Lord Ashcroft’s mega poll of key marginals, released yesterday, has been interpreted as showing three things:
1) Labour are doing very well against the Tories in the Tory-Labour battlegrounds
2) The Lib Dems are doing less well but still ok against the Tories in the Tory-LD battlegrounds
3) The Tories’ problems are a result of their voters defecting to UKIP
I agree with the first two interpretations, but the third looks to me to be a misreading of the data. Its extensive coverage in papers that would prefer Cameron to be more UKIP-like – the Telegraph, Mail & Express – suggests wishful thinking.
I’m going to focus on the poll of Tory-Labour marginals because that’s got more constituencies (32 vs 8) and a much bigger sample size – and it’s the one the coverage has focused on.
This is a poll of constituencies the Tories hold, so at the last election, Labour were slightly behind in all of them. Yet now the headline voting intent figure has Labour 13pts ahead*:
But UKIP’s vote is 14% and Labour’s lead is only 13pts, so that means UKIP are the reason Labour are leading in the constituencies, right?
Not even two in five of that 14% who would vote UKIP in the next election voted Tory in 2010:
If UKIP were to disappear after the EU elections and the Tories were to be reunited with their lost voters, they would gain just 5.3pts – not nearly enough to overhaul Labour’s lead. And of course if UKIP were to disappear, some of those Labour defectors could return, potentially adding 2pts to Labour’s score. Put those together and the UKIP damage to the Tories is just over 3pts: less than a quarter of Labour’s lead.
This isn’t to say Labour’s lead in these constituencies is secure. Out of the main parties it has kept the highest proportion of its 2010 voters, and of course it’s had a sizeable chunk of people who abandoned the Lib Dems after the coalition was agreed. But it also has the largest number of people who didn’t vote in 2010**:
These people have already said in the poll that they’re going to vote at the next election, but the high proportion of people who didn’t vote in 2010 is a risk if they regularly don’t turn up at elections (though some are probably first-time voters). And of course there are other reasons the Labour vote may fall.
But the idea that UKIP is the reason the Tories are behind in these key marginals is just not true – or at best it’s a quarter of the truth.
* There’s a separate voting intent question that encourages respondents to think about their particular constituency. This might be a better guide of how people will actually vote when push comes to shove, but I’m not using it here because I’d rather stick with something that’s comparable with other polls. Anyway, UKIP’s vote is smaller there and Labour’s lead is larger.
** The tables don’t show exactly what proportion of those “Other/Didn’t vote” were non-voters in 2010 (they need to be weighted by turnout), but the raw numbers suggest they’re overwhelmingly non-voters and that more than half of those who didn’t vote in 2010 but would now, would now vote Labour.
via Leo Barasi Liberal Conspiracy http://liberalconspiracy.org/2013/09/16/the-tories-have-a-bigger-problem-than-just-ukip/