The argument is what things should be government grant incentives to that wouldn't happen otherwise. By incentive, this could mean patents, or it could mean sub

The argument is what things should be government grant incentives to that wouldn't happen otherwise. By incentive, this could mean patents, or it could mean subsidies, tax-credits, etc. What we have to solve is which side is larger for each option: SocietalReturns : OpportunityCosts + USPTOCosts + EnforcementCosts If you did this for each patent and averaged this ratio, we should come to the right answer of which option(s) to pick. I'd be surprised if something like drastically raising the NIH/FDA funding budgets didn't solve the incentive problem better than patents - especially for drugs that are probably the best case for patents.