Amazon is a PBM
The Buy Box = Formulary placement. Seller Fees = Rebate payments. Amazon Prime Membership = Insurance membership
This morning, I read through Matt Stoller’s most recent article about Amazon. His newsletter, BIG, is an excellent read. You should subscribe to it. As I read the article, I thought to myself that “I’ve heard this story before.”
Amazon’s price game with “you can’t charge a lower price elsewhere” is very similar to the PBM-controlled pricing schema for pharmacy services.
In pharmacy, PBMs pay pharmacies on a “lesser of” scheme where the pharmacy is paid the lesser of 4 numbers:
1) the “AWP” (think MSRP- set by the manufacturer) minus a discount- typically a ~18+% discount for branded products and 25+% for generic products.
2) the pharmacy’s “usual and customary” price - the price they would charge outside of the insurance/PBM system.
3) the pharmacy’s submitted charge for the service delivered
4) a proprietary “MAC” price- a “maximum allowable cost” determined at the PBM’s sole discretion.
The impact of this scheme is that
1) AWPs for generic medications are grossly inflated. An example would be Ondansetron 8 mg tablets, which costs pharmacies <$4 for a bottle of 30 tablets, but bears an AWP of ~$1200 for that same bottle. Manufacturers know that pharmacies won’t buy their product if they bring the AWP in line with the actual sales price, so AWPs are insanely high and stay high. Generic manufacturers call this “selling the spread” - they sell their product to pharmacies based on the spread (or price differential) between the AWP and the actual price to the pharmacy.
2) pharmacy charges and usual and customary prices are inflated. Pharmacies know that if they set their price too low, they will miss out on potential revenue, so they set their prices very high. In particular, in the scenario where a rational U&C would be, say $50 for a prescription, but a PBM has a MAC price of $500, the pharmacy will have every incentive to set a U&C of >$500. More than 90% of all prescription business passes through a PBM, so setting a lower U&C to capture the remaining 10% of prescriptions doesn’t make much sense.
3) This situation of high submitted prices allows PBMs to profit by being in the middle through a scheme called “spread pricing”, in which they pay the pharmacy a MAC price and then charge the employer or insurer that is responsible for the claim a higher price. Say the pharmacy sets a U&C of $1000, the PBM sets a MAC of $500. The PBM will pay the pharmacy the lesser of those two numbers: $500. They might turn around and charge the employer $750, pocketing $250 simply by standing in the middle. They’ll even say they are saving the employer $250 off the pharmacy’s U&C!
4) This same scheme of high pharmacy submitted prices to capture the maximum revenue leads directly to the existence of GoodRx. GoodRx makes money by allowing pharmacies like CVS or Walgreens or Kroger with insanely high U&C prices to discount those prices for the 5% self-pay market without violating their agreements to submit their best “out of pocket” charge to insurers. GoodRx takes a fee from the pharmacy for this “service” of discounting their high price for the consumer, typically ~$8/prescription. GoodRx does this by processing the pricing through multiple PBMs who have different MAC prices in order to arbitrage the differential prices from each PBM. Adam Fein wrote a great article detailing how this works in more detail here.
There are companies out there who are trying to break this stupid system: the Cost Plus Drug Company, for example, sets an AWP only 15% above their sales price to pharmacies. Given the contract environment in which pharmacies operate, this means that their products will be sure losers for the pharmacies when sold through a PBM- a product with a cost of $1 with an AWP of $1.15 will fall under the “discount off AWP” scheme for payment- resulting in a maximum payment of $0.8625 to the pharmacy... so the pharmacy will buy for $1 and sell for $0.8625. That’s not sustainable.
Blueberry Pharmacy, operated by my friend Kyle McCormick, opts out of the PBM-controlled system and charges fair prices for all of their prescriptions. In fact, they charge a fixed markup on their actual cost to acquire their products. This works well for the 90% of prescriptions that are filled with generic products which bear low acquisition prices and therefore under a cost-plus system a low price to the consumer. It breaks down in the branded product space, because who would pay $450 out of pocket for their inhaler that their insurance will give them for a $50 copayment?
That brings up another point about branded products- their prices are inflated by PBM “rebate” shenanigans which are exceptionally similar to the Amazon seller fee system. Insulin for example bears an AWP of ~$400/box. Something like half its price is captured by PBMs in the form of a “formulary placement rebate”.
In the branded product space, coverage on a PBM’s formulary in a preferred position is very similar to the Amazon Buy Box. If you want to be the preferred product, you have to agree to pay the PBM a rebate for placement (or pay Amazon a seller fee and play by Amazon’s rules - like using Fulfillment by Amazon). This rebate is usually a percentage of your list price (in industry jargon this would be either the AWP or WAC). For insulin, these rebates typically exceed 50% of the sales price. The rational response to having to pay >50% of revenue to a PBM for formulary placement is that the manufacturer will raise their asking price to capture more revenue. As the manufacturer is in a monopoly position by law for their patented product, they are able to set their “list price” at whatever point they want, so increasing that list price in order to pay the “seller fee” or “formulary rebate” is a trivial exercise, which they engage in annually and sometimes multiple times a year.
In fact, the FTC sees issues with this rebate system (or at least Rohit Chopra does). I’d strongly recommend a read through the FTC’s report on rebates. The parallels between the Amazon system of seller fees, requirements to give Amazon Marketplace your best price, consenting to Amazon fulfilling your product using FBA and the pharmacy system of U&C charges, rebate setting, and the mail order PBM pharmacy poaching business from local pharmacies are strong.
The more I read about monopoly and monopsony positions in other industries, the more I realize that despite pharmacist’s complaints that our situation is uniquely bad, it’s not. Chicken farmers, Uber drivers, Amazon marketplace sellers, and journalists in the age of Facebook suffer similar plights to that of the independent pharmacist. We need a resurgence of local antitrust societies like we had a century ago to compare notes with folks in other industries and build a strong lobby to revitalize and enforce our antitrust laws. I think I’ll start with a book club reading the new (and old) antitrust canon.