries
  3. Allow higher out-of-pocket maximums. The ACA’s OOP maximum ($6,850) is substantially lower than the typical OOP max prevailing in the individual market prior to the ACA. By requiring plans to cover all costs above a lower threshold, Group 3 beneficiaries who might be willing to purchase less expensive coverage with higher OOP caps are effectively left out
  4. Merge a higher quality risk pool (or pools) into the HIE risks; or, make (proportional) participation on the HIEs a pre-requisite to being licensed to underwrite other, higher quality risks in a given jurisdiction
  5. Increase subsidies. Making Group 1 (coverage is essentially free) larger raises the proportion of HIE beneficiaries who are drawn from the ‘average’ risk pool. This option plainly isn’t cheap. Current premium subsidies are set in terms of the maximum percentage of 2014 income a beneficiary would have to pay for the second cheapest Silver plan in their area. If we reduce these maximum percentages by 25% (e.g. in the case of persons with incomes between 300% and 400% of FPL the maximum percentage would be reduced from 9.5% to 7.125%), total subsidies would increase roughly 47%, all else equal; and enrollment would balance to a projected MLR of 81
  6. Increase penalties. By increasing the certain costs (i.e. penalties) of being uninsured, the net costs of being insured effectively fall (because higher penalties lessen any savings that would have been captured by not paying premiums), bringing the net costs of coverage closer to the value of claims that are likely to be paid by coverage. If penalties were increased across the board by 50%, we again would expect enrollment to balance to a projected MLR of 81

Congress is likely to act, but only after things get much worse

There’s neither time nor political will to install any of these changes before 2016, so the only available option is for underwriters to raise premiums. All else equal (i.e. assuming no health cost inflation) premiums would have to rise by 16% to bring the HIEs’ MLR to a threshold 92% (roughly the non-profit standard, which we view as a theoretical minimum for a sustainable market). We believe per-capita health cost inflation is running at roughly 3-4%, which implies that average premiums need to increase by as much as 20% if the market is to reach a 92 MLR

Looking further ahead, the six options we’ve listed are in declining order of political feasibility. All of the options are politically challenging, but we believe that options 1 through 3 are at least politically feasible – and that all options are economically impactful. That said, we doubt the first three options are sufficient in and of themselves to eliminate adverse selection pressures, which implies that an effective re-structuring of the HIEs means enacting some or all of options 4 through 6 – and this is a very difficult political lift. Even accounting for the political uncertainties associated with the pending general election, we find it hard to imagine options 4 through 6 being enacted in the near- to mid-term

Adverse selection pressures tend to build very rapidly; as the costs of coverage balloon past the value of coverage the well leave and the sick remain, until the market collapses. Our belief is that adverse selection pressures are strong enough and building fast enough, that Congress will soon be faced with the binary choice of either passing comprehensive reforms, or allowing the HIEs to fail

Our best guess is that Congress ultimately will enact comprehensive reforms, but only after adverse selection pressures have compounded – things will get much worse before they get better. Two key reasons we believe Congress ultimately will act: 1) by the time the HIEs are in full crisis the national political frame will have shifted from the Obama Administration, to a new administration and new Congress. We believe this allows a corresponding tilt in the political framing at the level of a congressional district – away from a strategy of unified opposition to Obama Administration policies, and toward a strategy that is more immediately in tune with constituents’ interests. 2) As the political frame shifts from national R vs. D to local political economics, members of Congress are faced with the fact that very large average subsidies (about $3,200 per beneficiary) are being received by (or are eligible for receipt by) very large blocks of likely voters. To provide some sense of scale, there are as many current subsidy recipients as there are teachers, and as many subsidy eligible persons as there are living veterans

What this means for the HMOs

Details on the publicly traded HMOs’ relative exposures to the HIEs are spotty (Exhibit 5); however it’s reasonably clear that all have fairly minor exposures. As such a build-up of adverse selection pressures on the HIEs is less a matter of near-term earnings risk, and more a matter of strategic uncertainty

We continue to prefer the Medicaid-predominant HMOs (e.g. CNC, MOH, WCG) to the commercial-predominant HMOs (e.g. AET, ANTM, CI, UNH) for reasons largely unrelated to the HIEs. Specifically, we see the Medicaid-predominant HMOs gaining share and average contract price (because of duals) in an expanding market (because we see hold-out states expanding Medicaid to at least 100FPL). Conversely we see the employer-sponsored insurance (ESI) market as one in which the larger national-account oriented underwriters are losing share (as members go to private exchanges), and as one in which average contract values are falling (as private exchange beneficiaries choose plans with lower actuarial values)

  1. Net costs, after considering the effects of all subsidies and penalties, are less than 1% of annual income
  2. The idea here is that a significant proportion of health costs are in fact predictable, and that persons who expect to have significant health costs are likely to seek coverage if net premium costs are less than the costs of paying for necessary medical care out-of-pocket
  3. Throughout this analysis we use Medical Expenditure Panel Survey (MEPS) data to estimate total health spending for persons on an age and income-adjusted basis
  4. Throughout this analysis we assume the average plan purchased is Silver, which by extension means that HIE coverage pays 70% of the health costs predicted by MEPS
  5. Which includes not only the cost of coverage borne by the beneficiary, but also federal premium subsidies
  6. A subset of MEPS recipients are asked to self-assess the degree to which they agree with the statement “I do not need health insurance”
  7. See Section 1401(a)(ii), entitled “Indexing”, of the Affordable Care Act
  8. We realize that there are very inexpensive plans that putatively meet the minimum essential scope requirement. We’re not suggesting these plans be accommodated. Rather, we’re suggesting the loopholes that allow for these plans be eliminated, and that the minimum essential scope be narrowed for plans that are compliant with both the letter and intent of the requirement

©2015, SSR, LLC, 1055 Washington Blvd, Stamford, CT 06901. 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) and to Merck (MKGAY) on both securities-related and non-securities-related topics


Print Friendly, PDF & Email