A mailshot in the dark? The impact of the UK government's leaflet on the 2016 EU referendum

In this paper I explore the causal effect of exposure to the UK government’s mailshot on vote preference in the 2016 EU referendum. I find that exposure caused a drop in the probability of voting leave by 3 percentage points. The effect was stronger in individuals who were exposed to few other sources of referendum information. For instance, females and the risk averse were even less likely to vote leave after exposure. The effect was also larger for Conservative party supporters who consumed many other sources of information. The evidence is consistent with voters being liable to persuasion. On the mechanism, I show that exposed individuals experienced a “persuasion-through-knowledge” effect, which changed beliefs on topics of contention.