The tricky lessons to draw from a Labour win on Thursday

Arguably, the 2021 Westminster by-election in Airdrie and Shotts was competitive. The SNP’s Anum Qaisar won the seat by eight points, within the ten-point margin we typically use to define a marginal seat.

But it also took place on the same day as the Scottish Parliament elections, so if you wanted to understand the national political environment, you weren’t paying attention to Airdrie and Shotts.

Before then, our last competitive by-election was a Scottish Parliament by-election for the Cowdenbeath constituency in January 2014. Labour had won the seat in 2011 by just five points. But in the end, Labour’s Alex Rowley defeated the SNP’s Natalie McGarry by a whopping 27.4 points.

So, it’ll be understandable if the pages of Friday’s papers are filled with generalisations based on Thursday’s result, but we should be careful about how we interpret the outcome.

Firstly, as a rule, by-elections tend to be poor guides to what will happen in the next general election. A recent analysis by polling consultants Electoral Calculus found that seats gained in by-elections over the past 40 years are more likely than not at the next general election to flip back to the party that initially held the seat .

In the devolution era, just four Scottish by-elections have resulted in a gain by an opposition party. Of these, only one was then held at the next Holyrood elections – Ayr, won in a 2000 by-election by Conservative John Scott, who then held the seat until being defeated by the SNP’s Siobhan Brown in 2021.

Secondly, Rutherglen and Hamilton West is hardly a bellwether seat. Before the 2014 independence referendum, it was the sort of seat that Labour would comfortably win with stratospheric majorities. It had the third-highest Labour vote share of Scotland’s 59 seats in 2015 and the fifth-highest in 2019. Sandwiched between those elections, in 2017, Jeremy Corbyn’s Labour won the seat – while taking just seven seats nationally.

A Scottish version of the American Partisan Voter Index, measuring how much more or less in favour of each party each seat is compared to the country as a whole, would suggest that Rutherglen and Hamilton West is the fifth most Labour-supporting seat in the country. In contrast, it is the 36th most SNP-supporting seat.

Labour will almost certainly win this by-election. Even the most pessimistic models based on the most pessimistic current polls suggest that Labour would take Rutherglen and Hamilton West if a general election were held tomorrow.

With the SNP in retreat and Labour in the ascendancy, Rutherglen and Hamilton West is among the lowest hanging of the low-hanging fruit for Labour.

The vagaries of by-elections will further advantage Labour. Dissatisfaction with the Scottish Government and Labour’s status in the seat as the primary outlet through which voters can voice that dissatisfaction will likely have two effects: protest voting and differential turnout.

It would not be surprising to see many past SNP voters staying home or voting Labour in protest to give the government a bit of a kicking.

So a Labour win won’t, in and of itself, tell us much about the state of Scottish politics that we don’t already know from the polls. It could point to an environment in which Labour gains only a few seats at the next general election or to an environment where they can win over 20.

As far as interpreting the outcome, what will be more interesting is the size and the form of Labour’s victory.

Recent polls have suggested Labour could win between 15 and 25 seats in Scotland at the next general election. The upper end of those projections would put them neck-and-neck with an SNP deep in decline, but even the lower end would suggest significant progress.

In a general election in which Labour win 20 seats, you would expect them to win Rutherglen and Hamilton West by around 10 points. That would imply an election in which Labour wins around 32% of the vote and 20 seats, to the SNP’s 39% and 31 seats.

A narrower win, by six points or less, would point to a less rosy picture for Labour – such a result would imply winning fewer than 15 seats at the next general election and a similar SNP result to 2017.

A win by under four points would suggest that Labour would struggle to break into double-digits regarding seats won. Conversely, a win by 15 points or more would put Labour on course to overtake the SNP.

Then there is the form of Labour’s win. By-elections are fought under conditions in which turnout can collapse dramatically. A big question will be whether Labour wins despite also losing votes – such a win, even a significant vote share win, may prove to be built on a foundation of sand with those voters who stayed at home still up for grabs, setting the stage for an SNP comeback.

Labour will win on Thursday, and if they make significant gains in Scotland at the next general election, that victory will be remembered as the harbinger of a more ‘normal’ kind of electoral politics in Scotland – the first time since the 1970s that Scotland has not had one, singularly hegemonic party.

But we must be careful not to jump to conclusions on Friday morning. The conclusions we can draw from the result will be highly contingent and necessarily nuanced. Labour will herald a narrow win as a significant turning point in their fortunes north of the border, but it would not warrant such a response.

Likewise, the SNP will seek to minimise the meaning of a large and well-founded Labour majority in Rutherglen and Hamilton West, but such a result would presage disaster for Humza Yousaf’s party at the next general election.

Perhaps a new dawn will break for Scottish Labour. Perhaps not. A long road to the next general election still lies before us. Let’s avoid counting seats before a single vote has been cast.