Parties Are No Civic Charities: Campaigns, Demobilization, and the Changing Composition of the Electorate

Abstract

In contrast to non-partisan GOTV campaigns, political parties do not aim to increase turnout across the board. Instead, their principal goal is to affect the outcome of an election in their favour. This paper uses a randomized field experiment to test the effects of Conservative Party door-to-door canvassing and leafleting on the turnout of registered voters who support different parties in an English Parliamentary constituency during the 2014 European Elections. Using commonly-used light-touch campaign interventions, randomly assigned leaflets and door-knocks changed the composition of the electorate in favour of the Conservative Party, by decreasing turnout overall. Increasing turnout among Conservative and unattached voters went hand-in-hand with decreasing turnout among supporters of rival parties such as Labour. Moreover, in contrast to the non-partisan GOTV literature, we show that partisan door-knocks did not affect the composition of the electorate over and above what we would expect from impersonal campaign leaflets alone. We conclude that the campaign tactic of contacting supporters, opponents and unattached voters alike, was very effective as a campaign tactic, but had negative effects on political participation.

Keywords: parties, experiments, mobilization