Duverger's psychological effect: A natural experiment approach

Abstract

A large body of work suggests that electoral systems influence the strategic calculus of voters and parties. But electoral rules are endogenous as they originate in the decision making process of voters and parties, making empirical analyses vulnerable to reverse causality and contamination effects. This paper studies a unique case where an electoral system change is exogenously imposed by a supra-national body. I exploit an EU ruling which forced the UK to adopt proportional representation (PR) for European Parliament elections. Using a difference-in-differences design, I estimate the causal impact of the reform on electoral outcomes. I find that the shift to PR leads to more “effective parties” and thus to a more proportional electoral outcome. In line with Duverger's law, the reform causally increased the number of newly entering parties. Existent small parties did not appear to exert more effort. The findings survive several robustness checks.