We Must Pay More for Drugs
Patents alone are never sufficient
Knowledge has the unfortunate property of not being appropriable. Someone who discovers a new idea is not able to keep it to themselves, and charge to the extent it will benefit others. Instead, they will receive a small fraction of the benefits — the classic citation on the subject is Nordhaus 2004, who claims that only 2.2% of the benefits of innovation passed through to the innovators. The rest is just pure consumer surplus, which is defined as the difference between the price we would be willing to pay, and the price we do in fact pay.
Having more consumer surplus is great for us, of course, but not if it leads to innovation not being worthwhile to invest in. To this end, we grant inventors patents to their ideas. All other people are forbidden to produce a particular good, without the permission of the owner. By giving them a property right to the idea, you get more innovations.
Now, this property right is obviously not perfect. They do not get a claim on ideas which were merely inspired by the original invention, nor can they perfectly enforce the patent. The patent holder can discourage future innovation, especially when the idea is a technological component which can later be built upon (as opposed to a drug, which may inspire research but is unlikely to be literally built upon). Since only one firm can win a patent, the first one to find an idea renders the investment by other companies into finding that close to worthless.
The trouble for patents runs even deeper than this, though. Even if we focus on the use cases, such as pharmaceuticals, where patents are at their best, there is no way for patents alone to lead to the optimal rate of return on inventions. There is no way to assure the inventor all of the social gains. Outside subsidy is necessary. Let’s pull out our monopoly model.
In perfect competition, one seller’s actions lead to no change in prices whatsoever. The demand curve has a slope of 0, and is “perfectly elastic”. Buyers will simply go to another producer. In a monopoly, however, the individual seller faces a downward sloping demand curve. If they change their price, it affects the total amount of goods consumed. If we assume that they sell at the same price to all buyers, then the only way to sell more goods is to lower the price for everyone. The marginal revenue gained from selling an additional unit decreases faster than the demand curve does. As a consequence, the only way to maximize profits is to restrict the quantity sold, and price where that quantity intersects with the demand line. The monopolist produces where marginal cost equals marginal revenue (that is, where the cost of producing one more unit equals the revenue you would get from that unit).
The important thing is that the price of the good is tied to the cost of manufacture, and not to the cost of research. Even at the cost of waste, there is still consumer surplus. There are still projects for which we would be collectively willing to buy, but we cannot elicit through a patent mechanism. What are ways of getting around this?
An unintuitive one is that price discrimination is actually good. The whole reason that marginal revenue has a sharper slope than the demand curve is the requirement that the monopolist sell at only one price. If it can perfectly charge everyone exactly what they are willing to pay, not only will all worthwhile projects be attempted, there won’t even be distortion from the monopoly! The monopolist would produce until the point the marginal cost curve crosses the demand curve — same as a firm under perfect competition would. Even if the firm can’t perfectly price discriminate, it can charge different prices to people in different countries. It is absolutely necessary that people in the United States pay more for drugs for them to be found.
We can also pay a subsidy to whoever discovers an innovation satisfying certain requirements. How much to pay is not obvious. The government must differentiate between gold and dross. I am sympathetic to Michael Kremer’s proposal to require all firms to place their patents up for auction, and will cover it in a future post. The government should also create prizes for definable cures to diseases with particularly large externalities, like malaria or Covid. There is still room for private prizes to have great effect as well. Indeed, I think this is an excellent use of money for quite wealthy people interested in giving their money effectively — many effective charities cannot expand their operations at the same marginal cost forever.
It is an unwelcome prospect to many that we should pay more, not less, for drugs. Ah well. The world does not particularly care about our feelings. If we want to improve people’s lives, there is no other way.