What's new?
There hasn't been any new Holyrood poll recently. So this is a technical update as I phased older polls out of my super-sample and am now using only the two most recent ones. Super-sample size is 2,020 with a 2.1% MOE on both the constituency and list votes. Here is what the updated weighted average of voting intentions says:
These results are pretty good for the SNP. Of course not when compared to the 2016 results as the SNP is down on both votes. But they show a significant improvement on earlier Holyrood polling and also pretty much the same trend as House of Commons polls. The breakdown of list voting intentions by region also shows some interesting shifts from 2016.
Some polls include a regional crosstab of the list voting intentions and sometimes these stray far off what uniform national swing (UNS) would predict. Interesting case is North East Scotland where the SNP would enjoy the same lead over Conservatives as in 2016, while UNS on the list vote is roughly 3.5% from SNP to Conservatives. In South Scotland the SNP and the Conservatives would now be neck to neck with Conservatives slightly in the lead (which fits with UNS) while Conservatives won the list vote there only once in 2003.
How did we get here?
Not all polls since the last Holyrood election have been milk and honey to the SNP's ears (jings, another mixed-up metaphor 😬). Anyway here are the trends for both Holyrood votes based on the 17 polls fielded since June 2016.
Not surprisingly the patterns we see here are fairly close to those seen in Westminster voting intentions for the same period. Some hints at buyer's remorse in the months immediately following the 2016 Holyrood election with SNP 4-5 points up from the election results. Then a noticeable slump roughly during the 2017 GE campaign when the SNP's performance was far from stellar. Finally a gradual recovery from the end of 2017 until today.
Interestingly the trendlines also show that SNP going down/up is generally mirrored by Labour going up/down. Throughout the period the combined SNP + Labour voting intentions are roughly 65% ± MOE on the constituency vote and 56% ± MOE on the list vote, the balance being people intending to vote Green on the lists. I read this as evidence a number of voters are willing to switch back and forth between the two center-left parties, depending on which one appears the most convincing at the moment, but not (yet?) ready to throw the baby under the bus (crivens, another mixed metaphor 😁) and switch to the Conservatives.
What kind of Scottish Parliament would a snap election have delivered?
This is a rhetorical question as a Holyrood snap election is the unlikeliest of all possible political upsets these days. Per Part I, Section 3 of Scotland Act 1998 dissolution would require a 2/3 super-majority which is as likely as Ruth Davidson agreeing to an interview with a journalist from The National. Now let's say that for a moment we live in an alternate timeline where a snap election actually happened, and here's what the average of 2016 and pre-GE 2017 polls would have delivered:
67 seats for the SNP and 78 overall for the pro-Independence parties would have been quite a feat after a disappointing 2016 election but afterwards polling went south (quite literally as it swung towards Unionists) and the post-GE 2017 polls painted quite a different picture.
The SNP down to 54 seats and pro-Independence parties down to 62 seats overall would have strengthened the Unionists' narrative about 'too wee, too poor, too daft' Scotland. But this one too was never bound to happen and current polling is back to much more satisfactory results for the Yes movement.
So what do we get on current polling?
If election results duplicated the weighted average of the two most recent polls, there wouldn't be any major upset though ten seats are projected to change hands. Here are the projected constituency results by SLLM rating:
The SNP would lose two constituencies to the Conservatives: Edinburgh Pentlands (Gordon MacDonald) by less than 1% and more significantly Perthshire South and Kinross-shire (Roseanna Cunningham) by about 4%. Another four SNP seats would be in the danger zone (margin lower than MOE) in regions where the Conservatives did well at Council and House of Commons elections in 2017: Moray (Richard Lochhead), Perthshire North (John Swinney), Aberdeen South and North Kincardine (Maureen Watt), Angus North and Mearns (Mairi Gougeon). But the SNP would also have a fair shot at unseating Labour MSP Jackie Baillie in Dumbarton which was the most marginal Holyrood seat in 2016.
Projected list results are kind of a mixed bag with Conservatives bound to lose the most and LibDems to gain the most. The odd workings of AMS would make that some kind of musical chairs with eight seats projected to change hands.
Assuming all incumbent list MSPs would stand again and 2021 lists would have the candidates in the same order as 2016 lists, the following would notionally lose their seats: Andy Wightman (Greens), Maree Todd, Paul Wheelhouse (SNP), Elaine Smith (Labour), Jeremy Balfour, Dean Lockhart, Tom Mason, Maurice Corry (Conservatives). The one noticeable oddity would be the SNP unseating Andy Wightman in Lothian as the bizarre workings of AMS would automatically 'compensate' the loss of Edinburgh Pentlands with an unexpected SNP list seat.
Finally the full projected Scottish Parliament. Which would still have the SNP as the first party by a large margin and the only ones able to form a government. And the pro-Independence parties holding a two-seat majority which would extend the mandate for a second Independence referendum into the next term, if it hadn't happened already in the fallout of the Brexitocalypse.
Factoring in MOE and the way AMS often compensates changes in the constituency results with opposite changes in the list results, possible ranges of seats would be: SNP 57-61, Conservatives 29-30, Labour 24-27, LibDems 8, Greens 7. Meaning there still is a 30% chance that pro-Independence parties would be one seat short of a majority, though prevailing odds are they would hold a two-seat or three-seat majority.
Seat breakdowns by region
No comment needed on these I think. Just the detail of the projected results we have seen earlier.
Incidentally these results again show how AMS doesn't work as a genuine 'additional members' system but rather as 'compensatory seats for constituency losers', all parties included. Which could be the basis for some kind of electoral reform. More on this in a later blog entry.
The generation gap
Back now to Survation's most recent comprehensive Scottish poll. Here is the picture based on their vote-by-age crosstabs. National data are the poll's results and how they project into seats in Scottish Parliament, not the actual 2016 results or my 'current average polling' projection above. As Comrade Kevin would say the charts speak by themselves, or for themselves, or what-the-fuck-selves. Anyway let them speak and what we hear is quite astounding.
And here is how you can sum it up, showing a clear divide between Generation X and younger on one side and Baby Boomers on the other.
The 55-64s would actually deliver a hung Parliament with the SNP as the first party. But I'm pretty sure that it wouldn't be 2007 all over again. In the current climate Unionists would never allow another SNP minority government on only two more seats than in 2007, so some some sort of wobbly anti-Indy coalition would be the likeliest outcome.
The Generation Divide is even more obvious and striking when grouping parties by their stance on Independence. Millennials and Generation X would vote for a strong pro-Independence majority, close to the 81-seat super-majority mentioned in the Scotland Act 1998. Generation Z would go even further with an amazing 93-seat super-super-majority for pro-Indy parties. Only the Baby-boomers would stick to a pro-Union majority.
Based on current ONS statistics on Scottish population, this means that 43% of the voting-age population would squarely defeat AMS' built-in purpose and offer the SNP a majority; 18% would choose an SNP minority administration within a pro-Indy majority; and only 39% would stick to the plan and deliver some sort of coalition government. Demographics are squarely on the SNP's and Independence's side. Which doesn't mean the Yes movement should just sit on their hands and wait. Boldness should be the order of the day unless you want to whine about that ship having already left the station (help ma boab, yet another mixed-bag metaphor 🙈).
Strikingly too, no generation would pick Labour as their first choice, not even (or probably especially not) those who grew up under Labour's dominance of Scottish politics. This too should ring an alarm in Labour ranks. Even those who likely voted for them in their 20s or 30s have now mostly switched to the SNP, or the Conservatives for those whose foremost concern is to stay in the Union at all costs. Among younger generations the 'Corbyn factor' doesn't work too well either as all massively pick the SNP as their first choice. Scottish Labour is dead, they just haven't realized yet how funny they smell.
What to expect next?
I don't think 'Salmondgate' or any of the made-up 'SNP Bad' controversies will have any significant effect on voting intentions. Recent polls already show that a no-deal Brexit boosts support for Scottish Independence beyond 50%. This should in turn boost support for the SNP as the only party both willing and able to deliver Independence, even if the path to Indy goes through some convoluted twists and turns, and to a lesser extent for the Scottish Greens as the only other party with parliamentary representation openly supporting Indy.
Hopefully we will soon know for sure as further Scottish polls are published, so stay tuned for further briefings.
Cha togar m' fhearg gun dìoladh
No comments:
Post a Comment