Perm do both - it solves better

There are significant “public choice costs” under either of the optional prize plans. By this I mean there are risks of inappropriate transfers to patentees2 – that is, looting – and of confiscation of patentees, through the conduct of or through the omissions of government agents, sometimes working at the behest of private agents. I will argue that the confiscation risk appears to dominate.

The public choice costs point to a broader set of constraints that should be incorporated into any analysis of the patent system. There are rational actors operating at all levels, from patentees to licensees, and government actors. A discretionary prize system, where the government awards prizes and no enforceable patents, would invite lobbying and third-party interference in the determination of the award. Even in a prize process effectively shielded from third-party interference, government would confiscate patents on a regular basis.

The innovation regime I propose is a patent-plus-prize scheme. The patentee would receive the patent and a prize that approximates consumer surplus. Public choice costs are considerably lower than under optional or mandatory prize schemes: there would be no looting and no risk of confiscation under patent-plus-prize. In addition, private and social incentives to innovate are aligned. Private and social incentives align under the Kremer proposal too, but only if one ignores the risks of looting and of confiscation. Under the patent-plus-prize scheme, by contrast, private and social incentives align without requiring people to behave as angels.