Who wins preference votes? An analysis of party loyalty, ideology, and accountability to voters
| Author | Johanna Rickne,Olle Folke |
| Published date | 01 January 2020 |
| DOI | 10.1177/0951629819893028 |
| Date | 01 January 2020 |
| Subject Matter | Articles |
Article
Journal of Theoretical Politics
2020, Vol.32(1) 11–35
ÓThe Author(s) 2020
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/0951629819893028
journals.sagepub.com/home/jtp
Who wins preference votes?
An analysis of party loyalty,
ideology, and accountability
to voters
Olle Folke
Department of Government,Uppsala University, Sweden
Johanna Rickne
SOFI, Stockholm University, Sweden
Abstract
We study the distribution of preference votes across politicians with different behaviors and atti-
tudes. There are two main findings. First, preference votes are concentrated to politicians who
are more active in communicating their policy proposals and policy priorities. This suggests that
preference voting may incentivize more transparency and communication among politicians, and,
hence, be positive for accountability. Second, preference votes are concentrated to politicians
who are more—not less—loyal to the party in their voting decisions, and to politicians who’s
ideological and policy positions are mainstream—rather than extreme—compared to their party
colleagues. Together with the first finding, this suggests that preference voting can strengthen the
bond of accountability between voters and politicians without undercutting parties’ ideological
cohesion.
Keywords
Flexible-list proportional representation; local politics; party cohesion; political representation;
preference vote; Sweden
Corresponding author:
Olle Folke, Departmentof Government, Uppsala University, GamlaTorget 6, Uppsala, 75220,Sweden.
Email: olle.folke@statsvet.uu.se
1. Introduction
In the long term, the design of the ballotstructure does have the capacity to shape legislative
behavior in important ways. Reformers aiming to strengthen the local responsiveness and
accountability of legislators should consider the use of candidate-ballots. Alternatively,
reformers seeking to strengthen party discipline and cohesion should consider the adoption
of party-ballots.
Norris (2004: 245)
The electoral system dictates the rules of the democratic game. As argued by Pippa
Norris, the choice between party or candidate ballots is a particularly important
parameter of that system. With party ballots (closed-list proportional representa-
tion (PR)), voters choose between parties, and a politician who wants to be
re-elected needs to cater to the party selectorate that decides the ballot order. With
candidate ballots (open-list PR), voters choose between candidates, and politicians
who seek re-election need to behave in ways that attract the most personal votes.
A large body of research has documented behavioral differences between politi-
cians in closed- and open-list systems. Politicians in open-list systems do more con-
stituency service and introduce more particularized legislation, but work in a less
unified way to implement the party’s program (reviewed by Andre
´et al., 2017; but
c.f. Cheibub and Nalepa, 2020). These behaviors were at the core of a Swedish debate
on an electoral reform to introduce flexible lists, a debate lasting for nearly 30 years
until the reform was finally implemented in 1998. Critics of the status quo of closed-
list PR argued that politicians were too inwardly focused on the party, ignoring
accountability and responsiveness to the voters. Opponents of the reform argued that
preference voting would undercut party cohesion by loosening parties’ control over
candidate nomination. When preference voting was finally introduced, it was consid-
ered that the democratic value of ‘strengthening the bond between the voters and the
elected representatives’ outweighed fears of weakened abilities for policy aggregation
at the party level (Davidsson, 2006; SOU, 1993).
In this paper, we try to shed some light on this debate by analyzing the distribu-
tion of preference votes between politicians with different behaviors and attitudes.
We study how votes correlate with (1) the voter–politician bond, and (2) ideologi-
cal cohesion and attitudes towards party loyalty in voting. More specifically, we
want to investigate if preference votes are distributed in a way to suggest that vot-
ers reward politicians who provide a stronger bond, but who deviate ideologically
from their parties.
Our data come from a survey of Swedish local politicians in 2012, covering the
universe of 13,000 municipal councilors and reaching a response rate of 79% (N=
9725). Restricting the sample to local parties with eight or more elected politicians
gives us data for 504 local political parties and over 5000 elected politicians. Data
on preference votes come from the Swedish Electoral Agency and are not self-
reported. The number of preference votes for each politician in the 2010 election is
matched to him or her via the mandatory personal ID code and, hence, has perfect
coverage and no measurement error.
12 Journal of Theoretical Politics 32(1)
To continue reading
Request your trial