This paper has examined the short-run and the long-run dynamic interactions
between outward FDI and GDP per capita within bivariate and multivariate
frameworks. The obtained results in the bivarite framework support the suggestion
of Herzer (2008) which emphasizes that increased outward FDI is both a cause and a
consequence of increased domestic output in the long-run for Japan. The results also
indicate that Japanese GDP per capita has short-run effects on outward FDI.
However, after controlling for country specific effect with the introduction of
domestic investment and openness as control variables in a multivariate framework,
the obtained results indicates that outward FDI has positive effects on GDP per
capita only in the long-run. Based on the results in Table 9, these two control
variables are statistically significant. This suggests that the bivariate framework is
misspecified in terms of omitting important independent variables. Siliverstovs and
Herzer (2006) mention that the results of Granger causality test may not be valid if a
model suffers from the omission of important independent variable. Therefore, only
the results of the multivariate framework are discussed in this conclusion