FDI, ownership shares and domestic productivity

In this thesis the results are presented of a meta-analysis of the literature on Foreign Direct Investment and productivity spillovers. Special attention is given to the distinction between minority and majority ownership. First, this thesis provides an overview of the relevant theory on the existence and development of these spillovers. It distinguishes between three effects: competition, demonstration, and the labor mobility effect. Then, a dataset is established with ten empirical studies that take ownership shares into account when researching this topic. On this dataset meta-regression techniques are applied to find which study characteristics influence study outcomes. It was found that several study characteristics have a significant effect on the coefficient size. First, using different definitions for the dependent variable in the original study yields different results. Also, the proxy used for foreign presence has a significant influence on the original coefficient. Next, using cross-section data instead of panel data gives smaller results. This is related to the finding that the number of years included in the original study gives different results. No significant influence is found of the statistical method used, if spillovers are explicitly considered in the original regression or if the country is developed/developing. No evidence was found that study characteristics significantly influence the sign and significance of the coefficient. Finally, some evidence was found that results do differ when minority or majority observations are considered.