| Whether innovative city policies can be stimulated and how to stimulate the urban innovation is a major issue for policy makers and academics.Considering the persistence,high cost and arduousness of building innovative cities,it is very important to accurately evaluate the impact of approved Innovative cities on urban innovation performance and identify the impact channels of innovative urban policies,both in theory and in practice.However,few research has focused on the impact of the establishment of Innovative city on urban innovation performance and empirical testing of impact mechanisms.In particular,there is a lack of solutions to endogenous problems caused by two-way causality in the pilot policy.This paper aims to use the difference-in-difference(DID)identification strategy to empirically analyze the impact of approved Innovative cities on urban innovation performance and the influence mechanism through the quasi-natural experiments of innovative urban pilots started in 2008,using the 2003-2016 city-level database.The results show that the approved Innovative cities can significantly improve the urban innovation performance,and there is a significant "batch effect".In order to ensure the robustness of the basic regression,this paper also verified from the following aspects: First,the parallel trend test ensures that there is no significant time trend before the treatment group and the control group are affected by the approval of Innovative city.Second,in order to alleviate the endogenous problems that may arise from the sample self-selection,this paper uses the PSM-DID method to verify the basic conclusions again.Third,the placebo test ensures that the improvement in urban innovation performance is due to the impact of the policy of approved Innovation cities,rather than some other random factors.Fourth,considering that there are great differences between the four municipalities and other prefecture level cities in terms of system background,economic strength and autonomy,this paper excludes samples from municipalities to ensure the robustness of the basic regression results.Fifth,considering the possible reverse causal problems due to the “pilot policy”,this paper uses the instrumental variables of the ancient capital and controls the innovation performance of the city at the beginning and in the future,and the results are still robust.Further research finds that the impact of regional characteristics on urban innovation performance is heterogeneous: the approval of innovative cities promotes the urban innovation performance in the east,high-administrative cities and large-scale cities.The paper also finds that the industrial structure has a positive adjustment effect on the innovation performance of innovative cities.This study has the following three policy implications: First,approved innovative cities play a positive role in urban innovation performance,and the government should vigorously promote the construction of innovative cities.Second,in view of the innovation performance promotion effect of the approved innovative cities is more significant in the cities with better innovation foundation,priority should be given to setting innovative cities in the economically developed east,cities with higher administrative level,cities with larger scale and cities with higher industrial structure.Finally,cities with innovative initial resource endowments should implement urban innovation strategies according to the place and time.Instead of bidding for innovative cities,blindly developing urban innovation will be counterproductive. |
