Jeffrey A. EDWARDS, Cephas B. NAANWAAB, Alfredo A. ROMERO
Applied Econometrics and International Development Vol. 17-1 (2017)
Abstract: The research regarding the effect that foreign direct investment (FDI) has on economic growth has morphed into a quagmire that the economics discipline can’t seem to find a way out of, and in some ways, is sinking deeper into. In fact, the positive, negative, and dependent impact views many times blatantly contradict one another, leaving someone on the outside to wonder whether there may be ‘something else’ that is causing the differences in the empirical estimations of this effect. In this study, we employ a rolling window methodology across 60 countries that has been tailored for use with a perfectly balanced panel data set spanning 30 years, to investigate whether some of this variation in the effect FDI has on growth could be caused by shocks to the effect itself. What we find is that conservatively, 25% of our estimates are indeed dynamically sensitive, with this sensitivity continuing across economic sectors as well. In the end, at least some of this heretofore variation in the effect’s estimates may simply be due to investigators ignoring significant levels of parameter heterogeneity, and nothing to do with the inherent attributes of the economies being studied.