The ongoing review of Westminster constituencies isn’t the only boundary change excitement in Scotland right now. Though Ayrshire has no proposed changes in that batch, in May’s local elections voters in North Ayrshire will find themselves in the unique circumstances of electing councillors to the third set of wards in as many elections.
Although the Islands areas also have boundary changes versus 2017, they didn’t have any that year due to the upcoming Islands Act. Meanwhile, prospective changes for Argyll & Bute and Highland councils have been abandoned after a storm of local criticism, particularly in Highland. I’m afraid that in the latter case they are simply deferring inevitable change that will come a few years down the line, but that’s another story entirely…
Although STV is more proportional than FPTP, it is still a system that can be significantly impacted by where boundaries are drawn. I therefore thought it’d be a useful exercise to re-calculate 2017 results for these new boundaries. The rejection of Argyll & Bute and Highland proposals has saved me a lot of painstaking effort doing the same there, and the largely non-partisan nature of the Islands makes such estimates far less useful. That means it’s just North Ayrshire getting this special treat!
There weren’t any changes to the final boundaries compared to those I wrote about last year, so I didn’t need to make a new map. Just move from page to page to see the comparison of the new boundaries with the previous ones. Full detail on the changes is available in the linked post, but in short, there are substantial changes to the Garnock Valley, North Coast, Three Towns and Arran areas, whilst Kilwinning and Irvine have remained unchanged.
As long-time followers of BBS will be aware, council elections in Scotland are counted by machine, giving us a wealth of of data unavailable for other votes. That allows us to see how the in-person votes were distributed across polling districts within wards, and thus make a reasonable estimate of what results would have been like for these new boundaries.
For Independent and minor party candidates in split wards, they were associated with the new ward they had most support in, and their votes redistributed based on later preferences in the other. Where wards didn’t have a second party candidate to estimate in-party transfers, I’ve assumed a similar rate to the other ward(s) if available, or similar to other parties in the same area if not.
1. North Coast
The first of the new Ayrshire wards, North Coast is a slightly extended version of the present North Coast and Cumbraes, having grown to include the West Kilbride end of Dalry and West Kilbride. That also adds a councillor to existing tally of four, making this one of the first five-member wards in the country. Although this includes an inhabited island in the form of (Great) Cumbrae, it’s simply too small to justify a ward to itself.
This is quite a strong area for the Conservatives, and in theory they’d easily have enough votes for two councillors. The SNP too would do well enough for a double here, leaving the last seat for Labour. Notably, that means Ian Murdoch, who was elected as an Independent councillor in 2017, wouldn’t have made it in under these boundaries.
However, that works on the assumption the Conservatives would have had the sense to stand two candidates. They didn’t do so for the existing 4-seater, underestimating their strength. Had they done so they’d almost certainly have elected that candidate, and it was this lack of foresight that led to Murdoch’s election. With an extra seat up for grabs, they may well have been more tempted.
2. Garnock Valley
Next door is a much more substantially altered Garnock Valley ward. This combines the former Kilbirnie and Beith ward and its three councillors with the remaining Dalry portion of Dalry and West Kilbride which brings two, giving us another five-seater. This is, I think, a particularly good example of where the additional flexibility of a five councillor wards has shown its worth, as Dalry and West Kilbride was something of an odd creature.
The multiplicity of Independent candidates do collectively very well here and sum up to a plurality of votes, though the SNP are by far the most popular single choice. That is estimated to give the latter two councillors, whilst only one Independent, Donald Reid, makes it across the line. The remaining seats would go one apiece to Labour and the Conservatives. Again though there’s a bit of uncertainty in this estimate.
For Independent candidates, there isn’t any direct evidence of their potential support in areas they didn’t stand. The strength of local connection is highly evident, as seen with current Dalry and West Kilbride councillor Robert Barr, who drew about 90% of his vote from the Dalry end of the ward. If we assumed similar here, that is 10% of his vote coming from elsewhere, we could probably add another 90 votes to his tally. That would have made the difference for him to beat the second SNP candidate.
We now move southwards into the Three Towns area, where Ardrossan has been split off from the Ardrossan and Arran ward to stand alone. It’s also slightly expanded versus the mainland portion of that ward, taking in a bit of the former Saltcoats ward that was, strictly speaking, part of the Ardrossan locality. The rural bulb at the top will be lucky if it contains more than few a dozen voters, so pay no real heed to that bit! The urban expansion ensures this ward comes in at a standard size of three councillors.
The SNP had a big lead here, but it’s not enough for two councillors, meaning that Labour and the Conservatives join them with one seat for each party. Whereas Labour had been locked out of the previous ward, the purely mainland nature of this new version plays in their favour. The presence of the British Unionist candidate would also have helped keep Labour ahead of the Conservatives on first preferences, given we can probably assume that’s where the votes would have went otherwise.
There are a lot of interesting little bits about Arran, which is the other side of the former Ardrossan and Arran. For one thing, it’s the only single councillor ward in the country. Electing just one seat goes rather against proportionality, which may be why although it was a power granted Boundaries Scotland by the Islands Act, they proved reluctant to use it. Another sole seater for Eriskay and Vatersay in the Western Isles was proposed, but that became a two seater by the final output.
In addition, whereas the general aim of one councillor for an island is to reflect the unique circumstances, it’s Ardrossan that may be feeling better represented after this. Apparently there was some disquiet amongst the mainlanders that two of the three councillors elected in 2017 for the current ward were Arran based – though it’s worth remembering there’s no in-ward residency requirement anyway.
On first preferences, this shows an absolutely wafer thin lead for the Conservatives over the SNP. As their candidate was Arran based, it’d also be a pretty strong result by the SSP’s standards, as a little aside. Anyway, once we go through the transfer rounds, the lead flips to an equally bawhair win for the SNP. (You may have seen an earlier version of this on Twitter, but I realised I’d pointed the postal vote allocation calculation at the wrong cell!)
That makes this one of my uncertain estimates. There are just too few votes here to say for sure that it definitely would have went one way or the other, given the imperfections inherent to the estimation process.
One final note of interest here: that SNP councillor was one of those who defected to Alba after that party launched earlier this year. Working on the assumption she will be re-contesting the ward, this may be one of the more fascinating Alba-inclusive contests in the country, given personal votes are particularly strong in island communities. (Update: Since publishing this, the councillor in question appears to have left the party.)
5. Saltcoats and Stevenston
Crossing back over to the mainland the remaining two of the Three Towns, Saltcoats and Stevenston, are merged into a single ward of the same name – bar the chunk of the former transferred to Ardrossan. This actually largely recreates a ward that existed on the original 2007 boundaries, containing just a few extra streets. That version elected four councillors – this one elects five, perhaps because the council itself grew from 30 to 33 seats in total in the 2017 boundary changes.
There are extremely solid performances from both the SNP and Labour here, placing almost neck-and-neck on first preferences. That’d easily hand two seats to each party, with the fifth taken up by the Conservatives. Current Independent councillor for Saltcoats, Ronnie McNicol doesn’t make it through on this estimate. Unlike his colleagues further north, I don’t have any reason to caveat this one. Although previously a councillor for the joint ward, I think all the parties are just too far ahead for him to have managed it.
The remaining four wards are totally unchanged versus their 2017 boundaries. For one of them, Kilwinning, that makes perfect sense. It’s a self-contained town, electing four councillors (those went two Labour and one each SNP and Conservative), so it’s perfectly sized. What was interesting to me was that they haven’t redrawn any wards in Irvine, which remains three wards. Two of these have three councillors, and one is a four seater – so that’s ten in total.
It seemed like Boundaries Scotland were very willing to use the flexibility to introduce five-seat wards elsewhere in North Ayrshire – and indeed in their rejected Highland proposals, and I think that was in part due to proportionality. It therefore strikes me as a little odd they didn’t take the opportunity to turn Irvine into a pair of five-seaters. I’d actually made a suggestion as to how they could do so in my submission, but it didn’t really appear to be on their radar.
The (lack of) implications for seats there shows how STV can, by design, deviate a bit from ideal proportionality. In 2017, although the SNP placed first across the town, it was Labour who won the most seats, winning four of the ten. That left the SNP tied with third-placed Conservatives on three seats each. Had they redrawn the boundaries into two evenly sized wards, I expect we’d have seen 2017 estimates of two SNP, two Labour, and one Conservative for each.
You might be wondering why these estimates look so bad for the Independent councillors, even bearing some of the caveats in mind, and despite five-member wards being generally more proportional. Firstly, independents here tend to have highly concentrated vote shares, whilst STV necessitates relatively large wards. The example par excellence of this is Dalry and West Kilbride, where there were five independents in 2017.
Every single one of them drew at least 90% of their support from one or other of those towns – nobody had a split base of support. For the wards expanding to five-seaters, that means the base of support for these Independents becomes a smaller proportion of the overall vote, pushing their vote share down. For party candidates, that sort of thing is mostly countered by the reduced quota – but for Independents, that causes further damage.
A lower quota means that more votes stay within a given party if it has multiple candidates, and fewer transfer out. Given Independents are generally seen as acceptable to most party voters, not getting those surplus transfers means they remain behind second placed SNP and Labour candidates in some cases, and don’t overtake Conservative candidates in others.
Nonetheless, bear the caveats listed for each ward in mind. Whilst Labour are a stable 11 seats regardless, the SNP could be as low as 10 (less Arran and second in Garnock Valley), the Conservatives as low as 8 (no second North Coast) or as high as 10 (with Arran), and Independents could have suffered only minor losses with 3 (with Murdoch in North Coast and Barr in Garnock Valley).
All of this is of course looking backwards, to what the last election might have looked like. The results of the 2022 election could be very different indeed, not least if political shifts seen at other levels continue. In 2016, Labour placed narrowly ahead of the Conservatives at Holyrood, which held in 2017. This May, the Conservatives overtook Labour across North Ayrshire, and there’s every chance they could repeat the feat next year. That definitely marks this council out as one to watch.
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.