中文摘要：

创新投入是提高生产率的重要途径,而不同类型的创新投入对一国生产率改善的作用存在差异性。学理上的解释需深入研究创新活动的内部反应机制。从创新价值链视角将创新活动分为基础研究、应用研究和试验发展。基于研发资本存量的测算,构建反映创新投入结构与全要素生产率之间关系的理论模型,并以经合组织（OECD）23个国家2000-2013年的面板数据为样本,检验创新投入结构对生产率的异质性影响。结果表明：从时间维度来看,试验发展活动对当期全要素生产率增长具有显著的正效应,应用研究和基础研究功效的发挥存在一定时滞,且基础研究的滞后期更长;从影响程度来看,基础研究的偏效应最大,应用研究偏效应短期内为负值,而试验发展的作用虽然非常迅速,但其对生产率的带动效果相对较弱;就全要素生产率分解来看,基础研究和应用研究分别对技术效率和规模效率的偏效应最大,试验发展对全要素生产率三部分都有显著的正效应。因此,改善我国创新投入结构时应充分重视基础研究的投入和积累,以发挥其在促进生产率中的核心地位和长效作用。

英文摘要：

R＆D investment is an important way to improve productivity, different types of R＆D activities have different effects on the productivity of a country. This paper constructs a theoretical model which reflects the relationship between basic research, applied research, experimental development and productivity. Based on the panel data of 23 countries from 1996 to 2013 which belong to OECD, this paper examines the effects of different R＆D investment on productivity from the macro level. The results shows that the development activi-ties have a significant positive effect on the total factor productivity growth in the current period, and there is a lag in the application and basic research functions, and the lag of basic research is longer. From the view of effect, the partial effect of the basic research is the biggest, and the partial effect of application research is negative. Although the experimental development works very quickly, its effect on the productivity is relatively weak. In terms of total factor productivity decomposition, basic research and applied research have the greatest influence on technical efficiency and scale efficiency respectively, while the effect on the other two parts is not obvious, and the experimental development has significant positive effects on the three parts. As a result, too much rely on applied research and experimental development investment is not an effective way to improve the efficiency of a country’s production. We should pay more attention to the input and accumulation of basic re-search to exert its core position and long - term effect in promoting productivity.