The Achilles Heel of Plurality Systems: Geography and Representation in Multiparty Democracies

Abstract

Building on the unfinished research program of Gudgin and Taylor (1979), we analytically derive the linkage between a party's territorial distribution of support and the basic features of its vote‐seat curve. We then demonstrate the usefulness of the corresponding empirical model with an analysis of elections in postwar Great Britain, focusing in particular on the transformation of the Liberals from a territorially concentrated to a dispersed party in the 1970s. We show that majoritarian biases increase with the number of parties, and majoritarian systems harm small parties when their vote is more dispersed than average, and large parties when their vote is more concentrated than average. Moreover, the evolving experiences of Labour and Conservatives demonstrate how a party's territorial support, and hence its expected seat premium or penalty, changes with its electoral fortunes. This model has a wide variety of applications in multiparty majoritarian democracies around the world.