Adhering to innovation-driven development and cultivating new kinetic energy is an important strategy for China to promote stable economic development in the post-epidemic period. However, it is undeniable that China′s innovation has always faced the problems of difficult and expensive financing. The essential reason is the imbalance of China′s financial resource allocation. With the rapid development of China′s digital technology, the in-depth integration of the financial industry and digital technology has formed a new type of financial model—digital finance, triggering thinking: Can digital finance become a "new engine" for innovation and positively stimulate regional innovation? In order to verify this problem, based on the panel statistical data of 30 provinces in China from 2011 to 2018, this paper adopts methods such as individual fixed effects model, spatial measurement model, panel threshold model and structural equation model path analysis to empirically analyzes the incentive effect of China′s digital financial development on regional innovation. The main conclusions are as follows:
Firstly, the development of digital finance can positively stimulate regional innovation, but the incentive effect of the digital financial decomposition index is different. Among them, the depth of use of digital finance has the strongest incentive effect on regional innovation, followed by the breadth of coverage and the degree of digitalization of digital finance.
Secondly, in terms of heterogeneous characteristics, the innovation incentive effect of digital financial development has significant innovation subject heterogeneity and regional heterogeneity. From the results of the heterogeneity of innovation subjects, the development of digital finance has the strongest innovation incentive effect on scientific research institutions, followed by universities and industrial enterprises. From the perspective of regional heterogeneity, the innovation incentive effect of digital financial development in the eastern and western regions is significant, while the central region is not. At the same time, the western region has the strongest incentive effect, followed by the eastern and central regions.
Thirdly, in terms of spatial characteristics, the innovation incentives for digital finance development have significant direct effects and spatial spillover effects, indicating that the development of digital finance in a certain region can not only directly stimulate innovation in its own region, but also indirectly stimulate the innovation activities of other regions through spatial spillover effects. However, the contribution rate of direct effects to the total effect is higher than the spatial spillover effect, indicating that China′s digital financial development has insufficient spatial spillover power to stimulate regional innovation and there is room for further improvement.
Fourthly, in terms of non-linear characteristics, under the constraints of a single threshold of labor quality, government support and industrial upgrading, the development of digital finance has an incremental innovation incentive effect, while under the constraint of a single threshold of opening up, the development of digital finance has a diminishing innovation incentive effect.
Fifthly, in terms of transmission paths, the development of digital finance can indirectly stimulate regional innovation through internal factors and external factors. The transmission validity of internal factors is 0.128 and the transmission validity of external factors is 0.061, indicating that internal factors are the most important transmission path in the indirect incentive process. Although the transmission validity of external factors is low, the development of digital finance has a more significant impact on external factors. At the same time, R&D intensity plays a major role in internal factors, while economic development is dominant in external factors.
In addition, in order to ensure the consistency of the core estimation results, this paper carried out endogeneity discussion and stability test. In terms of endogeneity, this paper constructs dual instrumental variables of Internet penetration rate and mobile phone penetration rate to conduct IV-2SLS estimation test. The results show that the instrumental variables are effective and there is no problem of over-identification, the development of digital finance can still positively stimulate regional innovation, and the incentive validity has increased. In terms of stability testing, this paper uses three methods: dynamic panel model, replacing explained variables, and changing research objects to carry out stability testing. The results show that although the significance and validity of the innovation incentive effects of digital finance have changed, the direction of action is always consistent. It can be seen that based on the discussion of endogeneity and stability test, the core conclusion of this article is reliable.
Finally, based on the above research conclusions, this paper proposes policy recommendations from three perspectives: accelerating the digital transformation of the financial industry, building an interconnected digital financial ecosystem and improving the digital financial regulatory system.
