中国FDI水平溢出效应的Meta分析

本文通过从55篇研究文献中提取关于中国FDI水平溢出效应的相关信息，应用Meta回归分析方法考察中国FDI水平溢出效应的实证结论受何种因素的影响和所受影响的程度。本文Meta回归分析研究发现文献研究中指标的构建（被解释变量指标与FDI溢出效应指标）会影响FDI水平溢出效应的结论，区分FDI水平、垂直溢出效应后会降低得到FDI水平溢出效应是显著的结论的可能性；但文献研究中样本数据特征对FDI水平溢出效应的结论的影响更大，特别是文献研究中如果采用近期的数据会降低得到正向FDI水平溢出效应的可能性，采用的数据跨度越长（实际上是数据中包括的近期数据越多）会降低得到FDI水平溢出效应是显著的结论的可能性；文献研究中是否纳入控制变量对FDI水平溢出效应的正负及是否显著的影响不大，因此研究中国FDI水平溢出效应时模型的设定（主要是遗漏变量）问题并不严重。

A Meta-Regression Analysis on the Effect of FDI Horizontal Spillovers in China

This paper presents a meta-analysis of the empirical literature on horizontal technology spillovers from foreign direct investment(FDI) in China by collecting information from 55 published papers to test those factors and their degrees which affect FDI spillovers effect.Meta-analysis results suggest that the indices construction have significant effect on the conclusion of FDI horizontal spillovers effects in the literatures.The bias toward negative FDI horizontal spillovers is increased without distinguishing FDI horizontal and vertical spillovers in the literatures.The data characteristics also have influence on the conclusion of FDI horizontal spillovers.Meta-regression analysis results also highlight the omitted variables problem,which has incontestable effects on the FDI horizontal spillovers,of the model design induce the unpredictable bias toward FDI horizontal spillovers.