Labour’s win in the Chester by-election is not the walkover that people think it is

Labour did well in the Chester by-election yesterday, and congratulations to Samantha Dixon, the new MP. But the brutal fact is that a swing to Labour of 14 per cent falls short of where the party needs to be if Keir Starmer is to win a majority in the Commons.

This may seem surprising, given that Labour is so far ahead in the national opinion polls and Westminster is abuzz with news of Conservative MPs who are making alternative plans because they think the next general election is a write-off.

Those of us who learned our trade at the feet of Sir David Butler, the sultan of swing, as Michael Crick calls him in his biography, know that the swing is the thing. It is not the only thing – politics is about much more than numbers and is essentially unpredictable, thank goodness – but in politics you not only have to be able to count, you have to understand the concept of swing.

The number that matters at any election is the share of the vote that has switched between the two leading parties since the previous election. In the Chester by-election, the Labour share of the vote went up by 12 percentage points compared with the last general election, and the Conservative share went down by 16 points. That is the equivalent of 14 per cent of the electorate switching parties: the swing from Tory to Labour.

As it was already a Labour seat, the swing is only useful in telling us how good a victory this was for the opposition, and as a pointer to the next general election, because it is a number that can be compared with other by-elections and with what happened at subsequent general elections.

So, yes, the result in Chester was the worst for the Tories in the seat since the party was founded, whenever historians think that was because the seat has existed in roughly the same form since 1545. But bad things happen to government parties in by-elections that are usually worse than the subsequent result in a general election because by-elections are a vehicle for mid-term protest.

What matters is how bad the Chester result is for the Tories compared with other by-elections. A 14 per cent swing is not as bad, for example, as some of the 20 per cent plus swings won by Labour in by-elections before Tony Blair’s victory in the 1997 general election.

Well, no one is expecting Starmer to win by a landslide, it might be protested. But because Labour starts its general election campaign from such a low base, Starmer needs a bigger swing than Blair won just to win a bare majority in the Commons. The swing from the Tories to Labour from the 1992 election to 1997 was 10 per cent; Patrick English of YouGov calculates that Labour would need a swing of 13 per cent at the next election to win a majority of one.

The Chester result, if repeated across the country in a general election, would therefore be enough to give Starmer a majority. But by-election swings are never repeated at subsequent general elections. The question is how to compare the Chester result with other by-elections. Some of Blair’s best results were in Dudley West, SE Staffs and Wirral South, when Labour gained those seats from the Tories. Ed Miliband was also cheered by Labour gaining Corby from the Tories in 2012, with a swing of 18 per cent, although he went on to lose the general election and the Tories regained the seat.

But by-elections where seats change hands are different. There is more excitement about the prospect of gaining a seat than holding one with an increased majority. Even if the turnout in Chester was a respectable 41 per cent, the contest hardly set the country alight.

To keep up to speed with all the latest opinions and comment sign up to our free weekly Voices Dispatches newsletter by clicking here

So we ought to compare Chester with by-elections where Labour held the seat, and where there was no significant Lib Dem or nationalist vote. There were only three by-elections like that between 1992 and 1997, in Dagenham, Newham NE and Hemsworth. The swing was very different in each of them (23 per cent, 16 per cent and 5 per cent), so the comparison is not simple.

What is promising for Labour is that the swing in Chester was higher than the 13 per cent recorded in the Wakefield by-election in June this year, just before Boris Johnson announced his resignation. It suggests that the Conservatives have not rescued the situation by changing leader.

But the Chester result does not suggest that Labour is doing better than it was in Blair’s time – which is what it needs to do to win a majority. The party should be pleased if Starmer is prime minister of a minority government, which is likely in any hung parliament because the Tories have only the DUP as possible allies – but should not get the idea that a good win in Chester means it is heading for a majority government.