Some Skepticism on Patents

The gains from innovation are not fully captured by the inventor. Much of our innovation policy is around correcting this. We have several paths to creating innovations. We could hire people to work full time on finding stuff, as we do with professors and researchers, or pay them for specific projects as we do with grants. We could give people a prize for an innovation, as happens occasionally, and we could give people a monopoly over the product for a period of time — that is, a monopoly. 

We overwhelmingly use patents, rather than prizes, because it means that the money the innovator gets is proportional to the contribution. We cannot anticipate all the things which could be invented, nor is it possible to know in advance all the ways in which it might make the world better. What would the correct award for the cellphone be? Or how about for all the small inventions which came along the way? Any system of prizes would be unlikely to produce truly paradigm-shifting research, instead of iterative improvement upon established paths.

 Patents are imperfect. They give the innovator profits only through causing distortions. Being unable to perfectly price discriminate, the owner of the patent must reduce the quantity sold. Patents are hard to enforce, with the courts needing to adjudicate when an idea is just dissimilar enough to be legal. If the courts err too far in favor of the patent holders, then future innovation is prevented by regulatory tape. Perhaps most damningly, many industries don’t use patents at all. Instead, they move first, move fast, and keep it a secret. Heavy patent usage is mostly among old industries, and half of it is about tying up your opponents rather than making genuinely new things.

Patents are of overwhelming importance in the pharmaceutical industry, but it is here where my doubts arise. I have argued that patents are preferable to prizes when the goals cannot be well-defined, and the impacts cannot be easily anticipated or measured. The goals of medicine can be well-defined, and the impacts anticipated and measured. So why have patents, with their attendant distortions, instead of large prizes for drugs which reduce the mortality of such-and-such disease? To some degree, we do this. I see advance market commitments — where the government promises to buy at prices set by the market — as akin to prizes, and so think they’re good. Philanthropically minded billionaires could set up prizes of enormous quantities of money if, for example, a malaria vaccine with certain properties is made. 

I am now much more skeptical of patents being optimal. Where our goals can be precisely defined, we should favor prizes. The primary use of patents seems to be preventing reneging — I imagine they are more politically palatable than giving several billion as a prize. I’m not even convinced this is true, though. A very large part of compensation for pharmaceuticals is in essence a prize — Medicare and Medicaid will pay for the drug, so long as it is approved. If we can do that, we can make a prize for an HIV vaccine.