政府经费支持对企业R&D投入的效应分析——基于惠州企业动态面板数据模型的系统GMM估计

Effect Analysis of Government Subsidy on Enterprises' R&D Investment——System GMM Estimation Based on Dynamic Panel Date Model of Enterprises in Huizhou

文章旨在考察政府的R&D资助强度,是否会对受资助企业的R&D投入强度带来挤入或挤出效应,力图寻找出企业受资助的最优均值,以避免政府资助反向抑制企业R&D投入。基于2011-2014年惠州市工业企业数据为样本,通过系统GMM估计进行实证研究。研究发现,政府的经费支持在企业研发投入的初始阶段具有激励作用,而当政府的支持力度超过最优补贴值时,继续资助的行为反倒会挤出企业的R&D投入。在进一步研究中还发现,当考虑政府行为的滞后性等因素后,政府资助对企业的研发投入同样呈现出"先挤入、后挤出"的影响;企业新的研发投资决策会受过往投入水平的影响,形成一种"投资行为习惯"效应。

This paper aims to examine whether government subsidy on an enterprise＇ s R＆D exerts a crowding-in or crowding-out effect on the subsidized company, trying to determine the optimal mean value of R＆D subsidy, in order to prevent government subsidy from reversely impeding a company＇s R＆D investment. Based on the industrial enterprise in Huizhou during 2011-2014 as sample, and through the system GMM estimation for empirical research, it is found that government subsidy motivates a company＇ s R＆D input in the initial stage, however, when the amount of subsidy exceeds the optimal mean value, the continuing support would crowd out the company＇s R＆D investment. Further study discovers that when taking into consideration such facfors as time lag of government behavior, subsidy from government also has a ＂crowding-in first, crowding-out then＂ effect on the subsidized company. Furthermore a company＇ s decision on new investment in R＆D would be affected by the previous investment level, which is an effect called ＂habitual investment behavior＂.