Turnout: the EURef big unknown
Just how enthused is the general public?
For those inside the political bubble, whether as practitioners, supporters or lobbyists on the inside, or commentators and observers watching on, it’s easy to forget how little politics matters to an enormous number of people. One in three people didn’t vote at the last general election, more than half of Londoners and Welsh didn’t do so in May and only slightly fewer than half in Scotland and N Ireland.
Europe as an issue matters immensely to a small number but even now, after months of campaigning, only 28% listed it as an important issue in the May Mori Issues index (albeit that 38% mentioned immigration, which will to a greater or lesser extent be EU-related, and which will also presumably have quite a large overlap in those who mentioned both issues). All the same, we’ve heard a lot of noise but not seen much light this campaign: is it enough to drag the marginally- or profoundly-disengaged off their collective sofas?
It matters not only so as to increase the legitimacy of the result but also because of the effect it will have on that result. A low turnout will favour Leave, who are by far the more motivated side. A moderate turnout will depend on the relative strengths of two higher-turnout groups: the middle-class (more pro-Remain), and the elderly (pro-Leave). A high turnout reverses that, with the focus then on younger voters and the lower social groups.
On that question, unfortunately, we have to take the polls with a degree of scepticism – or perhaps with a discount. A golden rule of polling is that polls always report voters as more keen to go to the polls than they will in fact be. For example,
– In the Scottish Independence referendum, which recorded an unprecedented actual turnout of 84.6%, three eve-of-poll surveys from Survation, YouGov and Mori each found 90% or more saying they were ‘certain’ to vote.
– In last year’s general election (turnout 66.4%), four election eve polls each found well over 70% allegedly certain to vote and over 80% saying that they were 8/10 or more likely to do so.
There are some technical reasons that explain part of the gap, notably students and others registered more than once but only able to cast one vote. All the same, it doesn’t make up the difference.
The likelihood to vote figures this time are similar but subtly different from those for the General Election: the numbers for the weighted averages and the 8+/10 are close to identical to those for the 2015GE. On the other hand, those declaring themselves as ‘certain’ voters are notably lower: ORB’s previous poll (5 June) reported only 59% certain, though yesterday’s had it up at 73%.
If I had to guess, then based on the past gaps between declared likelihood to vote and actual turnout, and based on the current polling figures, I’d say that the EURef is heading for a turnout of around 62-64%.
What does this mean? One striking feature of ORB’s poll was that while the usual age group profile was apparent – 56% of 18-24s declared themselves 10/10s, rising to 88% of 65+’s – there was virtually no difference between social groups: all four ranged between 70% and 74%.
ORB may yet prove to be a rogue – it’s out of line with its previous finding, as well as those of other companies – and that social group voting likelihood profile is unusual. However, the evident signs of panic within Labour this last week suggest that even if the figures are a little overcooked, it may well have picked up a very genuine swing, one coming out of the traditional (or one-time) Labour working class deciding that this time it will vote and that it will vote Leave, and one which Labour has also identified. As an aside, the coincidence of the referendum with Euro2016 is unlikely to hinder that trend with that group.
Despite all this, the bookies still make Remain strong favourites, at no better than 4/11, with Leave best-priced at the time of writing at 12/5 with SportingBet and 9/4 widely available. That gap seems absurd to me. The pollsters are modelling something quite new to them and neither they nor we have a direct track record to go on. Certainly, some pollsters will again end up with egg on their face but do we have any more reason to believe it will be those who’ve given succour to Remain rather than those who’ve found Leave handily placed?
I’m finding this a very hard vote to call; every instinct and analysis comes with a ‘but’ attached. For all that, I can’t help but think that it’s close and that as such, Leave’s odds offer considerable value.