Co-Pay Cards: A Bottle for the Drug Pricing Genie

Print Friendly, PDF & Email

Richard Evans / Scott Hinds / Ryan Baum


203.901.1631 /.1632 / .1627 richard@ / hinds@ /


August 8, 2012

Co-Pay Cards: A Bottle for the Drug Pricing Genie

  • Drug manufacturers pay rebates to avoid exclusion from a formulary and/or to gain preferred formulary position. At retail, formulary position translates into co-pays; consumers pay higher ‘tier 3’ co-pays for non-preferred brands and lower ‘tier 2’ amounts for preferred brands. Co-pays commonly are fixed dollar amounts that do not vary according to drug price (a given tier’s co-pay is the same regardless of drug price), and over time these amounts have grown much more slowly than drug prices
  • Manufacturer-sponsored co-pay cards are a relatively new phenomenon; these subsidize consumers’ co-pay obligations, and serve as an alternative to paying additional rebates for preferred formulary status (particularly if the preferred rebate is more than the tier 3 – tier 2 spread). Last year 28% of commercial (non-Medicare / non-Medicaid) retail prescriptions were processed with a co-pay card, up from 14% in 2010; currently 69 of the top 100 US brands (by sales) offer co-pay cards
  • Before co-pay cards, formulary managers could not grow tier 3 co-pay amounts at (or anywhere near) the rate of drug pricing for (plan sponsors’ wholly accurate) fear that medically necessary consumption would be lost. Now that co-pay cards are commonplace, this constraint is vanishing. More to the point: because manufacturers can be relied on to use co-pay cards (or pay rebates) to limit growth in consumers’ out-of-pocket costs, we believe formulary managers are free to accelerate growth in co-pay amounts (particularly tier 3)
  • The optimal formulary strategy appears to be: fixed dollar tier 2 co-pays that grow slowly if at all; and percentage tier 3 co-insurance that grows at the rate of drug pricing. This accelerates growth in the tier 3-2 spread, creating the potential for larger preferred rebates (or forcing manufacturers to offer larger co-pay subsidies)
  • Very few claims are currently processed with percentage tier 3 co-insurance; most tier 3 co-payments remain fixed dollar amounts. If we assume the share of tier 3 commercial claims processed under a percentage co-insurance design grows to 80% over a five-year period (and that the current average tier 3 co-insurance rate of 37% holds), pricing power during this period (for brands subject to tier 3 status) is reduced by about one-third (i.e. a 10% rate of list price increase yields a 6.8% rate of net price gain after rebates). If we assume the tier 3 co-insurance rate grows to 50%, pricing power over this period for affected brands is reduced by more than two-thirds (i.e. a 10% rate of list price increase yields only a 2.9% rate of net price increase). For comparison, rebate growth in the trailing decade reduced brand list price growth by less than one-tenth (from 10% to 9.3%)
  • Since 1980, US real price gains explain almost half of US drugs’ total top-line growth; over the last several years US pricing is more than 100% of US drugs’ sales growth. Current valuations imply pricing power continues at or near current levels through the period covered by consensus estimates; we expect pricing power to erode
  • Manufacturers for whom price increases are a higher percentage of current growth, and who have higher average prices (the effect of the strategy we outline is more pronounced at higher starting prices), are most at risk; LLY, AZN, BMY are most at risk; ABT, BAY, GSK, NVS and SNY are somewhat less exposed
  • We recognize such a shift in formulary strategy is bullish vis a vis PBMs; however because generic dispensing spreads are more important than rebates, and likely to fall, on net our bearish PBM thesis is unchanged

Co-pays and co-opetition

Drug manufacturers negotiating formulary access face two distinct threats: 1) the threat of exclusion, and 2) the threat of ‘anti-preference’ (or conversely the opportunity to gain preference). The threat of exclusion generally is weak because of plan sponsors’ reluctance to exclude commonly used drugs outright. However the ‘threat of anti-preference’ can be quite strong: using tiered co-pays, formulary managers can steer patients toward preferred brands that offer higher rebates, without having to exclude any brand completely

Formularies with tiered co-pays emerged in the 90’s; early two-tiered formularies (tier 1 = generic, tier 2 = brand) had very little rebate leverage, as the threat of exclusion was relatively weak, and the single brand tier created no incentive for consumers to use a preferred brand[1]. Three-tiered (tier 1 = generic, tier 2 = preferred brand, tier 3 = non-preferred brand) formularies began emerging in the late ’90’s and became commonplace by 2003 (Exhibit 1). Again the threat of outright exclusion was weak; however brand preference was made more powerful by virtue of consumers facing higher out-of-pocket costs for non-preferred brands. The strength of the anti-preference threat was still constrained by relatively modest spreads between non-preferred (tier 3) and preferred (tier 2) brands (i.e. the ‘tier 3-2 spread’); however this spread has grown with time (Exhibit 2)

Crucially, up until now the formulary manager v. manufacturer competition has not been entirely zero-sum; rather this competition exists in a much broader framework of mutual benefit. Despite being in zero-sum competition over the proportion of total drug price growth that is either rebated or kept by the manufacturer, both the formulary manager and the manufacturer benefit when drug price grows[2]. Co-pays have grown more slowly than drug price inflation (Exhibit 3), and the same goes for the tier 3-2 spread (Exhibit 4); thus fixed dollar co-pays have almost wholly shielded consumers from drug price growth. Brand prices net of rebates have grown very nearly as fast as brand prices at list, whereas brand rebates have grown slightly faster than brand pricing (Exhibit 5). N.B. despite the fact rebates have grown at faster rates than brand price, brands still capture the lion’s share of total pricing gains (Exhibits 6, 7)


Rebates paid by manufacturers are the sum of two values, the first being an access rebate paid simply to be included on the formulary (tier 3), and the second being an additional rebate paid for preferred positioning on the formulary (tier 2)[3],[4]. Returning to the notion that the threat of anti-preference is stronger than the threat of exclusion, it follows that preferred rebates should be larger than access rebates

Mechanically, formulary managers’ leverage for driving brand share (and thus demanding preferred rebates) is the tier 3-2 spread; the larger the difference between consumers’ non-preferred (tier 3) and preferred (tier 2) co-pays, the larger the likely market share consequences of having your brand listed on either tier. Until recently, the dollar amount of preferred rebates paid for (tier 2) status was less than the tier 3-2 spread; however for multiple reasons[5] the dollar rebates paid for preferred access have grown to be equal to or slightly greater than the tier 3-2 spread[6]. This explains much (and perhaps all) of the explosion of manufacturer-sponsored co-pay cards (Exhibits 8, 9), as these are an obvious alternative to paying rebates larger than the tier 3-2 spread

For example, consider a drug having a $200 list price, and a formulary with a $30 tier 2 co-pay, a $50 tier 3 co-pay, and thus a $20 tier 3-2 spread. Assume the access rebate required to be listed on formulary (tier 3) is $10, and the incremental rebate required for preferred status (tier 2) is $30; together these leave the manufacturer with a net price after rebates of $160. Assume for the moment that co-pay cards are frictionless[7], in which case the manufacturer’s best option is to pay the $10 access rebate and distribute co-pay cards that can be used to offset consumers’ $50 tier 3 out-of-pocket obligation by $20, such that the consumer now has the same effective out-of-pocket burden ($30) as would have been the case had the drug been listed on tier 2. The manufacturer’s net price in this case is $170 ($200 – $10 access rebate – $20 co-pay offset), v. the $160 net if the manufacturer had agreed to the $30 incremental rebate (on top of the $10 access rebate) for preferred status

It follows that as long as co-pay cards are an efficient and widely accepted[8] alternative, manufacturers will not routinely pay rebates for preferred access that are larger than the tier 3-2 co-pay spread

Tipping point

This implies that unless the tier 3-2 co-pay spread begins growing at the rate of drug price inflation – which it never has (Exhibit 4, again) – that formulary managers will be unplugged from the benefits of drug price growth; and, correspondingly that manufacturers will capture a larger percentage of total list price gains

From the formulary managers’ perspective the obvious ‘mechanical’ response is to raise tier 3 levels, opening up the tier 3-2 spread and correspondingly increasing the size of preferred rebates. The practical limit on this option has long been consumer elasticity – despite having been protected from real price growth by fixed dollar co-pays, consumers are nevertheless maxed (Exhibits 10, 11). Thus at first glance raising tier 3 co-pays runs the risk of substantially reducing medically necessary consumption

Paradoxically however, the option to raise tier 3 co-payment levels has become more feasible as the availability of co-pay cards has grown. Manufacturers cannot allow consumers’ out-of-pocket burdens to grow in real terms[9], thus formulary managers and plan sponsors can generally rely on manufacturers to protect consumers from rising tier 3 co-pays – whether by paying the rebate required to be placed on tier 2, or by providing co-pay cards that reduce consumers’ out-of-pocket burden to (or nearly to) the tier 2 level

Pulling all of this together, broadly speaking we believe formulary managers’ best path is to exploit (rather than fight) the general availability of co-pay cards by expanding the spread between tier 2 and tier 3 co-pays. More specifically, we believe formulary managers’ best path is to:

  1. set tier 2 co-payments at a fixed dollar amount that grows very slowly (if at all);
  2. shift tier 3 from fixed dollar co-payments to a co-insurance percentage; and
  3. offer a ‘no-haggle’ preferred rebate equal to +/- 0.95x the tier 3-2 spread (calculated using tier 3 co-insurance and tier 2 fixed dollar co-pay)

This simple strategy achieves several key objectives for formulary managers:

  1. growth in the average size of preferred rebates accelerates;[10],[11]
  2. the percentage of manufacturers paying rebates rather than issuing co-pay cards rises;[12],[13] and
  3. formulary managers effectively put manufacturers on the hook (and take themselves off) in terms of bearing primary responsibility for shielding consumers from brand price inflation

What this means for drug pricing generally, and manufacturers specifically…

… depends entirely on how aggressively formulary managers press the strategy. As a conservative scenario, we assume nothing changes other than a steady shift to a percentage co-insurance rate for tier 3, growing from an estimated 5% of claims processed in 2012 to 80% in 2017. The tier 3 co-insurance rate is set at 37%, consistent with the average prevailing tier 3 co-insurance rate. Tier 2 co-payments (currently $30) remain set in fixed dollar terms, and inflate at the rate of CPI; average brand drug pricing is $180, and the average rate of change in brand drug list pricing is 10%. We further assume that manufacturers will not allow consumers’ average out-of-pocket (OOP) expenses to rise; accordingly they must either pay the rebates required to get their drugs on tier 2, or provide co-pay cards that reduce any tier 3 co-insurance dollar amounts to the prevailing tier 2 level

Under these assumptions, as brand list grows, actual tier 3 OOP payments for beneficiaries on the tier 3 co-insurance plan grow at the rate of brand inflation. Manufacturers, obligated by consumer elasticity to ensure that all consumers pay nothing more than the tier 2 rate, either pay a rebate (nearly) equivalent to the tier 3-2 spread, or issue co-pay cards that offer subsidies in the amount of the tier 3-2 spread

In the case of drugs that are legitimate candidates for tier 3 formulary status[14], manufacturers’ 10 percent list price increases are offset by correspondingly larger rebate amounts (or co-pay card subsidies) such that net pricing gains across a five-year period are reduced by one-third to a CAGR of just 6.8 percent (Exhibit 12). Importantly, because pressure on net pricing grows as more beneficiaries fall under the tier 3 co-insurance design, the 6.8% CAGR belies the fact that net pricing gains have been cut nearly in half by 2017 (Exhibit 12, again). This pricing headwind is much stronger than the trailing effect of commercial rebates on net pricing growth. From 2002 to 2011 growth in commercial rebates reduced pricing power by less than a tenth; the 10% list price CAGR drove a 9.2% net price CAGR (Exhibit 13)

If manufacturers can truly be relied on to protect consumers’ from rising average OOP spending, then formulary managers can press the tier 3 rate higher than the currently prevailing average of 37%, and this rapidly increases formulary managers’ strength. If the tier 3 co-insurance rate is raised 5% per year from 37% in 2012 to 50% in 2015, and all other assumptions are held equal, a brand list pricing pace of 10 percent is reduced by two-thirds to a 3.3% net gain (Exhibit 14). Importantly, in this scenario the pricing headwind is steadily growing, such that net pricing gains are reduced to just over 1 percent by 2017 (Exhibit 15)

So that we don’t overstate the longer-term headwind to pricing, it’s important to recognize that the headwind is greatest during the transition from the current fixed-dollar predominant strategy to the tier 3 co-insurance strategy we’ve outlined here. Once most patients have been shifted to tier 3 co-insurance, the pricing headwind stabilizes, though at a higher rate than before the strategy was put into effect. Assuming a constant tier 3 co-insurance rate of 37 percent, once the transition from fixed tier 3 to tier 3 co-insurance is complete, a list price rate of 10 percent would be reduced by just more than one percent. Assuming the higher co-insurance rate of 50 percent, a list price growth rate of 10 percent would be reduced by just under 1.5% (Exhibit 15, again). For comparison, rebates from 2002 to 2011 reduced a 10 percent gross pricing rate by 73bp (Exhibit 13, again)

Real pricing power in the US prescription market has long played a central role in sales growth; since 1980 real pricing gains explain slightly less than half of total sales growth (Exhibit 16). Because of weak volume growth and eroding mix, more recently US pricing gains have been substantially more than 100% of real US revenue growth

Company by company, the earnings relevance of diminished US pricing power varies according to several variables, which we summarize in Exhibit 17: 1) the product of US pricing and US as a % of worldwide sales; 2) average dollars per Rx (higher priced drugs are more susceptible to the strategy we’ve outlined); 3) R&D as a % of sales (higher R&D spending offers a source of quick cost offsets); 4) dividends as a percentage of gross profits; and 5) forward PE ratios (presumably higher priced stocks are more susceptible to loss of pricing power). In all cases higher values denote greater risk, and lower values less risk. We believe LLY, AZN and BMY are most at risk; conversely ABT, BAY, GSK, NVS and SNY are relatively less exposed

Exh 17: Comparison of metrics that may indicate exposure to real pricing risks

‘US pricing’ is measured for US branded Rx business across the period ’07 – present; ‘US as % of WW sales’ is US branded Rx business as % of total WW sales for trailing 12mo.; ‘Sales-wtd avg product price’ is measured as sales-weighted average of TRx $’s / TRx by branded product for trailing 4Q; ‘ratio of div / gross π’ is common dividends divided by gross profit for trailing 12mo.; ‘Availability of R&D offsets’ is 1 – R&D as a % of sales for trailing 12mo.; ‘fPE’ is the ratio of current share price to consensus EPS estimate for 2014

Sources: BLS; Bloomberg; Company filings; FDA Orange Book; IMS Health; S&P CapIQ; Wolters Kluwer Price Rx; SSR analysis & assumptions

  1. In reality two-tiered formularies relied almost wholly on threat of exclusion, the idea being that brands (all of which are tier 2) were on-formulary if they paid the rebate demanded and off-formulary if they did not. Consumers’ co-pays were the same whether the brand they’d been prescribed was preferred or non-preferred, thus consumers with prescriptions for non-preferred brands fought efforts to have these switched. Importantly, retail pharmacies were very much on the side of consumers – retailers needed to fill the Rx as written as quickly as possible to avoid losing that consumer to another store, so where overrides were available (and for a considerable period these overrides were commonly available) that allowed the non-preferred Rx to be filled, retail pharmacists routinely filled non-preferred brand scripts. Mail was a different story; switching to preferred from non-preferred occurred at much higher rates by virtue of 1) the mail order pharmacy being part of the formulary manager’s corporation; and 2) the amount of time the mail-order pharmacist had to call doctors and request that a prescription be switched. Mail share of dispensing was growing, but it soon became clear it would not displace retail, which made the need for additional formulary tiers very clear
  2. This is true even though most if not all rebates are now commonly passed through to plan sponsors. The fact that rebates are available through formulary managers but unavailable to sponsors who would choose not to use a formulary manager serves to drives sponsors to the formulary managers
  3. We’re not suggesting that these two values are defined separately in all brand rebates; rather we’re arguing that in the case of a brand with preferred (tier 2) formulary status, that the total rebate paid is the consequence of two distinct ‘pressures’, namely the threat of exclusion (access rebate) and the value of preference (or threat of anti-preference, tier 3 rebate)
  4. Obviously we appreciate that many formularies now are 4-tier; however with tier 4 generally being reserved for specialty and lifestyle drugs, the existence of tier 4 is somewhat peripheral to our main arguments regarding preferred (tier 2) and non-preferred (tier 3) status for more traditional retail brands. Accordingly, for simplicity we ignore tier 4 throughout this note
  5. Stronger formulary managers, reduced brand pricing discipline as Medicare Part D reduced Medicaid’s share of total drug sales, and drug price growth in excess of co-pay growth
  6. We think this became true for many drug categories in ‘08/’09; please see “Co-Pay Cards and the Stalling of Drug Rebate Growth”, SSR, LLC, January 5, 2011
  7. I.e. zero administrative costs, perfect availability to those consumers facing the higher co-pays, and no use by consumers facing only preferred co-pays
  8. Co-pay cards cannot be used for Medicare or Medicaid Rx’s. Until recently co-pay cards could not be used by MA residents, but this is no longer the case. CVS no longer routinely accepts co-pay cards for patients under its benefits filling prescriptions in its stores (or by mail); however CVS does accept co-pay cards at its retail outlets if the patient is covered under some other drug benefit; and irrespective of policy CVS (Caremark) beneficiaries can still use co-pay cards in non-CVS outlets, because outside of its own stores CVS cannot reliably see whether a co-pay card is used
  9. We’re explicitly assuming that if consumers’ out-of-pocket burdens begin to grow at historic drug price growth rates, that consumer demand elasticity is such that gross profits would actually fall
  10. Under the prevailing standard of fixed co-payments for tiers 2 and 3 that grow more slowly than drug inflation, where preferred rebates commonly are equal or nearly equal to tier 3-2 spread, preferred rebates will not grow more rapidly than the tier 3-2 spread, and accordingly are growing more slowly than (and wholly independent of) drug price inflation. By shifting to a fixed $ tier 2 co-payment / % tier 3 co-insurance, the tier 3 co-pay grows at exactly the rate of drug price inflation, the tier 2-3 spread grows faster than drug price inflation, and accordingly preferred rebates also can be expected to grow faster than drug price inflation
  11. Anticipating the ‘chicken or the egg’ question… we realize that by saying formulary managers can reduce co-pay card usage by discounting rebates relative to co-pay cards, that formulary managers are ‘re-exposing’ consumers to the out of pocket costs of tier 3. We don’t believe this is the case; co-pay cards are the nuclear genie that can’t be put back in the bottle – even if they disappeared entirely, manufacturers can always bring them back if the ‘ask’ for preferred rebates grows larger than the tier 3-2 spread
  12. If the rebate for preferred access is always at or near 0.95x the difference between tier 3 and tier 2, then paying a rebate is always preferable to issuing a co-pay card. Most obvious, the rebate is cheaper, being 0.95x the amount of co-pay subsidy the co-pay card would have to offer, to say nothing of the administrative costs associated with the cards. Less obvious, co-pay cards are hardly guided missiles – quite a large proportion of total use is off target, i.e. by consumers who would have consumed the drug irrespective of the co-pay card subsidy
  13. We appreciate that formulary managers pass most rebates through to plan sponsors; nevertheless formulary managers still should prefer rebates to co-pay cards. Strategically, if co-pay cards supplant preferred rebates entirely, then formulary managers’ market power declines – since no one needs a formulary manager to use co-pay cards. Tactically, formulary managers don’t pass 100% of rebates through – more like 88% — so the 12% rebate cut is real money
  14. The percentage of brand drugs that are ‘legitimate candidates for tier 3 formulary status’ frankly eludes us. Historically, we’d consider these to be highly interchangeable retail brands, e.g. Zocor v. Lipitor. The fact the brands were interchangeable gave the formulary manager the necessary power to have them bid against one another for the preferred tier 2 position, with the loser relegated to tier 3. In the glory days of interchangeable retail brands, about 34% (by dollar sales) of brands paid preferred rebates, but today that percentage would be somewhat (perhaps a lot) smaller. However looking ahead, we’re not so sure brands have to be interchangeable to be placed on (or threatened with placement on) tier 3. An aggressive version of the tier 3 co-insurance strategy we outline might simply take the position that all brands are on tier 3 by default, and only those that pay the necessary rebate are on tier 2. Even in the case of a drug for which there really is no ready substitute, the ‘auto-relegation’ to tier 3 policy makes the brand accountable for the elasticity effects of its pricing. If the manufacturer refuses the rebate and also refuses to issue a co-pay card, they lose business, but the sponsor loses medically necessary consumption. In all likelihood the percentage of brands that can be ‘defaulted’ to tier 3 will depend on the level of commitment taken by either side (formulary manager v. manufacturer) to its chosen strategy. If most formulary managers default a brand to tier 3, that brand will have to either pay rebates or issue co-pay cards – even if it’s a highly differentiated drug with no immediate competitors. However if most formulary managers default to the drug to tier 2, it will be more difficult for isolated managers to default it to tier 3. By the same token, if manufacturers resolutely refuse to offer co-pay cards, defaulting innovative brands to tier 3 – much less accelerating growth in tier 3 – becomes impossible. Formulary managers are more aggregated (consider ESRX’ share of prescriptions managed) and can take the pain of having an innovative brand on tier 3 longer than the manufacturer(s), accordingly we expect a high proportion of retail brands can be pressured under the tier 3 co-insurance strategy we’ve outlined. The simple fact that 69 of the top 100 selling US brands offer co-pay cards signals that most brand managers feel the need to offset co-pays already
Print Friendly, PDF & Email