Deciding on the Electoral System: Chiles Adoption of Proportional Representation in 1925

In 1925 a new electoral system was introduced in Chile. This reform changed the electoral formula from a cumulative voting system to a proportional one (d ́Hondt) and established new rules about district magnitude and form of voting. It has been argued that this reform was motivated by the emergence of new parties or the expansion of the electorate. This article offers an alternative explanation: in the case of Chile, the main reason for the electoral reform was the parties’ need to solve problems of strategic coordination stemming from the characteristics of the Chilean cumulative voting system. In this context, the Chilean case shows that there are many routes to proportionality. I any democratic country, decisions as to which electoral system is employed to elect authorities are of the greatest importance. Electoral systems are institutions defined by political players with incentives and interests.1 Therefore, they are not necessarily neutral for those who compete in the electoral arena. Though some of their components are mathematical formulas that convert votes into seats, they can be designed to safeguard or increase the power of the most influential parties. Through electoral rules, it is possible to favor certain groups or to prejudice others regarding access to parliamentary seats (IDEA 2005). As a result, electoral reform is a process in which the relative power of each political player is directly engaged. In September 1925 a new electoral system was introduced in Chile in a threestage process, in the midst of a profound institutional reform. The new Constitution of 1925 established a change in the electoral formula, providing that the cumulative voting system (CVS) for the election of members of Congress, in effect since 1874, be replaced by a form of proportional representation (PR). Two decree laws (DL 542 and DL 543), both of September 1925, specified the constitutional provisions, establishing new rules for the conversion of votes into seats (in accordance with the d’Hondt formula), district sizes, and the form of voting. Furthermore, in November 1925, DL 710 amended DL 542, enabling parties to draw up joint lists for competing in elections (electoral pacts) and introducing the “vote for the list.” © 2015 University of Miami DOI: 10.1111/j.1548-2456.2015.00269.x Ricardo Gamboa is an associate professor in the Institute of International Studies, University of Chile. [email protected]. Mauricio Morales is an associate professor in the School of Political Science, Universidad Diego Portales. [email protected] To date, little academic research in political science has been undertaken to examine in depth the origin and the goals of the adoption of this major electoral reform. This article is designed to fill that gap by answering two related questions: what explains the change from a majority system (CVS) to another with proportional representation and the d’Hondt formula? Within this framework, what explains the decisions that were made on three substantive points: district size, form of voting, and electoral pacts? In our attempt to answer these questions, we examine the argument of WillsOtero (2009), who maintains that the Rokkan-Boix hypothesis (Boix 1999) is applicable to the Chilean case. Although this thesis is well founded, we reject WillsOtero’s argument, since its premises are inapplicable in the Chilean case. We offer an alternative explanation, which is based on an empirical analysis of the arguments of those who favored a change from a CVS to a d’Hondt-style PR system from the decade of 1910 onward and the form in which the new system was structured and operated.2 We argue that the new electoral system emerged in the context of a national political crisis and during a period in which various countries had adopted PR systems. Both factors contributed to the reform. In the first place, the crisis generated a strategic political opening that provided the promoters of the reform with the opportunity to design and introduce it without significant opposition. Second, the adoption of the PR system by several European and Latin American countries served as an argument to justify the reform. While these factors played a facilitating role, the most important factor explaining the reform was the elite’s need to resolve problems of strategic coordination generated by the CVS, which differed from those underlying the Rokkan-Boix thesis. The CVS, combined with a low average district magnitude (M), generated a number of problems: how to determine the optimal number of candidates per district (particularly in districts where M was greater than 4); how to ensure that the voters “distributed correctly” their votes in the event that one party presented more than one candidate; and the internal tensions aroused by the parties’ need to exclude potential candidates who were important party figures. The new PR system largely resolved these problems. Because it was based on a list and larger districts (average M was greater), more candidates could be fielded in each district, which reduced the problem of excluding candidates. The votes of one candidate helped the party, and possibly the candidate’s fellow members, by lessening the problems caused by the concentration of votes for a single candidate and the difficulty of having to determine the optimum number of candidates per district; consequently, it also helped reduce tensions among candidates. The evidence is mixed as to whether the need to generate greater proportionality helped drive the reform. Its promoters maintained in 1912 that the change was necessary for that reason. Yet the lack of proportionality was not a substantial problem at that date. In 1925, moreover, the leaders of the reform (Arturo Alessandri and José Maza) did not consider the lack of proportionality to be a major problem to resolve. Effectively, the lower chamber had no proportionality problem at that 42 LATIN AMERICAN POLITICS AND SOCIETY 57: 2