As Britain’s election enters its final week, the polls remain firmly stuck. Labour and Conservative are tied on around a third of the likely vote each. The Liberal Democrats look trapped at around 8%, while UKIP hovers around 13%. No party – the SNP aside – shows any real sign of breaking the deadlock. A hung parliament and a complex round of negotiations as the rivals jockey to form a viable government is still by far and away the most likely outcome.
No surprise, then, that things are getting tense in the rival campaigns. To stand any chance of moving ahead parties need to squeeze out every last vote they can over the next few days. The battle, both in the constituencies, and nationally, is on and every vote counts.
But where are the extra votes going to come from at this late stage? Some will be gained by ensuring that as many existing supporters as possible actually turn out and vote. There is still scope to persuade supporters of rival parties to change their votes, too – either by changing peoples’ minds (though this is very hard to do so late in the campaign) or by convincing them to vote tactically in order to deprive an even less popular rival of victory.
But what of voters who have yet to make up their minds? Even at this late stage of the campaign, a sizeable chunk of the electorate remains undecided. The battle for their support could prove the wild frontier of the final days.
The great unknowns
How many voters remain undecided? We can get some idea from the pre-election wave of the 2015 British Election Study, which interviewed a large sample of voters across Britain in March this year.
Respondents were asked, on the eve of the official election campaign, whether they had yet made up their minds on how to vote: 68% said they had decided, 28% said they had still to make up their minds, and the remaining 5% said they had already decided to abstain. Even though some of these initially undecided voters will by now have made their minds up, many will not. A substantial minority of voters were still up for grabs as the election opened. All to play for.
What difference might it make?
The big question is what difference these undecided voters might make to the outcome of the election. One possibility is that they will prove to be a random cross-section of all voters, except that they make up their minds later than others. To the extent that this is the case, the electoral consequences of winning over undecided voters might be muted – and if all who are undecided at the start of the campaign go on to vote, the overall effect on the parties’ vote shares will be negligible.
But it is also possible that those who have yet to make up their minds at the start of the campaign will eventually come to different decisions to those who decide early. Early deciders are often strong partisans, who support their party through thick and thin. Late deciders, on the other hand, could be swayed by events late in the campaign. If this is the case, the party that can get an edge among the late deciders stands a good chance of improving its overall performance substantially – and potentially winning more MPs in the process.
While we don’t yet know for sure how different early and late deciders’ party choices will be in 2015, we have a few clues at our disposal which might help point the way.
Clue one: is the past a guide?
One source of clues is to look back to past elections. In 2010, for instance, similar proportions of respondents to the pre-election wave of the British Election Study said, a month before the election, that they had already decided how to vote (63%), were as yet undecided (29%) or had decided to abstain (4%). What, in the event, did they go on to do?
Of those who, at the start of the campaign, thought they had made their minds up, 95% went on to vote in the election. So too did 91% of the initially undecided – and 29% of those who began the campaign certain that they would not vote.
It is important to note that this substantially overestimates turnout for all groups: as this is an internet survey, it contains an especially strong bias towards those already interested in politics and voting. The 2010 study also carried out a face-to-face survey which had far fewer respondents, but was less affected by underestimation of turnout. This survey reported higher abstention rates among the undecided – 77% of this group voted, compared to 91% of those who claimed in the face-to-face survey to have made their minds up early.
One lesson, therefore, is that many of those who begin the campaign undecided never make up their minds sufficiently to cast a vote.
Among those who did cast a vote in 2010, both the internet and the face-to-face surveys agree. Those who decided late in the campaign voted differently from those who made their minds up early.
Those who began the 2010 election campaign undecided were considerably more likely – indeed, around twice as likely – to vote Liberal Democrat in the event than were those who began the campaign knowing what they intended to do.
This is not too surprising. Traditionally, the Liberal Democrats tend to benefit during the campaign period. In 2010, the first ever televised leaders’ debate excited a considerable amount of interest, particularly in the Liberal Democrats’ young leader. Cleggmania was one of the stories of the campaign.
But it wasn’t just about the leaders’ debates. The Liberal Democrats (and the Liberals and Liberal/SDP Alliance before them) had long been the main beneficiaries during the election campaign. Starved of media attention between elections (when the focus tends to be on the government and to a lesser extent the main opposition party), the extra publicity and media coverage generated by the campaign tended to remind voters that the Lib Dems were there (and also helped encourage tactical voting for the party).
It is likely, therefore, that this time late deciders will think and vote differently to those who made their minds up early.
It is much less likely that the Liberal Democrats will still be the main beneficiaries. Entering the coalition government with the Conservatives in 2010 ensured that the party remained much more at the centre of public attention through the 2010-2015 parliament than had been the case before then. Not all publicity is good publicity: the partnership with the Conservatives has proved deeply unpopular, not least among 2010 Liberal Democrat voters who thought they were casting a vote to keep the Conservatives out of office, only to discover their vote had achieved the opposite.
Clue two: current vote preferences
Let’s assume that people will actually vote on May 7 for the party which, at the start of the campaign, they say they are most likely to vote for, though they are still not sure. Analysis of the 2010 results suggest around 80% of the decided will go on to vote for that party, as will just over 50% of the undecided.
Here, too, the initially undecided at the start of the 2015 campaign look a bit different from the already certain. Relatively speaking, the Liberal Democrats and the Greens do better (and Labour and the Conservatives worse) among the undecided.
Whereas only 6% of those who had already made up their minds on how to vote on May 7 said the Liberal Democrats were the party they were most likely to vote for, it’s 14% of the undecided. This is not a huge number, but the Lib Dems are in no position to be too precious about this. So difficult is their current position that anything, no matter how small, which might boost their vote is welcome news for them. There could, therefore, still be a small campaign bounce for the Lib Dems as the initially undecided gradually make up their minds. If so, the election might not be quite as dismal for the party as current polls suggest.
That said, all the data for 2015 was collected before the official campaign began. Things have moved on since then. Thus far, Labour and its leader Ed Miliband are widely held to have had a good campaign, while the Greens’ surge never really began. We don’t yet know for sure how all of this has affected the undecideds. What we do know is that the battle for their vote could be one of the key factors in the election.