Dark nights and dismal campaigning will not enthuse waverers
There are four main factors in an election turnout: how important voters view the poll, how close they expect the result to be locally, how close they think it’ll be nationally, and what the prevailing local culture is towards voting.
On that basis, you’d expect the general election to have a pretty decent turnout. Of the two national variables, there is genuinely a major issue at stake with Brexit, as well as big gaps between the parties on policy; and polling suggests that while the Tories have a lead, it may not be enough to deliver an overall majority (though that depends on which company you use).
However, will that really be the case? I’m very doubtful that we’ll see a turnout of 69% like last time. Indeed, we could well be set for the lowest turnout since the Blair years and the 59-61% showings of the 2001/5 polls.
The most obvious reason is the timing. This will be not only the first election outside the April-June window since 1974 but the latest to be held in the year since 1918 (the last December election, that of 1923, was earlier in the month, on 6 Dec). This means it will be very dark on election day, with a decent chance of it being cold and wet into the bargain: not enticing to voters weighing up whether to bother.
To put some figures on that, on the Isle of Wight, on election day, sunrise is just before 8am and sunset just before 4pm: that’s seven hours of polling when the sun’s below the horizon. By contrast the times for the earliest recent election (9 April 1992) are that the sun was up well before polls opened and didn’t set until 7:50pm. The further north you go, the shorter the day will be. Up in Lerwick, nine of the 15 hours the polls are open will be in darkness.
That said, the evidence regarding December elections is mixed (and sparse, with by-elections our best guide). The Richmond Park by-election, held on 1 Dec 2016, and fiercely contested, produced a very impressive by-election turnout of 53.4%. On the other hand, the routine Labour hold in nearby Feltham & Heston in December 2011 generated a turnout of just 29%.
As well as the weather and darkness, there’s also the obvious distractions of Christmas – less than a fortnight away come polling day. Parties, works do’s, Festive shopping, nativity plays and any number of other things that don’t happen on the first Thursday in May will be competing for voters’ time.
In addition, 12 December looks to be the penultimate day of the term for many students. The greater swing to Labour in the 2017 election in student-heavy seats was one of the election’s key characteristics. That may be far harder to replicate this time is some students have already departed for home and many others have end-of-term Christmas events.
And then there’s the simple fact that voters are not particularly fond of politicians or politics at the moment. Leave aside the conditions in which the election will be held: any election where the Leader of the Opposition has a net satisfaction rating of -60, while the government’s rating is -55 (both figures from the latest in the Mori series) is unlikely to engender enthusiasm. True, those figures could change for the better but neither Tory nor Labour campaign has offered a first week to inspire confidence that they will.
Which brings us to the undecideds. These voters tend to be ignored in polling summaries as they don’t feature in the headline stats. How they break over the next five weeks though could be crucial. The most recent YouGov poll had 19% of 2017GE Lab voters saying that they Don’t Know how they’d vote this time or refusing to answer the question, against 15% for both the Tories and Lib Dems. The most recent ICM returned smaller shares for the undecideds but still with more Lab than Con.
The normal expectation is that the majority of these voters will end up voting (they did so last time), and most will return ‘home’ – which if so, would be a net gain for Labour. But will they? That surely has to depend on things like the issues already discussed. And of course, if they don’t all return, the question then becomes in what proportions they do turn out.
I don’t think it’s easy to predict which parties this will help at the moment. We can reasonably assume that any drop in turn out is likely to be lower among postal voters than those who vote the traditional way but postal votes are no longer confined in the main to the elderly – this is not an inbuilt Tory advantage. Likewise, it may well be that the over-50s will have fewer alternative calls on their time (which would benefit the Tories), but that’s speculative.
In reality, the party that will suffer most is likely to be the one that has the more demotivated voters. While that’s always true, the effects of the Advent Election will magnify it more than usual. The campaigns really will matter – which does beg the question why they’re quite so bad.