AN EVOLUTIONARY APPROACH TO PARTY SYSTEM STABILITY

Western democracies exhibit a remarkable continuity and stability in their party systems. Parties that dominate elections and governments now tend to be the same parties that domi-nated elections and governments as far back as the 1920s. I illustrate this using a new dataset that draws information from every democratic legislative election between 1930-2000 in 18 West European countries. It is an empirical puzzle as to why established party systems are so stable and why successful new parties are so rare. Existing models of party entry cannot explain why entering parties seem to be strongly disadvantaged with respect to established parties. This paper provides an answer to this puzzle by portraying elections as a coordination game amongst voters. Using a standard model from evolutionary game theory, I am able to show how voters might become coordinated on a set of established parties over time and why it is difficult for new parties to enter successfully once a coordination equilibrium is achieved. The model produces several hypotheses that are open to statistical and experimental testing.