Political Learning and the Number of Parties: Why Age Matters

Party system fractionalization was re-invented as an unsolved puzzle after the fall of the Berlin Wall. While scholars agree that the stability of the party system is imperative for the proper functioning of democracy, many note the high number of political parties in the East European states. Still, we lack a systematic analysis of party system development in those countries. A possible reason for this gap is that extant theories on the number of parties were written with established democracies in mind and are thus unequipped to explain the dynamics taking place in young democracies. This dissertation attempts to fill this gap providing at least preliminary answers for the variation in the number of parties between new and more established democracies. My theory proposes that learning the effect of institutions is crucial to whether they actually have an effect or not and is integral to understanding the number of parties in a given system. Furthermore, I argue that certain institutional arrangements, for example the translation of votes into seats, may play a more important role than the district magnitude when present. I view learning as coming from trial-and-error experience which elite members get by political participation over time, as well as experience with a changing institutional environment both within and external to the party system as such. To test my propositions I use a three-level hierarchical model on district data of 20 European democracies. The results show that at the district level, age of democracy has a positive effect on the level of party system convergence and the effect is stronger in young democracies. The analysis further reveals that pre-electoral institutional constraints such as signatures and deposits have a positive and significant effect on party system convergence, while in the presence of EU-related events the convergence index drops, likely due to the additional incentives for political competition that such events bring. Public funding does not prove significant and the effect of age of democracy on party system convergence in mature democracies remains inconclusive as alternate specifications elicit varying results