Religion and international trade: does the sharing of a religious culture facilitate the formation of trade networks?

American Journal of Economics and Sociology, The, Oct, 2007 by Joshua J. Lewer, Hendrik van den Berg

There is clearly much research to be done. This article addresses this need for further research on the economic role of religion in general and the specific role of networks on international trade by empirically testing the hypothesis that the sharing of a particular religious culture helps to build networks linking traders in different countries.

III

Testing the Relationship Between Religion and Trade Using the Gravity Model

To TEST THE HYPOTHESIS OF positive network effects of religion on trade, we use a gravity model to control for all other determinants of bilateral trade among a cross-section of 84 countries. Gravity models have been widely used in empirical studies of international trade because they normally explain well over 70 percent of the cross-section variation in world trade volumes. By controlling for so many of the influences on international trade, the model provides a convenient framework within which to test the marginal influence of other specific variables, such as religion.

As its name suggests, the gravity model of international trade specifies that trade between a pair of countries is negatively related to the distance separating the countries and positively related to the countries' "weight," namely, their national products. The specification of the model is almost always augmented to include variables that represent geographic, ethno-linguistic, and various economic conditions. A common form of the gravity model is:

[trade.sub.ij] = [a.sub.0] [a.sub.1]([gdp.sub.i][gdp.sub.j]) [a.sub.2]([pop.sub.i][pop.sub.j]) [a.sub.3][dist.sub.ij] [a.sub.4][cont.sub.ij] [a.sub.5][lang.sub.ij] [a.sub.6][bloc.sub.ij] [u.sub.ij], (1)

in which [trade.sub.ij] is the log of bilateral trade between countries i and j, [gdp.sub.i][gdp.sub.j] is the log of the product of i and j's gross domestic products, [pop.sub.i][pop.sub.j] is the log of the product of i and j's populations, [dist.sub.ij] is the log of the geographic distance between the capital cities of i and j, [cont.sub.ij], [lang.sub.ij], and [bloc.sub.ij] are dummy variables set equal to 1 when countries share a contiguous border, a common language, and the same trade bloc, respectively, and [u.sub.ij] is the error term. Recent applications of the gravity model include Frankel, Stein, and Wei (1995) and Freund (2000), who measure the effect of regional trade blocs, McCallum (1995) and Anderson and van Wincoop (2003), who test for the effect of political borders on trade, Eichengreen and Irwin (1998) and Deardorff (1998), who look at the effect of historical trade patterns and international relationships on current trade levels, Frankel and Romer (1999), who use the model to generate an instrumental variable with which to test the effect of trade on economic growth, Freund and Weinhold (2004), who look at the effect of the Internet on international trade, and Rose and van Wincoop (2001) and Frankel and Rose (2002), who test the impact of common currencies on bilateral trade. The gravity model's theoretical foundation is discussed by Linnemann (1966, 1969), Learner and Stern (1970), Anderson (1979), Deardorff (1998), and Anderson and van Wincoop (2003).