政府数据要素关注度对企业双元创新发展的促进作用及路径OA
The Promoting Effect and Pathways of Government Data Element Attention on Enterprises' Ambidextrous Innovation Development
数据要素市场发展需要政府关注和引导,政府对数据要素关注度深刻影响企业创新进步.为进一步揭示数据要素市场发展中政府干预在微观层面的传导效应,为政府政策传导和企业创新战略制定提供依据,文章基于中国31个省(区、市)的政府数据要素政策文件,结合程序化扎根与大语言模型构建政府数据要素关注度词频表,并突破准自然实验对政策事件的依赖性,运用2010-2023年上市公司数据构建"省份-企业"面板数据模型,采用双向固定效应模型实证检验政府数据要素关注度对企业双元创新的影响.研究结果表明:政府数据要素关注度能够显著促进企业探索式创新和利用式创新,具体通过政府补助、企业数字化转型和研发强度促进企业双元创新,且对企业双元创新的影响在新质生产力水平不同的省份和在资源类型不同的城市存在差异.其中,政府数据要素关注度在新质生产力高水平省份对探索式创新显著而对利用式创新不显著,在新质生产力低水平省份对双元创新均显著;在非资源型城市中对双元创新促进效应更显著.基于研究结论,提出以下建议:政府层面应建立数据要素政策评估机制并将评价结果纳入政绩考核体系,同时实施复合补助机制,推动数据要素市场规则从政府指导到生态共治转变;企业层面密切关注数据要素政策走向,依据政策标准进行自评,并且结合地方政策规划积极进行项目申报,提升政策红利转化效率.
As a new production factor,the market-oriented allocation of data elements has become a core engine driving high-quality economic development.As both holders of data resources and suppliers of institutions,governments profoundly influence the strategic choices and resource allocation efficiency of enterprise innovation behavior through policy orientation,resource allocation,and institutional design,via their attention to data elements.To further elucidate the micro-level transmission effects of government intervention in the development of the data factor market,and to provide a basis for government policy implementation and corporate innovation strategy formulation,based on government data element policy documents from 31 regions in China,combining programmatic grounding with large language models,this paper constructs a word frequency table of government data element attention points.Furthermore,to overcome the dependence of quasi-natural experiments on policy events,it employs listed company data from 2010 to 2023 to construct a panel data model,and using a two-way fixed effects model to empirically examine the impact of government data element attention on corporate ambidextrous innovation(exploratory innovation and exploitative innovation)and its mechanisms.The research findings indicate that:first,government data element attention significantly promotes corporate exploratory innovation(0.016 8,P<0.01)and exploitative innovation(0.024 0,P<0.01).Second,mechanism tests reveal that government data element attention stimulates corporate ambidextrous innovation through three pathways:government subsidies,enterprise digital transformation,and R&D intensity.Government subsidies provide financial support through direct funding and signaling effects,while corporate digital transformation optimizes resource acquisition and allocation efficiency through resource orchestration,and corporate R&D intensity influences the direction of corporate competition through the focused allocation of innovation resources.The impact of government data element attention on corporate ambidextrous innovation varies across provinces with different levels of new quality productive forces and cities with different resource types.Specifically,in regions with high level new quality productive forces,government data element attention has a significant effect on exploratory innovation(0.012 2,P<0.01)but an insignificant effect on exploitative innovation,while in regions with low level new quality productive forces,government data element attention significantly affects both exploratory innovation(0.026 8,P<0.05)and exploitative innovation(0.042 2,P<0.05);but in non-resource-based cities,the promoting effect of government data element attention on both exploratory innovation(0.012 9,P<0.05)and exploitative innovation(0.017 7,P<0.1)is more significant.Based on the results,the following suggestions are proposed:At the government level,an evaluation mechanism for the effectiveness of data element policies should be established and incorporated into the government performance assessment system,and a compound subsidy mechanism should be implemented to promote the transformation of data element market rules from government guidance to ecological co-governance;At the enterprise level,close attention should be paid to the trends in data element policies,self-assessment should be conducted based on policy standards,and active project applications should be made in conjunction with local policy planning to enhance the efficiency of converting policy dividends.
孙楠;何琦
上海工程技术大学管理学院,上海 201620上海工程技术大学管理学院,上海 201620
管理科学
数据要素政府关注度探索式创新利用式创新程序化扎根-大语言模型企业创新
data elementgovernment data element attentionexploratory innovationexploitative innovationproceduralized grounded large language modelenterprise innovation
《科技创新发展战略研究》 2026 (2)
33-48,16
评论