Can even small and seemingly innocuous changes to the electoral process affect voter turnout in democracies? New research, to be presented at the annual conference of the Royal Economic Society, examines this question.
Jean-Victor Alipour and Valentin Lindlacher examine whether reassigning voters to a different polling location affects voting behaviour and electoral turnout in Germany. The results show that reassignments permanently shift turnout from in-person to mail-in voting, and temporarily depress overall participation.
In their research paper, “No Surprises, Please: Voting Costs and Electoral Turnout”, the authors study elections in Munich. Election administrators aim to “facilitate voting as much as possible”, in particular, by avoiding overcrowding at voting locations and ensuring that they are accessible on election day. Concretely, these efforts include adjusting precinct boundaries and recruiting new polling venues when, for instance, a school undergoes construction work.
A by-product of these efforts is that some voters are assigned to a different voting location than in previous elections. This setting provides a natural experiment to study whether reassignments impose a burden on voters, and how individuals adjust their voting behaviour in response to a reassignment “shock”.
The statistical analysis is based on geo-referenced data on residential addresses and voting locations for the eight elections held between 2013 and 2020. Using this information, the authors determine (i) whether the assigned voting location changed relative to the previous election and (ii) the walking distance between eligible voters’ homes and their designated voting location. The final dataset contains the average distance to voting locations and the share of reassigned addresses at the precinct level, combined with official data on turnout (in-person, by mail, and overall) and socio-demographic characteristics (e.g., age structure, rent levels).
The statistical approach compares changes in turnout in precincts that experienced a voting location reassignment (the treatment group) with precincts that did not (the control group), while controlling for other determinants of turnout.
Reassuringly, the treatment and the control groups do not significantly differ in their observable characteristics, and do not show different trends in turnout preceding a reassignment. This indicates that the two groups are indeed comparable.
The results of the first statistical exercise show that voting location reassignments cause a partial substitution away from in-person to mail-in votes. Turnout at the new location declines by 0.75 of a percentage point and is only partly compensated by an 0.3 percentage point increase in mail-in voting. Thus, overall turnout falls by about half a percentage point on average.
This result is in line with the view that reassignments impose a burden on some voters, who then switch to mail-in voting or abstain from casting a vote. When the new voting location is relocated further away, these effects are even stronger: doubling the distance to the voting location reduces overall turnout by one percentage point.
A second analysis examines whether these effects persist in subsequent elections. The findings show that while the shift from in-person to mail-in voting is indeed persistent, the drop in overall turnout is fully recovered in the subsequent election.
This pattern is consistent with inattentiveness to voting location reassignments. In Munich, voters are not explicitly notified of changes to their voting location. Thus, inattentive voters may not notice the change until after the deadline for requesting mail-in ballots has passed. Some inattentive voters, who would have liked to vote by mail, thus abstain from voting in the current election and only switch to mail-in voting in the subsequent elections.
The results suggest that sending notification letters to individuals who are affected by reassignments may avoid temporary declines in voter turnout, by reminding inattentive voters to vote by mail in time. Otherwise, even seemingly modest changes to the electoral process can have material effects on turnout.
Authors and Contact:
Jean-Victor Alipour and Ludwig-Maximilian (University of Munich)
Email: alipour@ifo.de
Twitter: https://twitter.com/AlipourJv
LinkedIn: https://www.linkedin.com/in/jean-victor-alipour
Notes to Editors:
The press release is highlighting research papers presented at the RES Annual Conference 2022 (#RES20220) for further information, please contact j.randalledwards@res.org.uk on 07970 201456 if you want the link to the full paper.
The Royal Economic Society’s popular 2022 virtual annual conference with in-person conference hubs, provides a forum for research, debate, and networking. The RES provides resources for economists and support for education and the training of students, teachers and researchers.
The Royal Economic Society’s purpose is to promote the study of economic science. With over 6,500 members worldwide, RES is one of the oldest economic associations in the world. RES are a registered charity and membership is open to anyone who shares our aims and objectives.
The Society’s two peer-reviewed journals – The Economic Journal and The Econometrics Journal – contain high quality papers from an international authorship.
Please do tag RES on Twitter @RoyalEconSoc using #RES2022
