Background
It was quite the busy day for by-elections on the 26th of September, with a double bill of double bills. Coincidentally, both were effectively one urban, one rural ward within their respective councils. The second to declare, and thus second for my analysis, was Strathallan, giving us the rural portion of Perth and Kinross. Conservative Councillor Crawford Reid had resigned, finding he was no longer able to balance his duties as a councillor with his job in the NHS.
Although the Conservatives had a relatively strong win in 2022, I wasn’t entirely convinced they’d find it as easy this time around. I did think they were the favourites, but I pegged this as a “Lean” rather than “Likely” in my preview. One thing not captured in that piece, but which I stated in my polling day tweet, was that I’d heard from sources on the ground that the Lib Dems were really going for this one.
Headline Results
Councillors and Key Stats
1 Councillor Elected:
🟠Lib Dem: Alan Watt
Change vs 2022 (notional): Lib Dem Gain from Conservative
Change vs vacating: Lib Dem Gain from Conservative
Turnout: 34.0% (-17.5)
Electorate: 9657
Valid: 3258 (99.3%)
Spoiled: 22 (0.7%)
Quota: 1630
2 Continuing Councillors:
🔵Conservative: Keith Allan
🟡SNP: Steven Carr
Candidates
🔴Labour: Ken McCracken
🔵Conservative: Amanda Runciman
🟡SNP: Catherine Scott
🟢Green: Nettie Sutherland
🟣Reform UK: Ian Thomas
🟠Lib Dem: Alan Watt
First Preferences
First Preference History
Turns out the Lib Dems had indeed been giving this one welly, because although the Conservatives maintained a narrow lead in first preferences, the Lib Dems surged into a close second with a near-trebling of their vote. With the addition of Labour and Reform UK, everyone else who’d been on the 2022 ballot suffered enormously from the combination of a surge and greater choice. For both the Conservatives and SNP this is their worst result yet in the ward, and for the latter (who halved their vote) the first time they’ve fallen into third place.
Returning after their 2022 absence Labour end up with their second best share, whilst the Reform UK result is solid but a good bit lower than it was in the City by-election. That leaves the Greens in last place with their vote halving as well. The fact this is the only one of the four by-elections they lost share may suggest this was very much to do with the strength of the Lib Dems and return of Labour rather than anything specific. Still, given the Greens placed third in the list vote across Perth and Kinross in 2021, they might wish they were doing better here.
Even allowing for a significant difference between Westminster and Local elections, it’s also worth emphasising that Labour’s share here emphasises a point I made in the preview. Yes, Labour may have won the “Stirling and Strathallan” constituency in July, but they did not win the Strathallan portion of it, a reminder of how variable results are under the constituency-level.
Transfers
Two-Candidate Preferred
Given how narrow the Conservative lead was and the spread of votes amongst the remaining parties, it was all but guaranteed that transfers would hand this to the Lib Dems. Sure enough it was at the point of Labour’s elimination that the Lib Dems pulled ahead, with the SNP’s transfers turning a gap into a chasm. At the final head-to-head stage, the Lib Dems ended up a very comfortably 9.3% clear of the Conservatives.
For the two-candidate preferred measure, change numbers are versus a re-calculation of the 2022 election to be between these two parties, rather than the Conservative vs SNP head-to-head it actually would have been. The Conservatives would have easily won vs the Lib Dems in 2022, by a margin of about 18%, so this is a huge swing against them.
Detailed Results
Results by Polling District
If we look at where each party had their best result, for the Conservatives and the SNP that was Dunning, the only non-Auchterarder district to have enough voters to be identifiable on its own from the data. The Lib Dems then had a big lead in the districts covering Aberuthven and Blackford – whether that was a truly astronomical lead in one and placing second in the other, I’ve no idea, such is the data merging game. Labour and the Greens both peaked in the combined Braco and Greenloaning pile, leaving Muthill and Gleneagles for Reform UK.
For Auchterarder itself, the Lib Dems had a lead on in-person votes and it’s my apportionment of postal votes that tipped it narrowly Conservative. As postal apportionment is a (reasonable) estimate, it is possible that the Lib Dems led there too. This isn’t an exact science, but perhaps we need to think about how it could be if the aim is to be genuinely useful; maybe the count process should assign postals to the district they come from?
Second Preferences
Turning to transfers, as is so often the case the SNP and Greens have the strongest mutual flow. Interestingly, the relative size is somewhat reversed from 2022: then, SNP voters had been more likely to transfer Green than vice versa. This time around, Greens are more likely to go SNP than in return. Alongside a 10% decrease in transfers to the Lib Dems, that may confirm my suggestion earlier that the Greens lost some direct support to Labour due to them not having been around last time; you might expect a “well I’m normally Labour, but I’ll go Green without them” in 2022 to then have the Lib Dems marked second rather than SNP. With them going to their natural party here, that’d leave the remaining Green voter base a bit more SNP-friendly than in 2022.
Despite going head-to-head for the seat, the Conservatives and Lib Dems were actually one another’s most popular second choice, though with a notably lopsided flow. Labour were also most likely to prefer the Lib Dems, whilst Reform UK voters were by far most likely to go for Conservatives second.
It’s worth at this point talking about the pathetic temper tantrum some have been throwing over the Lib Dems winning this by-election. Unfortunately, UK politics is completely First Past the Post poisoned, which means some people think it’s terribly unfair that a party that couldn’t even muster the support of a third of voters in this election wasn’t just handed the win on that basis. If you consider that 62% of the vote here was for parties to the left of the Conservatives, it should instead be absurd to imagine we should ever use a voting system that would say the other 38% gets to win just because it had the single most popular candidate.
For one thing, people are making the mistake of thinking that if this was an FPTP election the Conservatives would have won. If it had been FPTP, voter behaviour may have been different. We know that tactical voting is absolutely rife under FPTP since only one person can win. Given Labour and the Greens were absolute no-hopers here, it’s entirely possible enough of their voters would gone Lib Dem under FPTP that they would have won even under that system.
Beyond simply the fact people will vote differently under FPTP for fear of vote splitting, the risk of that happening at all discourages no-hope parties from standing in the first place. STV largely removes that risk. Labour didn’t stand here in 2022 (likely due to candidate identification issues, not a vote splitting thing), and if they had again failed to do so, well we know nigh-on exactly what would have happened instead thanks to this chart: their voters overwhelmingly favoured the Lib Dems over the Conservatives. Labour’s recorded second preferences becoming first preferences would have given the Lib Dems a 27 vote lead.
I may (as I explained in the Perth City North analysis, after it went all-SNP) have qualms about using by-elections to fill vacancies under STV, but if we are doing that, thank goodness we use STV and not FPTP. This is a much fairer voting system that more accurately reflects what voters really want, without requiring them to compromise or do what is fundamentally a dishonest vote for fear of splitting the vote and letting the “bad guys” in, whoever they may view them as being.
We may only have three months left of 2024, but I’ve still got an absolutely staggering 22 further by-elections on my upcoming list. The first of these are coming up just this coming week on the 3rd of October in the form of another double bill. This one is in Dundee, with what should be an easy Labour win in Lochee and a harder fought one in Strathmartine that’ll make the difference between the SNP continuing their majority administration or slipping into minority, and thus risking losing control.
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)
