This is how Rishi Sunak could win the next election

As one of the bolsheviks at the Peter Mandelson Memorial Dim Sum Supper, I predicted that Rishi Sunak will be prime minister after the next election. The bolsheviks, Russian for those in the majority, argued that the economy is likely to turn round and that when it comes to a choice between Sunak and Keir Starmer, the voters could be evenly divided, which would be enough for the Conservatives to hang on.

The mensheviks, on the other hand, the minority at the annual predictions dinner named in honour of Lord Mandelson (but it must be stressed without his knowledge or approval), argued that the Conservative Party has broken something fundamental. Turning it off and on again has not restored it to working order.

Between them, Boris Johnson and Liz Truss have achieved the equivalent of the European exchange rate debacle of 1992, the mensheviks reasoned, and went on to argue that “14 wasted years” by the time of the election would be sufficient to propel Sir Competent Starmer to No 10 at the head of a majority Labour government.

As in Russian pre-revolutionary politics, everything is now seen through the lens of bolshevism or menshevism. We bolsheviks saw the 10 per cent swing to Labour in Thursday’s by-election in Stretford and Urmston, in Greater Manchester, as pointing to a hung parliament if there were a general election now, and consistent with Sunak holding on to power in 18 months’ time.

The mensheviks said the by-election was in line with national opinion polls, which point to a majority Labour government. Professor Sir John Curtice, who is like Karl Marx in this disagreement, was invoked by both sides.

The by-election was mainly interesting because it means there are now two MPs called Western: Andrew, the winner in Stretford and Urmston, and Matt, the Labour MP for Warwick and Leamington, who shall now be known as the south Western, while Andrew is the north Western. This led me to a search for MPs with points of the compass as names, in which I discovered, thanks to Andrew Gray, that there were four MPs called Dudley North, all descended from a member of the Long Parliament before the civil war (one of whose granddaughters was called Dudleya North), who I think are the only cases of MPs with the same full name as a (later) parliamentary constituency.

But the by-election was also interesting as a pointer to voting trends. Not that interesting, mind, as turnout was just 26 per cent and reporting of the contest was even lower key than that of the Chester by-election two weeks earlier. The fairest thing to be said about Stretford, though, for those who are neither bolsheviks nor mensheviks, is that the evidence it provided was inconclusive.

The swing to Labour was lower than the 14 per cent recorded in Chester, but Chester is more of a marginal, having returned a Conservative MP in the 2010 parliament. So although the swing in Stretford, if repeated nationwide in a general election, would produce a hung parliament, the result there was in line with seat-by-seat models using national opinion polls, which take into account the smaller vote available to “swing” in safe seats. And national opinion polls currently suggest a Labour majority.

Rob Ford, professor of politics at Manchester university, observes that by-elections in the whole of this parliament point to Labour losing the general election, while more recent by-elections suggest that Labour would – just – win.

However, I return to the speech that I delivered to the comrades at the dim sum supper, in which I argued: one, that Sunak is not in as bad a position as John Major was before the 1997 election; two, that Starmer is not in as good a position as Tony Blair was; and three, that the electoral terrain is more forbidding for Labour than it was then.

Not only does Labour start from a lower base than in 1992, when Major had a majority of just 21, but the loss of Scotland and the increasing concentration of the Labour vote in its urban strongholds means that a majority government is even further out of reach. A Labour majority would require a bigger swing than that achieved by Blair; with Labour about 12 percentage points ahead in share of the vote.

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

This is, incidentally, not because the voting system is inherently biased against Labour: in Blair’s time, it was biased in Labour’s favour, but now the distribution of Labour’s vote is “inefficient”, in psephology-speak. The new boundaries, coming into effect next year, which equalise constituency sizes and ought to make the system fairer, are worth six seats to the Conservatives, according to James Kanagasooriam, the guru who first identified the red wall.

Patrick English of YouGov reports that, on his calculations, if the shares of the vote won in 1997 were repeated in 2024, Labour would have a tiny majority, if at all, rather than Blair’s majority of 179.

It may be, then, that the sensible position would be to split the difference between the bolsheviks and the mensheviks, and to predict a hung parliament, in which Starmer would almost certainly be prime minister, leading a minority Labour government – because of the other parties only the DUP would consider propping up a Tory government.

But governments lose elections, and we bolsheviks are not yet convinced that this government has given up.