Abstract
Optimizing the allocation of innovation resources and strengthening the importance of innovation are the main engines of modern growth. Research and development (R&D) subsidy policy is an essential means to guide enterprises to increase R&D investment, enhance the efficiency of fiscal subsidy funds, and leverage higher-level corporate R&D investment at the core of policy design. This study uses data on applications for and acceptance of R&D subsidy projects and provides the first comparative analysis of the policy’s effects and fund-use efficiency between the resource-leaning (increased subsidy rate) and inclusive (increased number of funded enterprises) models of R&D subsidies. This study constructs a theoretical model based on the actual process of R&D subsidies, which includes three stages: enterprises’ subsidy application choices, government review decisions, and enterprises’ R&D behavior. It estimates the model parameters based on enterprise-level data regarding R&D subsidy applications and granted subsidy amounts. The empirical results demonstrate that the rules of current R&D subsidies reflect government preferences and selection of subsidy recipients. In this context, one unit of R&D subsidy can engender an increase of 4.51 units in corporate R&D investment and an enhancement of 0.98 units in net social benefits. Simultaneously, resource-leaning subsidies tend to attract leading enterprises, whereas inclusive subsidies may induce adverse selection, with the former exhibiting a higher fiscal expenditure efficiency. The conclusions of this research offer empirical support for the scientific formulation and optimization of R&D subsidy systems under various policy objectives, underscoring the complexity of designing measures that balance effectiveness with efficiency. By shedding light on the differentiated impacts of resource learning and inclusive subsidy designs, this study contributes valuable insights into the nuanced relationship between policy design and intended outcomes in the context of innovation-driven development.
JEL Classification: H29; L59; O31