Wednesday, January 9, 2019

Is the end NIGH for Ruth? SHOCK YouGov poll sets SNP on course for LANDSLIDE victory

The mystery I mentioned in the previous post has been solved much quicker than expected - and to be fair, it appears (just this once) that nothing sinister was going on.  The reason why the Scottish subsample was missing from the datasets of the YouGov mega-poll commissioned by the People's Vote campaign was, it seems, that it was being withheld for publication later on as a full-scale poll in its own right, which has now happened.  The results are devastating for Labour, and they're not much better for the Tories either.

Scottish voting intentions for the next Westminster election:

SNP 40% (n/c)
Conservatives 25% (-2)
Labour 21% (-2)
Liberal Democrats 8% (+1)
Greens 3% (+1)
UKIP 2% (+1)

The percentage changes listed above are from the last published full-scale Scottish poll from YouGov, which was conducted many months ago.  If anything, the SNP's 40% share might be seen as being on the lower end of expectations, given how well they've been doing in recent YouGov subsamples (never lower than 40% and reaching as high as 46%).  But of course what really matters in a first-past-the-post election is the gap between a party and its competitors - and the SNP's lead over both the Tories and Labour has essentially doubled since the June 2017 election.  On a uniform swing, the SNP would make a big jump from 35 seats at the moment to around 45 - and the big difference from some recent polls is that they'd be making significant seat gains from the Tories and not just Labour.  One of the constituencies they'd recover from the Tories would be Gordon - which might be of some interest to a certain Mr Salmond if he fancies taking his old Westminster seat back once he rejoins the SNP!

The main propaganda purpose of the Britain-wide poll from the point of view of the People's Vote campaign seemed to be to put pressure on the Labour leadership by demonstrating that they will lose massive support to the Liberal Democrats if they are seen to facilitate Brexit.  I wondered in the previous post whether the missing Scottish figures would show that Labour votes in Scotland would drift to the SNP rather than the Lib Dems, and that's broadly proved to be the case.

Hypothetical Scottish voting intentions for Westminster if a Brexit deal is passed by Conservative and Labour MPs, but opposed by SNP and Liberal Democrat MPs:

SNP 43%
Conservatives 27%
Labour 15%
Liberal Democrats 9%

People are generally bad at answering hypothetical questions, so those figures should be taken with a huge dose of salt.  It might well not be anything like that bad for Labour once the page is actually turned - although I also wouldn't rule out the slight possibility that it could be even worse for them if a legend really takes root that they were responsible for dragging Scotland into a hard Brexit.  Either way, it's heartening that voters in Scotland seem to instinctively see the SNP rather than the Lib Dems as the most natural repository for anti-Brexit votes.  That could be a sign that Nicola Sturgeon's post-2017 strategy is paying dividends.

*  *  *

Scot Goes Pop fundraiser: If you'd like to help this blog keep going strong over what could be an epic few months, just a reminder that last year's fundraiser is still very much open for donations, and hasn't reached the target figure yet.

YouGov and the mystery of the missing Scottish subsample

The remarkably well-resourced "People's Vote" campaign decided, for some reason, to commission a ridiculously enormous GB-wide YouGov poll of 25,537 respondents over the course of Christmas and New Year.  One possibility is that it was really a sequence of daily polls for private consumption, which were later aggregated for publication.  It shows Labour dropping to an unusually low 34%, six points behind the Tories.  In normal circumstances you'd say "it's just one poll, we have to wait for more information", but the reality is that this is the equivalent of an average of more than ten polls conducted over a span of two weeks, so it has to be taken very seriously.  The only saving grace for Labour is that polls taken over Christmas sometimes produce weird results which are later reversed, but there's certainly no guarantee that will prove to be the case on this occasion.

Oddly, and in a complete departure from normal YouGov practice, figures from the Scottish subsample are not provided in the datasets.  Even more oddly, the figures from the London subsample are there, which simply serves to make the omission even more blatant.  At first glance it almost looks like an accidental error, with the non-London subsamples being chopped off from the edge of the page.  But if it was an honest mistake, it's one that YouGov have now had plenty of opportunity to rectify.  So it seems more likely that the People's Vote campaign requested that the Scottish numbers should be withheld.

Although the SNP have pledged their support for a People's Vote, the campaign is presumably still very much under the control of "centrists" and "moderates" in the three main London-based parties.  You can certainly see why it might not be in the interests of such people to publish a Scottish subsample of a size equivalent to a full-scale poll, if by any chance it shows a handsome SNP lead.  And one of the obvious purposes of the poll is to put pressure on the Corbyn leadership by demonstrating that in the hypothetical scenario of a Brexit deal going through with Tory and Labour support, Labour would then lose a huge chunk of its voters to the Liberal Democrats.  If it shows those Labour votes would in Scotland actually be lost to the SNP rather than to the Lib Dems, that may be an inconvenient detail that the People's Vote campaign would rather we didn't hear about.  It's perfectly possible that's the case, because the SNP and Plaid Cymru have a combined GB-wide vote of 5% on the hypothetical question about a Brexit deal passing with Labour support, compared to 4% on the standard voting intention question.

This is all supposition, but what we do know for sure is that the SNP have been doing startlingly well in YouGov subsamples over the last few weeks.  The sequence of SNP results since mid-November has been 40 - 46 - 44 - 41 - 43 - 45.  That's the longest sustained run of 40+ results for a very long time.  There's only been one full-scale Scottish poll during that period (a Panelbase poll showing no change in SNP support at Westminster level), so it's difficult to be sure that the SNP have genuinely been performing better of late.  But if they have, what impact will the resolution of the Alex Salmond case have?  In the very short term, it might be a mildly negative impact, because anything that gives the impression of internal disunity is generally frowned upon by voters.  In the longer-term, though, I suspect the effect will be the complete opposite.  We still need to wait for the outcome of the police investigation, but if this proves to be the start of Alex Salmond's reintegration into the SNP, possibly with a view to him standing as an MP or MSP once again, that can only be a good thing from an electoral point of view.  The SNP is unlikely to be truly at ease with itself until the man who led it for almost one-quarter of its entire existence to date is back in the fold.  And that would certainly put a definitive end to controversial journalist David Leask's bizarre attempts to open up a rift between what he calls "the real SNP" and "alt-Nats" (the latter, incredibly, referring in part to the former First Minister and those most supportive of him).

*  *  *

I was interviewed on Radio Sputnik yesterday about the increasing likelihood of Nicola Sturgeon setting out a prospective timetable for a second independence referendum quite soon.  You can listen to the interview HERE.

*  *  *

Scot Goes Pop fundraiser: If you'd like to help this blog keep going strong over what could be an epic few months, just a reminder that last year's fundraiser is still very much open for donations, and hasn't reached the target figure yet.

Monday, January 7, 2019

Would a snap Holyrood election in 2019 really be such a risk?

I'm beginning to realise what it must have felt like to be a social democrat in about 1980, finding yourself by default in a shrinking middle ground as political debate becomes ever more polarised around you.  It seems to me that the two extremes of the debate on how to achieve independence are just as hopeless as each other at the moment.  On the one hand, I had people responding to my post on New Year's Day by saying that Nicola Sturgeon shouldn't even be asking for a Section 30 order, but should instead simply be "repealing the Act of Union".  I asked how that would even be possible, given that the constitution is explicitly reserved to Westminster.  The response was either pseudo-legal gibberish about how the Act of Union supposedly gives the Scottish Parliament a unilateral right to withdraw from the UK (hint: it really doesn't) or a link to Craig Murray's piece explaining that it is normal for countries to become independent without a referendum, and that it is usually done simply by securing recognition from other states.

In principle, I actually agree with Craig that independence is ultimately a matter for international law rather than UK domestic law.  But the snag is that none of the countries that bypassed the hurdle of domestic law in the manner that Craig suggests (for example Slovenia or Croatia) did so just four or five years after their own citizens rejected independence in a free referendum.  Few states, if any, are going to recognise an independent Scotland until it has been demonstrated that the No vote of 2014 has been unambiguously overturned by a fresh vote, which ideally would mean another referendum, or less ideally an election.  So even if you go down the road Craig wants, it just takes you straight back to the original problem of needing a clear mandate for independence.  There isn't any shortcut.

But equally unpromising is the position that some people are attributing to the SNP leadership, which implicitly recognises an absolute Westminster veto by accepting that a) independence cannot happen without a referendum, and b) a referendum cannot happen without a Section 30 order.  The only plan for getting around a veto would appear to be to shame London into backing down by securing mandate after mandate for a referendum, no matter how many years or decades that takes.  So if London says "now is not the time", we campaign some more for an independence referendum, and "take it to the people" in the 2021 Holyrood election.  And if we get another mandate but the reply is "now is still not the time", we campaign even harder and "take it to the people" in the 2026 Holyrood election.  And on and on into infinity: in other words a recipe for Scotland never becoming independent.  The example of Catalonia gives the lie to the notion that no central government would have the nerve to keep saying no indefinitely.

Is there a compromise position between these two extremes that might actually be more effective anyway?  I would guess that the "dissolve the union" camp would be less hostile to the notion of securing another seemingly needless mandate for an indyref if there were two assurances: firstly, that only one more mandate will be sought, meaning that if London are still intransigent after that an alternative course of action will be followed, and b) the new mandate will be sought very quickly.  If, for example, a snap Westminster general election was held this year, and the SNP decided to use that to have one last go at securing a mandate that London might actually respect, that wouldn't slow things down much at all.  The snag, of course, is that the timing of the next Westminster election is not in the SNP's control.

What is effectively in their control, though, is the timing of the next Holyrood election, because there is provision for an early election in certain circumstances.  Nicola Sturgeon can't literally "call" a snap election, but it would be relatively easy for her to engineer one for the purposes of securing another indyref mandate.  It's widely assumed she would never do that because of her instinctive caution, but I'm not actually sure the risks of an early Holyrood election would be as great as they appear.  Yes, elections can throw up surprise results, but that generally only happens when the underdog party has an inspiring leader capable of turning things around on the campaign trail.  Scottish Labour have...Richard Leonard.  OK, Ruth Davidson is a different proposition, but there appears to still be a natural ceiling of around 30% on Scottish Tory support, so any real danger would have to come from Labour, and at the moment I just can't see that happening.  Barring a very weird chain of events, I would suggest the worst-case scenario in an early Holyrood election would be the SNP returning to power once again as a minority government, but without a pro-independence majority.  That would clearly be a sub-optimal outcome, but I'm not sure it should be considered awful enough to deter us from chasing a potentially huge reward if the election went well.

*  *  *

Scot Goes Pop fundraiser: If you'd like to help this blog keep going strong over what could be an epic few months, just a reminder that last year's fundraiser is still very much open for donations, and hasn't reached the target figure yet.