We study the effects of two types of patent promotion policies (i.e. reward and subsidy) by constructing a tractable general equilibrium model of cumulative innovation, where new ideas strictly improve upon frontier technologies. The size of productivity improvements is drawn from a Pareto distribution. While both types of promotion policies increase R&D and patent quantity, reward policies lead to an increase in low quality patents, reducing the overall patent quality. On the other hand, subsidy policies do not reduce quality of patents. We find that there exist a welfare maximizing policy that improves the decentralized equilibrium to match the social planner's benchmark. Using patent level data from 1986 to 2005 from the State Intellectual Patent Office (SIPO) and China's pro...[ Read more ]

We study the effects of two types of patent promotion policies (i.e. reward and subsidy) by constructing a tractable general equilibrium model of cumulative innovation, where new ideas strictly improve upon frontier technologies. The size of productivity improvements is drawn from a Pareto distribution. While both types of promotion policies increase R&D and patent quantity, reward policies lead to an increase in low quality patents, reducing the overall patent quality. On the other hand, subsidy policies do not reduce quality of patents. We find that there exist a welfare maximizing policy that improves the decentralized equilibrium to match the social planner's benchmark. Using patent level data from 1986 to 2005 from the State Intellectual Patent Office (SIPO) and China's provincial patent promotion policies, we preformed empirical studies that verified the main implications of the theoretical framework. These conclusions are robust even as we adjust the model in several directions.