The coexistence of incentive and catering effects is a significant and complex phenomenon in the implementation of innovation-oriented industrial policies. As a widely applicable innovation-oriented policy, the R&D expense super-deduction policy has undergone several rounds of revision and improvement since its promulgation in 1996, and the tax incentives have been continuously strengthened. Given that this policy has been implemented for decades, how is the persistence in tax incentive application by enterprises? How do these two effects dynamically change with persistent application?

Based on the data of A-share listed companies in Shanghai and Shenzhen from 2008 to 2022, this paper quantifies the persistence in tax incentive application by enterprises and explores the impact of persistence on policy effects. The findings show that as enterprises persistently apply for R&D tax super-deduction, a “two-track” pattern emerges. Due to the motivation of “seeking innovation”, enterprises increase a large number of R&D investments. Meanwhile, they also boost strategic R&D investments out of the motivation of “seeking support”, leading to an intensification of both the incentive and catering effects. Heterogeneity analysis reveals that growth-stage and small-scale enterprises show a more significant catering effect when persistently applying for R&D tax super-deduction, while high-tech enterprises experience an enhancement of both effects. For enterprises facing severe financing constraints, these two effects are weakened. Further analysis shows that as the degree of persistence increases, the incentive effect gradually outweighs the catering effect, leading to higher-quality innovation outcomes.

This paper has the following contributions: First, from the perspective of tax incentive persistence, it focuses on how the incentive and catering effects dynamically change over time, expanding the research results of evaluating the effectiveness of relevant policies. Second, based on the complexity of enterprises’ motivations for application, it quantifies and distinguishes the incentive and catering effects of the policy, and extracts the “net effect” of the policy by examining the differences between the two, enriching the relevant conclusions on the evaluation of the effectiveness of innovation-oriented policy effects. Third, based on a large sample of data, it scientifically quantifies the persistence in application for R&D tax super-deduction by Chinese enterprises.