Unintended consequences of negative campaigning: Backlash and second-preference boost effects in a multi-party context
| Author | Cees van der Eijk,Annemarie S Walter |
| Published date | 01 August 2019 |
| Date | 01 August 2019 |
| DOI | 10.1177/1369148119842038 |
| Subject Matter | Original Articles |
https://doi.org/10.1177/1369148119842038
The British Journal of Politics and
International Relations
2019, Vol. 21(3) 612 –629
© The Author(s) 2019
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/1369148119842038
journals.sagepub.com/home/bpi
Unintended consequences of
negative campaigning: Backlash
and second-preference boost
effects in a multi-party context
Annemarie S Walter and Cees van der Eijk
Abstract
This study examines effects of negative campaigning by political parties on citizens’ electoral
preferences in the 2015 General Election in England. We do so by using a large Internet panel study
and an operationalisation of (perceived) negative campaigning that avoids social desirability. Our
study acknowledges England’s multiparty system by distinguishing between the campaign tones of
all parties. Potential problems of endogeneity are addressed by leveraging the panel structure of
the data and by extensive controls. We find that electoral preferences are weakened for parties
engaging in negative campaigning and that this backlash effect gets stronger over the course of the
campaign. We also find support for a second-preferences boost hypothesis: preferences for one’s
second-most preferred party are strengthened if its campaign is more positive than that of one’s
most-preferred party.
Keywords
2015 general election, backlash effect, electoral preference, multiparty system, negative
campaigning, second-order boost effect
Introduction
Negative campaigning is a widely applied campaign practice. It consists of attacking
opponents’ abilities, accomplishments and policies instead of focussing on one’s own
offering in these respects (Geer, 2006). However, contrary to popular belief there is
little empirical evidence that negative campaigning is particularly effective (Lau and
Rovner, 2009). Its use is not without controversy as it has been associated with unin-
tended consequences such as a decrease in political trust and voter turnout which
might endanger the health of representative democracy (Ansolabehere et al., 1994;
Ansolabehere and Iyengar 1995; Kamber, 1997; Krupnikov, 2011).
School of Politics and International Relations, University of Nottingham, Nottingham, UK
Corresponding author:
Annemarie Walter, School of Politics and International Relations, University of Nottingham, Nottingham
NG7 2RD, UK.
Email: Annemarie.Walter@nottingham.ac.uk
842038BPI0010.1177/1369148119842038The British Journal of Politics and International RelationsWalter and van der Eijk
research-article2019
Original Article
Walter and van der Eijk 613
Although used worldwide, research about its effects is overwhelmingly based on the
US context (exceptions are Ceron and d’ Adda 2015; Haddock and Zanna, 1997; Maier
and Maier, 2007; Pattie et al., 2011; Roy and Alcantara, 2016; Sanders and Norris, 2005).
The resulting insights are therefore potentially country-specific, and we have little firm
knowledge about its effects in other political systems (Fridkin and Kenney, 2012).
In this study we assess effects of negative campaigning on electoral preferences in
England. With it we contribute to the scholarly debate on negative campaigning and its
consequences by studying its effects on voters’ electoral preferences in a context with
more than just two (viable) electoral options. We theorise that negative campaigning can
be expected to be less effective in the presence of multiple parties as its benefits may
accrue to other parties than the attacking party. Our study focuses on the 2015 General
Election in England, in which voters generally had a choice of 5 parties. Our main inter-
ests are whether and how negative campaigning affects voters’ electoral preferences, and
our hypotheses involve so-called backlash and third-party effects which are discussed
below. This study contributes to the study of negative campaigning in three ways. First, it
gauges the effects of negative campaigning on preferences for multiple parties in ways
that go beyond the few previous studies that do so (Pattie et al., 2011). Second, the panel
structure of our data and the large sample size allow for an analysis design (including
extensive and refined controls) to ameliorate ubiquitous concerns about confounding fac-
tors and endogeneity better than other studies in the extant literature. Third, this study
looks at the heterogeneity of effects of negative campaigning over the course of the cam-
paign, which has not yet been considered in the extant literature.
We find that negative campaigning does affect English voters’ electoral preferences. It
depresses preferences for the attacking party, which is sometimes referred to as the ‘back-
lash’ effect, the existence of which has been suggested in the few other studies of negative
campaigning in a multiparty system (Maier and Maier, 2007; Pattie et al., 2011; Roy and
Alcantara 2016). This backlash effect becomes stronger over the course of the campaign.
Moreover, we report the novel finding that (in a multiparty system) negative campaigning
may be beneficial for voters’ second-most preferred party (a ‘second-preference boost’
effect).
We first discuss the literature on negative campaigning and its effects on voters’ elec-
toral preferences, resulting in our hypotheses. Next, we present our case (the 2015 General
Election in England), data, operationalisation and analytical design. We then present and
discuss the results, and end by drawing conclusions and implications for the study of
negative campaigning.
Negative campaigning and electoral preferences
In their election campaigns political parties decide on the balance between emphasising
their own abilities, accomplishments and policy stands and attacking their opponent(s) on
these grounds. The first is known as positive campaigning, the latter as negative cam-
paigning (Geer, 2006). The literature examining this choice generally uses a rational
choice perspective: political actors weigh the costs and benefits when deciding whether
to attack or not (Lau and Pomper, 2004; Riker, 1996). The goal to be realised is to win the
elections, and negative campaigning can help to realise this if it increases the attacking
party’s vote share or decreases the vote share of its main rival(s). Thus, a candidate or
party resorts to negative campaigning to become voters’ most preferred party by dimin-
ishing positive feelings for competitors (Budesheim et al., 1996; Lau et al., 2007). Yet
To continue reading
Request your trial