AMZN: A Pharmacy Role Beyond Rx Fulfillment

Print Friendly, PDF & Email


Richard Evans / Scott Hinds

203.901.1631 /.1632

revans@ /


December 4, 2017

AMZN: A Pharmacy Role Beyond Rx Fulfillment

  • A clinically ideal formulary is naturally inclusive, offering patients and physicians the maximum number of valid options for any given condition. Because the threat of exclusion is used as a lever to extract rebates from brand manufacturers, real-world formularies are more restrictive than the clinical ideal
  • Rebates paid by manufacturers are not closely reflected in patients’ out-of-pocket (OOP) costs. During their deductible periods patients pay list price for drugs that manufacturers pay rebates on; once their benefits are active, patients’ OOP costs are lower, but still are not calibrated to the true net cost of the drug
  • Intermediaries capture fees as shares of rebates paid, and as straight fees; the small number of large intermediaries arguably means average total intermediary costs are higher than would be the case with a larger number of smaller intermediaries
  • We present a theoretical construct in which AMZN serves as a ‘benefits platform’, a role larger than fulfillment, but well short of actual underwriting. Manufacturers provide coupons that approximate their average commercial rebates, plan sponsors provide credits, and patients’ OOP costs are AMZN’s selling price, less the manufacturer coupons, less the plan sponsor credit
  • Plan sponsors can set their subsidies in a manner so that the patient’s OOP mimics the current standard of fixed dollar co-pays or co-insurance, though we see a more creative alternative under which plan sponsors offer the same fixed dollar subsidy for every drug in a given class. This way patients are exposed to the entire net price difference for drugs in the class, and should naturally prefer the lowest cost clinically suitable option – thus potentially collapsing net price differentials to a range that reflects the true relative values of the options within the class. And, patients are exposed to any drug price inflation that exceeds the rate at which plan sponsors inflate their fixed dollar subsidies
  • If manufacturers will offer coupons on AMZN that approximate their commercial rebates, plan sponsors can effectively shift their prescription benefits to AMZN by making subsidies available to their beneficiaries who use AMZN. Formularies would be more inclusive (patient exposure to net price differentials across classes, and to inflation across time periods, replaces the threat of exclusion as a means of controlling prices); patients would benefit directly from manufacturer discounts; and, intermediary fees would be lower (the scope of services plan sponsors would need in order to provide benefits on AMZN is narrower, and more intermediaries could provide that scope, reducing fees; and, there would be few if any rebate dollars for intermediaries to share)
  • Our construct is theoretical, but nonetheless imminently practical; we believe the construct shows the potential vulnerabilities of the status quo in US pharmacy, and by extension its major incumbents (especially CVS, ESRX, FRED, RAD, WBA)

A role beyond fulfillment

In an earlier note[1] we outlined how AMZN might successfully enter, and ultimately transform, US pharmacy. That note focused on AMZN’s potential role in fulfillment, specifically ruling out a role for AMZN as an underwriter of prescription benefits, the logic being that an underwriting role restricts AMZN’s share of fulfillment to its share of underwriting, and puts AMZN in the incongruous position of telling its customers what they can and cannot have. In retrospect, we think that a black and white, binary distinction between underwriting and fulfillment is artificial and overly constraining. A middle ground exists between underwriting and fulfillment where novel approaches to pricing, funding, and management of payments affords AMZN further opportunity for ‘constructive disruption’ in US pharmacy

The basic construct …

In our theoretical construct, manufacturers sell to AMZN, or to AMZN’s wholesaler, at the same 2%/30, net/31[2] terms given to any other direct buyer. AMZN then sells to the public at their cost of acquisition, plus a margin – just like any other pharmacy, though perhaps at a lower margin

Where our construct diverges from the traditional is in its use of manufacturer ‘coupons’ and plan sponsor ‘credits’. Manufacturers provide AMZN with product-specific coupons that reduce the list price by the same or similar percentage as the manufacturer’s average commercial discount on each product; these coupons are valid for any AMZN Rx transaction except where restricted by the terms of the coupon (see later) or where not allowed by law (e.g. Medicare Part D, Medicaid). Coupons are applied to transactions at the point of sale. As a result, consumers (insured or uninsured) would be able to purchase at a price level close to the manufacturer’s average commercial discount

If we assume meaningful manufacturer discounts are available on AMZN, then plan sponsors can create a prescription benefit that flows directly through AMZN, without the use of a traditional intermediary such as a PBM. This can be accomplished by giving beneficiaries credits – on AMZN – which can be used specifically for prescriptions, provided those prescriptions are within the plan sponsor’s formulary (including limits that specify drugs allowable for specific conditions). Beneficiaries would register on AMZN for their specific plan sponsor’s credits, and would be approved for access to their plan sponsor’s credits by the plan sponsor (or the plan sponsor’s agent). Like manufacturer coupons, plan sponsor credits are applied at the point of sale

Transparency, with patients benefitting directly from manufacturer discounts

The net result of manufacturer coupons and plan sponsor credits is that patients’ out-of-pocket (OOP) costs for prescriptions equals AMZN’s selling price for the Rx, minus manufacturer coupons, minus the applicable plan sponsor subsidy. Ideally, consumers would be able to see all elements of the price, i.e. AMZN’s list, the manufacturer coupon amount, and the plan sponsor’s subsidy amount. Complete transparency

Transaction-level consumer elasticity, rather than the national threat of exclusion, as a pricing lever

The details of how plan sponsors calibrate and format their subsidies would meaningfully impact consumers’ OOP exposures, and by extension the effect of manufacturers’ net selling prices (and changes to those net prices) on consumer choice. Current benefit designs commonly fix patients’ OOP costs to a specific dollar amount (fixed co-pays), which means the variance in product costs across a given therapeutic category is borne by the plan sponsor, as is all net price inflation. Plan sponsors less commonly cover a percentage of costs, resulting in the patient OOP also being a percentage of costs (co-insurance); in this construct both the sponsor and the beneficiary are partially exposed to price differences across products, and across time. In the AMZN-based alternative we’re theorizing, plan sponsors could replicate these same dynamics. Variable plan sponsor subsidies that are equal to AMZN’s selling price, less the manufacturer coupon, less a fixed dollar co-pay amount obviously would fix the patient’s OOP cost at a specific level, mirroring the current market standard of fixed dollar co-pays. Similarly, setting the sponsor subsidy to a consistent percentage of either AMZN’s selling price or the net of AMZN’s selling price and the manufacturer coupon would result in the patient’s OOP being a consistent percentage of either the list or net price (effectively co-insurance)

Plan sponsors have a novel third option[3], which is to set their subsidies as fixed dollar amounts, which presumably would vary by therapeutic area. This shifts the difference in costs across alternatives (within a given therapeutic class for a given diagnosis) to the patient, and directly exposes the patient to net price inflation[4]. The result – all else equal – is that patients prefer the lowest net price option among suitable therapies. Manufacturers are thereby more directly exposed to consumer, as opposed to plan sponsor, price elasticity – with consumers being the more price elastic of the two. Exhibit 1 shows how this might work using 3 stylized scenarios involving a $100 prescription with 10 percent list price inflation. In scenario 1, the drug maintains its 35 percent commercial discount across periods, resulting in 10 percent net price growth. The plan sponsor similarly increases its subsidy by 10 percent, resulting in 10 percent growth in the patient’s OOP cost. Scenario 2 alters scenario 1 by assuming the plan sponsor keeps its $40 subsidy constant in both periods, in which case 10 percent list (and net) price growth results in 26 percent growth for the patient’s OOP costs. Finally scenario 3 alters scenario 2 by having the manufacturer increase its coupon (by 21 percent) to offset the consequences of the fixed plan sponsor subsidy on patients’ OOP cost growth; in this case manufacturer net price can grow only 4 percent if the goal is to keep patients’ OOP cost growth at 10 percent

Broader formularies, and a narrower role for intermediaries

In the current market, plan sponsors rely on intermediaries to perform 3 main tasks:

  1. develop and maintain a formulary that is clinically appropriate and up to date;
  2. develop and maintain a formulary which acts as a counterweight to drug price inflation; and
  3. adjudicate (determine whether the claim meets all necessary requirements) and process (manage payment and in the case of home delivery, fulfillment) claims

A plan sponsor providing the type of AMZN-based prescription benefit theorized here would rely on AMZN’s capabilities to manage payments and fulfillment, but would (except for HMO plan sponsors) need assistance with items 1 and 2 (and perhaps with claims adjudication in item 3)

We believe item 1 is easily dealt with; any of several prominent schools of pharmacy would be willing to develop and maintain formularies that reflect current clinical standards, and could in theory provide these to plan sponsors on the AMZN pharmacy platform

Item 2 – use of the formulary as a counterweight to drug price inflation – works differently in this theoretical construct than in the current market, but it’s still an item on which non-HMO plan sponsors are likely to need assistance. In the traditional drug benefit, intermediaries influence net prices by excluding (or by threatening to exclude) manufacturers from coverage. The resulting price pressure comes at the cost of a formulary that is narrower than the clinically ideal formulary. In our construct, plan sponsors apply price pressure by setting fixed dollar subsidies by therapeutic category (thus collapsing price differentials across the category), and by limiting real growth in the fixed dollar subsidy (thus pressuring manufacturers to keep net price growth at or near the growth rate in plan sponsor subsidies). In effect, our construct trades the threat of national exclusion for the reality of consumer price elasticity, which we believe can produce just as much of a counterweight to drug price inflation, while simultaneously allowing a more open formulary

Calibrating those sponsor subsidies by therapeutic area, and across time, is a complex and potentially fast-moving task best performed by a specialist. And, plan sponsors’ interests arguably would be served by at least somewhat coordinated management of their subsidies, as this increases their combined punching weight regarding drug prices. Yet because this task is much narrower than the scope of services plan sponsors currently buy from PBM and HMO intermediaries, plan sponsors’ total spending on intermediaries arguably is lower in our theoretical construct. This is true not only because the effort required of the intermediary is lower, but also because a larger number of intermediaries would be able to provide the service, putting competitive pressure on intermediary fees

Other considerations

Why might manufacturers go along with this?

Our construct obviously relies on manufacturers’ willingness to provide credits that approximate their typical commercial discounts – and to have these in plain view of anyone. And, while manufacturers would be reducing their exposure to the challenges they face with current intermediaries, they’d be participating in the development of an alternate setting in which they would face significant pricing pressure, albeit of a different form. So why bring this about?

Manufacturers would face fewer dominant buyers in the commercial benefits space, an obvious strategic benefit. And, beneficiary lives managed on the AMZN platform are likely to have more open and inclusive formularies, presenting manufacturers with less risk of outright exclusion. Patients would benefit directly from manufacturers’ price concessions, as opposed to the current standard under which patients pay full retail prices during their deductible periods, and have OOP cost shares that don’t necessarily tie to drugs’ net prices once their benefits become active. Limiting or even eliminating the share of price concessions captured by intermediaries would result in lower OOP prices for patients

This construct requires considerable pricing transparency, and manufacturers would have to decide whether this is a good or bad thing. The literature on pricing transparency is mixed, but to our reading tilts toward the conclusion that transparency actually firms up pricing. It’s not clear whether this evidence would be sufficient to overcome manufacturers’ natural reluctance to provide pricing transparency

Why might plan sponsors go along with this?

We believe this construct would allow plan sponsors to establish and manage effective counterweights against drug prices, but with formularies that are more open, and total intermediary costs that are lower. We also believe that beneficiary satisfaction would be higher in more open formularies that allow access to all clinically relevant options at price differentials that presumably reflect the relative attractiveness of these options[5], as compared to closed formularies that offer no coverage for some options, and where price differences for covered options are more arbitrary. Finally, we believe plan sponsors would be better able to simultaneously tailor formularies more to their liking and maintain effective pricing pressure, where under the current construct plan sponsors need to stick more closely to a one-size-fits-all national formulary in order to optimize pricing pressure

Why might AMZN offer this type of ‘benefits platform’, instead of limiting its role strictly to fulfillment?

If AMZN limits its scope to fulfillment only, it locks itself out of fulfillment for roughly half of current beneficiaries – specifically those that are covered within the benefits of intermediaries (e.g. ESRX, CVS/Caremark) that do not want AMZN to enter the pharmacy space, and would take any practical steps to limit AMZN’s role. Only by convincing plan sponsors who currently have lives with ESRX and CVS/Caremark to shift to another platform can AMZN hope to capture the fulfillment business associated with these lives. By offering an attractive alternative to traditional intermediaries, AMZN likely expands its share of prescription fulfillment

Why does it have to be AMZN that organizes this type of alternative benefits platform – why can’t the retail chains do the same thing?

If we could set aside the retail chains’ reliance on the larger intermediaries – particularly the larger PBMs – there would be nothing stopping the chains from offering the same type of ‘benefits platform’ construct we theorize for AMZN. The problem is, in truth we cannot set aside the retail chains’ reliance on PBMs, who would be directly hurt by the chains taking such a step – in other words the retail chains face a Catch 22. AMZN is not at all reliant on the PBMs, and has no realistic prospect of having a cooperative business relationship with the PBMs, thus they have nothing to lose by launching an alternative benefits platform that works against the PBMs’ interests

Under the construct we envision, the manufacturer coupons made available on AMZN would be limited for use only to persons who were either: 1) paying cash (i.e. applying no other cost offsets or subsidies to the transaction), or 2) purchasing the Rx with the assistance of a plan sponsor who attests that they receive no other manufacturer concessions. AMZN can be placed in a position – by its participating plan sponsors — to establish and verify that (2) is in fact true; traditional retailers are not in such a position. Because of this, we believe it’s possible for manufacturers to make their coupons freely available, to the extent this is or ultimately may be required by law – without the concern that their coupons could be used off AMZN by insured beneficiaries purchasing prescriptions with the help of traditional benefits wherein the plan sponsor obtains rebates from the manufacturer

Why doesn’t this construct ‘collapse’ back to the current state of affairs?

This construct mitigates, but does not necessarily eliminate, the role of rebates in drug pricing. The potential exists for manufacturers to put lower ‘standard discounts’ on the AMZN platform for any beneficiary, reserving supplemental rebates for plan sponsors who, through their management of what’s covered and/or the relative OOP costs of what’s covered, give manufacturers critical levels of market share. And, if plan sponsors subcontract the management of their AMZN-based formularies to agents, the potential exists for the agents to be paid in share of rebates captured, which takes us right back to where we are today – albeit with more ‘agents’ and intermediaries than we currently have. But we doubt this would happen

In a marketplace where manufacturer coupons directly, dollar for dollar, impact the OOP costs of patients, there’s no reason for a plan sponsor (that does not skim rebates) to prefer market-share based rebates over straight discounts in the form of manufacturer coupons

Consider a class with two drugs, both with $100 list prices, and both with $30 manufacturer coupons. The plan sponsor anticipates an incremental $10/Rx market share rebate on one of the drugs (‘drug A’), but not the other (‘drug B’). Because of this the sponsor offers a $40 subsidy on ‘A’, and a $30 subsidy on ‘B’. If the plan sponsor achieves the share threshold and earns the $10 rebate its net subsidy cost on drug ‘A’ is $30 – the same as for drug ‘B’. The plan sponsor achieved nothing in exchange for taking the risk of extending a higher subsidy on drug ‘A’. Alternatively, the plan sponsor might believe it can meet the share threshold and win the $10 rebate with only a $35 subsidy on drug ‘A’ ($5 more than on drug ‘B’). If this is true, the sponsor wins a $10 rebate in exchange for risking $5, but unless the plan sponsor is motivated to retain a share of the rebate (and in our construct plan sponsors generally are not), then the difference between the $5 risked and the $10 gained goes back to beneficiaries in the form of lower premiums. And, if a $5 difference in OOP is enough to hit the market share threshold, the manufacturer would be far better off offering $5 more in coupon than offering $10 in share-based incremental rebate


  1. “AMZN: How to Enter (and Dominate) Pharmacy”, SSR Health LLC, November 5th, 2017
  2. 2% cash discount on balances paid within 30 days; no discount on balances paid after 30 days
  3. This option is theoretically possible in the current system, but is rarely if ever used
  4. Unless plan sponsors raise their subsidies in line with net price inflation. We suspect that in the case of therapeutic categories with no innovative, higher priced new entrants, plan sponsors would raise subsidies infrequently (annually), and in amounts that approximate CPI
  5. This statement assumes that manufacturers whose products are more desirable can afford to let patients’ OOP costs in this construct drift higher, and vice versa; and that the net result would be pricing differentials across products that mirror the relative attractiveness of the options on offer


©2017, SSR, LLC, 225 High Ridge Rd, 2nd Floor, Stamford, CT 06905. 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), Gilead Sciences (GILD), Bristol-Myers Squibb (BMY) and Sanofi (SNY) on both securities-related and non-securities-related topics. One or more of SSR Health’s analysts owns long positions in the following stocks: AERI, AGEN, AKAO, ALKS, ALNY, ANAB, ARRY, BMY, DERM, DOVA, DPLO, ESALY, GILD, GWPH, INCY, IONS, KALA, KMPH, LJPC, NSTG, PFSCF, PGNX, PTLA, RARE, RHHBY, RIGL, TBPH, THERF, TRVN, TSRO, VRTX

Print Friendly, PDF & Email