and lower skills follows from the observation that investments abroad are more complex, costly and
risky operations than trade and, consequently, may require more skills and the availability of more
initial resources on the part of immigrants. Individuals with higher education levels are more likely
to possess the appropriate knowledge, to form business networks and to possess the required initial
capital.
Finally, we investigate whether the networks tied to the developing countries have stronger
effects on the bilateral FDI than those liked to the rich economies. It is a well known fact that the
biggest share of world’s FDI tends to remain within the group of rich economies and that the
smallest portion going to the poor economies falls short of what would be predicted by standard
theory. Some explanations of these “missing” international investments have focused, respectively,
on differences between the developed and developing countries in the relative resources of human
capital (Lucas, 1990), on disparities in culture and institutions (Alfaro et al. 2006) or on the high
set-up costs of foreign investments (Razin, Rubinstein, Sadka, 2004). These explanations, which are
not mutually exclusive, imply that the informal barriers that obstruct the international investments
into the developing countries are higher than those separating the developed economies from each
other. If this is so, then the impact of transnational networks could also be stronger on the
investments going to the poor countries. In other words, a positive effect of the world’s migration
flows South-North could be that of narrowing the FDI “divide” between the two groups of
countries.
The paper main results are that the aggregate migrants’ networks positively and significantly
influence the bilateral FDI of the four countries examined. More precisely, immigrants affect the
outward and inward FDI of Germany and France, the outward FDI of the U.K., and emigrants have
a positive and robust impact on the inward and outward bilateral FDI of Italy. The impact of
networks on investments abroad, however, depend entirely on the business networks of the skilled,
while the impact of the less skilled is negative. Finally, for two of our countries, Germany and Italy,
the influence on outward FDI of the networks tied to the developing countries is higher than those
linked to the developed economies. The databases of these two countries allow the utilization of a
dynamic version of the model that includes the lagged dependent variable. In both cases, the
stronger impact of the business links with the less developed countries is confirmed.
The paper is structured as follows. The following Section presents a simple model of
migration and international capital mobility that illustrates the different effects of skilled and
unskilled migration on developed and developing economies. In Section II the empirical model is
developed. The empirical analysis is based on four different databases, one for each country, with
time spans that vary from 1990 and 2005. Section IV presents the data and some descriptive