Generally speaking, the more individual candidates spend on their campaigns, the higher turnout in that constituency. Siim Trumm, Laura Sudulich and Joshua Townsley find that the money spent by viable contenders has a greater impact on voter turnout than spending by those who are unlikely to win.
During elections, constituency candidates from all parties and none routinely raise and spend money on their electoral campaigns. This spending covers, among other things, the printing and distribution of leaflets and letters, the use of a local party office and IT resources. In essence, money is spent with the purpose of mobilising voters, and to provide information to voters in the run-up to an election. Unsurprisingly, most studies show that the more candidates spend at election time, the more people vote.
But not all candidates are equal. While some are big players locally – likely to finish first or second place in the constituency – many are not. Is the connection between the amount spent on a campaign and the level of turnout conditional on how ‘viable’ the candidate is locally?
Studies that have compared the effects of spending by different parties have mostly focused on the traditional ‘big three’ parties – Labour, the Conservatives and the Liberal Democrats. But given the rise of other, so-called ‘minor’ parties in recent years – the SNP, Plaid Cymru, Greens and UKIP, among others – local electoral dynamics often involve different combinations of ‘viable’ and ‘less viable’ candidates in different seats.
Consider the marginal seat of Watford in 2010, where the three viable contenders were the usual suspects: Labour, Conservatives and Liberal Democrats. Between them, they secured 94% of all votes cast. Meanwhile, in the Na h-Eileanan an Iar constituency in Scotland, the viable contenders were the SNP, the Scottish Labour Party and an independent candidate, Murdo Murray, who received a combined vote share of 88%.
Instead of focusing on the traditional ‘big three’, we separate the effect of the ‘viable contenders’ in each constituency from the ‘other contenders’ – whichever party they represent. In so doing, we allow for different combinations of ‘viable’ and ‘minor’ candidates in each constituency.
We counted as ‘viable contenders’ those candidates who are in with a realistic chance of winning based on their performance at the previous election. For example, in a safe seat, the only ‘viable contender’ is the party that finished first at the previous election. In two-way and three-way marginal constituencies, the ‘viable contenders’ are the top two and three parties, respectively.
The Ochil and South Perthshire constituency, for instance, was a three-way marginal following the 2005 general election with Labour on 31.4%, SNP on 29.9%, and the Conservatives on 21.5%. In this case, the viable contenders were Labour, SNP and the Conservatives, with the other candidates representing ‘other contenders’. Whereas the conventional approach would only capture the campaign spending of Labour and the Conservatives in this seat, our approach also captures the efforts of the SNP.
Campaign spending and voter turnout
We take the amount spent by candidates in constituencies in Britain, and divide it by the legal spending limit in each constituency. This is because candidate spending in Britain is limited by law and varies according to the size of the electorate and the geography of the constituency. Therefore, we test the impact of relative spending on turnout.
Table 1 presents two regression models, both with voter turnout (%) as the dependent variable. We control for marginality (the bigger the margin of victory, the lower turnout is), and a series of other constituency-level factors that also influence turnout (proportion of single occupancy households, social class, size of the electorate, and whether the constituency is a county or a borough/burgh).
Table 1: Explaining variation in voter turnout with campaign spending
Model 1 shows that the more candidates spend, the higher turnout is in that seat. For every 1 percent that spending (as a proportion of the legal maximum) rises, turnout rises by around 0.6 percent.
Model 2, meanwhile, distinguishes between the spending of ‘viable contenders’ and ‘other contenders’. By separating out the two, we can see that the former has a stronger impact on turnout. When the spending of viable candidates goes up by 1 percent, turnout rises by 1.22 percent. Meanwhile, when the spending of other candidates rises up 1 percent, the associated turnout boost is much lower (0.48 percent).
In order to illustrate this, Figure 1 shows the predicted turnout at all levels of campaign spending by viable contenders and other contenders. The difference in the steepness of the lines indicates that the effect associated with spending by viable and other contenders varies notably.
Figure 1: Effect of viable contenders’ and other contenders’ spending on turnout
Campaign spending increases turnout. We show that candidates also differ in their capacity to mobilise voters, according to their prominence in the constituency. When candidates spend more on their local campaigns, the level of participation in that constituency increases – and the higher the local standing of the candidate, the stronger their ability to mobilise voters.
In further analysis presented in the paper, we also find that these effects are not conditioned by how competitive the race is in each constituency. In other words, these mobilising effects are consistent in ‘safe’ constituencies, where the winner is hardly challenged, and in marginal constituencies, where the result is likely to be close.
The study therefore testifies to the relevance of local electoral campaigns when it comes to mobilising voters. We also present a template by which future studies can ensure the substantial sums spent by candidates representing other political parties are captured. Given the complex nature of Britain’s party system, the campaign efforts of parties and candidates beyond the traditional ‘big three’ may become increasingly important.
Note: This article was originally published on Democratic Audit. It is based on the authors’ paper, ‘Information effect on voter turnout: How campaign spending mobilises voters’, published in Acta Politica.
Siim Trumm is Assistant Professor in Politics at the University of Nottingham. His areas of research include British politics, party politics, parliaments and legislative behaviour, electoral campaigns, and public policy.
Laura Sudulich is Senior Lecturer in Politics at the University of Kent. Her research focuses on electoral campaigns, digital media and voting behaviour, public opinion and the effects of information on political behaviour.
Joshua Townsley is a PhD candidate in Politics at the University of Kent. He researches electoral campaigns and political behaviour. Joshua also runs the LSE and Democratic Audit’s Democratic Dashboard voter information project. He tweets @JoshuaTownsley.
All articles posted on this blog give the views of the author(s), and not the position of LSE British Politics and Policy, nor of the London School of Economics and Political Science. Featured image credit: Pixabay (Public Domain).