By Marc Guinjoan, Universitat Autònoma de Barcelona
What is the story?
In real-life elections, it is not uncommon for some parties to compete even if they have no chance of winning or gaining parliamentary seats. According to most theories of party competition, the existence of a cost of entry (i.e., the cost of running a campaign or to even to qualify as a candidate) should deter non-viable parties from competing. How can we thus explain the presence of non-viable parties?
In a recent book, I address both the organisational and institutional factors that explain the decision of political parties to compete without chances. In a recent blog post, I dealt with the organisational factors; today, I address the institutional explanations of this phenomenon.
Explaining votes in favour of non-viable parties
Most theories of party competition assume that voters do not vote for a party if they think that this party has no chance of being elected in their district. They do so as to maximize the chances that their single vote will make a difference in the final electoral outcome (we talk about a strategic vote, in contrast to a sincere wasted vote for a non-viable party). However, in reality, we observe that parties that stand no chance of being elected receive votes. This might be an explanation for why these non-viable parties decide to compete.
As I mentioned in my previous blog post, many democracies are composed of various electoral arenas that overlap on a given territory (e.g., local, regional and national elections). One can assume that the vote share received by a party in a given arena is at least partly a function of the vote share this party receives in other arenas of the same territory, or even of the same arena but in other territories. Thus, the institutional architecture of the State and the way the territory is divided into electoral districts are very important when it comes to explaining the persistence of non-viable parties.
To study this issue, I constructed an original dataset that covers 46 democracies, 240 elections, and around 26,000 districts. I calculated, at the national level, an aggregated measure of the number of votes cast in favour of non-viable parties using the difference between the effective number of electoral parties and the effective number of viable parties (the parties that obtain at least a seat + the first runner-up party) that compete in each district.
How the institutions affect votes for non-viable parties
My empirical analyses show that, in countries using a mixed-member system combining proportional representation and single-member plurality in overlapping constituencies such as Germany, an increase in the share of seats a party gets in the proportional tier has a positive effect on the votes received by this party in the plurality tier (even if this party is not viable in this latter tier).
I also demonstrate that the degree of power granted to a sub-national entity such as a region is a strong determinant of the vote share obtained by non-viable parties. For example, regional parties that are successful at the sub-national level still receive votes at the national level even if they are not viable in this electoral arena. In the same vein, I find that the presence of an ethno-linguistic cleavage in a territory increases the number of votes received by parties that have no chance.
Finally, I show that there is an effect of district magnitude, and particularly of the variation in the number of elected candidates by district in a given electoral arena. In countries where there are very large and very small districts, the parties that are non-viable in small districts take advantage of the fact that there are viable in large districts. There is thus a contamination effect.
For more information, see the book “Parties, Elections and Electoral Contests: Competition and Contamination Effects” (Ashgate Publishing, 2014).
