Incumbent parties, incumbent MPs and the effectiveness of constituency campaigns: Evidence from the 2015 UK general election
| Author | Charles Pattie,Todd Hartman,Ron Johnston |
| Published date | 01 November 2017 |
| DOI | 10.1177/1369148117718710 |
| Date | 01 November 2017 |
| Subject Matter | Articles |
https://doi.org/10.1177/1369148117718710
The British Journal of Politics and
International Relations
2017, Vol. 19(4) 824 –841
© The Author(s) 2017
Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/1369148117718710
journals.sagepub.com/home/bpi
Incumbent parties, incumbent
MPs and the effectiveness
of constituency campaigns:
Evidence from the 2015 UK
general election
Charles Pattie1, Todd Hartman2
and Ron Johnston3
Abstract
Parties’ local campaign efforts can yield electoral dividends in plurality elections; in general, the
harder they campaign, the more votes they receive. However, this is not invariably the case.
Different parties’ campaigns can have different effects. What is more, the particular status of a
candidacy can also influence how effective the local campaign might be. Analyses of constituency
campaigning at the 2015 UK General Election reveal inter-party variations in campaign effectiveness.
But looking more closely at how a party was placed tactically in a seat prior to the election, and
at whether sitting MPs stood again for their party or retired, reveals distinct variations in what
parties stand to gain from their local campaigns in different circumstances.
Keywords
campaign effects, constituency campaign, incumbency, political parties, United Kingdom, voting
Local election campaigns matter. A vast international literature shows that, on the whole,
the harder parties work on their local campaign in an election, the greater the dividends in
terms of vote share (e.g. Denver and Hands, 1997b; Green et al., 2016; Jacobson, 2015;
Fisher et al., 2014; Benoit and Marsh, 2003). For instance, campaigning can increase turn-
out among supporters and swing undecided voters to a party. Yet, there is little evidence
that those who previously voted for another party are won over by the doorstep canvassing
characteristic of local British campaigns. As such, a persistent topic for analysis within the
literature concerns the relative efficacy of campaigns conducted by incumbents and their
challengers. In a now-classic analysis of US Congressional elections, Jacobson (1978)
demonstrated that incumbents standing for re-election received much smaller benefits
1Department of Politics, University of Sheffield, Sheffield, UK
2Sheffield Methods Institute, University of Sheffield, Sheffield, UK
3School of Geographical Sciences, Bristol University, Bristol, UK
Corresponding author:
Charles Pattie, Department of Politics, The University of Sheffield, Sheffield S10 2TN, UK.
Email: c.pattie@sheffield.ac.uk
718710BPI0010.1177/1369148117718710The British Journal of Politics and International RelationsPattie et al.
research-article2017
Article
Pattie et al. 825
from their campaign expenditure than did challengers. Since Jacobson’s pioneering paper,
his claim has been subject to repeated scrutiny, with some studies corroborating his find-
ings and others challenging them. Consensus remains elusive.
What is more, much of that research effort has focussed on campaigns in systems domi-
nated by two political parties (or, in multi-party polities, on the two most competitive par-
ties in an area). A question mark hangs, therefore, not only over the relative effectiveness
of challengers’ and incumbents’ campaigns but also over whether all challengers in a seat
gain similar advantages. Moreover, incumbent races can be split into those which involve
an incumbent candidate seeking re-election versus those where only party incumbency is
involved (such as where the incumbent representative has retired before the election).
In this article, we extend the literature on challenger and incumbency campaign effects
by examining how they vary across parties and different types of candidates in Great
Britain. Using data from the 2015 UK General Election,1 we show that not all challengers
were equally advantaged by intensive campaigns and not all parties defending a seat were
equally disadvantaged by such campaigns (much depended on whether their incumbent
MP stood again). But surprisingly, the effects of the campaign also differed significantly
depending on which party we consider – the effects vary, both in size and (in some cases)
direction from party to party.
Campaign effects for incumbents and challengers: Past
research
Much of the debate over the relative merits of incumbents’ and challengers’ campaigns
stems from Jacobson’s (1978) paper on US Congressional elections in the 1970s, in which
he showed that incumbents enjoyed far smaller electoral gains from their campaigns than
did challengers. The difference, he argued, reflected the candidates’ relative positions.
Incumbents were already relatively well known within their districts as a result of their
legislative records, work for their districts, and local media coverage. In contrast, most
challengers were much less well known initially. Their campaigns provided an opportu-
nity to introduce themselves to the electorate and to increase their name recognition.
Incumbents, therefore, had relatively little to gain from campaigning hard as they already
enjoyed wide recognition; challengers, meanwhile, had much to gain as the campaign
helped raise their local profile. Indeed, for incumbent candidates, having to campaign
hard might even represent weakness rather than strength, as it reflected a closer-than-
expected race generated by a particularly strong challenger.
Some have challenged Jacobson’s core finding, arguing that incumbent campaign
spending actually has a substantial influence on voting (see for example Benoit and
Marsh, 2008; Compston, 1993; Erikson and Palfrey, 1998; Gerber, 1998; Green and
Krasno, 1988, 1990; Magee, 2012; Samuels, 2001; Thomas, 1989). But other work,
across a range of elections and polities, supports Jacobson’s claim, suggesting that incum-
bents’ campaigns are less effective than their challengers’ (see Benoit and Marsh, 2010;
Gerber, 2004; Jacobson, 1990, 2006; Kenny and McBurnett, 1992; Palda and Palda,
1998); note, however, Levitt’s (1994; Levitt and Dubner, 2005) claim that the only ‘true’
test of the impact of campaign spending is where the same pair of candidates contested
two successive elections (Jacobson, 2006; Johnston and Pattie, 2006a). In sum, a consen-
sus has yet to emerge.
A notable feature of the literature, however, is how dominated it is by the particular
example of elections in the United States. While there are studies in countries such as
To continue reading
Request your trial