Starmer on the brink, and Labour's Reform problem

Sir Keir Starmer will deliver a speech to Labour MPs on Monday designed to head off an open leadership challenge, after the party’s drubbing in last week’s English local elections produced a cascade of public rebukes from senior figures over the weekend. His former deputy Angela Rayner issued a “last chance” warning and named Andy Burnham as a credible replacement; the Labour MP Catherine West threatened to challenge Starmer herself if no minister would; and the BBC’s politics desk now runs seven scenarios for what happens next, all of which begin from the assumption that the Prime Minister’s authority has been broken. The proximate cause was Reform UK winning seats from Sunderland to Swansea, with the Greens picking off Labour councils in inner London, while the Conservatives lost ground to both flanks at once.

The received wisdom

The standard reading inside the Westminster lobby is that this is a normal mid-term wobble that has been amplified by an unusually fragmented party system, and that Starmer’s instinctive response — bring back old hands, promise “bolder action,” widen the policy aperture on immigration and welfare — is broadly the right one. On this account, the local-government results were always going to be ugly because Labour inherited a fiscal mess, the Reform vote will recede once it is asked to govern anything serious, and the Greens are a London-bourgeois epiphenomenon that does not threaten Labour’s industrial heartland. The further version of this reading — most clearly stated by Bridget Phillipson over the weekend — is that any leadership change now would simply hand the next election to Reform on a plate, and that party discipline therefore requires MPs to swallow their misgivings until the polling improves. It is an argument from prudence, and it is not a stupid one.

A different read

What the lobby framing misses is that the Labour Party is not having a wobble. It is being squeezed by two different electorates that no longer believe the same party can speak to both of them, and the first-past-the-post system has stopped insulating it from the consequences. The graduate-progressive vote in Lambeth and Lewisham moved to the Greens over Gaza, housing, and a sense that Labour in office has become managerial in the dullest sense. The post-industrial vote in Sunderland and the Welsh valleys moved to Reform over immigration, energy bills, and the suspicion that the Labour leadership regards their cultural preferences as a problem to be managed rather than a constituency to be served. Both groups concluded, on the same Thursday, that the official party of the British centre-left had nothing distinctive to offer them. That is not a wobble; that is the architecture of New Labour finally cracking.

The conservative point worth making — and it is a point about institutions rather than personalities — is that Starmer’s predicament is the predictable cost of a particular theory of how to govern Britain. The theory, inherited from the Blair-Brown period and ratified by the Starmer leadership in opposition, was that competence, fiscal restraint, and a mildly progressive social agenda would assemble a stable majority in the centre. That theory presupposed a politics in which Reform did not exist on the right and the Greens were a fringe on the left. Both presuppositions have failed within twelve months of the general election. The historical parallel is not 1997 but 1992-94 in Canada, when the Progressive Conservatives were squeezed simultaneously by Reform and the Bloc Québécois and reduced from 156 seats to two; it is not impossible for a governing party to be filleted by flank parties even before its term is up, and the British system, contrary to its own folklore, now allows that to happen once the two-party share falls below about 65%.

The further point, which the Labour Party will resist hearing, is that the Burnham option does not solve this problem. A move to a more redistributive economic register might recover some of the Reform vote on bread-and-butter grounds, but it will not buy back the cultural ground lost on immigration, policing, and family policy, where the gap between the Labour activist base and the Labour-voting working class is now enormous. Nor will it satisfy the Greens-curious metropolitan vote, which is alienated by exactly the tonal compromises that the Reform-recovery strategy would require. The Liberal Democrats’ Ed Davey understood this clearly enough over the weekend when he positioned his party as “the alternative to the extremes.” That is the move of a leader who has noticed that the centre is now empty and is racing to occupy it. Whether his party has the capacity to do so is another matter; that he has noticed at all is the news.

The deepest version of the argument is that British politics has now caught up with continental politics, and that Labour is going to have to decide whether it wants to be a French-style social-democratic rump in coalition arithmetic or an Italian-style party of government willing to police its own cultural left. Starmer has neither the temperament nor the mandate to make that choice, and Burnham, for all his political talents, would face the same vice from the opposite direction. The honest answer is that the next Labour leader, whoever it is, will inherit a party that no longer has a coalition; the work is rebuilding one from the wreckage, not preserving the appearance of one in office.

What to watch

First, whether any cabinet minister moves against Starmer this week — a single resignation on principle would change the arithmetic immediately. Second, the precise terms of the Burnham return talk: a parliamentary seat would have to be found, and any MP volunteering to stand down becomes the story. Third, Reform’s behaviour in its newly-won councils — competent administration would consolidate the realignment; visible chaos, as with the racism investigation in Sunderland, would slow it. Fourth, the Liberal Democrats: if Davey’s “alternative to extremes” line tracks in the polling, the centre is back in play.

— J