King's College London

The global increase in extreme weather events in recent years has spurred political scientists to examine the potential political effects of such phenomena. This paper explores effect of flooding on electoral outcomes and offers evidence that the impact of adverse events varies with changes in political context. Using a difference-in-differences identification strategy to analyse three consecutive general elections in the United Kingdom (2015, 2017 and 2019), the paper finds variability in partisan electoral benefit from one election to the next that calls into question the blind retrospection and rally-round-the-leader explanations which are often advanced to account for electoral reactions to natural disasters. Instead, changing party positions on environmental issues appear to account more convincingly for shifts in electoral support in response to flooding. This suggests that parties can derive benefit from, or be punished for, the positions they take on environmental issues when extreme weather events affect citizens.