The social value of vaccines is enormous. I cannot even find an estimate for the value of all vaccines since the beginning of time. Drawing from estimates for one disease at a time, I would conservatively estimate it at $100 trillion. The social value just of a covid vaccine equally effective against all strains is around $1.5–2.6 trillion greater than prior strain specific vaccines.
Yet, vaccines are less profitable than drugs. This doesn’t, on the face of it, make sense. Two solutions to a problem should have the expected value. Arbitrage will happen until profits are the same everywhere. If a drug company could make more money with either, then they would shift their research funding into the more profitable one. It shouldn’t matter that drugs are administered more frequently. Think about it — does a super effective pill you have to take once cost half as much as a less effective pill you have to take twice? Quite the opposite, as each pill of the effective drug will cost twice as much. This reminds me of a result from Peter Swan on durable goods and planned obsolescence; if people are buying a flow of output from a product, like light from a lightbulb, the optimal length of each product is totally independent of extracting maximum consumer surplus. It is determined only by technological constraints.
So what gives? Michael Kremer and Christopher Snyder have an interesting paper discussing two primary reasons why vaccines are less profitable than drugs, to which I will add some additional reasons of my own. The first is that drug companies are unable to observe your propensity to catch a disease before you do. Since vaccines are before, and drugs are afterward, you catching the disease reveals information about your type. To borrow an example from their paper, suppose 100 risk neutral consumers are all willing to pay $100,000 to cure the disease, conditional upon them having the disease. Ten people have a 100% chance of catching it, while the remaining 90 have but a 10% chance. The vaccine monopolist can charge $100,000 to ten people, or $10,000 to a hundred, with a profit either way of one million dollars. If they’re selling a drug, they can just charge $100,000 to all 19 people who eventually contract the disease, which is much larger.
The second reason is that vaccines tend to interfere with the spreading of the disease more than drugs do, so the demand for the intervention is reduced. This has applications for what sort of drugs are developed, not just vaccines. It’s a common strain of cynicism to say that pharmaceutical companies would rather develop an antidepressant you take every day for the rest of your life, than antibiotics you take for two weeks and then hardly ever again. As we’ve already seen, the focus on number of doses is incorrect — the ratio of expected utility to price should be the same — but if antibiotics inhibit the spread of diseases, they will be less profitable than a drug for a noncommunicable disease. Unmentioned by Kremer and Snyder, but I think relevant, is that very high prices will encourage adverse government action to cap prices. Have we not seen how high prices are demonized by politicians? I am willing to credit the idea that people will be especially moved by one big headline price, than by a series of small prices over time.
Kremer and Snyder are pretty cynical about what can be done. Obviously, we should subsidize those medicines with externalities, but if the price of medicine to the government is determined by the ability of drug companies to sell elsewhere, then the same problem will be replicated in government prices. We saw this during the pandemic. Somehow, for actions that saved millions of lives and prevented hundreds of billions in economic losses, the drug companies got only tens of billions, and were reviled for it anyway. Nevertheless, we must try. There will be future pandemics; if we are not prepared to swiftly act to produce vaccines, we could go through Covid again, or worse. Companies have to be incentivized to prepare by government action. To not spend a few billion dollars preparing to mitigate the next catastrophe would be inconceivably foolish.