University of Iowa

Strategic Behavior After Electoral Reform: Cross-National Experiments

Following the collapse of communism in Eastern Europe, several of these new democracies experimented with different voting rules borrowed from established, industrialized democratic counties. Casting a vote for a candidate other than the communist party choice was a new and meaningful experience. The political leaders designing the constitution - typically former communist or socialist party members - generally chose election rules based on what would maximize the probability of a successful re-election bid for their own party members. For example the leaders of the Polish United Workers Party (PZPR) expected a reform in 1989 allowing multi-party competition to accommodate the demands of the Solidarity movement, and yet, maintain their complete control of all avenues of power. However, Solidarity won in a landslide victory; thus triggering a complete change in the political system and causing the irrevocable decline of the PZPR. Scholars have developed several theoretical arguments explaining the causes and consequences of changing the rules of the political system. Juxtaposed with this literature is research on the strategic behavior of voters under different voting rules. Still, we are only beginning to understand how reforms to pre-existing voting arrangements can translate into important, and sometimes unexpected, outcomes. In order to overcome this deficit, I develop a theory of strategic learning under different electoral rules to explain three interrelated questions: 1) How do voters learn new election rules? 2) Are different rules easier to learn? 3) Does prior democratic experience lead to greater strategic behavior? The general theoretical argument presented in this project is that changing the election rules is a shock to the political system, and thus to the prior knowledge and heuristics voters relied upon to make voting decisions in the past will be of limited help with the new voting rules. After a few elections, voters will have the prior experiences and knowledge to rely upon to change their vote if they deem their candidate hopeless and a wasted vote.
I test the theoretical expectations using a combination of empirical and experimental techniques. I begin by using agent-based modeling developed in computational economics and design a computer-based simulation performed in an artificial election environment. This artificial world is similar to a laboratory in that I interrupt the status quo with new voting rules to examine how the agents (i.e. voters) behave. Based on the patterns from the computer simulation, I design a second experiment using human agents voting under different election rules. After conducting both the simulation and experiment, I use post-election surveys in Germany, France, and Hungary - each with histories of significant election reforms.