Flexible Election Timing and International Conflict

While there are a large number of studies examining the differences in conflict behavior due to varying institutional arrangements, scholars have not effectively addressed the differences in electoral cycles between systems with fixed versus flexible election timing. At the same time that parliamentary regimes give the Prime Minister the power to dissolve parliament, they give the parliament the power to remove the government with a vote of no-confidence. Together, these institutional attributes make the precise timing of the electionsand thus the public's opportunity to hold the government accountablelargely uncertain. I develop a theory that expectations of an upcoming election constrain the foreign policy decision making of executives, and induce pacific behavior. I estimate the probability of an election in a sample of 17 advanced parliamentary democracies from the 1950s to 2001. I find convincing support for my theory, as the probability of hostile dispute initiation is a function of those characteristics that influence the timing of elections, including majority support and the electoral cycle. More specifically, majority and minority governments face varying incentives to initiate disputes because of the different risks of an immediate election.