Measuring Duvergerian Effects of the French Majority Runoff System with Laboratory Experiments: Duverger’s Laws Under the Microscope

Abstract

The study of the effects of voting systems started before Maurice Duverger, but the credit goes to him for systematizing the analysis (Riker 1986) and for stating firmly the “laws”, which are still used by contemporary political scientists to describe the relationships between election rules and party system: the plurality system favours bipartism (Duverger 1951, p. 306); the runoff system and proportional representation tend to favour a multiparty system (Duverger 1951, p. 331). He can also be credited for bringing into light the theoretical foundations on which these “laws” are based, by making a clear distinction between the “mechanical effects” of voting systems, i.e., the conversion of votes into seats, and the “psychological effects” of voting systems, i.e., the tendency of voters to anticipate the mechanical effects of electoral rules and to adapt their behaviours to the chances of winning of the various parties running, to maximize the utility of their votes (Duverger 1951, p. 315). In this case, we talk about “strategic voting” (Downs 1957; Cain 1978; Cox 1994, 1997) of “sophisticated voting” (Banks 1985; Shepsle and Weingast 1984; Abramson et al. 1992) or even of “tactical voting” (Johnston and Pattie 1991; Niemi et al. 1992) or, in France, of “vote utile” (Parodi 2002).