The purpose of this chapter is to apply the conceptual discussion and qualitative data analysis in preceding chapters to econometric tests of the determinants of Japanese FDI in the East Asian economies. The existing econometrics literature consists almost exclusively of studies of European and U.S. FDI. It is surveyed in UNCTC (1992b) which summarizes results from studies using cross-sectional and time series data. Very few econometric tests exist of Japanese or other Asian outward FDI because of the paucity of comparable data. As will be seen, such problems arise in this study as well.

Cross-sectional studies reported by UNCTC test the significance of factors such as the intensity of a firm's use of R&D, skills and advertising. Time series studies, most of which are focused on FDI in industrialized countries, examine the impact of dynamic variables such as market size in the host country (as a proxy for potential economies of scale), tariffs (to reflect barriers to trade), capital controls (as influence on FDI), financial variables such as long bond yields (as a measure of the opportunity cost of capital); trade balance; measures of institutional arrangements such a free trade areas and common markets (the latter as proxies for trade barriers); and currency changes.

One of the few attempts to study determinants of Japanese outward FDI to East Asia was carried out by Urata (1992, p. 188) on data for the 1977-86 period. He analysed flows of Japanese FDI to Asia across 8 manufacturing sectors. Using Dunning's eclectic framework, he tested the influence on FDI of establishment size, product differentiation, and R&D-intensity as well as measures of export dependence and import penetration. Urata found that neither product differentiation nor technological superiority were important determinants of outward FDI in Asia during the study period. Trade variables both provided significant explanatory power, however: both export dependence and import penetration were positively associated with FDI.