Wu Xiaokang; Tie Ying
2023, 46(10): 170-195.
This paper creates a novel dataset on patent trading between cities, exploring boundary barriers to knowledge trading and assessing potential efficiency losses. Empirical evidence shows that crossing boundaries significantly suppresses patent trading. On average, patent trading in the same city is approximately 17.654 times higher than between neighbouring cities in the same province, while, on the other hand, patent trading between neighbouring cities in the same province is approximately 2.406 times higher than between neighbouring cities across provinces, indicating that“provincial boundaries” are more difficult to cross than “municipal boundaries”. Mechanism analysis indicates that the institutional transaction costs constitute an important source of boundary barriers. Through new institutional arrangements, such as patent trading centres and intellectual property courts, it is proven that improving institutional quality can effectively reduce boundary barriers. Finally, the paper constructs a theoretical framework to describe the patent trading behaviour, structurally assessing efficiency losses caused by boundaries and ultimately breaking down their sources. The results indicate that approximately 78.8% of cities experience efficiency losses, with a mean and median of 3.5% and 4.2% respectively. The explanatory power of institutional costs on the difference in efficiency losses is more than 5.5 times that of distance. The conclusion of this paper suggests that deepening supply-side structural reform, constructing a large unified market for knowledge and supporting relevant policies are crucial for optimising knowledge allocation and improving innovation efficiency.