SNP lead by 19% in richly satisfying new YouGov poll

With less than two months to go until the general election, The Times have released a new full-scale Scottish poll from YouGov...

Scottish voting intentions for the May 2015 UK general election (YouGov, 10th-12th March) :

SNP 46% (-2)
Labour 27% (n/c)
Conservatives 18% (+3)
Liberal Democrats 4% (n/c)
Greens 3% (n/c)
UKIP 2% (-2)

[UPDATE : The following analysis was written before it emerged that YouGov have made a big methodological change, which means that the poll is not directly comparable with previous ones.  You can read more about this in a fresh post HERE.]

The percentage changes are measured from the last comparable poll way back in late January/early February.  Clearly there are only two possibilities - either there has been very little movement since then, or there has been no movement at all.  But which is it?  On the face of it, YouGov are in agreement with the last Survation poll in suggesting that the humungous SNP lead has edged down just slightly.  But in fact that's a false comparison, because the Survation poll gave the SNP their lowest lead since the referendum, whereas tonight's lead of 19% falls bang in the middle of YouGov's post-referendum range.  The smallest gap they've reported since September has been 16%, and the biggest has been 21%.  The SNP's vote is actually 3% higher now than it was in the YouGov poll conducted at the same time as the famous Ipsos-Mori poll in the mid-autumn, which gave them a 29% lead.

The other recent piece of evidence we've had was from a TNS-BMRB poll which flatly contradicted the direction of travel in Survation by suggesting that the SNP's lead had shot up by 6% (albeit from a suspiciously low base).  So the explanation that best fits the pattern we've been seeing is that the state of play has remained pretty much static since October, and that any apparent small changes reported by the polls have been meaningless margin-of-error "noise".

Here's something that's slightly weird - there have been four YouGov polls since the referendum, and Labour have been on 27% in every single one.  Given the 3% margin of error, you'd expect a little bit of fluctuation, so that sequence of results speaks rather eloquently to their complete failure to achieve any lift-off at all since Jim Murphy replaced Johann Lamont.  Although we'll have to wait a while for the datasets from tonight's poll, we've already been told that Murphy's personal satisfaction rating has slumped from 33% to 26%.  In truth, that may just be the inevitable effect of him now having had sufficient time to get on people's nerves - but it certainly doesn't leave much room for optimism that Labour will be able to achieve over the next two months what they've failed to achieve over the last three or four.

The die-hard Scottish Tory "surgers", who have spent the last twenty-eight years looking for any possible portent of that elusive recovery, will doubtless feel a little frisson of excitement tonight, because this is the first post-referendum poll from any firm to suggest that they might slightly increase their share of the vote from 2010.  But one swallow does not a summer make, and this could just be a freakish effect of the margin of error. 18% is obviously within 3% of 15%, which is roughly what they've been averaging of late.

*  *  *

I was a touch bemused by the consensus among the panel on This Week that Labour should rule out a deal with the SNP.  That would be perfectly understandable if the people concerned were motivated by the racist attitudes towards Scots that we've seen in the London press over recent days, and by a sense that it's unconscionable to work constructively with a distinctively Scottish party.  But that wasn't the rationale being put forward - instead we were invited to believe that a rejection of a deal would prevent harm being done to the Union.  It's difficult to know whether to laugh or cry - have the Westminster establishment truly learned nothing over the last twenty years?  If you want to prevent Scottish self-government (although why you'd actually want to do that is anyone's guess), you have to argue the case against it, and make it less popular in Scotland.  What you can't do is say "no matter how many people in Scotland vote for Home Rule, we'll make sure we block it using English votes at Westminster".  That's what Margaret Thatcher and John Major attempted, and it was self-evidently counter-productive - by the end of those eighteen years, a slim majority in favour of a weak Scottish Assembly had been transformed into a "settled will" for a powerful Scottish Parliament.  We could be right in the midst of a similar transformation, and if Labour create a reverse 1979 legend by letting in a Tory government specifically because they couldn't bear to work with a left-wing party that Scotland had just voted for in huge numbers, a tipping-point in favour of independence could be reached very quickly.

The other line I can't quite get my head around is "Labour can't do a deal with the SNP because Scottish Labour MPs won't put up with it".  The whole point about what's happening is that the opinions of Scottish Labour MPs are about to count for much less than at any time in the last century, because there simply won't be very many of them.  It's the views of English Labour MPs that matter - and more to the point, what their views are after the election, rather than right now.    It's easy to be gung-ho about rejecting a deal until you're faced with hard, unyielding parliamentary arithmetic that forces you to make a very tough choice between being in government with the help of others, and not being in government at all.