land border (border) and a dummy for landlocked countries (locked). In particular, the
coefficients on these two additional variables are largely insignificant, thus suggesting
that the extent of trade of emerging economies with the EU-15 is mostly explained by
economic size, geographical distance and financial openness.
The results for the full group of 44 countries highlight an interesting difference:
financial openness does not appear to increase trade to any significant extent. In fact,
the estimated coefficient on z has a p-value of slightly more than 0.1. Still, it does not
pass a zero restriction test at usual confidence levels. Moreover, its point estimate is
about ¼ of the value estimated for the group of emerging economies. Taken at face
value, this would mean that the role of financial openness as a trade-facilitating factor
is strong in the formerly centrally planned economies, but not in western economies.
The other results in columns VI, VII and VIII partially mitigate this conclusion. The
two measures of international financial integration exhibit significantly positive
estimated coefficients. Even more importantly, when the gravity equation includes the
two dummies for common border and landlocked countries, the coefficient on z
becomes significant, albeit the corresponding marginal effect remains much smaller
than that estimated for the emerging economies. Overall, the evidence does suggest
that the trade promoting-effect of financial openness is stronger and more relevant for
the emerging economies than for the rest of the sample. A deeper analysis of the
reasons why this is the case is certainly an interesting avenue for future research.
Robustness checks analogous to those performed for equation (3) are carried out for
the gravity model (i.e. changes in the definition of q and inclusion of additional
variables on the r.h.s. of the model). Of some specific interest is the inclusion of a
dummy variable to control for the existence of preferential trade agreement between a
country and the EU-15. This dummy turns out to have a large and positive coefficient.
Furthermore, the variable D, distance, has been recomputed using different cities as
the EU-15 reference. Overall, results on financial openness are qualitatively
unchanged.
5. Conclusions and directions of future research.
The main result of the empirical analysis is that financial openness facilitates the
economic integration of emerging economies with the EU-15. This integration-effect
takes the form of faster per-capita income catching-up and greater bilateral trade in
goods and services. Furthermore, the effect of financial openness occurs over and
14