The odds of a Labour victory have just got a lot better

In the past three months, Keir Starmer’s party has solidified its lead and made a majority government look a lot likelier

October 11, 2023
A protester threw glitter over Keir Starmer before his keynote conference speech. Image: PA Images / Alamy Stock Photo
All that glitters: a protester disrupts Starmer’s keynote conference speech. Image: PA Images / Alamy Stock Photo

Time for an update. Regular readers may know I have been cautious about Labour’s prospects. Three months ago, I said the party had a 50-50 chance of securing an overall majority at the next election. Since then, the odds have moved in Keir Starmer’s favour. He is now likely to lead a majority government and enjoy a full five-year term in Downing Street.

Before we look at what has happened, it’s worth reminding ourselves why such an outcome would be remarkable, and why it has made sense to be cautious. Britain’s electoral geography has tilted in the Conservatives’ favour. After 2019, Labour needed a swing of 12 per cent across Britain to reach 326 seats and a majority in the House of Commons. This is more than any party has achieved since 1945. In Tony Blair’s landslide victory in 1997, the swing was 10 per cent. His personal ratings were stratospheric; Starmer’s are barely OK. Achieving a bigger swing in 2024 than 1997 was, shall we say, not inevitable. 

So why the change in the odds?

Here are three factors.

First, Rishi Sunak has lost his early shine. When he took over from Liz Truss, many voters were happy to give him the benefit of the doubt. His party’s ratings were catastrophic; his own ratings were not great, but not nearly as bad as those of the Tories generally. The gulf between the two was never likely to last. Parties and their leaders usually have different ratings, but not very different. The question was: would Sunak lift his party, or would his party drag him down? 

Now we know. Sunak’s ratings have steadily declined, while his party has stayed on the floor. Starmer has established a sustained lead as voters’ preferred prime minister.

Second, the SNP’s decline, and the scale of Labour’s victory in the Rutherglen and Hamilton West byelection, make an overall majority significantly more achievable across Britain. The need for a 12 per cent swing was based on a straight Con-Lab swing across Britain. As long as the SNP stayed dominant north of the border, such a Britain-wide shift would give Labour two probable gains in Scotland, and close SNP-Labour contests in five others.

The SNP-Labour swing in last week’s byelection would restore Labour’s traditional dominance in Scotland, with around 40 seats. Given that byelection swings tend to be much greater than general election swings, a more realistic target is 20-25 seats. (Because of the distribution of votes across Scotland, there are likely to be around 30 very tight SNP-Labour battles in individual seats.)

Here’s how this affects the Britain-wide arithmetic of the next general election. Every extra seat Labour wins in Scotland means that it is one seat closer to its target of an overall majority across the UK. We can apply a rule of thumb of five to one: a variation of one per cent point in the gap in the popular vote between Labour and Conservative across Britain leads to a five-seat variation in the number of MPs each party will have in the new House of Commons.

Apply that five-to-one rule to a 20-seat bonus for Labour in Scotland: it knocks four points off the lead Labour needs in overall votes across Britain. Instead of requiring a 13-point lead to win an overall majority, Labour would need a lead of around nine points: still high, but no longer exceptional.

The third factor brings the winning post even closer. I have discussed tactical voting in past blogs. It is becoming increasingly clear that this will have a significant impact on the next election. The pattern from byelections in the past two years, starting with the Liberal Democrats’ triumph in Chesham and Amersham, is remarkable. The Conservatives have defended eight seats, in four of which Labour was the clear challenger, and the other four in which voters decided that the Liberal Democrats were the challengers. In the four Con-Lab contests, Labour’s share rose by an average of 11 percentage points, while the Lib Dems shed four points. In Con-Lib Dem contests, the averages were: Lab down 12 points, Lib Dem up 33.

This is reminiscent of the pattern in byelections leading up to Blair’s victory in 1997—indeed, the force of tactical voting has, if anything, been even greater this time. A perfectly reasonable guesstimate at this stage is that the impact of tactical voting will be similar to that in 1997. Labour would gain around an extra 20 seats from the Tories and the Lib Dems at least 10.

An extra 20 seats Labour brings the winning post even closer. The one-to-five rule applies here, too. It reduces the lead Labour needs in the popular vote by another four points. The target for an overall majority falls from a 9 per cent lead to just 5 per cent.

Indeed, the combined impact of Scotland and tactical voting could wipe out the Conservative bias that has built up in Britain’s electoral geography. Without these two factors, Labour needed up to two million more votes than the Conservatives to become the largest party in parliament. Include these two factors, and parity in votes is likely to mean something close to parity in seats.

It must be stressed—and this is more than my simply covering my backside—we can do nothing more at this stage than to give rough estimates of how Scotland and tactical voting will play out over the coming months. The wide range of possible election outcomes looks like a shallow Bell curve. But the mid-point of that curve has surely moved towards a clear Labour victory—at least for now.