Four weeks out from the General Election which is approaching terrifyingly fast a popular topic is, as always, tactical voting. One of the many flaws of First Past the Post is that with only a single seat up for grabs, most individual votes would have as much effect being dropped into a shredder as they do a ballot box. Voters are well aware of this reality and many will seek to maximise the possible impact of their vote by voting for some they think is more likely to win their constituency than who they would otherwise naturally support. Hence, tactical voting.
Whilst in England and Wales a lot of the discussion on this front very much echoes 1997, as voters keen to remove the Conservatives from office vote for whichever of Labour or the Lib Dems is locally best placed, Scotland has a very different dynamic. In addition to an undoubted degree of dissatisfaction with the Conservatives here, we also have the constitutional dimension. Tactical sorting has already happened to a significant degree over the past two elections, with pretty much every seat in Scotland having become the SNP versus one of the Pro-Union parties – absolutely nowhere has two Pro-Union parties meaningfully competing for the seat.
Yet tactical voting is an often very poorly understood phenomenon, as is voter behaviour in general. Many people, including politicians and commentators who should know a little better, imagine voters to form highly rational, regimented blocs, moving unanimously from one party to the next most closely aligned. That simply isn’t the case, and you only need to look at how voters in Scotland use their later preferences at local elections (as I did in depth here) to see that they can be very messy, and often downright incoherent.
With a significant defeat for the SNP seeming likely for the first time in over a decade, there has been a lot of speculation around how tactical voting amongst the Pro-Union parties might play out. The overly simple assumption often made is that all Pro-Union voters naturally favour other Pro-Union parties over the SNP or other Pro-Independence parties. In reality, not only are some voters more than willing to cross the constitutional divide (in both directions), the flow of votes between the parties is decidedly uneven, and in a way that may mean trouble for the Conservatives.
Method
To get a sense of how tactical voting might be impacting the Pro-Union parties, I’ve gone back through the 2022 data for the wards making up key constituencies. For each ward, I first eliminated all of the candidates bar those from the SNP and the three Pro-Union parties to get a baseline vote. I then set each constituency as a head-to-head between the SNP and the most likely Pro-Union winner to see where the transfers went. Emphasis here is that this is mapping 2022 local election data to Westminster constituencies: it is not 2019 data!
An imperfection of this process is Labour and the Lib Dems end up with some Green preferences sitting with them that may then have went SNP ahead of another pro-Union party, effectively returning to their own camp. Quite a lot of those probably wouldn’t really be “Pro-Union voters”! It’d also be a pain to do this for absolutely every seat in the country, so instead I’ve focused on the new versions of the seats the non-SNP parties won in 2017 and 2019. Labour have a lot more seats in play than just these of course, but we can pretty safely assume that seats across the Central Belt will show similar patterns.
I’ve also broken down the transfers in three ways. “Transfer gain by % of total available transfers” tells us the proportion of votes sitting with the other two Pro-Union parties that went to the third, but there’s obviously a big difference between getting 50% of 6,000 votes versus of 600. The “transfer gain by % of total votes” effectively tells us how many actual votes moved over. The final “net transfer gain vs SNP” is probably the most useful measure as it tells us how much the gap widens (or narrows) by after transfers.
It’s important to remember that there is a difference between a second preference and a tactical vote. Not everyone who transferred their vote in a particular way at the local elections would be inclined to do a tactical vote at Westminster. Likewise, many of the people who did tactically vote in 2019 wouldn’t have felt the need to do so under STV in 2022. This isn’t intended to suggest the number of votes actually up for grabs in each constituency, but instead a likely reasonable possible depth for the tactical pool.
Labour
Transfer Gain by % of Total Available Transfers
Transfer Gain by % of Total Votes
Net Transfer Gain vs SNP
Starting with the expected big winners of the election, Labour were a very popular transfer choice for Pro-Union voters. In fact, in only one of the seven seats the party won in 2017 are they beaten by “Didn’t Transfer”, and even then only barely. Overall, just over half of the vote sitting with other Pro-Union parties had Labour over the SNP when you force the count to a choice between the two, compared to just over 7% going to the SNP; a ratio of about 7:1.
Some seats like Coatbridge and Bellshill or Glasgow North East have a sub-10% net gain in votes in 2022 terms which may look modest. That simply reflects the very limited support for the Conservatives and Lib Dems in those seats rather than anything else, and could still make the difference in a tight race.
Lib Dem
Transfer Gain by % of Total Available Transfers
Transfer Gain by % of Total Votes
Net Transfer Gain vs SNP
No surprises at all that the Lib Dems too fare very well in terms of picking up transfers. In theory they are the middle ground party between Labour and the Conservatives, and so should be pretty attractive to both of those voter pools. They win significantly more than half of the transfers they have available to them across their marginals overall. That said, there is a notably higher rate of transfers to the SNP here than there was for Labour contests, which perhaps speaks to a lingering distrust of the Lib Dems left over from the Coalition era. It’s still quite a small proportion overall, the Lib Dems coming out with a roughly 5:1 advantage.
The sheer size of the Mid Dunbartonshire bar likely reflects the fact that a lot of Westminster Lib Dem voters in that seat were already even more tactical than average and thus reverted to their “normal” party preference at the locals, rather than there being an abnormally large pile of brand new tactical votes they’ll be able to draw upon at the election. They will however still be able to get quite a few and don’t need many to take the seat.
Conservative
Transfer Gain by % of Total Available Transfers
Transfer Gain by % of Total Votes
Net Transfer Gain vs SNP
Straight away, we can see a huge difference in the character of Conservative v SNP head-to-heads compared to Labour and the Lib Dems. For those parties, the SNP’s share had been so small I didn’t even manually label it on the charts as it wouldn’t display properly. Here, the most popular option in every case was not to transfer at all. Of those that did, not only do the SNP pick up a much greater proportion of transfers than against the other two parties, but at the top of the table here you can see some constituencies where they actually get more of those transfers than the Conservatives.
It’s worth bearing in mind that these may have been influenced in part by particular local context in 2022. In all of Dumfries and Galloway, East Renfrewshire, South Ayrshire and Stirling, the SNP and Labour had spent the previous five years in coalition running those councils. In addition, these are the seats that spent a significant period of time in Labour hands historically and still maintain some Labour voter base, which nationally in 2022 was the most likely to transfer to the SNP out of the three Pro-Union parties. Everywhere else has never really been in contention for Labour, bar Aberdeen South where the local context had instead been five years of Conservative-Labour coalition.
The one other interesting standout is Berwickshire, Roxburgh and Selkirk. That is however missing a total of three wards from 2022; two because the SNP didn’t stand at all, and a third because only the Conservatives, SNP and Greens stood so there were no other Pro-Union parties to transfer from. Given these include some of the strongest Conservative wards in the Borders, I’m inclined to believe that in reality this seat would tilt net-Conservative transfer wise. The flipside is the Aberdeenshire North and Moray East seat, in the news today for some really grubby political nonsense, is missing a highly SNP-favourable ward that went uncontested, which may have helped move the dial back towards the SNP.
Scottish Conservatives vs. the World
Given that I’d compiled comprehensive, national second preference data from the 2022 elections, I already knew for sure the idea of a unified Pro-Union voter bloc was an exaggerated myth. Voters are, always have been, and always will be a lot more complex, strange and sometimes, yes, silly than terminally online hyper-partisans assume. What framing it this way shows however is that the big divide in Scottish politics isn’t necessarily Independence versus the Union, but instead the Conservatives versus Everyone Else.
Whereas Conservative voters in 2022 did overwhelmingly fall in behind their constitutional fellow travellers, Labour and Lib Dem voters were far more hesitant to return the favour. Although on balance the Conservatives do pick up more transfers than the SNP, it’s not by a huge margin, as many people are happier to cross the constitutional chasm than they are to vote Conservative. Simply put, the Conservatives remain a deeply unpopular party in Scotland.
Their revival in 2016 was undeniably remarkable, but it capped out at not even 30% of votes the following year as the party seemingly hit the limit of willing support. Though they remain third overall in party polling, a vote for a party is not a vote against all others, and on balance they are the most disliked party. Recent party favourability polling by Ipsos actually has them come out as the most net-unfavourable party, even ahead of Reform UK and Alba.
Party Favourability (Ipsos, March 2024)
Net Party Favourability (Ipsos, March 2024)
The Conservatives have almost certainly felt the consequences of this already, falling short of gaining seats in 2017 and losing some in 2019 they might have won were they as popular to other Pro-Union voters as the other two parties. It’ll have cost them plenty of councillors in 2022. It may also be a factor if they lose some of their remaining seats this year, or fail to gain seemingly easy pickups from the SNP. Beyond the impact on election results, moving away from Westminster it may also impact government formation in the longer term.
Even at the peak of Davidson Mania I never thought it likely the Conservatives could lead a Scottish Government. That would have required active support from Labour, something it was hard to imagine them giving, vocally opposed as they were to both the SNP and Conservatives. Even with Labour back in the lead or at least the largest party in a Pro-Union majority and potentially taking power at Holyrood come 2026, if they are left reliant on Conservative support to govern, that may quickly prove more difficult to navigate than they’ve found it at local council level.
It is important to keep in mind though that back in May 2022, the SNP remained very popular, these being their best results at local elections so far. By contrast the Conservatives were already on the slide, as simmering tensions with Boris Johnson approached their July boil-over that would lead to the short but disastrous tenure of Liz Truss. In this year’s election, the SNP are in a lot more difficulty, and so people who might have preferenced them ahead of the Conservatives in 2022 may be less likely to do that now, never mind go so far as to tactically vote for them. “A pox on both your houses” may be how voters in SNP-Conservative marginals are currently feeling, and the winner determined by who prompts the least distaste overall.
What About the Pro-Independence Parties?
The overwhelming thrust of this piece has been about the three major Pro-Union parties, but obviously tactical voting will be at play for the Pro-Independence bloc as well. There’s a lot less to write about on that front though because it’s of a fundamentally different character. Whereas on the Pro-Union side three separate parties have cultivated strength in separate seats, the Pro-Independence side is completely dominated by the SNP everywhere. It is worth a quick look at this side of the equation though for completeness.
Greens
Although the Greens have a significant vote at Holyrood – and one that contrary to popular misbelief only has a limited tactical element itself – and other proportional elections, that has never been replicated for Westminster. Even if it had, it’s not concentrated enough to put them in contention for an MP. The same is largely true for the Green Party of England and Wales, who perform much better at local elections than for Westminster, precisely because of the tactical voting squeeze on smaller parties.
With the Greens claiming they will contest more constituencies than ever before however, people may be looking to see whether their presence negatively impacts the SNP. The answer to that is simply yes, it will, but not by nearly as much as people think from headline figures. It has long been the tendency for people to assume a given party’s voters will all transfer en bloc to one of their counterparts, but as demonstrated at length for the Pro-Union parties above, that’s simply not the case. Voters go in all kinds of weird and wonderful directions. The same thing is true for the Greens, as we can see from their transfer patterns in 2022:
Effectively, in 2022 a ward that had both a Green and SNP candidate generally saw between 40-60% of Green voters giving the SNP their second preference. Nearly 20% overall opted for Labour, and around 12% the Lib Dems. In other words, for every 1% a Green candidate wins at Westminster, it’s unlikely that much more than 0.6% of it will be at the SNP’s expense. It could even be less than that, as the voters sticking with the Greens regardless may therefore be those least likely to vote for the SNP anyway, especially following their ejection from the government.
That absolutely could make the difference in very tight constituencies, but in other cases people will exaggerate the impact. For example, when Conservative MP David Mundell held his seat in 2015 by 798 votes when the Greens won 839, a lot of SNP supporters were furious. Yet to overturn that majority would have required 95% of Green voters to go to the SNP in their absence, a completely unrealistic expectation.
Alba
Alba may consider themselves to be defending two MPs headed into this election, but neither of those are genuine. Mid-term defections absolutely do not count the same way that elected MPs do, and despite defections Alba are still yet to win a single seat at any level of election. There is precisely zero chance of that changing on the 4th of July. Like the Greens however, you would expect their presence to drain support largely from the SNP, so even if they continue to perform poorly, they could still have an impact. Again, if we look at their 2022 transfer patterns overall:
Notably, Alba voters were less likely to second preference the SNP than Green voters were. The dissatisfaction Alba’s limited pool of supporters feel with the SNP may have set some of them so firmly against the party they didn’t want to transfer, and oddly enough they were the most likely of all Pro-Independence voters to transfer to the Conservatives, albeit still quite limited. Similar to the Greens then, whilst we can expect any Alba presence to most significantly impact the SNP, it won’t be as simple as seeing their candidate get 500 votes and declaring all of that would have been behind the SNP otherwise.
Overall then, the biggest impacts of tactical voting in Scotland are going to be felt amongst the Pro-Union parties. Rather than firm constitutional comrades however, Conservative voters’ fondness for Labour and the Lib Dems is largely unrequited. The party may frame themselves as the only option for Pro-Union voters in key seats, but they can’t escape the reality that this election is fundamentally a verdict on their government at UK level for the past 14 years. If Scotland is indeed moving on from intense constitutional polarisation and looking more at the bread-and-butter issues, that is unlikely to be the benefit of a highly unpopular Conservative party.
And all of this is to say nothing of the potential impact of any Reform UK candidates, with the party claiming it will stand in every seat in Scotland. Right-of-Conservative options have consistently been far less popular in Scotland than down south, but just like the Greens and the SNP, it could make a difference in very tight seats. Even if their Scottish result is better in relative terms than for their English counterparts, the party simply cannot expect a flood of goodwill from the Labour and Liberal voter bases.
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)
