GE24: Mayhem in the Marginals

Having covered off the headline results already, and then the detailed spread of party votes and swings across the country, what do you get at the intersection of those two? Everyone’s favourite thing for when it comes to talking about future elections – marginals!

As usual, before the election I went through the existing marginals in my Ballot Box Battlegrounds series to give a flavour of what might be the most hotly contested seats in the country. At the time, I acknowledged that most of the seats that’d change hands would likely be outside that pool given where polling was going, but also that I couldn’t very well write “this is a Central Belt seat so Labour will probably win it” a couple of dozen times over and keep that even vaguely engaging!

Now that the dust has settled and we know the shape of things in every seat, what does that mean in terms of marginals? What seats are still on the list, what seats have come off it? Which of those many big Central Belt wins Labour racked up are most vulnerable to being unpicked again at the next election – or where is the room for further growth?

Scotland's Shifting Marginals

Before the election, Scotland was home to 17 seats that would traditionally be considered marginal, i.e. where the winning majority was below 10%. Most of these were between the SNP and the Conservatives, reflecting just how badly Labour had done back in 2019, when they slumped to their worst ever Westminster result in Scotland. 

Unlike the last election, which significantly reduced the marginality of seats across the country, this election has largely just reprofiled what seats are marginal. A total of 19 seats fall below that 10% marker, split almost evenly between 9 in the extremely marginal 5% or below band, with the remainder between 5-10%. Of these, just 7 of the pre-election battlegrounds remained marginal, with the rest going solidly for one party or another. There’s a complete list of marginals (and their margins) below, with those new to the list in bold.

Despite that reprofiling, it’s still effectively the case that the biggest number of marginals are between the SNP and Conservatives. As the map above shows, much of the Central Belt has some of the safer green tint, and most of the areas that don’t are still sat in the 10-15% majority band I have in grey. By and large it is rural and northern Scotland that was unsettled at this election, with much clearer mandates handed to MPs in Labour’s old heartlands – though, following on from the points made in yesterday’s piece about vote shares, those aren’t necessarily as big as they were in the party’s past.

Complete list of marginals
  1. Dundee Central, SNP over Lab 1.7% (wasn’t marginal, SNP over Lab 32.6%)
  2. Arbroath and Broughty Ferry, SNP over Lab 1.9% (wasn’t marginal, SNP over Con 20.7%)
  3. Gordon and Buchan, Con over SNP 2.0% (was 12th, Con over SNP 6.8%)
  4. Dumfries and Galloway, Con over SNP 2.0% / over Lab 3.8% (was 3rd, Con over SNP 3.0%)
  5. Moray West, Nairn and Strathspey, SNP over Con 2.2% (was 8th, SNP over Con 5.2%)
  6. Aberdeenshire North and Moray East, SNP over Con 2.5% (was 9th, Con over SNP 5.2%)
  7. Stirling and Strathallan, Lab over SNP 2.8% (wasn’t marginal, SNP over Con 14.8%)
  8. Aberdeen North, SNP over Lab 4.2% (wasn’t marginal, SNP over Con 28.3%)
  9. Inverness, Skye and West Ross-shire, LD over SNP 4.5% (wasn’t marginal, SNP over Con 25.3%)
  10. West Aberdeenshire and Kincardine, Con over SNP 7.0% (was 2nd, Con over SNP 1.6%)
  11. Livingston, Lab over SNP 7.9% (wasn’t marginal, SNP over Lab 25.2%)
  12. Aberdeen South, SNP over Lab 8.1% / over Con 8.4% (wasn’t marginal, SNP over Con 10.9%)
  13. Edinburgh East and Musselburgh, Lab over SNP 8.2% (wasn’t marginal, SNP over Lab 22.6%)
  14. Glenrothes and Mid Fife, Lab over SNP 8.2% (wasn’t marginal, SNP over Lab 22.1%)
  15. Perth and Kinross-shire, SNP over Con 8.2% (was 5th, SNP over Con 4.2%)
  16. North Ayrshire and Arran, Lab over SNP 8.4% (wasn’t marginal, SNP over Con 17.7%)
  17. Glasgow South West, Lab over SNP 9.2% (wasn’t marginal, SNP over Lab 12,9%)
  18. Dumfriesshire, Clydesdale and Tweeddale, Con over SNP 9.6% (was 14th, Con over SNP 8.5%)
  19. Glasgow South, Lab over SNP 9.8% (wasn’t marginal, SNP over Lab 18.8%)

Of course, marginality has meant next to nothing in Scotland for the past few general elections. Being a distant second, third or even fourth place in most seats in 2010 didn’t stop the SNP winning all bar three in 2015. Having swept all opposition aside that year didn’t stop them from then losing two-fifths of those seats in 2017 and finding their hold on almost all of the rest very shaky indeed. They then set themselves up with what would classically be seen as huge leads again in 2019.

This election was very much same genre, different song on that front. The most dramatic example is that what was notionally the safest seat on the new boundaries is now the most marginal in the country. In 2019 terms, Dundee Central was estimated to be a whopping 32.6% lead for the SNP over Labour. That was whittled down to an absolutely miniscule 1.7% this year. Indeed, Dundee is a marginal double-whammy, as the Arbroath and Broughty Ferry seat which contains the rest of the city was the second closest at just 1.9%, having been the 18th safest seat last time.

Dundee wasn’t alone in just how dramatic the SNP’s reversal of fortunes was, as many of their other theoretically very safe seats swung much harder against them than average. Aberdeen North was their number two in 2019 (on the old boundaries it was top), but now it’s the 8th most marginal seat in the country. What was the 8th least marginal last time, Livingston, is 11th closest now. In Glasgow, the West constituency had been their safest seat in the city, and now it’s Labour’s.

One final thing worth noting here is that we once again have a couple of three-way marginals. We’d had some in 2017, but although there were a few seats with close Labour vs Conservative contests for second place, none of those were actually marginal for first in 2019. Last week’s SNP deficit of 2% in Dumfries and Galloway wasn’t all that far ahead of Labour’s 3.8%. Meanwhile in Aberdeen South, Stephen Flynn’s 8.1% over Labour isn’t much more than his 8.4% over the Conservatives. That bodes ill for him if voters coalesce more firmly around one of those next time.

Overall, the Conservatives have ended up as the most theoretically vulnerable party, with 4 of their 5 seats (80%) classed as traditional marginals. The SNP aren’t far behind at all, as they’re on 7 of 9 (77.8%), another throwback to their pre-referendum era when very few of their MPs were elected on significant majorities. Labour also account for 7 marginals, but that’s a much smaller share of their total number of 37 (18.9%). Absolutely no surprises given the way they play the game, but only 1 of the 6 (16.7%) Lib Dems seats is a marginal, having won everything but that late-declaring Inverness seat by blowout margins. 

That's (almost) a wrap...

That brings to an end all my formal analysis of the 2024 UK General election. A big thank you to everyone who has supported Ballot Box Scotland through this election, including the very generous support for the crowdfunder in May that helped fund a poll just before the election (part 1 and part 2 here). I’ve effectively been working flat out since the polls closed on Thursday night until I got back to my actual day job this morning, so my brain is now composed of about 43% election data. 

There are still some bits for me to do this week that aren’t part of the headline analysis. I’ll be piecing together a full PR result for the whole UK using my preferred Scandinavian model, just for a bit of illustrative fun, and will aim to have that up later this week. I’ve also let council by-elections slip with all the focus on Westminster, so will have to catch up on writing a preview for the (at least) one due next week.

In other words, still much to do. If you can afford to, please do think about chipping in a wee donation to support all my work here on BBS. Fortunately I will actually be taking a proper break next week, as I booked a little holiday shortly after the election was called as something to look forward to, though much like when I want over to Northern Ireland last year there’s a very slight element of a busman’s holiday about this one. If nothing else, your donation might cover something to eat in one of Wales’ many fine dining establishments!

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)