Political Behaviour/Sociology

Session: F10 - Electoral System Characterics and Voter Behaviour

Date: May 31, 2017 | Time: 03:45pm to 05:15pm | Location: POD-484 (Podium Building)| iOS / Outlook

Chair/Président: Dieter Stiers (Katholieke Universiteit Leuven)

Discussant/Commentateur: Patrick Fournier (Université de Montréal)

Participants & Authors/Auteurs:

Filip Kostelka (Université de Montréal), Alexander Wuttke (University of Mannheim) : Election Frequency and Voter Turnout in Federal Democracies: Canada and Germany
Abstract: From a normative standpoint of contemporary participatory democratic theory frequent opportunities to cast ballots at elections or referenda could be seen as desirable. However, empirically a number of arguments suggests that a higher frequency of participatory options could decrease the overall turnout of political contests. The present article tackles the effect of election frequency on voter turnout. The theoretical section describes the various mechanisms through which frequent elections reduce voting rates. The empirical analyses focus on Canada and Germany and are conducted at the regional level (i.e. the level of provinces and federal states). They draw on an exhaustive dataset of all electoral contests and referenda held in the two countries since 1960 and an original measure of electoral frequency. The results confirm that a high frequency of elections is detrimental to voter turnout and estimate how this factor has contributed to the observed trends in voter turnout in the two countries.

---

Ignacio Lago (Universitat Pompeu Fabra) : Distal or Proximal? The Psychological Effect of Electoral Systems
Abstract: In this research note I challenge the conventional wisdom about when the psychological effect of electoral systems is observed. I rely on data from founding presidential and legislative elections in 47 third-wave presidential and semi-presidential regimes to show that the psychological effect of electoral system manifests itself in founding elections. The effective number of legislative parties is significantly higher than the effective number of presidential candidates in founding election when the electoral system in legislative elections is more permissive than in presidential elections.

---

Marc-André Bodet (Université Laval), Melanee Thomas (University of Calgary) : How Much Electoral Politics is Local? Making the Most of a Natural Experiment
Abstract: Even though the fundamental unit of Canada's representative democracy is the local electoral district, the most oft-cited explanations for Canadian electoral behaviour are national in focus. We seek to bring the local back in through innovative research methodologies that mobilize the literature on natural experiments and Geographic Information system (GIS). In our case, the natural experiments occur in the (non-partisan) districting process, especially near delimitations between districts. Voters being assigned to a given district is, then, the closest we can get to a random assignment within the general population. GIS allows us to test whether there is a stronger covariance within electoral districts than between them by pooling together all polling divisions bordering other districts in a given election. In this paper, we focus on the 2012 and 2014 Quebec elections. We want to measure how much of the variation in turnout and party competition is explained at the local level relative to the national level. We expect that local campaigns have non-trivial effects on electoral outcomes, and more specifically the largest effect on party competition, then turnout. Also, we expect that incumbency and party support stability condition the importance of local campaigns relative to national campaigns. This paper also provide a methodological contribution by presenting an original empirical method that shed new light on the importance of local politics.

---

Jean-François Daoust (Université de Montréal), Damien Bol (London King's College) : The Spatial Dimension of Political Representation: Evidence from Canada
Abstract: Good political representation supposes that deputies have frequent contacts with their constituents. However, in large countries the frequency of these contacts is negatively affected by long distances. Transports inevitably take money and time. In this paper, we explore this topic in analyzing the case of Canada. We combine CES data from 1988 to 2015 and various geographical sources to test two hypotheses: (1) the size of the district (and quality of transport links) decreases the frequency of contacts between deputies and constituents, and (2) the distance from the capital (in kilometers and in transport time) decreases it. To further explore this last hypothesis, we exploit a specificity of the Canadian federal legislation: deputies elected in a district located at 100km or more from the federal parliament receive a specific budget for staying over in Ottawa some days a week during the parliamentary year. This concerns 95% of all deputies, but the remaining 5% are supposed to come back to their constituency every night. We can thus exploit the discontinuity created by this 100km threshold to analyze the causal effect of the deputy having the possibility to stay over in Ottawa on the frequency of contacts with constituents. We expect our analysis to bring an important contribution to the literature on political representation. Our anticipated conclusion is that geographical features tend to create natural inequalities between citizens, as those living close to the federal capital and/or in a small district are better represented by deputies than others.

---