Professor John Curtice on how the outcome of the General Election in Scotland adds to the evidence that Brexit is undermining the foundations of support for the Union.
First published in December 2019.
The outcome of the 2019 UK general election has added a new twist to the debate about Scotland’s constitutional future. Thanks to the Conservatives’ success in winning an overall majority, it is now clear that the UK will leave the European Union. However, in Scotland, the Conservatives fell back while the SNP enjoyed an eight-point increase in its share of the vote. As a result, the stage has seemingly been set for a constitutional clash between a Scottish government that is emboldened in its demand that, following Brexit, Scotland should be able to hold a referendum on independence, and a UK government that is determined that no such ballot should take place.
Almost inevitably, the reasons for the SNP’s advance have themselves become a bone of political contention. Some in the unionist camp have argued that the increase in SNP support was simply generated by the party’s ‘Stop Brexit’ message, and that this meant that much of the increase in the party’s share of the vote did not necessarily come from those wanting independence. In contrast, SNP spokespersons have argued that the election result indicated that there is now a renewed demand from the Scottish electorate that they should have the opportunity to vote on independence. Here we consider what light the polls cast on this debate.
To examine the extent to which, if at all, the SNP may have secured support in particular from those who are opposed to independence, our first table shows how people voted in 2017 and in 2019, broken down by how they voted in the 2014 referendum.
Note that in the absence of any post-election polling of how people in Scotland voted as yet, we are relying on data provided by the last three polls to be conducted during the 2019 campaign by Panelbase, Survation and YouGov. On average, these polls somewhat underestimated the SNP advance and overestimated the strength of the Conservative position – all put the Conservatives on 28%, three points above what they actually achieved, while they estimated support for the SNP to be between 38% and 44% when, in the event, the party won 45%. This means that, in practice, our figures may underestimate whatever change in the pattern of support occurred between 2017 and 2019. However, one advantage of using these three polls is that all three companies conducted a very similar poll shortly before the 2017 election, thereby providing us with a robust basis for making a comparison between the two years.
There is little sign in our table of any substantial change in the relationship between how people voted in the independence referendum and how people voted in the two UK general elections. True, support for the SNP was three points higher among No voters in 2019 than it had been in 2017 – but it was also a couple of points higher among those who voted Yes. This pattern does not constitute clear evidence that the SNP disproportionately advanced among those who are opposed to independence. Meanwhile, we can see that the pattern of support for the Conservatives was also much the same in 2019 as it was in 2017, with the party’s vote coming predominantly from those that voted No.
Our second table undertakes the same analysis, but this time distinguishing between those who voted Remain and those who backed Leave. Here the link with party choice is not as strong as it is in respect of independence. For example, whereas in Table 1 there was as much as a 35-point difference between the proportion of Yes supporters who voted Conservative and the proportion of No supporters that did so, the equivalent gap in respect of Remain and Leave voters was just 23 points. A similar observation applies in respect of the other three parties. Overall, how people voted in the 2019 election was more likely to reflect their views on independence than their opinion of Brexit.
However, apart from Labour’s support, the link between EU referendum vote and party choice is stronger now than it was in 2017. The Conservative vote was nine points higher in 2019 among Leave voters but four points lower among Remain supporters. In contrast, the SNP saw its support increase by seven points among Remain voters (thereby denying Labour a wellspring of support from that quarter on which the party was reliant south of the border) while it dropped by eight points among Leave voters. Remain voters were, of course, nearly twice as numerous as their Leave counterparts, so the increased concentration of its support in the former group was to the SNP’s advantage, while the increased reliance of the Conservatives on the backing of Leave supporters was to their disadvantage.
So, while still less important than independence, Brexit did structure the level of support for each party in Scotland to a greater extent than it did in 2017 – much as was the case across Britain as a whole. To that extent, the unionist claim that the increase in SNP support owed more to a wish to stop Brexit than a greater demand for independence has some merit. However, as we have already seen, it does not follow that this meant the SNP advance was built primarily on the backing of those who are opposed to independence.
At this point, we should also remind ourselves of the observation we have made previously that the polls have consistently registered a higher level of support for independence this year than last, and, moreover, that this increase has occurred entirely among Remain voters. With one exception (from YouGov), this finding was also largely corroborated by the five polls published during the election campaign. These on average put support for independence on 48%, a point down on previous polls this year but still three points up on the figure in the second half of last year. Meanwhile, at 55% support among Remain voters for leaving the UK was five points up on the polls from last year, while (at 30%) it was four points down among Leave voters.
Rather than being irrelevant to the argument about Scotland’s constitutional status, therefore, the fact that the SNP’s vote increased primarily among those who voted Remain is better regarded as further evidence that Brexit appears to be eroding the foundations of support for the Union, albeit not yet to the extent that a majority are in favour of independence – or indeed of holding another independence referendum within the next year as the First Minister is proposing.
What, however, will now be crucial is whether or not any evidence emerges, as Brexit becomes a reality in the coming weeks and months, that support for independence has increased further – such that the polls begin consistently to report a majority in favour. Any such development – which Panelbase’s final campaign poll suggested might become a reality – would add yet further fuel to the now rekindled flames of Scotland’s constitutional debate. But equally, if that does not happen, the SNP’s renewed ardour for Indyref2 might begin to cool as rapidly as it has arisen.🔷
Share this article now: