Election Laws, Disproportionality and Median Correspondence: Implications for Two Visions of Democracy

TLDR

Comparative studies of election rules focus on vote–seat disproportionality as a measure of poor representation. The study argues that starting from citizens' preferences rather than votes better serves a majoritarian vision of democracy. The authors analyze how election rules affect vote–seat and median left–right correspondence across 70 elections in 17 countries. Theoretical analysis shows stringent conditions to reduce vote–seat disproportionality in high‑threshold systems, yet empirical data reveal high variance and distortion; moreover, proportional representation systems outperform single‑member district systems in achieving good median correspondence.

Abstract

Comparative studies of election rules and legislative representation have focused intensively on vote–seat disproportionality as an indication of poor representation. Beginning with citizens' preferences, rather than votes, has important advantages and is especially more appropriate for a majoritarian vision of democracy. We analyse the effect of election rules on both vote–seat correspondence and median left–right correspondence in seventy elections in seventeen countries. We show theoretically the stringent conditions necessary to reduce vote–seat disproportionality in high threshold systems and empirically their high variance (and higher levels) of distortion. Although good median correspondence could be created, in theory, under a wide range of electoral systems, our empirical results suggest that proportional representation (PR) systems tend to outperform single-member district (SMD) systems by this criterion also.