Hedge Hospital Pricing Risks with Non-Rx Consumables

Print Friendly, PDF & Email

Richard Evans / Scott Hinds / Ryan Baum


203.901.1631 /.1632 / .1627

https://twitter.com/images/resources/twitter-bird-blue-on-white.png richard@ / hinds@ / baum@ssrllc.com


March 11, 2013

Hedge Hospital Pricing Risks with Non-Rx Consumables

  • Hospitals’ share price performance (v. SP500) is highly correlated with changes to real revenue; Hospitals tend to outperform on accelerating revenue, and vice versa. On this basis, pending volume gains (rising employment, reform-related expansion coverage) argue for owning Hospitals
  • Hospitals’ aggregate pricing is the wildcard; if prices fall as volumes increase share price performance arguably will be muted
  • If the 2 pct Medicare rate cut (sequester) were the only change to hospital payment rates, pricing would be more or less stable through the first years of the reform-related expansions; the Medicare cut and the dilutive effect of lower Medicaid rates would be offset by reduced costs for charity care
  • More realistically, disproportionate share (DSH) reductions, discretionary reductions to Medicaid rates by states, discounts to exchange-based plans following the initial enrollment period, and rising bad debt exposure as average plan generosities fall will lead to falling aggregate prices for Hospitals
  • Given current low valuations and the high likelihood of accelerating unit demand we continue to recommend an overweight position in Hospitals. However because Non-Rx Consumables (e.g. CFN, OMI) see very nearly all the benefit of rising admissions, but have more stable pricing; we recommend pairing Hospitals and Non-Rx Consumables rather than a more concentrated position in Hospitals alone

The share price performance of Hospitals as a sub-sector appears to be highly correlated with changes to revenue; Exhibit 1 compares real change in revenue (black, dotted line) to Hospitals’ relative share price performance v. the SP500 (green, solid line) since 1985. Accelerating revenue generally leads to outperformance, and vice versa. There is no clear pattern with respect to whether share prices anticipate or follow changes to revenue

Changes to volume (admissions) have been far more tightly correlated with share price performance than changes to real price; we believe this is true simply because volume changes have dominated changes to revenue, particularly since the early 1990’s (Exhibit 2). Ultimately we believe changes to revenue growth brought about by changes in price are just as relevant to share price performance

Hospital volumes are likely to accelerate substantially, because of both rising employment and the pending coverage expansions associated with the Affordable Care Act (ACA). This argues that Hospitals are likely to outperform, especially given the relatively cheap valuation starting point (Exhibit 3), assuming real pricing remains stable

This latter assumption is the most crucial in the Hospitals investment case – volume growth is very nearly a given, but the outlook for net pricing is far less certain. There are four major moving parts to aggregate price:

  1. Medicaid

The expansion of Medicaid inevitably reduces average pricing; Medicaid hospital payment rates are 37 pct below commercial rates, and states are likely to put further pressure on these rates as Medicaid grows as a percentage of state budgets. Physicians’ Medicaid payment rates have been increased to Medicare levels for 2013 and ’14 by the ACA, placing hospital payment rates under additional pressure. Finally, Medicaid disproportionate share (DSH) payments are being reduced per the ACA, which should lower aggregate Medicaid payments by an additional 1.4 pct

  1. Medicare

Hospital payment rates will fall 2 pct as a result of the sequester; and as with Medicaid, DSH payment reductions should reduce Medicare rates by as much as an additional 1.4 pct

  1. Exchange-based plans

We believe hospital pricing to exchange-based plans generally will parallel commercial rates before and during the exchanges’ initial enrollment period; after the initial expansion has settled down we would expect exchange-based plans to run narrower networks than many other commercial plans, with lower hospital pricing a likely result

  1. Charity care / bad debt

In economic (as opposed to accounting) terms we define charity care as the actual cost (not billed charges) of caring for persons who cannot reasonably be expected to pay. We define bad debt as the actual cost (not billed charges) for care extended to persons who, ex ante, could reasonably have been expected to pay. In the real world we recognize the distinction between charity care and bad debt is not always clear; and, we emphasize that the common practice of accounting for charity care and/or bad debt at billed charges (rather than at costs) is distorting, particularly when charges are billed at unrealistically high rates

Changes to tax reporting recently have made cost-based estimates of charity care and bad debt more accessible[1]; these data show that hospitals’ costs of charity care and bad debt are on average 2.5 pct and 2.3 pct of revenues, respectively

We expect coverage expansions under the ACA to reduce costs of charity care in proportion to the reduction in the nation’s uninsured. Using CBO estimates of 56M uninsured before coverage (Medicaid and exchange) expansions and 26M after, we estimate a reduction in charity care costs of 46 pct (26/56), or from 2.5 pct of revenues to 1.2 pct

We do not assume a reduction in bad debt. We believe a substantial proportion of bad debt is the consequence of insured persons not meeting their co-payment obligations, and the ACA’s coverage expansions should have little if any direct effect on this. In fact, because we believe the emergence of exchanges leads to lower average actuarial values (AVs) across the commercial market, we see more risk of bad debt (eventually) rising than falling[2]

In Exhibits 4 and 5 we attempt to put all of these effects into motion using reasonable assumption ranges. In both cases we’re comparing expected aggregate pricing in 2013 and 2016, assuming 2013 is the last pre-expansion year, and 2016 is a point at which expansion effects have largely played out[3]. We rely on the CMS actuary’s National Health Expenditure projections for the percentage of hospitals’ third party payments received from any given payor, and on the American Hospital Association’s member survey for an estimate of starting-point payment : cost ratios. Baseline numbers for charity care and bad debt are from the Modern Healthcare article cited in footnote 1

In Exhibit 4, we assume payor mix changes in line with the CMS actuary’s assumptions, and that payment : cost ratios remain static across payors, with the single exception of a 2 pct cut to Medicare hospital payment rates as a result of the sequester. The effects of the Medicare rate cut and Medicaid’s increased share of payments are offset by a reduction in charity care costs as a percentage of revenue, and the effect is that aggregate payment rates net of uncompensated care costs are virtually unchanged. More realistically, we would expect Medicaid rates to fall further as states react to expansion-related budget pressures, DSH payment reductions to lower both Medicaid and Medicare average payment rates by 1.4 pct[4], and for hospitals’ pricing to exchange-based plans to be at least marginally (for arguments’ sake we assume 5 pct) below prevailing commercial rates; taken together these assumptions would lower hospitals’ aggregate payment rates by 1.6 pct in 2016 (Exhibit 5)

To our minds, the basecase scenario in Exhibit 4 is too optimistic – meaning the very best case is that hospitals’ aggregate payment rates remain flat or degrade slightly. We believe the revenue pressures in Exhibit 5 are far more realistic, and that further downside risks (particularly larger Medicaid rate cuts, larger discounts for exchange-based plans after the initial enrollment period, and a rise in bad debt as average AVs in the commercial insurance markets fall) exist. Thus it’s our view that hospitals’ aggregate payment rates are more likely to fall than remain stable

For this reason, we recommend maintaining exposure to rising volumes but mitigating exposure to Hospital pricing pressures by holding Non-Rx Consumables as part of a broader healthcare portfolio’s ‘volume-levered’ positions. Non-Rx Consumables’ share price performance v. the SP500 has been in essence a lower-beta version of Hospitals’ (Exhibit 6). Names with more US-focused businesses (e.g. OMI, CFN) should benefit from rising US hospital admissions very nearly as much as Hospitals themselves, but with more stable pricing. Suppliers with more commodity-oriented lines (again OMI and CFN) should have the most stable pricing, as prices are at a stable competitive equilibrium, and thus less subject to further negotiation or regulation

  1. Please see: “Out in the Open”, by Melanie Evans and Joe Carlson, Modern Healthcare, December 19, 2011
  2. Actuarial value (AV) is the average percentage of allowable charges paid by the insurer. Current employer-sponsored (ESI) AVs average about 82; we believe average AVs of exchange-based plans will be +/- 65, and that the existence of lower AV plans in the exchanges will have the effect of pulling down average AVs in ESI. A person with an 82 AV plan pays 18 cents on the dollar (out of pocket) for hospital care, whereas a person with a 65 AV plan pays 35 cents on the dollar (again, out of pocket) for hospital care. Current collection rates on co-payments for insured persons are roughly 30 to 40 pct. Thus a unit of hospital care priced and billed at $1.00 to a person with an 82 AV plan would net about $0.88 ($0.82 collected from the plan, plus 35pct of the remaining $0.18 collected from the beneficiary). All else equal the same unit of hospital care, priced at $1.00, and billed to a person with a 65 AV plan would net about $0.77 ($0.65 collected from the plan, plus 35pct of the remaining $0.35 collected from the beneficiary)
  3. For the moment, we’re ignoring the crucial question of when the exchanges actually begin operating. We see very real risks of a delayed and/or staged roll-out
  4. A note of caution – because DSH payments are not shared equally across hospitals, and because it is not entirely clear how smaller post-ACA DSH payments will be allocated, we cannot assume that all hospitals will see a 1.4 pct change in DSH payments. Some will see larger reductions, and some smaller. In the case of the publicly traded for-profit hospitals, because these are generally less exposed to poorer beneficiaries than the non-profits, the DSH reductions are likely to have less than a 1.4 pct effect on Medicaid and Medicare rates
Print Friendly, PDF & Email