Social Networks and Brexit: Evidence from a Trade Shock

Abstract

Regional exposure to Chinese import competition has often been linked to support for the Leave option in the 2016 UK EU membership referendum. Looking at 143 harmonised International Territorial Level 3 (ITL3) regions covering England and Wales, and using data on the density of online social ties between them, I show that regional support for leaving the EU was also associated with exposure in socially connected regions. I first delineate 18 commuting zones based on interregional flows over three Census years. For each region, I then construct a measure of own exposure to Chinese import competition and a measure of exposure in a set of social neighbours located outside its commuting zone. Exploiting variation within commuting zones, and using an instrumental variable approach, I find that the two measures have comparable positive effects on the regional share of the Leave vote. In a series of checks, I do not find evidence that the effect of social neighbours’ exposure is driven by an economic channel or a relationship between import competition and social ties. I also corroborate the regional results using survey data on vote choice. I interpret these findings as indicative of social spillovers between local labour markets: information flows from social neighbours are a likely channel behind the estimated spillover effects on voting outcomes.