Reexamining Determinants of FDI Using Dynamic Panel Quantile Regression Analysis: Evidence of Korea

Abstract

The primary objectives of this study are to identify the determinants of FDI inflows to host countries with selected variables by using a panel quantile regression model, and to examine whether the different levels of FDI inflows to host countries are differently related to the determinants of FDI inflows. The time-series annual panel data from 1995 to 2012 regarding Korea's FDI inflows was gathered from several secondary databases. The previous empirical results from the conditional mean-based models for FDI inflows to South Korea are not informative: most of the estimated coefficeints are statistically insignificant. It implies the determinant of FDI inflows can be quite heterogeneous; that is, the determinants may differ depending on the level of FDI Thus, the dynamic panel quantile regression model is used to show that the determinants of FDI would differ based on the level of FDI inflows. To be more specific, the GDP, employment, and human resources education levels of the host country are significant determinants of FDI inflows only when the level of the inflow is low. However, corruption and anti-environment investment levels are statically significant determinants for middle- and high-level FDI inflows. This result suggests that the government should have policies to decrease the degree of corruption and increase anti-environment pollution investment in order to realize increases in FDI inflows.