Evidence from abroad suggests that mixed legislative systems have much to commend them, but close attention must be paid to national contexts

Different countries use different electoral systems, with one key variant being the number of elected representatives in each district or constituency. Timothy S. Rich argues that context is important in assessing the desirability of mixed legislative systems, which on paper have much to commend them to electoral reformers.

Scotland, which uses both kinds of constituency system (Credit: emily horne, CC BY NC 2.0)

Most legislative systems in democracies use one of two general electoral designs: majoritarian systems, usually where one legislator is elected to a geographically bound constituency in a single member district (SMD), and proportional representation (PR) systems where legislators are elected via a national or regional party list. SMDs are commonly associated with the US, UK, and former British colonies, with PR commonly associated with the rest of Europe. Political science research going back to Duverger’s Law (1954) expects two-party competition in SMDs, with more than two viable parties the norm under PR. While outliers occur, especially in newer democracies and ethnically diverse populations, most countries using SMDs see two large parties in district competition. Furthermore, as there can only be one winner in SMDs, strategic voting away from candidates beyond the top two is expected. In contrast, proportional representation generally discourages strategic voting, leading to a multitude of parties both diverse in ideology and size.

Mixed legislative systems that include both SMDs and PR seats within the same legislative chamber challenge the expectations of two-party competition in SMDs. Whereas West Germany was the only stable democracy to use this system prior to 1990, its use has spread to East Asia, Eastern Europe, Latin America, and Africa and regional parliaments in Scotland and Wales. Early proponents viewed these systems as the best of both systems, providing local level representation in SMDs but diverse representation through PR. Similarly, proponents expected two-party competition in SMDs and multi-party competition in PR should emerge. Others questioned whether the mere presence of PR seats encouraged small parties to run in SMDs despite having marginal chance of success, thus creating a situation in which SMDs function differently than their SMD counterparts in pure majoritarian systems.

In a recent paper of mine, I tackle the question of which factors influence electoral competition in SMDs in mixed systems. Earlier work found many countries averaging more than two viable district candidates in SMDs, conflicting with Duverger, while other countries like Taiwan seem to be dominated still by two parties. Through a quantitative analysis of over 15,000 district elections in 90 national elections in 23 countries from 1990-2012, my findings suggest several factors influence the number of parties. Within mixed systems, the sub-type of mixed system matters as does the number of ballots voters receive and the percentage of the vote necessary for PR seats.

In other words, mixed systems that require the overall distribution of seats to be determined by the PR vote (e.g. Germany, New Zealand) rather than treating both seat types as separate (e.g. Taiwan, Japan) have more parties. Secondly, those that only allow voters to choose a district candidate without a separate vote for the party list (known as a fused ballot, as seen in Mexico and previously in South Korea) have more parties. This is in part due to supporters of smaller parties voting for non-viable candidates in SMDs in the hopes that their votes across districts aggregate enough to win PR seats. Third, as the electoral threshold for PR seats (the minimum percentage of the vote to receive any seats) increases, the number of district candidates declines. Finally, compulsory voting laws, even where poorly enforced, encourage additional parties.

Several other patterns emerge. Presidential systems tend to have fewer district competitors while federal systems have more. Legislatures with more seats also have a very minor positive correlation with the number of competitors. Finally, post-communist countries tended to still exhibit more competitors in district elections than other countries. All of these patterns endured even after controlling for the number of elections held under mixed system rules, assuming voters and parties learn over time which candidates have a realistic chance of winning.

The results suggest that while electoral reformers may find a mixed legislative system to be attractive, failure to consider the many moving parts within these electoral systems makes predicting their outcomes far more difficult than in either purely majoritarian or PR systems. Furthermore, relying solely on evidence from early adopters (e.g. Germany, New Zealand, and Japan) may lead reformers to overlook broader patterns. Furthermore, while some countries adopted a mixed legislative system as a means to promote turnout, the empirical record is less clear. Finally, it remains unclear to what extent the public understands how mixed legislative systems work. Like with any electoral reform, public and party understanding of the rules remains largely a function of trial-and-error experience.

—

Note: This article gives the views of the authors, and not the position of Democratic Audit UK, nor of the London School of Economics. Please read our comments policy before commenting.

—

Timothy S. Rich is an assistant professor of political science at Western Kentucky University. His research focuses on electoral politics, with a focus on mixed legislative systems in East Asia (Japan, South Korea, and Taiwan).