中国工业企业自主创新的影响因素和产出绩效研究  

An Evaluation of the Efficiency of Chinese Industry Enterprises' Innovation Performance

作　　者：周亚虹[1] 贺小丹[1] 沈瑶[2] 

机构地区：[1]上海财经大学经济学院 [2]上海大学经济学院

出　　处：《经济研究》2012年第5期107-119,共13页Economic Research Journal

基　　金：国家自然科学基金项目(70871073,71171127,71001061);教育部“新世纪优秀人才支持计划”(NCET-09-0538)资助;上海市重点学科建设项目(B801);上海财经大学“211工程”四期重点学科建设项目的资助

摘　　要：在当前经济转型的大背景下,中国工业企业要想在国际竞争中求得生存和发展,自主创新是关键。本文主要围绕我国企业自主创新绩效进行分析。关于企业自主创新的讨论在现有文献中并不少见,但仍存在诸多不足之处。本文提出了较为全面的研究思路来克服这些不足:在使用广义Cobb-Douglas生产函数作为企业创新绩效研究的基本方程的同时,将R&D的累积投入代理企业知识资本;着重讨论R&D内生性问题,结合我国工业企业的特点,对R&D行为决策进行建模分析找出合适的工具变量,从而避免了参数估计值的不一致性。本文实证部分使用了中国2005—2007年度近3万家产值在500万元以上的工业企业数据。研究表明:企业通过产品与技术革新等创新活动,提高了企业的产出水平,其投入产出弹性达到5.5%,修正了常规方法偏低的估计值。In this Post-crisis era, China＇s economy is now facing a big challenge. With the change of the global economic environment, it is a critical period now for China＇s economic transition. This situation strongly requires Chinese industry enterprise＇s innovation. And only in this way, China＇s enterprises could survive in international competition. In this paper we will discuss the performances of innovation of China＇s enterprises. As we know that this is a hot research topic in recent years, and there are a lot of relevant literatures, both in foreign and domestic journals, on the R＆D performances. However, there are still some main drawbacks in the existing literatures. For example, self-selection in the firm＇s R＆D inputs is seldom discussed, and R＆D＇s cumulative effect on production has not yet been emphasized. These drawbacks motivate us to propose an appropriate way to evaluate the efficiency of the innovation performance. In the theoretical part of this paper, we establish models of firms＇ innovation determination and production function. To avoid possible model misspecification, we adopt Chamberlain＇s random effect approach to estimate the binary probit choice model which determines whether an enterprise engages in R＆D activity, and then use the probit fitted value as the IV of R＆D input in the linear regression of production function. By using a dataset containing more that 30000 firms in different economic sectors in China from year 2005 to 2007, our empirical result shows that the elasticity of R＆D on the added value is 5.5% , which implies that traditional approach underestimated the efficiency of R＆D input for the China＇s industrial enterprises.

关 键 词：自主创新 研发投入 Cobb—Douglas生产函数 工具变量 Chamberlain随机效应模型 

分 类 号：F425[经济管理—产业经济]