Friedman wrote, in Capitalism and Freedom:
there are many “inventions” that are not patentable. The “inventor” of the supermarket, for example, conferred great benefits on his fellowmen for which he could not charge them. Insofar as the same kind of ability is required for the one kind of invention as for the other, the existence of patents tends to divert activity to patentable inventions.
As I noted in The Forgotten Costs of the Patent System, the patent system “skews resources away from theoretical R&D and toward practical gizmos and applications,which surely has some cost as well”. Friedman recognized this, but still, bizarrely, supported patents (though he wanted shorter terms). Rothbard also recognized this, but, unlike Friedman, opposed patents.