The transfers game: A comparative analysis of the mechanical effect of lower preference votes in STV systems
| Author | Hannah Schwarz,Stephen Quinlan |
| DOI | 10.1177/0192512120907925 |
| Published date | 01 January 2022 |
| Date | 01 January 2022 |
| Subject Matter | Articles |
https://doi.org/10.1177/0192512120907925
International Political Science Review
2022, Vol. 43(1) 118 –135
© The Author(s) 2020
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/0192512120907925
journals.sagepub.com/home/ips
The transfers game: A comparative
analysis of the mechanical
effect of lower preference
votes in STV systems
Stephen Quinlan
GESIS – Leibniz Institute for the Social Sciences, Germany
Hannah Schwarz
Universitat Pompeu Fabra, Spain
Abstract
Debates about electoral reform revolve around giving voters more choice. Consequently, reformers often
favor adopting the single transferable vote, a candidate-based system that allows voters to rank order
candidates. Nonetheless, studies about whether lower preferences (transfers) influence STV election
outcomes remain scant. To address this gap, our comparative multivariate approach tests transfers’ impact
on election results in Ireland, Malta, Northern Ireland, and Scotland. We find that, on average, transfers are
pivotal in the election of about one in 10 elected candidates. Hence, their impact is the exception rather
than the norm. We show that when lower preferences are decisive at the candidate level, they benefit
smaller and moderate parties and non-incumbents and, in Malta and Ireland, female candidates. Our results
have implications for understanding the extent to which multiple preferences influence election outcomes
and for debates on electoral reform.
Keywords
Preferential voting, STV, electoral systems, vote transfers, electoral reform
Introduction
Citizen disenchantment with politics is en vogue. Participation in elections cross-nationally is
lower today than in recent decades and citizens express low levels of trust in political institutions.
Concurrently, we see the rise of anti-establishment parties globally and increased voter volatility.
The question of whether institutional change might address these challenges has generated research
Corresponding author:
Stephen Quinlan, GESIS – Leibniz Institute for the Social Sciences, B2,1 Mannheim, Baden Wuerttemberg 68159,
Germany.
Email: stephen.quinlan@gesis.org
907925IPS0010.1177/0192512120907925International Political Science ReviewQuinlan and Schwarz
research-article2020
Article
Quinlan and Schwarz 119
on electoral reform (e.g. Bowler and Donovan, 2013; Renwick and Pilet, 2016). Among the central
planks of reform is choice: giving citizens more options in choosing elected representatives. This
comes in many forms, including increasing citizen involvement in selecting certain positions and
enabling voters to choose parties and candidates (e.g. Renwick and Pilet, 2016). Another option is
offering voters preferential voting, which permits electors to rank-order candidates. Our focus is on
this. Our interest is to understand whether allowing voters to express several preferences impacts
election outcomes decisively. We break new ground by focusing on the classic preferential voting
system: the single transferable vote (STV) or choice voting as it is referred to in the United States.
Our analysis builds on a wide-ranging literature exploring the impact of electoral systems on voter
behavior and election outcomes (e.g. Carey and Shugart, 1995; Gallagher, 1991).
STV merits focus as scholars have evaluated it as the second ‘best’ electoral system, just behind
mixed-member proportional systems (Bowler et al. 2005). Moreover, many proponents of electoral
reform herald it as the preeminent system to adopt (e.g. British Columbia Citizens’ Assembly on
Electoral Reform, 2004). Not surprisingly, therefore, STV has been an option offered to voters in
several plebiscites on electoral reform—for example, the 2009 referendums in British Columbia,
the 2011 vote in New Zealand, and the 2016 vote in Prince Edward Island, Canada. Additionally,
the majoritarian version of STV—the alternative vote (AV)—was offered as an option to British
voters in 2011. STV’s prominence in the electoral reform arena has coincided with a growth in
candidate-centered electoral systems in Europe (Renwick and Pilet, 2016). STV could legitimately
be described as a classic candidate-centered system as voters have no option but to vote for candi-
dates. Therefore, understanding whether a central plank of this system, namely being able to cast
lower preferences for candidates influences the results, can offer new insights into the debate con-
cerning electoral reform.
Additionally, many have pointed to the vast choice STV multi-preferences offers to voters
(e.g. British Columbia Citizens’ Assembly on Electoral Reform, 2004: 5; Farrell et al., 1996),
its promise to provide broader representation of the electorate, and for reducing wasted votes
compared with non-preferential systems. STV voters favor its retention (Farrell et al., 2016), and
research also shows that preferential voting systems (including STV) foster a greater sense of
fairness among citizens, which leads to higher voter satisfaction with democracy (Farrell and
McAllister, 2006). Putting all of this together, STV is of interest to electoral scholars and is perti-
nent to the debate on electoral reform, thus warranting further investigation.
In STV systems, the voter votes in a multi-seat district. A list of candidates is presented to the
voter, which they rank in descending order of preference. Voters have one vote which is transferred
from one candidate to another, depending on the rank order. A candidate is required to reach a
quota to guarantee election.1 The counting of votes begins with the tabulation of voters’ first
choices. A candidate who reaches or exceeds the threshold is elected. If a candidate exceeds the
quota, their surplus votes are transferred following the voters’ next preference. If no candidate
reaches the quota, the elimination of candidates with the fewest votes occurs, and these votes are
distributed to the voters’ subsequent preference. This process continues until all seats are filled
(for a review of STV, see Farrell and McAllister, 2000 & Farrell and Sinnott, 2018).
Thus, there are two types of preference votes2 in STV: the first preference, which we assume
reflects the voters’ sincerest wish about the outcome, and lower preferences, which reflect their
subsequent choices (often termed transfers).3 Political actors make great efforts to earn lower
preferences (e.g. Ahern, 2010: 257; Burke et al., 2016). Conventional wisdom, especially in the
media, holds that these votes have a decisive role in deciding elections (Coghlan, 2002; Collins,
2015). However, we lack a systematic test of this vital component of STV. While literature exists
examining their impact on party and campaign behavior (e.g. Katz, 1985) and how voters use them
(e.g. Marsh, 1985; Marsh et al., 2008), remarkably few studies have explored the mechanical
To continue reading
Request your trial