Pharmacy Benefit Managers, the Middlemen You Never Hired, Explained

Let me tell you about a business that has a hand in nearly every prescription in America, handles billions of dollars, decides which drugs your plan will cover, and has a name that sounds like the least interesting department at a mid-sized insurance company. It is called a pharmacy benefit manager, and you have never met one. You did not hire one. You did not choose one. And somehow it has an opinion about what you pay at the counter.

Folks, I have no objection to middlemen as such. A good middleman earns his keep: he finds buyers, finds sellers, reduces cost and takes a fair fee. What I object to is a middleman whose pay depends on a number you cannot see. So let me explain how this business works, step by step, in plain English, and then tell you what I think should be done about it, which is less than Washington thinks and more than the industry would like.

Who they are and what they do

A pharmacy benefit manager, or PBM, is a company hired by a health plan, an employer or a government program to manage the prescription drug benefit. It does several jobs:

1. It builds the formulary. The formulary is the list of drugs the plan covers, usually sorted into tiers. A drug on a low tier carries a low copay. A drug off the list may not be covered at all, or requires special approval.
2. It negotiates with drug makers. The PBM bargains for rebates, payments from the manufacturer back to the PBM or plan after the sale, in exchange for favorable placement on the formulary.
3. It builds the pharmacy network. The PBM contracts with pharmacies to fill prescriptions for its members and sets what it will reimburse them.
4. It processes claims. When you hand your card to the pharmacist, the PBM's computer decides in seconds what the plan pays and what you pay.
5. It often runs its own pharmacies. Many PBMs own mail-order and specialty pharmacies, and the largest PBMs are owned by, or part of the same corporate family as, large health insurers. The three biggest, CVS Caremark, Express Scripts and OptumRx, are tied to CVS Health and its insurer Aetna, to Cigna, and to UnitedHealth Group respectively. Three companies handle most prescription claims in the country.

There is nothing sinister about any one of these functions. A plan needs somebody to run the benefit, and large buyers have bargaining power that a single patient lacks. That is the legitimate argument for the PBM, and it is a good one. The trouble is in the details of how money moves.

The four pieces of the puzzle

The list price. Every drug has a list price, formally the wholesale acquisition cost. It is the manufacturer's published price before discounts. It is not what most payers pay, but it is the starting point for many calculations.

The rebate. After the plan's members fill prescriptions, the manufacturer pays a rebate, a percentage of the list price, to the PBM, which may keep some and pass some to the plan. A larger rebate buys a better place on the formulary.

Spread pricing. In some contracts the PBM charges the plan one price for a drug while reimbursing the pharmacy a lower one, and keeps the difference, the spread. The plan sees one number, the pharmacy sees another, and the PBM sits between them.

The patient's cost. A person's deductible and coinsurance are often calculated on the price at the counter, which is tied to the list price, not the rebated net price. Here is where the trouble starts.

A worked example, clearly labelled hypothetical

Let me walk you through the arithmetic with made-up numbers, because arithmetic is the best friend a skeptic has.

Imagine two drugs that treat the same condition. Drug X has a list price of $600 a month, and the manufacturer pays a rebate of $250. Drug Y has a list price of $320 and pays no rebate.

What does the plan actually pay, net of rebate? For Drug X: 600 minus 250 is $350. For Drug Y: $320. So the plan comes out ahead with Drug Y, by $30 a month.

Now suppose the PBM keeps 20 percent of the rebates it collects as part of its compensation. On Drug X, 20 percent of $250 is $50. On Drug Y there is no rebate and so no such income. Which drug does the PBM have a reason to put on the formulary's preferred tier? Drug X, which earns it $50, though it costs the plan more. Nobody has to be a villain for that to happen. The incentive is built into the pay structure, and incentives are patient.

Now look at the patient. Suppose he is in the deductible phase of his plan, which means he pays the full cost of the drug until he has paid a certain amount out of pocket. For Drug X he pays based on the $600 list price. For Drug Y, $320. The rebate that brought the plan's net cost down to $350 does not reach his pocket; it goes to the plan, or to the PBM, depending on the contract. So a patient can be paying $600 for a drug that the system nets at $350, while a cheaper, rebate-free alternative sits one tier off the formulary.

That is how a list price can rise while someone else pockets the difference. The manufacturer raises the list price, which raises the rebate it can offer, which buys a better formulary spot. Everyone in the chain but the patient at the deductible can see a benefit. Many commentators call this a perverse incentive; I call it a feature of paying people a percentage of a number somebody else controls.

I have to be fair on one point. PBMs say, with some justification, that rebates reduce the plan's overall cost and, through the plan, premiums. They also say that most of the rebates they collect are passed through to their clients. Both claims may be true in many contracts, and I cannot verify any particular one, which is the problem. The contracts are private. That is precisely what makes it hard to say who gets what.

The pharmacy side

Spread pricing and reimbursement also affect pharmacies, particularly independent ones. A PBM sets what it will reimburse a pharmacy for a generic drug through a list called a maximum allowable cost list. If the number on the list is lower than what the pharmacy paid to buy the drug, the pharmacy loses money on the sale. Many small-town pharmacists say that the reimbursement rates are not negotiable, since a pharmacy that wants to serve a plan's members must accept the network's terms.

That is the setup for the Supreme Court case that every state legislator working on this issue should know.

Rutledge v. Pharmaceutical Care Management Association

In 2015 Arkansas passed a law, Act 900, to regulate how PBMs reimburse pharmacies. It required that a PBM reimburse an Arkansas pharmacy at a price at least equal to the pharmacy's acquisition cost, and it gave pharmacies a way to appeal reimbursement rates that fell below that cost. The Pharmaceutical Care Management Association, the trade group for the PBMs, sued, arguing that the Employee Retirement Income Security Act of 1974, ERISA, pre-empted the state law.

Why ERISA? Because ERISA governs employer-sponsored benefit plans and contains a broad preemption clause, overriding state laws that "relate to" such plans. The PBM argument was that because PBMs administer drug benefits for ERISA plans, state regulation of PBM reimbursement effectively regulates the plan.

In Rutledge v. Pharmaceutical Care Management Association, decided on December 10, 2020, the Supreme Court ruled unanimously, in an opinion by Justice Sotomayor, that ERISA did not preempt Arkansas's law. (Justice Barrett, newly seated, took no part.) The Court reasoned that Act 900 was a form of cost regulation, one that affects what a PBM pays pharmacies. It did not dictate plan choices, and it did not require anyone to structure a benefit in a particular way. A state law that merely increases costs, even costs passed through to plans, is not necessarily one that "relates to" a plan in the sense the statute forbids.

Here is why a conservative should like that decision. It restores a measure of power to the states, which is where the Founders expected most regulation of local commerce, and especially of pharmacy practice and insurance, to take place. The Tenth Amendment keeps to the states the powers not delegated to the United States, and the police power to regulate the health and economic affairs of their residents is the heart of what is left. A federal statute passed in 1974 to protect pensions was being used to stop a state from checking how a middleman pays a pharmacist. The Court said no. That was the correct result, and an unusually clean one: eight votes out of eight.

And here is the lesson: if you want to know what a PBM does to a small-town pharmacy, you do not need a federal agency. You need a state legislature that is willing to look.

What not to do

The tempting response to a murky industry is to create a federal agency, a price board, that decides what drugs should cost. I argue against it, for three reasons.

1. A price board repeats the mistake of treating the symptom. The problem is not that the number is too high in some abstract sense. The problem is that no one can see how the number is built. A board that sets a price does not solve the opacity; it adds one more opaque actor with its own incentives and its own lobbyists.
2. A price board sets a number and ignores the incentives that produced the problem. The incentive is payment as a percentage of rebates. If the structure stays, the behavior stays.
3. It does not fit our system. Setting prices for private goods by federal decree is a task the Constitution does not assign to the federal government, and one that, as I argue in Price Controls on Medicines Today Mean Fewer New Medicines Tomorrow, tends to shrink the supply of what is being priced.

What a transparent market looks like

Here is what I would do instead, in order of how much trouble each causes the people who like the current system.

• Require disclosure to the payer. Plan sponsors, meaning employers and unions and states, should see exactly what the PBM collects from manufacturers and what it pays to pharmacies. An employer who is paying the bill is owed the receipt. Congress has taken some small steps toward this, and it should finish the job.
• Delink compensation from list price. Pay the PBM a flat, disclosed fee for its services rather than a share of rebates. That removes the reason to prefer a high-list-price drug. The PBM then earns money by saving the plan money, which is what we thought we were paying it to do.
• Pass rebates through at the point of sale. If a drug is rebated, let the patient's deductible and coinsurance be calculated on the net price. The patient pays what the system actually pays.
• Ban spread pricing, or require it to be disclosed to the plan. A plan that agrees to pay a PBM more than the pharmacy receives, knowing it, is making a business decision. A plan that does not know is being charged a hidden fee.
• Protect the patient's right to know. Congress already acted on one piece of this. In 2018 it prohibited the so-called gag clauses that kept pharmacists from telling a customer that paying cash would cost less than paying the copay. That law, signed in October 2018, passed with bipartisan votes and is an example of a small, sharp reform that works.
• Let the states act. After Rutledge, a state can regulate how PBMs reimburse pharmacies within its borders. Fifty legislatures experimenting is a better research program than one federal board.
• Break up the conflicts. A PBM owned by an insurer that also owns the pharmacy that the PBM steers patients toward has a conflict of interest that disclosure may not cure. Whether to require separation is a harder question, and conservatives disagree about whether it is a job for antitrust law or for Congress. I would at least make the conflicts visible.

None of that sets a price. All of it lets you see one. Milton Friedman's argument about the social responsibility of business, in Friedman's 1970 Essay on Business Responsibility Still Answers Its Critics, was that businesses should operate within the rules and the law. A good rule here is that the people paying a bill are entitled to see how it was built.

Why this matters to a family

The reason I care about this is not the industry, but the family at the counter. Take a hypothetical family in Ohio with a high-deductible plan. The father needs a daily medication. At the counter, the system charges him the list-based price, and he pays it because he must. Nobody in the chain has done anything illegal. But he paid more than the plan's net cost, and the person who could have told him a cheaper alternative existed could be barred by contract from saying so until 2018. The solution to that story is not a federal board. It is sunlight.

For the shopper's side of the same problem, how to find a price before you commit, see A Shopper's Playbook for Getting a Real Price Before a Medical Procedure. And for a parallel story about an industry that went to court to keep its prices out of view, see Hospitals Went to Court to Keep Their Prices Secret, and They Lost.

The case in brief

1. Pharmacy benefit managers build formularies, negotiate rebates, set pharmacy networks and process claims, and three large companies tied to insurers handle most prescriptions.
2. When a PBM is paid a share of rebates, it can prefer a high-list-price drug that costs the plan more, and a deductible patient pays on the list price while the rebate goes elsewhere. The hypothetical of Drug X and Drug Y shows the arithmetic.
3. In Rutledge v. Pharmaceutical Care Management Association (2020), the Supreme Court held unanimously that ERISA does not preempt Arkansas's regulation of PBM reimbursement to pharmacies, which confirms a state role.
4. A transparent market, with disclosure, flat fees, point-of-sale rebates and state action, fixes the opacity at the root, while a federal price board adds one more opaque decision-maker.

What you can do: ask your employer's benefits office for a plain explanation of how your plan's PBM is paid, and tell your state legislator whether your state has a PBM transparency law, and if not, why not.