A social networks framework to study the influence of language on international trade in services.

Abstract

Social network theory has been overlooked as a tool to examine relationships in economics whereas all economic interactions create networks, some more complex than others. Specifically, attempts to estimate drivers of bilateral trade fall in this category. A significant number of studies explaining trade focus on the gravity model and estimate varieties of transformations. The literature widely measures the effect of language in gravity models using a dummy variable indicating the presence of common language between trading countries, and recently developed measures of linguistic distance. However, there has not been an attempt to estimate the effect of language as a factor that connects countries and allows one country to influence other countries in the network. This thesis constructs a network of countries whose interconnections are the similarities of the languages of the countries. I first estimate the linear effects of eigenvector centrality on per capita income, and then on exports. The results reveal that there is a positive relationship between linguistic influence and exports.