Keir Starmer announced an additional £15 billion of defence spending power over four years on 30 June.
The Treasury's funding table published the same day identified where £10.3 billion would come from.
Against the remaining £4.7 billion, it said only:
"To be funded at Budget 2026."
That is 31 per cent of the £15 billion package without a specific financing measure attached when the plan was announced. £1.8 billion of it falls in 2026/27, followed by £1.1 billion, £1 billion and £900 million in the following three years.
The £15 billion commitment itself should not be reduced to £10.3 billion.
The government has increased Ministry of Defence spending power by £15 billion. Its own breakdown says that includes £11.6 billion of additional cash and another £3.4 billion generated by shifting costs away from the MoD, asset sales and reprioritising existing defence spending.
Total MoD spending is set at £68.3 billion in 2026/27, rising to £79.1 billion in 2029/30. Core NATO defence spending is due to increase from 2.6 per cent of GDP this year to 2.7 per cent from 2027/28.
The incomplete part was the financing.
The government's own funding explainer uses two formulations which sit uneasily together. It says the package "has been funded" through reprioritising public spending, then immediately says only £10.3 billion has been identified and that another £4.7 billion will be confirmed at the Budget.
Starmer had already announced his intention to resign as Labour leader and Prime Minister on 22 June. Labour's published leadership timetable said parliamentary nominations would not even open until 9 July. He would remain in office until a successor was chosen.
The defence plan was therefore announced by a Prime Minister who had already said he was leaving, with almost a third of its additional funding still requiring decisions at a later fiscal event.
Defence minister Luke Pollard added an unusual detail the following morning.
He told Sky News that he had first seen the detailed £15 billion funding breakdown when the Treasury published it on 30 June.
Asked about Andy Burnham, then the clear favourite to succeed Starmer, Pollard said he understood Downing Street had been keeping Burnham's team involved and had told them when the Treasury statement and financial breakdown were published.
That is not evidence that the Defence Investment Plan had been devised without the Ministry of Defence.
It does show that one of its ministers had not seen the final Treasury financing breakdown before publication.
Starmer's answer was that governments do not have to announce every financing measure at the same time as a spending commitment.
At Prime Minister's Questions on 1 July he pointed to roughly £22 billion of fiscal headroom created at the previous Budget and compared the defence announcement with earlier spending decisions made between Budgets.
He cited the Conservative government's announcement in 2018 of NHS funding over five years, when ministers similarly said the detailed financing would be set out later.
That is a legitimate defence.
Fiscal headroom means the margin by which the government's forecast satisfies its fiscal rules. Having £22 billion of it can give a government scope to increase spending without immediately breaching those rules.
It is not the same thing as identifying where £4.7 billion will come from.
The Treasury document does not say that £4.7 billion has been allocated from a £22 billion pot. It says the money is "to be funded at Budget 2026".
The difference can already be seen in the £10.3 billion for which the government has identified measures.
Every department is being required to surrender 1 per cent of its capital budget, producing £4 billion over four years.
The Department for Energy Security and Net Zero must find another £2 billion, including £400 million of financial transactions.
Transport is contributing another £800 million, including up to £700 million from roads funding. Asset sales and changes to costs borne by the Treasury make up other parts of the package.
Those are identifiable choices.
Some have identifiable consequences.
On 26 March, the government published its third Road Investment Strategy and said work would begin during the 2026 to 2031 period on the A38 Derby Junctions and A46 Newark Bypass.
Three months later, the defence funding document said both could be cancelled.
The Department for Transport is consulting on removing the schemes because neither has entered its main construction contract and they are less advanced than other projects. No final decision had been taken by 4 July.
That should not be reported as £700 million already saved by cancelling those two roads.
The government's document says only that they are among the schemes being considered as part of wider roads reductions of up to £700 million.
The defence plan had already been through a bruising funding dispute.
John Healey resigned as Defence Secretary on 11 June. Contemporary reporting said the MoD had sought about £18 billion to close the gap in its programme and that the package available when he left was about £13.5 billion.
Dan Jarvis replaced him that day. When the final plan was published, Jarvis said it contained £1.5 billion more than when he had taken the job.
The eventual £15 billion should therefore not be described as an imaginary spending commitment or reduced to the £10.3 billion already matched to financing measures.
The problem is narrower.
The government committed the defence spending.
It identified departmental capital reductions, roads savings, energy savings, asset sales and Treasury changes to finance £10.3 billion of it.
For the other £4.7 billion, including £1.8 billion in the current financial year, the published answer was Budget 2026.
Starmer announced that commitment eight days after saying he would leave office.