Population Relatedness and Cross-Country Idea Flows: Evidence from Book Translations

Abstract

This paper uses book translation data as a measure of idea flows to document a robust empirical relationship between these data and two measures of population relatedness: linguistic and genetic distance. I find a significant negative relationship between linguistic distance and book translations that is robust to a variety of controls, fixed effects specifications and sample restrictions. Genetic distance similarly exhibits negative correlation with book translations, but after conditioning on linguistic and geographic distance the sign is reversed. This conditional positive relationship between genetic distance and book translations is statistically significant and robust to numerous robustness checks. The benchmark estimate indicates that a one standard deviation increase in linguistic distance reduces book translations by 12 percent, while a one standard deviation increase in genetic distance increases book translations by 10 percent. I argue linguistic distance reflects a cost on idea flows via the translatability of a language pair, whereas genetic distance captures a concomitant incentive to communicate when dissimilar countries have more to learn from each other.

Keywords: Book Translations, Idea Flows, Linguistic Distance, Genetic Distance

JEL Classification: F10, O47, O50, Z10