主管:中国科学院
主办:中国优选法统筹法与经济数学研究会
   中国科学院科技战略咨询研究院

中国管理科学 ›› 2024, Vol. 32 ›› Issue (2): 221-230.doi: 10.16381/j.cnki.issn1003-207x.2022.2728

• • 上一篇    

政府多工具组合优惠对企业创新行为的影响研究

赵凯1(),李磊2   

  1. 1.青岛大学经济学院,山东 青岛 266075
    2.华侨大学统计学院,福建 厦门 361021
  • 收稿日期:2022-12-20 修回日期:2023-01-11 出版日期:2024-02-25 发布日期:2024-03-06
  • 通讯作者: 赵凯 E-mail:kzhao_kai@126.com
  • 基金资助:
    国家社会科学基金项目(20FJYB017)

Study on the Influence of Government Multi-tool Combination on Enterprise Innovation Behavior

Kai Zhao1(),Lei Li2   

  1. 1.School of Economics, Qingdao University, Qingdao 266075, China
    2.School of Statistics, Huaqiao University, Xiamen 361021, China
  • Received:2022-12-20 Revised:2023-01-11 Online:2024-02-25 Published:2024-03-06
  • Contact: Kai Zhao E-mail:kzhao_kai@126.com

摘要:

为提高政府创新政策运作的协调性与实施的高效性,挖掘出能够有效激励企业创新的最优政策工具“组合拳”策略,本文基于2009—2019年我国沪深A股上市企业数据,着重考察和比较直接补贴、税收优惠和政府采购这三种创新政策工具及其组合对企业创新行为的影响,揭示政府多工具组合优惠策略对异质性企业创新的效果差异。结果显示,由直接补贴、税收优惠和政府采购构成的多工具组合优惠策略,无论是在激励企业创新投入还是提升企业创新质量方面,皆为政府最优的实施策略,并且该策略对于激励国有企业创新的效果尤为明显。

关键词: 多工具组合优惠, 多水平处理效应模型, 政府采购, 税收优惠, 实质性创新

Abstract:

To further improve the coordination of government innovation policy operation and the efficiency of implementation, and dig out the optimal innovation “combination boxing” strategy that can effectively stimulate enterprise innovation behavior, it is urgent to deeply explore and evaluate the actual effects of different innovation policy tools and their combinations. When evaluating the effectiveness of innovation policies, especially the multi-tool combination, both self-selection of enterprises for policies and government agencies’ picking-the-winner strategies often lead to endogenous issues. The multi-level treatment effect model is used which can effectively identify the multi-tool combination, and the actual effect of different innovation policy tools and their policy-mix on the innovation behavior of enterprises in China is evaluated, under the premise of effectively solving the endogenous and selection bias issues. To satisfy the random distribution conditions of causal inference, the multi-level treatment effect model should meet the Conditional Independence Assumption (CIA) and the Overlap Assumption (OA), and the functional expression of conditional expectation value of enterprise innovation can be written as Εyj|x=Εyi|Ti=j,x=β0j+xβ1j. Thereby, the Average Treatment Effect (ATE) when the treatment level changes from Ti to k{0,1,,J}can be estimated. Based on the data of listed enterprises in China from 2009 to 2019, it is found that the multi-tool combination consisting of direct subsidy, tax incentive and government procurement tools may become the government’s optimal implementation strategy of innovation policy, both in stimulating the R&D input and the innovation quality of enterprises. From the perspective of stimulating substantive innovation, government multi-tool combination is more suitable for state-owned enterprises. From the perspective of stimulating strategic innovation, the optimal strategy does not exist. It not only helps to evaluate the effect of innovation policy combination scientifically and reasonably, but also provides empirical basis and practical guidance for improving the design of innovation policy system in this study.

Key words: multi-tool combination, multi-level treatment effect model, government procurement, tax incentive, substantive innovation

中图分类号: