Recent literature finds that informal networks are quantitatively important in explaining the cross-country patterns of trade. We check the robustness of this result by looking at the Jewish community. Analyzing the effect of the Jewish Diaspora on trade is particularly interesting: (i) we build and uncover the Jewish population data from publications of the American Jewish Committee from 1899 to 2005, what allows a rich analysis both in panel and cross-sections; (ii) the creation of the State of Israel is an interesting exogeneous variation that has not been empirically exploited yet. We apply a theory-based gravity model, address the issue of heteroskedastic errors in log-linearized models and use the pseudo-maximum likelihood estimation as our main econometric strategy. Our results confirm the trade creation effect of social networks found in cross-sections by the previous literature, even though the effect we find is much smaller in comparison. Moreover, we show robust trade creation effects in a panel: in our benchmark specification for the period 1951-2000, the jewish networks lead to a trade creation of 0.85%, compared to 23.6% of trade creation due to free trade agreements. We divide networks between direct and indirect networks with Israel, and show that most of the network effect is captured by the direct links. Finally, we use the tariff equivalent to reafirm the trade-creating effects of the Jewish networks and find no evidence of a decrease of the network effect over time.
Dieser Eintrag ist Teil der Universitätsbibliographie.
Search Authors in