Where's the bias? A reassessment of the Chilean electoral system

This paper presents exploratory data analysis of Chilean electoral results from 1989 to 2001 to show that contrary to the conventional wisdom, the ‘binominal’ electoral system does not benefit the second largest coalition. Rather, it benefits the two largest coalitions at the expense of the smaller lists, as is the case in almost all electoral systems. Previous analyses have erred by extrapolating predictions that are correct at the district level into the aggregate multidistrict setting without accounting for the fact that there is considerable variation on how well lists perform across different districts. Using simulations I show that the variation of vote shares across districts is, in fact, the driving force behind this result.