Will Keir Starmer be the UK Prime Minister at 23:59 BST on 30 June 2026?

TL;DR

I estimate a 78% chance that Keir Starmer is still the UK Prime Minister at 23:59 BST on 30 June 2026.

Context

Starmer is still the formal officeholder. GOV.UK lists The Rt Hon Sir Keir Starmer KCB KC MP as the current Prime Minister and says he became Prime Minister on 5 July 2024; the same GOV.UK role page also shows fresh official PM activity, including PM remarks dated 5 June 2026 (GOV.UK Prime Minister; GOV.UK PM remarks, 5 June 2026).

The crisis is real. The Mandelson vetting affair damaged Starmer before the 7 May elections, and Labour then suffered a severe election shock. The current hinge is the Makerfield by-election on 18 June 2026, where Andy Burnham is the Labour and Co-operative candidate; Burnham has now said he would seek to enter a Labour leadership contest if he wins, while No. 10 says the Labour process has not been triggered and Starmer will not walk away (Wigan Council voter information; ITV, 5 June 2026).

Evidence

The historical backbone points to survival through this exact deadline. LabourList says no sitting Labour Prime Minister has ever faced a Labour leadership election since Labour began holding leadership elections in 1922 (LabourList, May 2026). The House of Commons Library, published 31 October 2025, says Labour MPs cannot formally hold a confidence vote in the leader; a challenger needs 20% of Labour MPs, and the incumbent is automatically on the ballot if the challenge reaches members (House of Commons Library, 31 October 2025).

The rules also make speed hard. The Labour rule book says a challenger where there is no vacancy needs support from 20% of Commons PLP members; all nominees must be Commons members of the PLP; valid nominations are published; and the election timetable must be approved by an independent scrutineer (Labour Rule Book 2025). LabourList translates the current threshold as 81 Labour MPs out of about 403 and notes that the 2020 Labour leadership election gave members 38 days to vote while the 2016 contest lasted 33 days (LabourList, May 2026).

Modern UK handovers support the same point: political defeat and formal departure are different things. Theresa May announced her resignation plan on 24 May 2019 and remained Prime Minister until 24 July 2019 (GOV.UK May statement; GOV.UK Theresa May). Boris Johnson announced he would step down as Conservative leader in July 2022 but remained Prime Minister until 6 September 2022 (GOV.UK Boris Johnson; GOV.UK Johnson final speech). Liz Truss was the fast outlier: she announced resignation on 20 October 2022 and Rishi Sunak became Prime Minister on 25 October 2022, but that used an emergency Conservative process, not Labour's member-ballot rules (GOV.UK Truss statement; GOV.UK Rishi Sunak).

The anti-Starmer bloc is large but not yet a working removal mechanism. ITV reported on 13 May 2026 that it had tracked 95 Labour MPs prepared to oust him, more than 23% of the 403-member PLP, but the same report stressed that a contest is triggered only if a named candidate gets the 81 supporters needed to challenge Starmer (ITV, 13 May 2026). That distinction is central. Many MPs wanting Starmer gone is not the same as 81 MPs publicly backing Burnham, Streeting, Rayner, or anyone else before the deadline.

The public and member polling say Starmer is vulnerable if a contest reaches the electorate. Ipsos fieldwork from 8-12 May 2026, with 2,191 British adults, found 66% thought Starmer should not lead Labour into the next general election, but only 35% wanted him to resign immediately (Ipsos, 13 May 2026). YouGov fieldwork from 14-18 May 2026, with 706 Labour members, found Burnham beating Starmer 59% to 37% in a head-to-head and taking 47% first preference in a wider field; it also found 56% of members thought Starmer should stand in any leadership election, which supports the idea of a contest rather than an immediate quit (YouGov article; YouGov tables).

Makerfield is the live catalyst. Wigan Council says the by-election is on Thursday 18 June 2026, with the count after 10pm and results expected on Friday 19 June 2026 (Wigan Council voter information). Survation's 15 May pre-poll model put Burnham-on-the-ballot Labour at 45%, Reform at 42%, and Labour winning in roughly two-thirds of 10,000 simulations (Survation initial estimate, 15 May 2026). A later Survation telephone poll, fielded 26 May-1 June 2026 among 518 Makerfield adults, had Burnham on 49% and Reform's Robert Kenyon on 39%, with a 4.8-point margin of error (Survation update, 4 June 2026; Independent, 4 June 2026).

I model the deadline as an event tree, not as a generic judgement on whether Starmer is doomed later. My central inputs are below.

The calculation is: P(NO) = 0.045 + (1 - 0.045) x [(0.76 x 0.21) + (0.24 x 0.10)] = 22.0%. So P(YES) = 78.0%.

What's non-obvious

The obvious story is that Starmer is in grave danger. That is true. The less obvious point is that a Labour leadership contest, a public resignation timetable, or a political concession does not by itself resolve this question NO. If Starmer is still the King's Prime Minister at 23:59 BST on 30 June, the question resolves YES, even if he has already announced that he will leave later.

Burnham's route both helps and hurts the rebels. It gives them their strongest candidate, but it burns nearly all the remaining time. If Makerfield reports on 19 June, there are only 11 calendar days left to get Burnham into the Commons, assemble 81 public nominations, force Starmer not merely to lose authority but to stop being Prime Minister, choose an interim or successor, and complete the King's appointment. That can happen in a true Cabinet collapse. It is not the base case.

Limitations

The biggest unknown is the private PLP and Cabinet count. Public reporting can count MPs who want Starmer gone, but the decisive variable is 81 firm nominations for one named challenger plus enough Cabinet pressure to make Starmer leave immediately. I found no public evidence current to 7 June that Burnham, Streeting, or Rayner already has that package.

The second uncertainty is Makerfield. The latest public poll is only one constituency poll with 518 adults and a 4.8-point margin of error, and by-election turnout can be strange. A large Burnham win would raise the chance of a late-June cascade. A Reform win would also be destabilising, but in a different way.

The third uncertainty is scandal timing. The Mandelson issue has already produced severe damage. A new document, police development, or Cabinet-level allegation could compress the timetable in a way rulebook-based models miss. That tail is why I keep NO near 22%, not in the low teens.