All the prime minister’s glory? Leader effects and accountability of prime ministers in parliamentary elections
| Author | Jan Berz |
| Published date | 01 November 2020 |
| Date | 01 November 2020 |
| DOI | 10.1177/0263395720907054 |
| Subject Matter | Articles |
research-article2020
Article
Politics
2020, Vol. 40(4) 444 –459
All the prime minister’s
© The Author(s) 2020
Article reuse guidelines:
sagepub.com/journals-permissions
glory? Leader effects and
https://doi.org/10.1177/0263395720907054
DOI: 10.1177/0263395720907054
journals.sagepub.com/home/pol
accountability of prime
ministers in parliamentary
elections
Jan Berz
Leuphana University of Lüneburg, Germany
Abstract
Are prime ministers held accountable for their government’s performance? The personalisation of
parliamentary elections and subsequent voting behaviour based on the personality of party leaders
questions the accountability of elected governments. In this article, I analyse the confounding
of prime ministers’ leader effects by voters’ evaluation of government performance to examine
whether prime ministers are held accountable for the performance of their government. I use
individual-level data from British, Danish, and German elections and a natural experiment at the
German state level to show that voters hold prime ministers personally accountable. The findings
constitute an important extension of electoral accountability and have implications for the study
of personalisation and presidentialisation in parliamentary democracies.
Keywords
leader effect, party leader, presidentialisation, personalisation, prime minister, voting behaviour
Received: 12th June 2019; Revised version received: 18th November 2019; Accepted: 22nd January 2020
Introduction
Do voters hold prime ministers (PMs) accountable for the performance of their govern-
ment? The personalisation of parliamentary elections and electoral impact of party
leaders1 has received growing attention in electoral research (Aarts et al., 2011; Berz,
2019; Bittner, 2011, 2018; Ferreira da Silva, 2018; Ferreira da Silva and Costa, 2019;
Garzia, 2014, 2017a, 2017b; Lobo and Curtice, 2014; Poguntke and Webb, 2005;
Wattenberg, 1991). However, these findings raise a major concern: Do parliamentary
elections become dominated by voters’ perception of party leader’s personality which
Corresponding author:
Jan Berz, Institute of Political Science, Leuphana University of Lüneburg, Universitätsallee 1, 21335 Lüneburg,
Germany.
Email: jan.berz@leuphana.de
Berz
445
would endanger the accountability of elected governments (Curtice and Hunjan, 2011;
Huber, 2014)? Or do perceptions of these leading candidates also depend on their per-
formance in office? Parliamentary elections dominated by voters’ perception of party
leaders may allow parties and elected politicians to avoid public scrutiny for their actions
in office. In this article, I address these concerns by examining whether government
performance affects both vote choice (Downs, 1957; Fair, 1996; Fiorina, 1981; Jacquart
and Antonakis, 2015; Key, 1966; Przeworski et al., 1999; Shabad and Slomczynski, 2011)
and perception of PMs, and whether leader effects are consequently confounded by voters’
evaluation of government performance, because voters hold PMs personally accountable
for their government’s work in office.
I use survey data from British, Danish, and German parliamentary elections to test for
a confounding of leader effects of PMs by voters’ evaluation of government perfor-
mance. I find that in almost all studied elections, leader effects of re-running PMs are
partially confounded by voters’ evaluation of government performance. Consequently,
leader effects on vote choice are between 10% and 40% lower once government perfor-
mance is accounted for. Furthermore, the article uses a case of prime ministerial replace-
ment at the German state level as a natural experiment to ensure that the investigated
causal relationship does not work in the opposite direction. These findings suggest that
even if the nature of parliamentary elections has become more and more personalised,
voters are likely to hold PMs accountable for their behaviour in office. In its conclusion,
this article also discusses how voters’ perception of the PM’s influence over governmen-
tal decision-making could differ between PMs and therefore moderate personal account-
ability of individual PMs.
I will begin with a review of the electoral personalisation literature and subsequently
discuss findings on performance voting to formulate my hypothesis. Thereafter, I analyse
the confounding of PMs’ leader effects and discuss the robustness of my findings. I con-
clude by suggesting some further avenues for research on electoral personalisation and
electoral accountability.
Electoral accountability of PMs
Why should voters hold PMs personally accountable for government performance? In
this section, I argue that voters make their vote choice in light of past government perfor-
mance and either reward or punish incumbents. In addition, voters will change their eval-
uation of the PM – whose status as agenda-setter serves as a heuristic for responsibility
assignment under joint decision-making in government – depending on the government’s
performance in office. Consequently, the effect of voters’ evaluation of PMs on vote
choice is confounded by government performance and likely to be overestimated when
performance is omitted. Figure 1 shows this relationship graphically.2 Previous electoral
research has clearly established the electoral impact of voters’ perception of party leaders
(Bittner, 2011, 2018; Ferreira da Silva and Costa, 2019; Garzia, 2014, 2017a, 2017b;
Lobo and Curtice, 2014; Mughan, 2015). The direct effect of voters’ perceptions of party
leaders on vote choice is usually labelled as leader effects and I use this term throughout
the study. These findings are also relevant towards PMs since they have often been ana-
lysed as party leaders, alongside other competing prime ministerial candidates in parlia-
mentary elections.
From the perspective of electoral accountability, the increasing electoral impact of party
leaders has raised concerns (Curtice and Hunjan, 2011). If vote choices are increasingly
446
Politics 40(4)
determined by the personality of party leaders, elections may fail to fulfil their democratic
function to hold elected governments accountable for their actions in office. This is espe-
cially true if leader effects are unaffected by actions PMs take in office. I argue that the
leader effects of PMs do depend on their actions in office. King (2002) outlines the possi-
bility that party leaders influence vote choice because they are affiliated with the govern-
ment, as well as their party, rather than due to the character or personality of party leaders.
While the literature has primarily analysed such a confounding effect of party leader’s
party affiliation (Garzia, 2012; Huber, 2014), the government affiliation of party leaders
has received scarce attention by the literature. Studying whether voters hold PMs person-
ally accountable will not only address a crucial concern of electoral accountability, but also
contribute to an understudied aspect of leader effects. Formichelli (2014) argues that leader
effects may be moderated by the type of government coalition prior to the election, arguing
that the fewer parties form the government, the easier it is for voters to process decisions
made by the government and link them to parties and subsequently to party leaders. The
study offers important comparative evidence of leader effects under varying government
contexts. However, recent findings on the assignment of responsibility under coalition
governments refine the seminal study of Powell and Whitten (1993) who find differences
in the clarity of responsibility between government types. Current evidence (Angelova
et al., 2016; Debus et al., 2014; Duch et al., 2015; Duch and Stevenson, 2013; Plescia and
Kritzinger, 2017; Williams et al., 2017) strongly suggests that in coalition governments,
voters hold the agenda-setter accountable for government performance. To identify the
agenda-setter, voters can rely on the party affiliation of the PM as central heuristic which
signals control over the government’s agenda (Duch et al., 2015; Duch and Stevenson,
2013). This means that voters will focus on the PM’s party – and as I argue also the PM
herself – when assigning accountability under coalition governments.
That voters punish or reward incumbent parties for retrospective performance on poli-
cies and economic dimensions when casting their vote is a well-established finding
(Anderson, 2000; Duch et al., 2010; Fair, 1996; Fiorina, 1981; Green and Jennings, 2012;
Key, 1966; Lewis-Beck and Stegmaier, 2009, 2013). Recent research furthermore empha-
sises that voters behave in this way across European countries regardless of differences in
electoral institutions (Dassonneville and Lewis-Beck, 2017). While scholars have focused
on the accountability of the government parties, Plescia and Kritzinger (2017) and Stiers
(2019) show that the mechanism of performance voting extends to opposition parties.
Why will voters apply the mechanism of performance voting to the PM and not only
hold the PM’s party accountable? I argue that electoral accountability extends directly to the
PM, because voters infer behaviour in governmental affairs and build expectations of per-
formance from their evaluation of candidates (Clarke et al., 2004, 2011; Miller et al., 1986).
PMs matter to voters not simply due to their personality or appearance, but due to their
influence on future government decision-making and in...
To continue reading
Request your trial