SP21 – What Happened to the SNP’s Majority in 2016?

Keep tabs on all the latest polling, articles and information ahead of the 2021 Scottish Parliament election in the Ballot Box Scotland Holyrood Hub!

Polling day is looming, and postal ballots are already beginning to pile up in Council mailrooms across the country. There isn’t much time left for parties to change the shape of the next parliament, and we’ll soon have the answer to one of the big questions of this election – will the SNP regain the majority they lost in 2016? Most polling points in that direction, but a few polls suggest otherwise, and either way it looks to be on a knife edge.

This fixation on whether there would be a single party majority would have been unimaginable in the early days of the Scottish Parliament. With a largely proportional voting system it was widely assumed such majorities were all but impossible, and the first three elections bore that out. That all changed in 2011 when the SNP swept to an unforeseen majority with 69 of 129 seats, triggering a dramatic referendum on Independence and an equally dramatic realignment of longstanding Scottish voting patterns.

Ahead of the 2016 election, old assumptions were turned on their head, and instead the expectation was the SNP would easily hold their majority. Polling was looking very healthy for them, and it seemed likely they’d repeat their Westminster feat of the previous year and win almost every constituency. Instead, with a net loss of 6 seats, the SNP fell just short of a majority. With current polling not a million miles off 2016, it seemed worth looking at exactly what happened last time that led to a minority government.

Was it the Greens' Fault?

Probably the most widely accepted theory on social media is that the relative success of the Scottish Green Party was to blame. As the only other party at Holyrood who had supported Independence, they’d experienced a substantial increase in profile as a result of the referendum, and in 2016 trebled their MSPs from 2 to 6. This theory is rooted in two key arguments – the first around tactical voting, and the second what looks like a direct swing between the two parties based on national results.

Tactical Voting

Starting with the “Greens as a tactical vote party” side of the theory, although voters don’t mark their reasons alongside their votes on their ballot, we can look at other evidence to say that tactical Green voting is highly exaggerated. Just as an initial point I’ve made a few times, those of us who are hooked directly into the Scottish politics social media bubble aren’t reflective of voters overall. Fevered debate around tactical voting in that bubble isn’t really taking place to anywhere near the same degree amongst ordinary voters. It’s also hard to square this theory with the fact the Greens have always had MSPs at Holyrood.

Though it’s never been a huge group, it clearly required a Green-specific base of support to exist prior to 2016. Remember that the “maximise the pro-Independence bloc” motivation didn’t, and mathematically couldn’t, exist before the SNP looked like they would win almost every constituency. The SNP won list seats in every region in 1999, 2003 and 2007, and in all but one region in 2011. If you’d told someone at any of those elections the way to get the largest group of pro-Independence MSPs was to vote for anyone other than the SNP, they’d have rightly looked at you like you’d grown a second head!

Beyond Holyrood, the Greens have had councillors across Scotland since a semi-proportional voting system was brought in for Council elections in 2007 – most notably with sizeable groups in Edinburgh and Glasgow. Although, or perhaps because, it’s the lowest turnout election, the Greens have also had their best results in % terms in European elections. Neither of those things could be true if their support was entirely or even largely tactical and confined to Holyrood.

The Non-Smoking Gun

Okay, that’s all well and good, you might say – but don’t the 2016 results show an obvious shift of regional list votes from the SNP to the Greens? After all, the SNP were down 2.3% and the Greens up 2.2%. You can’t argue with that, surely? You can, because the notion of a direct shift between the two parties falls apart as soon as you look below the national figures.

What looks like an equal-but-opposite swing at national level simply doesn’t match up when you look at the individual regions. The Greens’ biggest gain was in Glasgow where they were up 3.5% – but that’s also where the SNP had most growth, up 4.9%. Similarly, the North East had the weakest Green boost at just 1%, but it saw the heaviest swing against the SNP at 8.1%. There’s close to parity in Lothian, but with all of the other regions as they are, that’s clearly coincidence rather than correlation.

That’s not to say that if some Green voters hadn’t instead plumped for the SNP they wouldn’t have won more seats. In both South and West the increase for the Greens could have instead landed an SNP seat, and those two seats would have just squeaked the majority number of 65. It’s just that, with the evidence of other regions and elections, it’s vanishingly unlikely enough of those votes were tactical votes gone wrong, as the theory would imply.

With all of the above in mind, there’s a very simple answer for why the Green vote went up in 2016 – they genuinely convinced more people to support them as a party in their own right. As a result of the referendum, they’d gone from 1,500 to 9,000 members. That gave them more activists, money and resources than they’d ever had. They had more media attention than ever before. With five years since the last election and Votes at 16 coming into play, there were seven years of young voters added to the electorate, and we see globally that young people are more likely to vote for Green parties. Of course that was going to lead to more votes!

It says rather a lot about the weird, wild world of the Scottish Politics social media bubble that this extremely obvious explanation is largely disregarded for less credible, but more politically convenient, theories. It also says a lot about it that I’m actually bracing myself for “BIASED BOX SCOTLAND!” nonsense for drawing the most logical conclusion possible from data publicly available to everyone…

Anyway, if it wasn’t the Greens’ fault the SNP lost their majority, what really happened?

A Headache in the Heartlands

Given the size of the bars, you probably picked up a big clue from the chart above. The SNP experienced big swings against them in the North East and Highlands & Islands regions. Though it’s a lot smaller, their third worst negative swing being in Mid Scotland & Fife also points to something – that the SNP were losing support in their long established heartlands.

Across these three regions, the SNP lost a total of five seats versus 2011 – two in each of H&I and NE, plus one in M&F. That alone accounts for most of the net loss, as had they held steady they’d have won a total of 68 seats. The beneficiaries weren’t the Greens, but instead the Conservatives.

Losses in Core Areas...

By the time the Scottish Parliament was established in 1999, the SNP had embedded themselves into a few core areas. These formed a bit of an arc up from Perthshire through Angus, stretches of Aberdeenshire, and into Moray. They consistently won UK Parliament seats in these areas, and they also emerged as the winners around the southern Highlands, likely building on Winnie Ewing’s long service as an MEP for that region in the European Parliament. That meant that the party consistently did well in the Highlands & Islands and North East regions, plus the Perthshire bit of Mid Scotland & Fife.

Although they remained the largest party in those core areas in 2016, they lost a huge amount of support across them, amounting to -8.3% overall. At the same time the Conservatives surged, gaining 12.5% in the same areas. As with the Greens, this is unlikely to be an entirely direct transfer between the two parties, especially given the increase in turnout – but it’s also very hard for there not to have been a large number of direct SNP to Conservative shifts, given the level of swing involved.

... Especially Core Constituencies

Of course, the SNP weren’t ever equally strong in every constituency within those areas. We can also look just at the constituencies the party has held since 1999, or their rough successors following the 2011 boundary changes.

It was an even worse story for the SNP across most of these constituencies, with an overall swing of -9.8% versus +13.5% for the Conservatives. Losses were particularly steep in Moray and in Banffshire and Buchan Coast, where Conservative gains were correspondingly highest. The seats of Inverness and Nairn plus Skye, Lochaber and Badenoch actually bucked trend with losses less severe than their region overall, but were still substantially worse than the national figure.

Absolutely none of the above should be a surprise looking back, given that in both of the 2017 elections these are exactly the areas the SNP had their biggest setbacks. In May’s Council Elections, they lost seats against a Conservative advance across Perth & Kinross, Angus, Aberdeenshire and Moray. Then in June’s snap UK election, they famously lost most of their seats there to the Conservatives, and even their 2019 bounceback wasn’t enough to reclaim them all. In that respect, 2016 set the stage for what would happen the following year.

Other Constituency Clumsiness

Though that 2011 SNP majority is often thought to have “broken” AMS, it was actually the least incorrect result according to AMS’ own rules since devolution, and 2016 would be similar. At both elections, the SNP’s successes in the constituency portion of the system only gave them one seat more than they would have won had AMS worked flawlessly. On that basis it is fair to say it was the SNP’s decline in the regional list vote in the core areas above that did most to cost them their majority. However, there was still a path open if they’d won a few more constituencies and induced more overhangs.

There are four constituencies that the SNP won in 2011 but lost in 2016, which if they had won would have given them an overhang, and where there wasn’t a Green on the constituency ballot – Aberdeenshire West, North East Fife, Edinburgh Southern and Edinburgh Western. In addition, they should have found Dumbarton an easy pickup based on national swings, which would similarly have given them a boost. Had they won any two of these five constituencies, they’d have held onto their majority, and all of them would have had them barely down on 68 seats total.


The 2016 election was part of Scotland’s political realignment following the referendum, as the SNP’s centre of political gravity continued to shift away from the North East, Highlands and Perthshire down to the western Central Belt. Severe negative swings in their historic heartlands weren’t entirely counteracted by gains in their new strongholds. Though they gained constituencies overall, they failed to hold or win certain key seats which would have allowed them to distort AMS just enough to retain a majority.

But contrary to social media theorising, those losses map extremely poorly to Green gains. Instead, the Conservative resurgence was felt most strongly in those SNP heartlands. That would bear further (and juicier) fruit for the Conservatives the following year, as they gained large numbers of councillors across those areas, and flipped UK Parliament seats in the snap election. 

Though the SNP continue to ride high in the constituency polling for May’s election, they’ve slumped in the regional list polls. That could yet mean they experience a repeat of 2016 – and, uncomfortable though it may be for them to face, it’s more likely that is because FPTP artificially inflates their constituency vote, rather than AMS deflating their list vote. After 14 years in government the SNP remain unchallenged for the position of Scotland’s dominant party, but the form of proportional representation used has laid down a fine line between outright majority and ongoing minority.

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)