Subnational determinants of foreign direct investment: The case of Vietnam

This dissertation studies subnational determinants of FDI inflows with the case of Vietnam's provinces. We analyze the relative significance of 6 major groups of determinants: agglomeration, market size, labor and human capital, infrastructure, policies and institutions, and geography. Two models are estimated using a panel data set covering 57 provinces over nine years. Models of FDI value and project count are estimated by feasible generalized least square (FGLS) and fixed-effect Poisson estimators respectively. Both levels and changes of the explanatory variables are included, as are both national and provincial market variables. Overall results of our models show reasonable fitness and robustness to different specifications. Some variables are found to have consistent effects on FDI flows in count and value. National but not provincial market size affects FDI flows strongly and positively, in contrast to findings of previous studies. FDI flows are strongly discouraged by high labor costs. FDI flows are also significantly and positively determined by agglomeration effects from local economies. Some determinants affect FDI value and project count models differently. FDI agglomeration is significant in value model but not in project count model. Human capital variables are significant in both models, however, their signs are different. Market institution variables strongly affect FDI project count but not FDI value. Both level and change of infrastructure variables are significant determinants of FDI project count but only the change is a significant determinant of FDI value. These differences can be explained by different responses of FDI projects with different sizes. Smaller FDI projects are more sensitive to market institutions, infrastructure and agglomeration effects than large ones. Our findings of significant agglomeration effects imply that the extreme patterns will continue into the future. Whether over-concentration, infrastructure bottleneck, and costly negative externalities will eventually work to disperse FDI flows to other provinces remains a question for future work. We find supports for policies in affecting regional FDI inflows by maintaining stable and favorable macro economic environments, promoting private sector and market institution, and providing public goods such as education, labor training and other social policies.