Communication Effort and the Cost of Language: Evidence from Stack Overflow

Abstract

The transmission of information is crucial for productivity and growth, but language differences may limit its effectiveness. In this paper, I empirically investigate how the exogenous cost of language affects communication effort, and the trade-off faced by knowledge platforms in implementing their website in multiple languages. I exploit the introduction of websites for languages different from English on a question-and-answering platform and compare the behavior of non-English speaking users before and after the introduction. Results show that the quality of communication improves by more than 24% when writers use their first language, rather than English, and answers are 7% more likely to solve the questioner's problem, a 20% increase from the baseline. In addition, the size of the effect increases when the sender is more incentivized and when the questioner's effort is higher. With the introduction of other languages, the community size increases, but the quality of the contribution of the new joiners is lower. Finally, information is more dispersed. These results show that the platform should adopt multiple languages to maximise the quality and quantity of the information collected unless the size of the communities using those languages does not justify the cost.