Update, 2 August 2026. Arithmetic unchanged; politics not: Burnham PM from 20 July, and wartime-scale food and climate asks will test fiscal credibility sooner.
You have read nine chapters describing things Britain could do. Each came with a direction and a human stake. This chapter answers the question waiting since chapter two.
Can we afford it?
The honest answer: not from current tax and spending without change. Affordability is not the same as impossibility. Britain borrows in its own currency. The constraint is whether government can borrow credibly, raise revenue progressively, and spend in ways markets and voters will tolerate, including in years when some of the rents that funded easy revenue packages are fading.
This is the load-bearing wall of the series.
The scale
Add the delivery commitments together and you reach roughly GBP 72-91 billion[1] per year at central estimate, depending on how fast housing scales. Against UK GDP of roughly GBP 2.8 trillion[4], that is about 2.6-3.3%[2]. Large, but compare it to the GBP 40 billion energy cap package in a single year in 2022[5], about 1.4%[3] of GDP in that same year, or GBP 100 billion in debt interest already on the books[6].
Industrial strategy capital sits outside that annual total because it is project-by-project investment over years, not a standing benefit line.
What is in the headline total, and what is not. The roughly GBP 72-91 billion[1] per year figure sums the seven delivery chapters in the deep-dive table (food through justice). Companion chapters on councils delivery, skills and FE, and the Resilience Corps architecture in Governance are outside that headline until a later costing pass, deliberately: council delivery support (low billions, phased with housing and retrofit), skills (order of GBP 2-4 billion per year as industrial projects commit), and Corps running costs (largely within council and civil-service capacity lines, not a separate billion-pound vanity line). Do not double-count them against the delivery total. Large adaptation capital (flood defences, water storage, managed retreat from zones insurers are already leaving) is also outside the programme total; it is treated below as standing fiscal load the framework must acknowledge, not as hidden headroom.
The question is not whether the number is big. It is whether doing nothing is cheaper. Managed decline has its own invoice: rising housing benefit, emergency NHS spending, crisis packages, and a state that cannot respond to the next shock without another improvisation. When the US accelerates long-lived fossil infrastructure abroad, that invoice grows in categories the delivery chapters only partly address:
- Adaptation and flood capital: defences, drainage, water storage, and retreat from zones insurers are already leaving.
- Insurance and mortgage retreat: uninsurable coastal and flood-prone stock becomes a public liability (emergency housing, blighted council tax, forced sales) unless adaptation is funded.
- Defence and patrol: Hormuz, the Arctic, and sea-lane protection as permanent load, not a one-off Gulf spike.
- Climate-migration settlement: casework, housing, and integration capacity whether or not the politics admits the label.
These are not fully priced in the GBP 72-91 billion[1] delivery stack. They are why honest revenue and borrowing architecture cannot assume a growth rebound pays for them later.
How you pay for it
This programme does not pretend tax rises are optional. Revenue should fall mainly on those with the greatest capacity to pay: hardened windfall levies on energy profits, a corporate surcharge on very large UK profits, stronger HMRC enforcement against offshore wealth, inheritance and high-value property reform, and carbon pricing with a household dividend.
Fully implemented, honest revenue yields roughly GBP 20-35 billion[7] per year. Real money. Not enough alone.
Post-peak revenue stack. Be explicit about what survives after extraction rents and property booms fade. Durable under contraction: progressive tax on incomes and profits that still exist, inheritance reform, high-value property banding where values hold, HMRC enforcement against avoidance, carbon pricing with a household dividend. Cyclical or fading: energy windfalls timed to price spikes, financial-sector and extraction surcharges that assume 2010s-style rents, council-tax uplifts that depend on ever-rising paper wealth. Stack the Budget with the durable lines first; treat windfalls as upside, not structure. Do not write the programme as if 2022-style energy profits or permanently rising property paper wealth fund half the state forever.
Do not assume US climate finance returns. The largest historical contributor to global climate finance is not a reliable partner under current US doctrine. Loss-and-damage flows, Just Energy Transition partnerships, and cooperative clean-tech supply chains cannot be counted in the revenue or borrowing plan. If they return after 2028, that is upside; the framework must stand without them.
The gap is closed by borrowing, and by being clear what borrowing buys. Homes, grid capacity, demand reduction in the built stock, and reserves are assets. Borrowing for them is not the same as borrowing for unfunded tax cuts. NHS workforce, social care, and social security are current spending with diffuse but real returns. Markets care less about diffuse returns, which is why the revenue side must be credible.
Borrowing for investment is harder when energy use, food systems, and industry are flat or falling than when they are expanding. The tax base that serviced past debt does not automatically grow to match new borrowing. Markets still distinguish assets from consumption. They also ask whether the state can service debt if those three stay stuck. So borrow for assets that cut future crisis spend: retrofit, grid, reserves, social housing that replaces rising housing benefit. Sequence it. Stress-test it. Build adjustment triggers that treat harder revenue years as normal. The deep dive models climate-shock, insurance-retreat, and higher-defence paths on top of the gilt scenarios.
Housing uses infrastructure borrowing headroom; health uses revenue expansion. Sequenced honestly, they are not in competition.
The Truss question
Every serious fiscal plan since September 2022 lives in Truss's shadow. That mini-budget failed because markets concluded Britain was borrowing for consumption without credible revenue or growth.
This programme is structurally different: progressive revenue, infrastructure borrowing with identifiable assets, and pre-committed adjustment triggers if forecasts miss or gilt yields spike. Protect social security from cuts; defer non-urgent capital; accelerate highest-yield revenue. Markets are told what happens if their fears materialise. Under a tighter operating environment those triggers matter more, not less: they are how you keep credibility when some extraction revenues fade.
The Bank of England is not a backstop for weak plans. Gilt credibility rests on fiscal substance.
Who pays and who benefits
A framework that funds progressive spending with regressive revenue collapses politically in year two. Inheritance reform targets concentrated wealth. Council tax reform hits high-value property, not ordinary semis. Corporate surcharge falls on firms with pricing power.
Households that benefit most from food, energy, housing, and health spending need to see a reciprocal deal: we fund this together, progressively, and life materially improves, even when the national cake is not growing the old way.
Full programme tables, revenue yield assumptions, line-by-line inaction costs, and stress tests are in the Fiscal Framework: Deep Dive.
The Next Piece
Arithmetic is necessary but not sufficient. Crises still get routed through committees designed for calm Tuesdays. Governance covers decision architecture; Civil Service covers whether anyone is available to execute.
Read next: Governance.