Recategorising Electoral Systems Analytically and Judging Some Undemocratic (Or Learning from an Australian Mistake.

Abstract

Political science has distinguished between plurality, majority and proportional representation electoral systems. The argument is that this categorization is insufficiently analytic: it mixes outcome considerations, about how votes are translated into seats, with issues of the internal mechanics of electoral systems. A more analytic approach is developed in which proportional representation systems are divided into two groups following internal mechanics; a highest average (quotient) group and a quota and largest remainder group. The quotient group is argued to be an adaptation of the plurality electoral formula to multimember districts, while the quota group is a similar adaptation of the majority formula. The paper thus constructs taxonomic trees of just two analytic families of electoral systems in common use in both single and multi-member districts: the plurality and quota families. The paper also identifies numbers of votes given to electors as a missing analytic variable. Single vote systems are defended as reasonable adaptations to multimember districts in both analytic families. But adaptations which give electors multiple votes – such as the block vote, the limited vote and preferential block (majority) – are condemned as mistaken and undemocratic. An Australian example of absurd outcomes from preferential block (majority) is used to substantiate this judgment.