从创新行为到生产率效应的“黑箱”解构——来自投入产出视角的全过程检验  

"Black Box"Deconstruction from Innovation Behavior to Productivity Effect

作　　者：翟钰 赵博[2] 杨琛胤 孙晓华[2] ZHAI Yu;ZHAO Bo;YANG Chenyin;SUN Xiaohua(School of Economics and Management,Harbin Institute of Technology at Weihai,Weihai 264209,China;School of Economics and Management,Dalian University of Technology,Dalian 116024,China)

机构地区：[1]哈尔滨工业大学(威海)经济管理学院,山东威海264209 [2]大连理工大学经济管理学院,辽宁大连116024

出　　处：《科学学与科学技术管理》2024年第7期68-87,共20页Science of Science and Management of S.& T.

基　　金：国家社会科学基金重点项目(22AZD129);国家自然科学基金面上项目(72074043)。

摘　　要：“创新是第一动力”,技术创新是提高企业全要素生产率的根本途径。与现有研究将创新过程看作一个黑箱不同,基于投入产出视角分析了企业技术创新的动态过程,从理论上创造性地构建了由研发选择到创新进而影响企业生产率演化的两阶段模型。实证研究中,在Dixit-Stiglitz框架下提出收益生产率的估计方法,以2008—2016年中国制造业上市公司微观数据为样本,从产品创新专利和工艺创新专利两方面深度挖掘创新产出的差异,按照低技术、中低技术、中高技术和高技术的行业分类,检验了企业技术创新的动态过程及其对生产率的异质性影响。结果发现:(1)当企业从事研发活动时,实现以发明专利为形式的创新产出的概率显著增强,但研发投资并不是创新产出的充要条件。比较而言,低技术和中低技术行业完成工艺创新专利的可能性更大,中高技术和高技术行业则会致力于产品创新。(2)创新产出对于企业的生产率具有显著的提升作用,但存在一定的行业异质性。在低技术和中低技术行业中,生产率的提升效应只来源于工艺创新专利,产品创新专利没有起到应有的效果。在中高技术和高技术行业中,产品创新专利的作用开始显现,对高技术企业生产率的贡献尤为明显。(3)产品创新专利与工艺创新专利之间不存在交互影响,反映出中国制造企业新产品开发和工艺流程设计改造尚未紧密结合的事实,导致不同类型创新产出对于企业生产率无法产生理想的联动效应。Technological innovation,as a fundamental way to improve total factor productivity,is the fundamental driving force for the economy to improve quality and increase efficiency.In recent years,China's innovation-driven development strategy has made remarkable achievements in the world,with the scale of R&D investment and the number of patents ranking second and first in the world respectively.However,at the same time,Chinese enterprises are not willing to carry out R&D activities,the patents are"many but not excellent"and the total factor productivity is slow to improve.How to stimulate the motivation of enterprises'R&D innovation and improve their innovation efficiency is the key to achieving high-quality economic development.To rationalize enterprise R&D investment decisions,optimize enterprise innovation selection,and provide a theoretical basis for the high-quality development of Chinese enterprises,this paper attempts to deconstruct the"black box"of the innovation process from the perspective of the whole input-output process by combining enterprise R&D selection,product and process innovation,and productivity.In the theoretical study,a two-stage model from R&D selection to innovation and then to productivity evolution is constructed;in the empirical study,the estimation method of earnings productivity is proposed in the Dixit-Stiglitz framework,and the dynamic process of enterprise technological innovation and its heterogeneous impact on productivity are further examined with the data of Chinese listed manufacturing enterprises from 2008 to 2016.The study finds that,first,the probability of achieving innovation output in the form of invention patents will be significantly higher when firms engage in R&D activities,but R&D investment is not a sufficient condition for innovation output.In comparison,low-and medium-low-tech industries have a higher probability of achieving process innovation,while medium-high and high-tech industries will be committed to product innovation.Second,innovation output has a positive

关 键 词：研发投资 工艺创新 产品创新 生产率 

分 类 号：F273.1[经济管理—企业管理] F425[经济管理—国民经济] F832.5