The present paper conducts a meta-analysis of literature on intra-industry spillovers from foreign direct investment, using 97 different outcomes from 67 empirical studies. Apart from the traditional approach, robust meta-regression, random-effects model, and normal probability regression are employed. Results of combined significance analysis are mixed but it is evident that studies published in leading academic journals tend to report rather insignificant results. Our findings suggest that the outcome of an empirical work is, in general, dependent on its design, although this pattern seems to weaken over time. Contrary to previous studies, evidence for publication bias was not detected.
meta-analysis productivity spillovers technology transfer foreign direct investment multinational companies