Susceptibilities of Democratic Electoral Systems

The two most common families of electoral systems (ESs), defining the rules used to elect assemblies and legislative institutions, are proportional representation (PR) and plurality (Bormann and Golder, 2013; Farrell, 2011). When they are evaluated, most often the arguments come from social choice theory and political science. The former overall uses an axiomatic approach that includes a list of mathematical criteria a system should fulfill (Urken et al., 1995; Sen, 1995). The latter predominantly focuses on the tradeoff between proportionality of apportionment and governability (Monroe, 1994; Carey and Hix, 2011). However, there is an ongoing discussion about which ES is the best (Bowler et al., 2005; Farrell and Gallagher, 1999) and which set of indexes and measures would be the most important in such assessment (Pennisi, 1998). Although previous research addressed various perceptions of fairness related to the proportionality of different ESs (Blau, 2004; Plescia et al., 2020), the sensitivity of ES to efforts that influence opinions has been neglected. Here, we address this research gap with a framework that can measure ESs' susceptibility to different means of influence. Using a simulation study, we show that plurality ESs are less stable than PR. They are more susceptible to coordinated efforts to influence opinions, for example, by political agitators and media propaganda. A review of real-world ES reveals possible improvements in their design, leading to lower susceptibility. Additionally, our simulation framework allows the computation of popular indexes, such as the Gallagher index and the effective number of parties, in different scenarios. Our work provides a new tool for dealing with modern threats to democracy that could destabilize voting processes (Hyde, 2020). Furthermore, our results add an important argument to a longstanding discussion on the evaluation of ES.