2021 in Review: Constitutional Polling

Although 2020 was the third year of running Ballot Box Scotland, it was the first to feature an annual review of the year’s constitutional polling. I’d noted at the time that was because the previous two years had been pretty status quo and thus boring, whereas last year had seen the first ever sustained polling lead for Independence and was therefore remarkable. I also said I’d probably have to do it every year from then on, and so I’m keeping that pledge.

Fortunately, from the perspective of a nerd that runs a polling website, this year has been another genuinely interesting year for polling on the all-consuming constitutional question. As with the parliamentary polling review, we’ll be looking at both the year as a whole and comparing Q4 2020 with Q4 2021. In this case we’ve got six polls (from five pollsters), so remember the quarterly average will be slightly different from the latest point on the rolling average chart, which is a flat “last five polls.”

Updated 27th of December: Just to spite me personally, Opinium and the Sunday Mail published some polling data on Boxing Day. This included Westminster and Independence data, but not Holyrood. This article has been updated accordingly.

As is always the case with polling on this question, this is going to look at figures both with and without Don’t Knows included. That comes with the caution that Don’t Knows are particularly important on this sort of binary question, and faced with a real referendum may not split neatly between options.

Independence Polling Average Through 2021 (Including Don't Know)

This year began with a continuation of last year’s lead for Independence, around 6% ahead of the Union, alongside around 11% of voters who were unsure. Over the course of the Holyrood campaign, No first took a brief lead in mid-March, fell back behind Yes again for a few weeks, and then pulled consistently ahead by the election. Perhaps unsurprisingly, this somewhat mirrors the SNP’s polling over the year – they’d started with expectations of a stonking majority and finished just short of one.

The peak of the pro-Union lead came immediately after the referendum when it was just shy of 5%, but is back to barely half a percent ahead now. Note that the second last poll of the year was the Ipsos MORI with a big lead for the pro-Independence camp, which really contributed to a narrowing. At the same time, Don’t Knows have also dropped to 8%, indicating voters as a whole have become a little bit more certain of their constitutional position this year.

Independence Polling Average Through 2021 (Excluding Don't Know)

Caveats about not simply assuming neat splits amongst Don’t Knows firmly in mind, what this chart shows more clearly is that whilst the polling trend has pretty firmly moved back towards the Union this year, it remains very close. Even if you were to discount the Ipsos MORI poll as being too out of the normal range, we seemed pretty settled around the Brexit-esque 52% to 48% point. 

Independence Polling Average Q4 2021

Changes here are versus Q4 2020 and, for the pure No:Yes figures, the 2014 referendum.

Compared to the end of last year, No has had a substantial gain of 6.2%. That comes almost equally from a decrease in Yes and Don’t Know figures, which suggests both a direct dip in support for Independence whilst some undecided voters firm up for the Union. On the simple head to head measure, it’s effectively a game of fives. Each side has shifted around 5% from their position in Q4 2020, but also about 5% from their 2014 result.

That means I’m wrapping this section up with something I feel I’ve said a few times lately. There may be cause for relief on both sides of the debate, but there’s no justification for confidence or complacency.

For the Pro-Union camp, the fact polling has turned around from an unprecedented Independence lead will be very welcome indeed. Whatever else has happened in recent years, a compelling and enduring case for Independence has not yet been made. However, with support below the 2014 watermark, it seems much the same can be said of the case for the Union. Merely being in the lead isn’t enough for people who want the constitutional issue to cease being so prominent – support for the Union has to actually grow.

On the Pro-Independence side, although the dissipation of the lead they held last year will sting, they can at least take solace from being ahead of where Yes was seven years ago. Anyone assuming Independence is dead on the basis of within margin of error polls would be absolutely daft. Independence has taken the lead once, and could do so again. But the caution here is that it’s not guaranteed to, and contrary to what many true believers think, starting from a higher base than the last campaign doesn’t mean an easy waltz to victory in the heat of a new one.

Council Area Projection

Please see this page for how projections work and important caveats.

On this purely for indication, purely for fun projection versus 2014, No would have a lead in 20 council areas (down from 28 at the referendum, but up from 10 in last years estimate), compared to 12 for Yes (up 4 from the last vote, down 10 from 2020).

At this point I have to admit to being very disappointed, because I realised when writing this piece I’d missed a poll in October from my initial average. The original version of this map had one particular council area come out at a tie, which would have allowed me to quip that Scotland was on a Fife edge. I enjoyed that terrible gag so much I refuse to let it go entirely unused, even if it no longer applies.

Multi-Option Possiblities

Something that often comes up in discussions around the constitution is the prospect of having multiple options, rather than a simple binary Independence vs Union referendum. Typically this involves some form of “more powers” option, though generally in the form of further devolution rather than federalism. The issue of a federal UK is another long conversation entirely, but suffice it to say, whilst Scotland has the right to change its own systems of governance it can’t unilaterally change those of the UK as a whole.

Polling on the multi-option side of things is unfortunately infrequent, but we did have a couple such polls this year that I think are worth a little look back at.

Back in January, Panelbase asked probably the classic form of this question, which is placing all options directly in competition with one another. They also asked what the preferred format of any future referendum would be, which we’ll look at first.

Right from the outset this doesn’t make for particularly cheery reading for those whose preference is more powers short of Independence. Although we’re nearly a year on from this poll, and further debate on the topic could change minds, an overwhelming majority of Scots said that in the event of a future referendum they would prefer a straight binary question. But maybe if asked it, they might at least prove amenable to a third option?

Ah, no. Independence ends up the most popular single option in this scenario, then the current status quo within the Union has twice as much support as more powers. As this was back in January when Independence was at a peak, at a guess I’d say the Independence versus Status Quo options would be closer to a tie now. Regardless, when actually asked, folk just did not seem keen on further powers as an option in and of itself.

This highlights one possible flaw with a multi-option referendum, which is that a single option may not win a majority. Nobody credible would think it sensible or possible to declare Independence on less than half of the vote.

Yet, if we did this as AV rather than FPTP, we run the risk of a majority voting for further powers in some form, but another majority opting for unmodified Union over Independence after more devolution drops out as the most popular option. That’d leave most people very unhappy, to say the least.

Much more recently, YouGov have been asking about one possible solution to the dilemma outlined just above – instead of a single question with multiple options, how about multiple questions with binary options? The first question was given as “Do you want to change the powers of self-government in Scotland?” 

This shows a stonking two-to-one majority in favour of more powers in some form. That shouldn’t come as a particular surprise to anyone, given voting patterns in Scotland. The mainstream constitutional position has long been to expand Holyrood’s competencies, even if there is a lot of disagreement as to by how much.

Interestingly I had a little bit of pushback when I was tweeting about this on the basis that some voters might think this allows them to express a vote to reduce Holyrood’s powers, or even abolish it entirely. They’d then be blindsided by having to choose between further devolution or Independence. I understand why some might think that, but it isn’t really an issue.

For one thing, reductionist and abolitionist sentiment about Holyrood is a completely fringe pursuit. All five Holyrood parties, between them winning about 95% of the vote in May, lie somewhere on the spectrum of “Status quo to More Devolution to Federalism to Independence”. Parties calling for outright abolition of Holyrood barely mustered 0.6% of the vote. Even allowing for the fact voting is rooted in far more than constitutional policies, there’s no evidence such sentiments are particularly prevalent.

Even if they were, whilst they might impact slightly on the findings of a poll, they’d have no impact on an actual vote. You’d have to be staggeringly un-or-misinformed about a major referendum in order to turn up at the polling station thinking you could vote for fewer powers. Then when looking at the ballot itself, it would make crystal clear what your options were.

Anyway, that particular diversion aside, the second question was “If there is to be change which would you prefer?” This pits More Powers within the Union with Independence – thereby allowing those who support the Union to express a preference on further powers within it, separate to the issue of Union versus Independence.

This question is much more closely run – indeed, it’s not very much different on the straight More Powers versus Independence figures to the plain old normal Union versus Independence polling. There’s a relatively small lead for maintaining the Union with further powers over going the whole hog with Independence. It’s also an interesting comparison with the Panelbase poll from earlier in the year – More Powers within the Union may be the third most popular option in terms of people’s first preference, but it’s possibly the one with most consensus overall.

Where it differs from binary polling is in the size both of Don’t Knows and Wouldn’t Votes (not shown on chart) – at this point in time, voters seem a lot more unsure about how or even whether to vote if these were the options facing them. That may reflect the currently unspecified nature of further powers – YouGov didn’t give an indication, and there’s also currently a lack of clarity around this from proponents of the idea. Many voters may feel unable to choose between the two without a clearer idea of how extensive the changes would be.

The Wouldn’t Votes are even more interesting. As I noted in my analysis piece for the poll overall, whereas 6% of respondents said they wouldn’t vote on the first question, 15% said they wouldn’t vote on the second. That could suggest 9% of voters are so opposed to extending Holyrood’s powers that they’d refuse to choose between options for doing so.

On the assumption those people would much prefer for Scotland to be part of the United Kingdom, it has to be noted that would be potentially self-defeating if they did it in a real vote. If things were close between More Powers and Independence, Status Quo (or abolitionist) voters deciding they couldn’t stomach any further devolution could end up handing Independence a majority on the second question. That’s a reminder that whilst the two-question format may have some advantages over the multi-option, it isn’t without its own democratic risks.

Onwards to 2022

That puts a very neat little bow on Ballot Box Scotland’s coverage for 2021. Christmas is just around the corner, when I’ll be taking a much needed break, as I hope readers will be able to as well. 2022 is going to be another exciting year here on BBS, as all 32 of Scotland’s local councils go the polls in May.

Not only will I have the most detailed coverage of the final results you’ll find anywhere, but in the run up to the election I’ve got a huge pile of content queued up or in development. Not least of these are the various exciting bits of content that I’m currently crowdfunding for (link below!).

I’ve already raised enough for a full poll which I’m expecting to commission for late March or early April, thanks to everyone’s incredible generosity. Support throughout the year for BBS has been unbelievable, and I am immeasurably grateful to everyone who has been a part of that. Here’s to an exciting and hopefully better 2022, with more polling and less pandemic.

Have a very Happy New Year, and I’ll be back again in January to get the local election ball rolling!

If you find this or other Ballot Box Scotland output useful and/or interesting, and you can afford to do so, please consider donating to support my work. I love doing this, but it’s a one-man project and takes a lot of time and effort. All donations, no matter how small, are greatly appreciated and extremely helpful.
(About Donations)