Abstract:

The rapid development of FinTech is transforming traditional financial systems and driving global economic growth through the integration of new technologies. Effective FinTech innovation system is crucial for the high-quality development of China's digital economy. While existing literature focuses on FinTech as an explanatory variable for the development of the real economy and financial industry， it lacks exploration of the micro-level source of FinTech， leaving the question of how to promote FinTech innovation unanswered. It aims to measure the level of FinTech innovation for financial firms at the individual level and analyze the impact of executives with IT backgrounds on FinTech innovation. The patent data used in this study comes from the National Intellectual Property Administration and China Intellectual Property Right net， which includes all patent application information and current status of patents filed in China. The financial data related to the enterprise comes from China Stock Market & Accounting Research Database. Using a deep learning-based Chinese text classification method to identify FinTech patents， and combining the patent data with A-share listed financial enterprises' data from 2013 to 2020， it is found that executives with IT backgrounds can indeed promote FinTech innovation. The robustness of the conclusion is confirmed after considering possible endogeneity. Mechanism tests reveal that this effect is achieved through increased R&D investment and is also influenced by myopic behavior and the direction of the firm's innovation search. Further research demonstrates that the differences in executive positions， FinTech categories， and implementation of FinTech industry policies are important heterogeneity factors affecting the relationship between executives with IT backgrounds and FinTech innovation. The main contributions of this paper are threefold. First， it fills the research gap by exploring how to promote FinTech innovation at the micro level from the perspective of innovation research， treating FinTech as an innovation output. Second， it uncovers the differences and mechanisms of IT background executives' influence on FinTech innovation， providing decision support for financial enterprise managers and policymakers. Third， the method proposed in this study， which utilizes character-level convolutional neural networks for identifying FinTech patents， contributes to the research on FinTech innovation at the micro-level.

Key words: FinTech, executives background, enterprise innovation, deep learning, text mining