Compensatory vs. Parallel: Comparative Strategic Impacts of System Choice

• In Event: Electoral Systems and Voting Rules

Abstract

Parallel and compensatory electoral systems share many similarities in design. By combining electoral formulas, these systems create something of a middle ground between proportional representation and single-member district plurality systems. To date, however, there is little formal understanding of how the differences in design between the two systems produce different strategic behavior on the part of voters and political parties. Generally, previous work seems to suggest that given the lack of a proportional tie-in, parallel systems should provide more incentive for strategic voting than compensatory systems. In this paper we propose to fill this gap in the literature with a unique approach to the study of strategic voting. Using electoral data from Japan, Korea, Germany and New Zealand, we will compare the performance of political parties at the district level in the SMD and PR tiers in each country, evaluating the rate at which parties are able to maintain (or grow) their PR vote total in the SMD tier. Using panel regression models, we are able to not only compare strategic voting behavior at the district level, controlling for specific district level characteristics such as turnout and candidate pool, but also to compare strategic voting behavior across different systems. By focusing on countries with a relatively high level of economic development and a history of democratic governance, we hope to control for some potentially confounding variables. This approach also allows us to evaluate strategic behavior on the part of political parties, particularly in the choice of which districts in which to run candidates.

Authors

• Geoff Allen, University of California Santa Barbara

• David Matthew Jenkins