At Stepping Hill Hospital in Stockport, ten of the fourteen lifts no longer work. Staff carry meals up flights of stairs because the alternative is not available. Families with mobility problems cannot reach relatives on the higher floors. The repairs backlog stands at £138 million, a figure that has become almost routine in parliamentary debate, repeated by MPs describing hospitals built decades before anyone imagined the NHS would still be patching them together in 2026. Across the whole estate, the bill for high and critical maintenance work has nearly tripled since 2015, reaching £15.9 billion. It is against this backdrop, largely unmentioned in Westminster's current fiscal argument, that John Healey's search for borrowing flexibility should really be judged.

The chancellor has twelve weeks until his first budget, and the immediate pressures are well rehearsed: a VAT cut on energy bills promised by Andy Burnham, a £5 billion hole in the defence investment plan left behind by Rachel Reeves, and a Treasury headroom figure that may not have survived the Iran war intact. But the more consequential question, the one that will shape the next decade rather than the next quarter, concerns how far Healey is willing to stretch the borrowing architecture Reeves built before she departed for the backbenches.

Reeves's redefinition of debt, measuring public sector net financial liabilities rather than the older, blunter metric, already lets the government borrow more without breaching its own rules, provided the borrowing buys something that counts as a financial asset. Economists at the Resolution Foundation argue the public financial institutions created or expanded under this framework, including the National Wealth Fund and the new National Housing Bank, could raise a further £9 billion a year within the existing rules. Others go further still. Thomas Aubrey of Cambridge's Bennett school wants public corporations allowed to borrow directly from markets, tapping pension funds and other buyers who are not already absorbing gilt issuance, in the way infrastructure is financed in most comparable economies.

This argument has been treated, so far, as one about roads, energy grids and housing. It should not be. The Department of Health and Social Care published its own ten year capital plan in July, committing at least £6.75 billion over nine years to an Estates Safety Fund and confirming a further £2 billion for eradicating RAAC concrete from hospital buildings. Delegated spending limits were raised so that individual capital projects can proceed without Treasury sign off unless they exceed £300 million. On paper, this reads as a genuine attempt to break the NHS out of a cycle in which capital budgets exist mainly to be raided whenever day to day spending runs short, a pattern that successive Public Accounts Committee reports have criticised for years without producing a published capital strategy that survives contact with a difficult budget.

Whether that plan holds now depends substantially on decisions Healey has not yet made. If the Treasury remains wedded to financing large NHS infrastructure entirely through conventional gilt issuance, competing for headroom against energy policy, defence and housing in every fiscal event, then hospital trusts waiting years for wave three of the New Hospital Programme, North Devon among them, have every reason to expect further delay. If instead ministers allow health infrastructure to draw on the kind of market borrowing Aubrey describes, distinct from gilts and backed by credible revenue and cost projections, the arithmetic underpinning the ten year plan becomes considerably more durable.

Helen Miller at the Institute for Fiscal Studies has warned against getting too absorbed in what the rules technically permit, since borrowing is borrowing whatever it is called, and the real question is whether the investment itself is worthwhile. On hospital estates that argument barely needs making. A ceiling that has already collapsed, a lift that has already failed, does not require a case to be built for its repair. What it requires is a chancellor willing to treat capital commitments to the NHS with the same seriousness now being extended to growth and devolution, rather than allowing them to become the first casualty of whichever crisis dominates the weeks before 28 October.