Constituency size and turnout in mixed electoral systems

Scholars of democratic representation argue that polity size affects political representation in a multitude of ways. Studies of elections consistently show a negative correlation between population size and political participation, particularly in first-past-the-post (FPTP) elections. Less research has investigated the effects of size scaling in mixed electoral systems. I posit that tier linkage is critical in determining whether or not mixed systems using FPTP are subject to the “size effect”. To test this hypothesis, I study recent legislative election returns from 10 national assemblies. I find that unlinked systems using “parallel voting” tiers are vulnerable to the size effect on turnout. In five out of the six assemblies that use parallel voting, there is a negative association between turnout and constituency population size. By contrast, in the remaining four assemblies that use tier linkage, there is not a negative correlation between turnout and size with the exception of South Korea, which recently adopted reforms. The results underscore the potential of seat and/or vote linkage to improve representational outcomes in mixed systems.