One thing is clear: support for the SNP and for independence is now consistently lower in the polls than it was at the beginning of the year.

In the last half dozen polls support for Yes (after leaving aside Don’t Knows) has averaged 49%, with No narrowly ahead on 51%. This represents a five-point fall in support for independence as compared with the position in the last half dozen polls conducted before the end of last year. It is the first time that No have been ahead in a running average of the polls since this time last year.

Meanwhile, the equivalent comparison suggests that support for the SNP now stands at 49% on the Scottish Parliament constituency vote compared with 54% at the turn of the year, while on the list vote support for the party has dropped from 44% to 42%.

It is not surprising that support for the SNP has fallen alongside support for Yes. The most recent Opinium, Panelbase and Savanta ComRes polls confirm the evidence of previous polls that support for one is now close to being synonymous with backing for the other.

According to Opinium 86% of current Yes supporters but just 6% of No supporters propose to vote SNP on the constituency ballot in May. At 86% and 7% respectively the figures in Panelbase’s most recent poll are almost identical, while Savanta ComRes’ most recent poll puts the figures are 89% and 6% respectively.

It therefore makes sense to consider the two falls in tandem. But how might we best account for them?

One place to start is to look at those phenomena that have previously been established as sources of increased support for independence.

The first of these is Brexit. There is little sign here of any substantial change of opinion. Four polls this year have on average put support for joining or remaining in the EU at 64% (after leaving aside Don’t Knows) similar to the 62% support registered for Remain in the 2016 EU referendum.

Meanwhile there is little sign that Brexit is having less impact on voters’ attitudes towards independence. In polls conducted since the beginning of February, 55% of those who voted Remain have said they would vote Yes to independence, compared with 33% of those who supported No. That 22 point difference is little different from the 25 point gap that was in evidence in the polls a year (and more) ago.

The second is perceptions of the handling of the pandemic, where the much higher levels of approval for the Scottish Government’s handling of the pandemic than for the UK government’s role has previously appeared to persuade some people of the case for independence.

Recent months have seen the introduction of a vaccine programme, where the vaccine itself has been procured by the UK government while the roll-out north of the border has been the responsibility of the Scottish Government.

The vaccination programme is very widely applauded. According to YouGov, 82% believe it is being rolled out well and just 11% badly.

However, it is far from clear that the UK government (and by implication, perhaps, the Union) are getting the credit for this success.

Ipsos MORI found that people were just as likely to think the Scottish Government (78%) are doing a good job at ensuring the public are vaccinated in a timely fashion as they were the UK government (75%). Meanwhile YouGov found that 43% believe that ‘the way that the vaccine is being rolled out in Scotland’ is the result of decisions made by the Scottish Government, while only 17% believe it is the result of decisions taken by the UK government. Even those who voted No in 2014 are more likely to credit the decisions of the Scottish Government (37%) than those of the UK government (25%).

That said, there has been some improvement in evaluations of how well the UK government and the Prime Minister are thought to have been handling the pandemic. YouGov report that 30% now feel that the UK government is handling the pandemic well, up from 21% in November. Over the same period Panelbase report a rise from 19% to 26% in the proportion who believe Boris Johnson is performing well.

Even so, this still means the Prime Minister and the UK government are badly trailing their devolved counterparts. Panelbase estimate that 62% reckon that Nicola Sturgeon is responding well to the pandemic, down just four points on November, while YouGov report that as many as 67% believe the Scottish Government is handling the pandemic well, down just a single point.

Nevertheless, there are some signs that the pandemic is no longer as much of a recruiting sergeant for independence as it was. Back in August last year 20% of those who voted No in 2014 told YouGov that Scotland would have handled the pandemic better as an independent country, while only 28% believed it would have done so worse. Now the figures are 18% and 42% respectively. At the same time, there has been an increase – from 4% to 9% – in the proportion of 2014 Yes voters who feel that an independent Scotland would have handled the pandemic worse.

Meanwhile, although the company has no previous measure with which it can be compared, Opinium’s poll this week paints a similar picture of where the public stand now on this issue. They report that just 13% of 2014 No voters now feel that an independent Scotland would have handled the pandemic better, although they still outnumber the 8% of Yes voters who believe it would have been handled worse.

However, the explanation for the fall in SNP support necessarily does not necessarily lie in the same place as the sources of the increase. The last few weeks have seen the theatre and associated media coverage of lengthy appearances by Alex Salmond and Nicola Sturgeon at the parliamentary inquiry into the Scottish Government’s handling of the harassment complaints against the former First Minister. Perhaps this serious row between the two leading characters of the nationalist movement has undermined support for independence and for the SNP?

There seems to be little doubt that voters – and especially SNP voters – have reacted more favourably to Ms Sturgeon’s side of the story than they have to Mr Salmond’s. Back in January 30% of voters told YouGov that Ms Sturgeon has generally been telling the truth about the Salmond inquiry, while 36% reckoned she was not and one in three said they did not know. After her appearance before the inquiry the proportion saying she had been telling the truth increased to 44%, while the proportion saying they did not know fell to one in five. Meanwhile the swing in her favour (from 49% saying in January she was telling the truth to 72% now) was even bigger among those who had voted SNP in December 2019.

Mr Salmond’s appearance, in contrast, did little to advance his case in the eyes of voters. Although the proportion who felt he was generally telling the truth rose from 13% (in January) to 20% (after his appearance), those saying he was not also increased, from 50% to 53%.  Among those who backed the SNP in 2019, those thinking the former First Minister was not telling the truth rose from 42% to 61%, leaving just 12% who felt he was doing so.

A similar pattern among SNP voters at least emerges from a question on the impact of the inquiry that Savanta ComRes first asked in December and have since repeated on their most recent poll. This asks voters whether what they have heard about the inquiry has affected their trust in Ms Sturgeon and Mr Salmond. In December, 24% of those who voted for the SNP in 2019 said that they now trusted Ms Sturgeon more, while 16% indicated that they now did so less – with most (54%) saying it had not made much difference either way. However, in the company’s recent poll, 38% now said that they trust her more (a 14 point increase), while 19% (up three points) indicate they do so less.

In contrast, the proportion of 2019 SNP voters who say they now trust Mr Salmond less has increased from 47% to 64%, while only 12% (down two points) say that they trust him more.

At the same time, when Savanta ComRes asked voters directly in their most recent poll whose account of the events being investigated by the inquiry they believed, 52% of all voters – and 77% of 2019 SNP supporters – said Ms Sturgeon’s, while just 26% of all voters – and only 14% of 2019 SNP supporters – said Mr Salmond’s. At the same time,. Opinium found that 61% of SNP supporters feel that Ms Sturgeon was telling the truth but not Mr Salmond, while only 6% expressed the opposite view.

However, while Ms Sturgeon may have largely outpointed Mr Salmond in the court of public opinion, that does not necessarily mean that the events of recent weeks have not undermined support for the SNP and for independence among some previous supporters. After all, the loss of just a relatively small minority of these voters could be enough to occasion a notable drop in support.

Many a polling question suggests that there is a minority of former SNP and Yes supporters whose willingness to vote that way again may be at risk of being undermined by the Salmond inquiry.  For example, according to YouGov, 13% of 2019 SNP voters still do not think Ms Sturgeon has been telling the truth, according to Panelbase 17% believe she has not been ‘entirely honest’, while Opinium’s data suggest that 18% doubt her word. Meanwhile, Savanta ComRes find that 23% of former SNP supporters now say they trust the SNP less, while 21% of 2014 Yes voters state that the Salmond inquiry has made then less likely to vote for independence. Might these perceptions dissuade some of these voters not to support the party in May’s Holyrood contest?

In the meantime, one approach to addressing what impact the inquiry may have had on voting intentions so far is to see whether support for the SNP and/or independence fell in the three weeks or so leading up to the inquiry appearances of Mr Salmond and Ms Sturgeon, that is after the inquiry began to get substantial media attention.

In truth, it is far from clear how much support did fall over this period.

Savanta ComRes polled at the very beginning of this period in early February, and again twice shortly after Ms Sturgeon’s appearance. Using their estimates that do not take account of reported probability of voting (which had a particularly marked impact in the February poll), support for Yes stood at 49% in early February, and then at 48% and 49% after the inquiry.

True, Panelbase did register a two-point fall in support for independence (from 52% to 50%) over a somewhat longer time period (between the third week of January and immediately after Ms Sturgeon’s appearance). However, between them the two companies that polled not long before the Salmond inquiry hit the headlines and then after the testimony of the two leaders do not point to the fallout from the inquiry having had a substantial impact on support for independence.

The same, though, may not be so true of support for the SNP. Savanta ComRes registered a three-point drop (from 51% to 48% after excluding their turnout weighting) in support for the party on the constituency vote during the four weeks to early March, while Panelbase registered a five-point fall since January. It may be that the inquiry has had rather more impact on support for the SNP than it has on that for Yes.

In any event, what we do have to bear in mind is that support for Yes appears to have been falling before the Salmond inquiry was hitting the headlines. Between October/November and January/early February support had already dropped on average by four points according to polls conducted by Savanta ComRes, Panelbase and Survation. In short, much of the fall in support in independence might well have predated the emergence of the Salmond inquiry into the media spotlight.

So what else might have been happening to account for this fall? One possibility we should not discount is that whereas for much of last year much political argument was stifled by the focus on the pandemic, now that the election is approaching politics at Holyrood has become more partisan, as politicians of all stripes become more concerned to appeal to their potential voters than dealing with the minutiae of the legislative process.

As a result, perhaps some of the voters who last year were saying they might vote for the SNP or for Yes maybe for the first time in their lives – are now returning to their more familiar political loyalties.

If that has been happening then what we might see in the polls is that support for the SNP and for Yes has declined more sharply among those who did not vote for the SNP in 2019 and among those who did not vote Yes in 2014 than it has done among those who have previously voted that way.

Such a pattern is in evidence in some of the recent polls. For example, in their most recent poll Panelbase show only a marginal decline since January (from 95% to 92%) in the proportion of 2019 SNP voters who now say they would vote for the SNP on the constituency vote, whereas the proportion of those who voted for other parties in 2019 who now say they will vote for the party in May has nearly halved from 12% to 7%. Meanwhile, YouGov’s polling suggests that while support for independence has held up since last November among those who voted Yes in 2014, support among those who voted No seven years ago has fallen from 22% to 17%. In short, some of the decline in SNP and Yes support may simply be a consequence of voters returning to previous loyalties.

The original cause of the rise in support for independence – Brexit – is still with us. However, the pandemic may no longer be bolstering support as much as it was, the Salmond inquiry may have done some damage to SNP support, while a more partisan environment may have served to undermine support for both independence and the SNP. But perhaps one of the key unknowns now is whether any issues other than these come also start to matter to voters between now and polling day.

About the author

John Curtice is Professor of Politics at Strathclyde University, Senior Research Fellow at ScotCen and at 'UK in a Changing Europe', and Chief Commentator on the What Scotland Thinks website.