A Modern Britain That Works

Foreword

We gave the world the Industrial Revolution, the NHS, the World Wide Web, the English language, the common law, and more Nobel Prize winners per capita than almost any nation on earth. Our financial markets, our universities, our creative industries, our life sciences sector and our legal system are genuinely world-class. London remains one of the two or three most consequential cities on the planet. These are not myths. They are facts.

And yet Britain has talked itself into a malaise. The 2008 financial crisis, a decade of austerity, the divisiveness of Brexit, the trauma of Covid, the cost of living squeeze, and the slow erosion of public services have left the country frayed and the political conversation bitter. Too many politicians have exploited that bitterness rather than addressed its causes. Too many have offered grievance where they should have offered solutions.

We reject the idea that Britain is broken. But we accept that it is underperforming — badly and unnecessarily. The gap between what this country is capable of and what it is currently achieving is not inevitable. It is the product of political choices: choices to tax work and movement instead of the things that cannot leave, to prize complexity over simplicity, to borrow instead of reform, to manage decline instead of driving growth, and to tell people what they wanted to hear rather than what they needed to know.

Future Forward Foundation was created to offer something different: comprehensive, costed policy ideas built not on class or envy but on opportunity, growth, and an honest assessment of Britain’s genuine strengths and real weaknesses. Every figure in this policy paper is drawn from public data. Every policy has been modelled against current baselines. We invite challenge and refinement — that is what serious policy debate looks like.

Our politics is not left. It is not right. It is just forward. It is not nostalgic and it is not revolutionary. We believe Britain has too much going for it — too much history, too much talent, too many global connections and too many unrealised opportunities — to keep having the same tired arguments about the same tired ideas.

This policy paper is our contribution to a different kind of conversation. One that starts from Britain’s strengths, is honest about its challenges, and offers a coherent path to a country that works better for everyone who lives and works in it.

It is time to get back on our feet.

Executive Summary

Twenty-two chapters. Costed from public data and subject to independent verification. Honestly presented. Here is the whole programme in plain English — every number in this summary is backed by the full fiscal framework that follows.

The Goal

Growth. Everything in this document serves it — because growth is the only honest route out of Britain’s fiscal hole, and the only one that does not run through austerity or decline.

• Tax that is simple and transparent, built to fund growth — not to engineer how people live
• Immigration built to benefit the economy — open to talent, closed to entry that bypasses the rules
• A state that builds the conditions for wealth creation — and asks a fair return from it

The Tax Philosophy

One idea runs through every tax change here: stop taxing work and movement; tax what stays put and what is earned but escapes tax today. We abolish the taxes that punish doing the right thing — National Insurance on working, the employer tax on hiring, business rates on premises, stamp duty on moving, council tax on simply living somewhere — and lean instead on the taxes that do least damage to growth: profits, a floor under the income of the very wealthy, and consumption. Britain should be the easiest place in the world to work, hire and build, and should raise its revenue from the things that cannot pick up and leave.

The Five Big Ideas

• A tax system that is simple, fair, and only cuts when Britain can afford it: nothing on your first £15,000, one honest rate on your payslip, and National Insurance abolished. Every tax cut comes in stages, and only when the public finances can bear it — no unfunded giveaways, no promises we cannot keep.
• The NHS and social care, funded properly and built to last: a dedicated health contribution, half of it ring-fenced for named NHS improvements and half replacing general-taxation health funding — and we show both halves, and a National Care Guarantee that caps lifetime care costs at £86,000 for everyone — so the health service has stable money and no one is forced to sell their home to pay for care.
• Britain back at the heart of the European market: back in the Single Market — but not the EU, keeping our own trade policy, currency and foreign policy, and our domestic law outside the scope of the agreement. We should be precise about the part that is not ours: EEA law is dynamic, relevant EU acts are continuously incorporated, and Single Market matters would be supervised by the EFTA Surveillance Authority and adjudicated by the EFTA Court. The friction of Brexit removed from trade with our largest market, and the freedom for Britons to live, work, study and retire across Europe again.
• The industries of the future, built and owned here: a serious plan for growth — Britain investing in its own companies through a national wealth fund, an end to paying some of the highest energy prices in Europe, and a real, sovereign stake in the industries that will define the century: AI, life sciences, and the capacity to build our own defences. Wealth and security created here and owned here, not just bought from abroad.
• An immigration system that works — legal routes open, illegal routes closed: migration built around what the economy needs, paired with a serious plan to stop illegal Channel crossings by negotiating back the right to return arrivals. We welcome the migration that builds Britain and end the migration that bypasses the rules.

One Simple Tax

• First £15,000 of income: completely tax-free
• One flat rate above it: a 38% income tax, gated down toward 35% as the gates allow, after debt reduction — and one published combined rate of 42% (income tax, local income tax and the Health Levy together), falling to 39%, so the figure on your payslip is never more than the one we print
• Employee National Insurance: abolished outright
• Employer NI replaced by an 8.5% Employer Levy — roughly half the old rate, with a training rebate built in; the structure arrives in Year 2, and the core rate step-down that halves the charge is taken in certified increments as headroom appears, mostly in Years 4 and 5, rather than promised to a date
• Council Tax replaced by a hybrid that takes the best of both worlds — a 2% local income tax that rises with ability to pay, plus a 0.3% annual property charge that gives local services a stable base, with low-income owners in modest homes exempt outright; the 2% Health Levy is the largest new charge on income, and it is not the only one — the individual care contribution above £150,000 and the Landowner Levy on commercial freeholds are new too, and all three are named in the scorecard rather than folded into a replacement
• Corporation Tax 30% with permanent full expensing — new investment effectively untaxed — and the global minimum-tax floor closing artificial profit-shifting; Business Rates abolished for every occupier, with a 1% Landowner Levy on large commercial sites so the gain reaches businesses, not their landlords
• Stamp Duty abolished: your home’s gain is deferred to death and never taxed for moving; second homes and shares are taxed as the investments they are
• The ISA maze replaced by a universal £15,000 tax-free gains allowance and £5,000 savings allowance — the same £15,000 that starts the income tax scale, so there is one number to remember and no wrapper to open; existing ISAs protected
• Every threshold CPI-indexed — fiscal drag ends

The Honest Numbers

• The deficit falls from 4.4% of GDP in Year 1 to 1.6% by Year 5 — a fall of £84bn in cash, from £142bn to £58bn, and a bigger fall than the OBR's own £56bn from a lower starting point. It ends level with the official forecast rather than below it, because the £47bn-a-year growth dividend is spent on the reform programme instead of on a smaller headline number
• Year 1: the package costs about £27bn on static policy, and F³ borrows about £27bn more than the OBR's published path once the first-year growth dividend is netted off — borrowing is higher in Year 1, not lower, because the Health Levy's spending half lands before growth arrives and every additional rate reduction beyond the initial structural package waits for certified headroom. The gap closes through Years 2 to 4 and the central case ends level with the official forecast
• Year 5: borrowing of £60bn (low), £58bn (central), £41bn (high) against the OBR's published £59bn — the central case essentially level with it — £58bn against £59bn, the high case well below, and the low case £1bn above it. We claim no more than that
• A glide path enforced automatically: if growth disappoints, the next tax cut waits
• Everything quantified is in the scorecard — defence, the EEA fee, the NHS, the courts — and the liabilities we have not been able to cost are identified there too, not omitted. Find a quantified promise missing from the table and we correct it in public

The National Wealth Floor

• Anyone worth over £10 million pays at least 25% of their economic income in tax — counting the borrowing-against-assets the very rich live on. Pay your share already and nothing changes. Not a wealth tax — a floor under contribution
• No exemptions and no forced sales: defer against illiquid assets, or pay in shares
• An exit charge and trailing liability stop people leaving to dodge it; a ten-year runway keeps arriving talent
• Scored at an honest £5bn a year — a third of campaign-group claims

Europe — Not Reversing Brexit

• Rejoin the Single Market — not the Customs Union, so Britain keeps its own trade policy and deals. Stay out of the EU: our own currency, foreign policy and trade deals, and domestic law outside the agreement’s scope — with EEA-relevant rules supervised by the EFTA Surveillance Authority and the EFTA Court rather than the European Court of Justice
• We distinguish throughout between legal migration that supports economic growth and illegal migration that bypasses the rules — welcoming the first, ending the second
• Free movement returns, and we say so plainly
• Before Brexit, small-boat crossings were measured in hundreds: 299 in 2018. By 2022 they reached 45,774, and around 41,000 in 2025. Brexit did not create illegal migration, but it removed Britain’s most effective legal route for returning many Channel arrivals — the EU’s Dublin system, with nothing put in its place
• Our central negotiating objective is an EU returns agreement — Single Market accession is the leverage that makes it likely, not a guarantee. No gimmicks, no Rwanda
• Trade deals continue through EFTA; CPTPP and existing deals kept, though we are candid they are worth a fraction of single-market access
• The membership fee — about £3bn a year — is in the scorecard, against a dividend worth many times more
• If Brussels says no or moves slowly, the tax and domestic reforms that come first need no one’s consent but Parliament’s, and even our no-deal scenario beats doing nothing

Health and Care

• NHS free at the point of use, with the 2% Health Levy raising around £20bn a year — about half of it additional, ring-fenced money for named NHS improvements, about half replacing general-taxation health funding; the one tax rise, named as such, and split openly rather than counted twice
• Australian-style model: support to go private if you choose, public capacity freed if you do not
• NHS dentistry rebuilt; free degrees for doctors, nurses and dentists in return for five years of UK service
• Social care pre-funded at last: a National Care Guarantee capping lifetime care costs at £86,000 for everyone, financed by a pooled National Care Fund rather than individual accounts

Education and Families

• Free tuition for medicine, engineering, computing, maths, physics and life sciences — five-year UK service obligation
• The 18–22 Guarantee: an apprenticeship or further-education place for every young person — leaving school into nothing stops being an option
• A fully transferable tax-free allowance between married parents — up to £30,000 sheltered in a single-earner family with a child
• Parental leave made fully shareable and flexible by the day: same pay, more freedom
• One childcare entitlement replacing three schemes — providers funded directly at independently audited delivery cost, delivered through school buildings where possible
• Child Benefit delivered as food, uniforms and holiday support through schools
• A four-week summer holiday; a £5bn capital programme for SEND schools

Housing, Planning and Land

• A presumption in favour of development; scheme-killing affordable-housing quotas scrapped
• A 2% Land Banking Tax on permissioned-but-unbuilt land, ring-fenced to build social housing
• 300,000 homes a year by Year 3 — with the honesty that immigration openness is reviewed if supply lags

Immigration

• EU free movement via the Single Market — and, for talent from beyond the EEA, £120k+ earners fast-tracked in 14 days plus a US Talent Welcome programme
• Illegal entry: the negotiated right to return small-boat arrivals (see Europe), plus first-safe-country reform within the Conventions — tightening the system as comparable European countries do, not walking out of the treaties

Energy and Transport

• Small modular reactors, deep geothermal, gas as the bridge — and locational pricing so we stop paying wind farms to switch off
• North Sea oil and gas taxed but not wasted: a stable 78% on existing fields, 43% on new production to keep it flowing — and every pound from new fields saved in the British Future Fund, Norway-style, never spent. We don’t bank a penny of it in our deficit numbers
• One flat Road Charge replaces car tax and pay-per-mile (£450 cars and vans, less for bikes) — funding free parking at hospitals, town centres and railway stations, and clearing the pothole backlog in four years
• A £75/month national Travel Pass covering rail and bus — a labour-mobility measure, honestly part-subsidised. 80mph motorways. Fuel duty frozen, no mileage tracking

Defence and Security

• 3% of GDP this parliament, on a statutory path to NATO’s 3.5% core commitment by 2035 — the day-to-day cost (£3bn rising to £10bn a year) in the scorecard, equipment on Defence Bonds — which are government borrowing and score as debt, whichever spending rule they sit outside; the instrument changes who lends and on what terms, not whether the liability exists
• Buy British through the treaty exemption France and Germany already use

Pensions

• The Pensioner Guarantee: a state pension that always rises with inflation, protected in law so it never falls in real terms — plus a five-yearly earnings review. Inflation-proof for life, replacing the Triple Lock’s unaffordable ratchet
• Employer pension and care contributions raised from a 3% minimum to 5% — 3% to the pension as now, and 2% ring-fenced to the National Care Fund; with employee contributions on top, total retirement-and-care saving reaches about 10% of qualifying pay
• £60,000 annual allowance, no taper, no lifetime cap; salary sacrifice ends
• The 25% tax-free lump sum can be taken once at any age — so the pension can fund a first home as well as retirement

Welfare

• PIP assessments back to face-to-face; awards time-limited by default
• Universal Credit taper cut from 55% to 45% — scored as the cost it is
• Mental-health treatment expanded BEFORE eligibility tightens — sequencing in statute
• Net savings of £4–6bn by Year 5 from PIP alone — half the usual headline claim, because we scored it properly; with UC and work-capability assessment the combined scorecard row reaches £8bn. That £8bn is our own estimate and subject to independent verification: no published costing exists for raising face-to-face assessment from 6% to 50%, which is our central mechanism. The nearest published comparators are for a different lever — a minimum four-point score in one daily living activity — which the OBR costed at £3.9bn by 2029-30 and called a highly uncertain judgment, against a £5bn Green Paper package and around £6.4bn including the Universal Credit health element. Our figure sits above that range because our package is wider, not because we think the OBR is wrong

Justice

• £0.7bn a year criminal justice recovery — £2bn of court capital over the parliament, legal aid rates up 20%, and courts sitting to the maximum capacity the judiciary can supply. Police back to 2010 strength
• Medical cannabis on the NHS. No to decriminalisation

Technology and Society

• Algorithmic accountability, not source-code publication — auditable risk assessments, disclosure of material ranking factors, independent researcher access, testing for discrimination and manipulation, incident reporting and a user right of appeal; chronological feeds available by default; firm under-18 protections
• Platforms jointly liable for what they are paid to promote and what their algorithms amplify — amplification is publication; neutral hosting keeps protection, with teeth. Liability for foreign-state interference. Digital literacy in every school

Economy and Markets

• VAT two-tier for small firms — full exemption to £90k, a simple 8% flat rate to £350k — so growth never hits a cliff edge
• A British Growth Exchange — a dedicated London market for high-growth companies, with listing rules built for scale-ups, so the next generation floats here instead of New York
• Inheritance tax: nothing below £3m; above it, the flat-tax rate — 38% falling to 35% with it. Most family homes and ordinary farms fall below the threshold; and where the bill falls on genuinely illiquid business or farm assets it can be paid over ten years rather than all at once, as Business Relief becomes a deferral rather than a total exemption — so no family firm or farm is broken up to pay it
• Clean rivers through binding pollution limits and real enforcement on every operator, public or private — with failed companies like Thames Water taken into public ownership through insolvency, at their true worth
• Food standards non-negotiable in trade deals. Tourist VAT-free shopping restored. Farming moved from the ELMS cliff edge to simple public-goods payments

Every quantified measure shown. Every material uncosted liability separately identified. Every loser named. Every critic answered before they speak.

Not left. Not right. Just forward.

What Britain Is Good At

Every claim in this paper rests on one proposition: that Britain grows fastest by removing the obstacles in front of the things it is already the best in the world at. Not by picking new winners. By stopping the penalising of the ones we have.

The comparative advantage we already have

Britain is a services economy, and not marginally. Services are 72.4 per cent of UK value added — one of the highest shares in the G7 — and in 2024 Britain exported roughly £508bn of services to the rest of the world. That is not a weakness to be corrected by reindustrialisation nostalgia. It is the thing we do better than almost anyone.

The numbers are worth stating precisely, because they are the foundation of everything that follows. Britain is the world’s largest net exporter of financial services, with a surplus of $127bn in 2024 — ahead of the United States on $64.2bn, and larger than Singapore, Switzerland and Luxembourg combined. Add legal, accounting and business advisory and the combined financial and related professional services trade surplus reaches £119.1bn. The same industry contributed £281bn of gross value added, roughly £12 in every £100 of national output, and paid £110.2bn in tax — the largest of any sector.

It is not a London story, and the paper will not pretend otherwise for rhetorical convenience. Around 2.5 million people work in it — one worker in thirteen — and two thirds of those jobs are outside London, in Birmingham, Edinburgh, Leeds and Manchester. Britain is also the world’s largest exporter of insurance and pension services, the second largest exporter of professional business services, and London handles 38 per cent of global foreign-exchange turnover.

And here is the fact that makes this a growth strategy rather than a boast. Financial services productivity is £104.10 an hour against a whole-economy average of £39.30 — two and a half times higher. In an economy whose central problem is flat productivity, growth in this sector does not merely add output; it raises the average. A percentage point of growth here is worth more than a percentage point almost anywhere else.

The constraint is access, and it is self-imposed

If that is what we are good at, the question is what stops it growing. The answer is not talent, capital or law — Britain has all three, and the common law itself is an export. It is market access for services, and it is the one obstacle we built ourselves.

The Trade and Cooperation Agreement is a goods agreement with a services chapter attached as an afterthought. It does not deliver mutual recognition of professional qualifications, the right of establishment, or anything resembling passporting. For a country that sells goods, that is survivable. For a country that sells services, at two and a half times the national productivity rate, it is the single largest self-inflicted constraint on growth in the British economy. The EU still takes 31 per cent of financial and related professional services exports; the barrier sits directly across our largest concentration of comparative advantage.

This is why Single Market accession is not a Europe policy in this paper. It is the growth policy. It is the only available measure that removes a first-order constraint on the sector where Britain is world-leading and most productive — which is why it carries the largest single number in our fiscal model, and why we phase it honestly rather than book it early. Chapter 4 sets out the terms, the costs we accept and the sovereignty we do not pretend to keep.

Taxes that fall before profit

The second constraint is domestic and simpler. Britain taxes business activity before it taxes business success. Business Rates fall on premises whether or not a penny is earned in them. Employer National Insurance falls on jobs whether or not the job is yet profitable. Both are levied on the inputs to growth rather than its results, and both are why a shop can be busy and still fail.

That is the reasoning behind two measures the reader will meet later and might otherwise read as simple giveaways. Abolishing Business Rates and replacing employer National Insurance with a lower flat Levy are not tax cuts for their own sake; they are the removal of taxes that are charged before a business knows whether it has made anything. Corporation Tax rises to 30 per cent in exchange — a tax on profit actually earned, which a growing business can pay and a struggling one cannot be killed by.

The effect is not marginal. A high-street shop with £400,000 of turnover, eight part-time staff and £35,000 of profit is better off by around £7,700 a year — about a fifth of its previous profit, because abolishing Business Rates is worth £14,000 to it while abolishing the Employment Allowance takes £6,000 of that back — almost all of it from killing two taxes that took no account of whether it was making money. A twelve-site hospitality group gains roughly £1.3m. Those examples are set out in full, with the Corporation Tax rise netted off, in the business chapter.

Simplify, because complexity is itself a tax

The third constraint costs nothing to remove and is the one politics never touches. Britain’s tax and benefit system imposes an enormous private cost in accountants, advisers, software, forms and time — a deadweight that produces no revenue for the state and no value for the payer. One income tax rate above one threshold, one infrastructure charge instead of two negotiated ones, one childcare entitlement instead of three overlapping schemes, one savings allowance instead of a wrapper regime: none of these raises money by itself. All of them return time and certainty to the people and businesses making decisions.

Certainty is the part usually omitted from growth papers. A business does not invest against a tax code it expects to change in eighteen months. That is why every threshold in this programme is indexed to CPI, why every gated step is pre-legislated, and why the sequence is published in advance and certified by the Office for Budget Responsibility rather than announced at a Budget. Taking the politics out of the schedule is itself a growth measure, and it is free.

What we claim, and what we do not

We should be exact about the size of the bet, because the whole fiscal case turns on it. Strip the growth dividend out and this programme leaves borrowing about £36bn a year higher than it would otherwise be by Year 5 — £35.4bn of static policy cost, shown as £35bn in the table after rounding, and £1bn of additional interest — on top of the £10bn of threshold-freeze receipts we hand back by indexing. With the dividend, the central case ends the parliament level with the OBR's published path. There is no version of this paper in which the numbers work without growth, and we would rather write that sentence ourselves than have it written for us.

So we book the conservative end, and we should be precise about what the central case assumes, because an earlier draft of this passage gave a decomposition that did not match the model. The central case assumes 4.3 per cent cumulative additional GDP by Year 5, and it breaks down as the bridge in Chapter 2 sets out: 1.0 point from employer tax reform, 0.9 from planning, housing and infrastructure, 1.8 from Single Market accession phased from Year 3, and 0.6 from labour supply and free movement. Those four figures sum to 4.3 and are the only decomposition this paper uses. An earlier draft attributed 2.2 points to the Single Market and phased them 0.7, 1.2 and 1.6, which was inconsistent both with itself and with the £47bn the table books; it has been withdrawn. Published estimates of the long-run cost of leaving the Single Market cluster around 4 to 5 per cent of GDP, so recovering 1.8 points of it within five years is at the cautious end of that range, and it is a trade-and-investment figure only — the labour channel is counted separately and not twice. We are less cautious about the domestic component, and a reader is entitled to discount it: 4.3 per cent in five years is an ambitious central case, and the low scenario exists precisely to show what happens if it does not arrive. That scenario assumes accession never happens at all, so it books no Single Market growth and no free-movement labour supply, reaches 1.9 per cent from the unilateral reforms alone, and still ends close to the official forecast — because the tax cuts that depend on the dividend simply never fire.

And we do not claim the things we cannot measure. The productivity gain from a simpler tax code is real and unbooked. The output effect of clearing the road-maintenance backlog is real and unbooked. The health saving from private insurance uptake is real, small, and booked at zero. A growth paper that books its own optimism is not a growth paper, it is a prospectus — and the difference is exactly what this document is trying to establish.

Fiscal Framework: The Numbers

Every policy in this policy paper is costed against public data — including the items that are usually left out: the defence spending commitment, the EEA membership contribution, the NHS capacity investment that welfare reform requires, and the realistic first-year yield of every phased measure. The scorecard below shows two columns: the honest Year 1 position and the steady-state position once all phased measures are fully in force. The rows sum. Check them.

Current Position (2025-26 Baseline)

Total government receipts: ~£1,140bn | Total spending: ~£1,270bn (receipts plus borrowing) | Deficit: £129.8bn — the ONS provisional estimate for the financial year ending March 2026, 4.2% of GDP, £2.9bn below the £132.7bn the OBR had forecast | National debt: 94.1% of GDP (ONS, end July 2026) | GDP: ~£2,950bn

The distinction every budget debate muddles, settled once: the DEFICIT is this year's borrowing — the gap between what the state spends and what it raises — most recently £129.8bn, the ONS provisional estimate for the financial year ending March 2026, or 4.2% of GDP. The DEBT is the accumulated stock of every past deficit: about £2.8 trillion, or 94.1% of GDP. One is the overdraft; the other is the mortgage. Our rules bind the deficit, year by year. Our destination — debt below 60% of GDP within a generation — is what falling deficits eventually deliver.

The deficit is one year’s borrowing; the debt is the accumulated stock. F³'s rules bind the deficit.

The Architecture: Pre-Legislated Steps, Each Independently Gated

F³ is not a single fiscal event. It is a sequence of pre-legislated steps, each of which proceeds only when the Office for Budget Responsibility certifies, in its published forecast, that the public finances are on track — with the National Audit Office auditing after the event whether the gate was in fact honoured:

• Step 1 (Year 1): Employee National Insurance abolished entirely. Employer NIC begins its replacement — rate cut from 15% to 12.25%, including the 1-point earn-back Training Point. Business Rates abolished for retail, hospitality, leisure and schools. The 38% flat rate, Health Levy and all revenue measures take effect.
• Step 2 (Year 2): Employer NIC is abolished outright and replaced by the Employer Levy. Two things happen here and we separate them, because only one of them is a promise. The structure changes on Parliament's timetable: NIC goes, the Levy arrives, the Training Point starts earning back. The core rate is a different matter. Moving it from 11.25% to 7.5% costs about £32.8bn a year, and together with the flat-rate cut it makes up £52.4bn of gated giveaway. It is therefore taken in certified increments as headroom appears, not delivered to a date. Our own scenario tables show what that means in practice, and it is slower than the label 'Step 2' suggests: in the central case almost nothing is certified before Year 3, and £29bn of the £52.4bn has fired by Year 5, taking the core to about 7.9% — a headline of about 8.9% with the Training Point — rather than the 7.5% and 8.5% that full implementation would deliver. The balance follows in the next parliament as headroom allows. In the low case none of it fires at all and the Levy stays at its Year 1 rate for the whole parliament. Earlier drafts of this paper described this step as sequenced rather than gated, and then said the step-down arrives in Year 2; both were inconsistent with our own tables. Second tranche of Business Rates abolition. The National Wealth Floor begins.
• Step 3 (Year 3): Business Rates fully abolished. Single Market accession completes; EEA contribution begins.
• Step 4 (Year 4): Flat rate cut from 38% toward 35% — and the cut need not arrive in one jump. It descends gradually, in increments as small as the public finances require, each step proceeding only on OBR certification, so the rate falls exactly as fast as the headroom allows and no faster. In time, and on the same gated basis, it can fall further still, provided the deficit stays on its statutory glide path.
• What 'certified headroom' means, and one circularity we have to close. Headroom is certified by the OBR in its published forecast, and the test is applied on a PRE-MEASURE basis: the OBR must find the headroom before, and without counting, any growth attributable to the measure being activated. Without that restriction the mechanism could partly certify itself — an assumed hiring response to the Employer Levy cut would appear as growth, that growth would create headroom, and the headroom would license the cut that generated it. That is not a gate, it is a loop, and a reader is entitled to have seen us close it. So: the dynamic effects of a gated measure may be counted in the forecast AFTER it has fired, and toward the next increment; they may never be counted toward certifying the increment that produces them. Each increment is tested separately, on the forecast as it stands before that increment is assumed. The NAO audits after the event whether the headroom the OBR certified actually materialised.
• Beyond this parliament: once the 1% structural deficit target is met, the core Employer Levy reduces in further gated steps — funded from certified headroom, never from borrowing — with full abolition as the long-term destination.

If growth disappoints, the next step pauses — automatically, by statute, without a political crisis. This gating is the enforcement mechanism for the fiscal rules in Chapter 2, and the reason the LOW scenario below cannot run away: in that world, Step 4 simply does not fire until the numbers allow it.

We are deliberate about how this is presented, because an opponent will call a conditional tax cut a fictional one. The honesty is the point. Every other manifesto promises tax cuts as certainties and then quietly abandons them when the money is not there — breaking faith after the election. F³ does the opposite: it tells you in advance that the 35% rate is earned, not gifted, and names the exact test it must pass. A promise you can audit is worth more than a promise you cannot. If you want the rate cut guaranteed regardless of the public finances, no honest party can offer that — and the ones who say they can are the reason Britain keeps electing governments that miss their own targets. We would rather under-promise the rate and over-deliver the discipline.

Which Cuts Come First, and What Each Costs

Because the cuts cannot all happen at once, honesty requires saying which come first and what each one costs the Exchequer when fully in. The ordering is not arbitrary: the day-one measures are the ones that do the most for work and growth and are funded by the revenue measures arriving alongside them; the gated measures are sequenced by how much fiscal headroom each needs and how much growth each unlocks per pound forgone. The table below is the priority list an opponent can hold us to.

Two principles govern the order. First, nothing that costs money jumps the queue ahead of the deficit glide path. Certified headroom is applied to debt reduction first; only then do the gated cuts fire — and in strict order, the Employer Levy step-down before the flat-rate cut, which comes last of all. Each fires only when the Office for Budget Responsibility certifies, in its published forecast, that the headroom exists — the OBR because headroom is a forecast judgement and that is its statutory job, with the National Audit Office auditing after the event whether the rule was in fact complied with, which is the NAO's. That gate is why the LOW scenario simply never reaches them. Second, within that constraint we front-load the measures with the highest growth return per pound: cutting the cost of employing people and occupying premises does more for output, sooner, than a headline rate cut, which is why the rate cut waits until last, after debt reduction and the payroll and premises reforms. The 35% rate and the Levy step-down are the genuine luxuries — desirable, costly, and explicitly last in line.

One point of honesty the table makes explicit: because F³ aligns the headline rates, the single cut from 38% to 35% is really three cuts at once — income tax, capital gains tax and inheritance tax all step down together, since the CGT taper top and the IHT estate rate are both pinned to the income rate. That is deliberate: aligned rates are what kill the avoidance games of disguising income as capital or sheltering it in an estate. But it also means the gated Step 4 carries a slightly larger price than the income-tax figure alone suggests, and we would rather show that here than let a reader discover it as a hidden extra. The same statutory gate covers all three: if the headroom is not certified, none of them fire.

F³ Tax and Spending Changes — Full Scorecard

The scorecard is presented in two layers so that both jobs get done. Below is the summary: nine blocks, each of which is the sum of its own measures, and a net line that is the sum of the nine. Every figure in this table is generated from the fiscal model, and the column adds up — you can check it with a pencil. The full line-by-line table, all eighty measures with their notes, is in the Scorecard Annex at the back; the block nets here and the measure lines there are the same numbers, because they come from the same file.

One reconciliation the reader will want, because the two tables in this paper do not show the same number and should not. The scorecard above nets to +£58.8bn in Year 5, and +£26.6bn in Year 1; the central scenario shows a static policy cost of +£35.4bn, shown as £35bn in the table after rounding. The £23.4bn difference is not an error — it is the gates doing their job. The scorecard prices every measure at full implementation, which is the only honest way to publish a costing: it shows what the programme costs if all of it happens. The scenario shows what actually fires under debt-first sequencing, where certified headroom pays down debt before it funds giveaways. The two measures that account for the whole of the gap are the Employer Levy core step-down from 11.25% to 7.5% (£32.8bn at full implementation) and the flat-rate cut from 38% toward 35% (£19.6bn). Together they are £52.4bn of gated giveaway; in the central case £29bn of that has been certified and fired by Year 5 and £23.4bn has not. What that means in practice is set out plainly rather than left in a table: the Levy core reaches about 7.9% rather than 7.5%, and the flat rate does not fall below 38% at all within this parliament. Each scenario table below carries a line showing how much has fired in each year. So: the scorecard is the price list, the scenarios are the purchase orders.

Against the OBR's published path, the High case ends about £18bn below it, the central case essentially level, and the Low case about £1bn above. F³ borrows more than the official path in the first two years in every scenario. By Year 3 the High case has already crossed below it (£82bn against £86bn) while the Low and central cases are still above. The shaded gap is the Base-case growth dividend.

Year one borrowing is higher, not lower: revenue lands in full but so does the Health Levy's spending half, while the growth dividend has barely started and every gated cut is still waiting for certification. Against the OBR's 2026-27 forecast of £115bn, F³ borrows £142bn — £27bn more. That gap closes through Years 2 to 4 and the central case ends the parliament level with the official forecast.

The Path Forward — Five Year Projections Across Three Scenarios

All scenarios in this paper are now run against the OBR's own published path, not against a baseline of our own construction. Earlier drafts used an internal counterfactual in which the deficit rose to £157bn by Year 5. We have retired it. It sat £33bn above the official forecast in Year 1 and £98bn above it by Year 5, and while the threshold freeze we hand back explains roughly £10bn of that gap, the remainder rested on assumed spending pressures we believe in but cannot source to anybody's published figure. A baseline that flatters us by £98bn and cannot be checked is not a baseline; it is an advantage we awarded ourselves. So the comparison throughout is the OBR's March 2026 forecast — borrowing falling from £132.7bn in 2025-26 to £59bn by 2030-31 — with one adjustment stated on the face of the table: we add back the receipts from the threshold freeze, because F³ indexes the £15,000 threshold to CPI from day one and therefore cannot claim money it is deliberately giving away. That adjusted line is our starting point, and every scenario builds from it. The elasticity we use is the standard ~£11bn of receipts per 1% of GDP — consistent with the OBR's published fiscal-forecast relationships and HMRC's own ready-reckoner sensitivities, and deliberately treated as a round approximation rather than a precise coefficient, since the true figure varies with the tax mix, inflation and the composition of growth. The Single Market dividend is phased honestly: accession completes in Year 3, so no SM growth is booked before Year 3 in any scenario — the 2-4% literature describes a long-run effect, and we treat it that way. We are deliberately conservative about the pace: the central case books 1.8% by Year 5 against a long-run effect the literature puts at 2-4%, a partial realisation precisely because front-loading a fifteen-year effect into a five-year window is the commonest way fiscal plans flatter themselves. The Low case books no Single Market growth at all, because in that scenario accession never happens; it reaches 1.9% of GDP by Year 5 from the unilateral measures alone. If even the Low trajectory is too quick, the gates do their job — the rate cuts that depend on the dividend simply wait until it arrives. The plan does not require the growth to be fast; it requires it to be real.

The Bridge: From the OBR’s Published Path to Ours

A reader is entitled to insist on this table. It starts where the official forecast starts and shows every step to our own numbers, so that nothing is hidden inside a baseline. It is also the table from which the three scenarios below are built: their opening line is this table's third row.

Read the bottom line before anything else, because it is the honest shape of this programme and it is not the shape a manifesto would choose. Against the OBR's own published path, F³ borrows about £27bn more in Year 1, £16bn more in Year 2 and £7bn more in Year 3, is £2bn below in Year 4, and ends essentially level in Year 5 — £58bn against £59bn. The high case ends about £18bn below it; the low case ends about £1bn above it. Revenue measures land first, the Health Levy's spending half lands next, and the gated giveaways and the growth dividend arrive last. We make one claim for this programme's fiscal effect and no more than one: it holds the deficit level against the official forecast while pre-funding social care, cutting business taxes and rejoining the Single Market. It does not beat that forecast on the central case, and it does not beat it at all in the low case, which ends £1bn above — and we are not going to pretend otherwise by choosing a friendlier baseline.

One consequence of dropping the old counterfactual should be stated rather than left for a reader to notice. Measured against a baseline that rose to £157bn, this programme looked as though it cut borrowing by tens of billions. Measured against the official forecast, it does not: the central case ends essentially level with it — £58bn against £59bn — while delivering the reform. That is a materially weaker fiscal claim than earlier drafts of this paper made, and it is the true one. What the programme buys with that money is set out in the chapters that follow; what it does not buy is a smaller deficit than doing nothing on the central case.

One further thing the cash comparison hides, and it is the answer to the obvious question — if this is a growth programme, where did the growth go? It went into the reform. The deficit does fall, and it falls further than the official forecast's does: from 4.4% of GDP in Year 1 to 1.6% by Year 5, or £142bn to £58bn, a reduction of £84bn against the OBR's own fall of £56bn from a lower start. The growth dividend is real and large — £47bn a year by Year 5 in the central case, without which borrowing would end at £105bn rather than £58bn. It is the single biggest number in this chapter. What it does not do is produce a smaller deficit than doing nothing, because it is almost exactly consumed by the £35bn of static policy cost and £1bn of additional interest that the programme carries. That is a choice, not a failure. We could have spent the dividend on a lower headline number and delivered less; we have spent it on Single Market membership, a pre-funded care guarantee, employer payroll tax substantially reduced and Business Rates abolished — none of which the official forecast delivers. So the honest claim is not that F³ borrows less than doing nothing. It is that F³ buys the entire reform programme for the same deficit as doing nothing, and shows the receipt.

Debt is the part of this that runs against us, and we would rather put it here than have it found. Because F³ borrows more than the official path in Years 1 to 3 and only draws level by Year 5, it adds about £47bn of cumulative borrowing across the parliament. The OBR's March 2026 forecast has public sector net debt at 93% of GDP at the end of 2024-25, rising to 96% by 2028-29 and easing to around 95% by 2030-31. On our central case the stock is roughly £47bn higher than that, which is about 1.3 percentage points of GDP, so F³ ends the parliament with debt a little above 96% rather than a little below it. The gap closes only after the window, as the reformed deficit runs at or below the official path while nominal GDP keeps growing. An earlier draft of this chapter implied debt falling to around 90%; that was wrong — it decayed the ratio with nominal growth and ignored that the official path itself has debt rising before it falls. The honest summary is that this programme buys the reform for the same annual deficit and a slightly larger debt stock, and recovers the difference in the parliament after.

LOW SCENARIO — cumulative extra GDP of 1.9% by Year 5

Assumes: Single Market accession does not happen — Brussels says no, so no EEA fee is paid, no Single Market growth dividend accrues, and no free-movement labour supply arrives either; the unilateral measures still deliver, reaching 1.9% of GDP by Year 5 through employer tax reform and planning; welfare savings held to £4bn by legal challenge; gilt yields +75bp. The gated cuts — Step 4 (the 35% rate cut) and the Employer Levy step-down — never fire, so their cost is never incurred. An earlier draft of this narrative said 0.7% by Year 5, which contradicted this scenario's own table; 1.9% is the figure the table has always used.

Low scenario: with every assumption against us — no Single Market dividend, no Step 4, the Employer Levy core never stepping down, welfare reassessment delivering £4bn rather than £8bn — borrowing runs above the official path for the whole parliament and ends about £1bn above it. That is the honest floor, and it is lower than an earlier draft of this paper claimed. Two things are worth reading carefully. The first is that the gates work: in this scenario the headroom test is never met, so none of the £52.4bn of gated giveaways fires, and their cost is never incurred. That is what keeps the Low case from running away. The second is that this no longer beats doing nothing. An earlier draft showed the Low case ending furthest below the official path, on the reasoning that a scenario which never spends must end lowest. That was wrong: it credited Low with never paying for the giveaways while not charging it for the weak growth and half-delivered welfare that define it. Corrected, and after the savings and family-allowance re-costings, the Low case ends slightly above the official forecast. The honest claim for it is therefore narrower than the one we used to make: even with no Single Market, no rate cuts, no Levy step-down and welfare reform half-delivering, the plan delivers the entire core programme — excluding the explicitly gated rate reductions, which by definition do not happen in this scenario — for about £1bn a year more borrowing than doing nothing. That is a floor, not a triumph.

Stress-Testing the Design: What a Recession Does to It

The three scenarios above test whether the plan holds if growth is weak. A fair reader will push further and ask a different question: not whether the aggregate numbers survive a slow decade, but whether the individual mechanisms survive a sharp shock — a proper recession, of the kind that arrives once or twice a generation. We think a plan should be able to answer that, and we would rather set out the vulnerabilities ourselves than have them found. Three mechanisms carry the most weight, and each deserves an honest look.

Start with the mechanism that replaces Council Tax: the hybrid of a 2% Local Income Tax and a 0.3% residential property charge. The standard objection to any income-based local tax is real and we state it plainly: income is cyclical — in a recession, earnings fall, bonuses vanish, self-employment income drops — whereas Council Tax was stable, because a house does not stop existing in a downturn. The hybrid is designed around exactly that objection. Roughly half the replacement revenue — the property component — is as recession-proof as the Council Tax it succeeds: property values move slowly, the ±5% annual cap smooths what movement there is, and the charge is levied on a stock, not a flow. The income component that remains is collected nationally and distributed to councils by need, so the cyclical risk it carries sits with the Treasury, not with individual council budgets — and absorbing cyclical risk across the whole tax base and the whole economic cycle is precisely what central government is for. A council’s settlement is set by need, not by last year’s local receipts; the broad base — pension and investment income as well as earnings — dampens the swing further. We are candid that a deep recession still reduces what the income component raises, and that cost shows up honestly in the deficit. But the design has moved beyond answering the objection: half the base does not fall at all, and the residual risk is carried where a modern state can actually absorb it.

The second load-bearing mechanism is the Employer Levy, the largest single pillar of the revenue side. It is a payroll-based charge, and payroll taxes are cyclical: when employment and wage growth fall, so does the yield. We do not pretend otherwise. Two things contain the risk. First, this is not a new exposure we are introducing — the employer National Insurance the Levy replaces was cyclical in exactly the same way, so the plan is no more sensitive to the cycle here than the system it succeeds, and rather simpler. Second, and more importantly, the plan does not spend the Levy’s good-year buoyancy as though it were permanent: the further reductions in the Levy beyond this parliament are gated, funded only from certified headroom, so a downturn that dents receipts simply means those cuts do not happen yet. The mechanism cannot over-commit in the good years and then blow up in the bad ones, because it is structurally forbidden from doing so.

Which brings us to the third mechanism, and the one that makes the whole design robust rather than merely hopeful: the gating itself. No revenue source can be made recession-proof — income, payroll, consumption and profits all fall in a downturn, whatever you tax. A plan cannot promise that a recession will not raise the deficit; anyone who tells you otherwise is not being honest. What a plan can do is guarantee that a recession slows it rather than breaks it. That is exactly what the gating delivers. Every discretionary tax cut in this paper — the move toward the 35% rate, the Employer Levy step-down, the reliefs beyond this parliament — fires only when the Office for Budget Responsibility certifies the headroom in its published forecast, and pauses automatically, by statute, when it does not — with the National Audit Office reporting after the event on whether the pause was respected. So a recession does not force a political choice between abandoning the fiscal rules and slashing services in a panic; it simply pauses the giveaways until the economy recovers, while the revenue measures and reforms carry on. The commitments that cost money are contingent; the disciplines that protect the public finances are not. That asymmetry — costs gated, discipline permanent — is the single most important robustness feature in this paper, and it is why the Low scenario above does not spiral even when every assumption runs against it.

None of this makes the plan invulnerable, and we will not claim it does. A severe recession would reduce revenue, widen the deficit in the year it struck, and slow the pace of reform — as it would for any government of any stripe. The honest claim is narrower and, we think, more valuable: that the plan is built to bend rather than snap. Its costs are conditional on the money being there, its disciplines apply whether it is or not, and the cyclical risk in its revenue sits where a modern state can actually absorb it. A fiscal framework that has thought through its own worst days is worth more than one that has only rehearsed its best.

BASE SCENARIO — cumulative extra GDP of 4.3% by Year 5

Assumes: Single Market uplift reaching 1.8% of GDP by Year 5, phased from accession in Year 3 — a partial realisation within five years of a long-run effect the independent literature puts at 2-4%, and a trade-and-investment figure only; employer tax reform and planning together adding 1.9%; labour supply and free movement adding 0.6%; welfare delivering £8bn (the combined PIP and UC assessment row, of which £4-6bn is PIP); gilt yields +40bp. The first three structural stages proceed, while £29bn of the £52.4bn gated Employer Levy reduction is certified, and the income-tax-rate glide does not begin within this parliament. An earlier draft of this line stated the Single Market uplift as 0.7%/1.2%/1.6% and separately as 2.2% cumulative, which were inconsistent with each other and with the £47bn booked below. The bridge that follows replaces both.

Base scenario: borrowing is about £27bn higher than the official path in Year 1, converges through Years 2 and 3, dips £2bn below in Year 4, and ends the parliament at £58bn against the OBR's £59bn — level. The central case does not beat the official path. It matches it, and buys the reform programme with the growth dividend. By Year 5 £29bn of the £52.4bn of gated giveaways has been certified and fired; £23.4bn has not, which is why the Employer Levy core reaches about 7.9% rather than 7.5% and the flat rate stays at 38% throughout.

Read the bottom two lines together, because they are the whole growth case. The uplift is a LEVEL effect on GDP, not an addition to the annual growth rate — the distinction matters and conflating them is how growth claims get inflated. By Year 5 the economy is 4.3% larger than it would otherwise have been, which at the standard ~£11bn of receipts per 1% of GDP is the £47bn carried in the table above. Two components deserve their caveats stated rather than buried. The Single Market figure is trade and investment only: the independent literature puts long-run membership at 2-4% of GDP, we book 1.8% by Year 5 as partial realisation within a five-year window, and we do not double-count the labour channel inside it. And the labour supply line is booked gross of two effects that run in opposite directions and which we have not attempted to net. More people mean more demand for public services, which is a real cost. More workers also mean lower unit costs in exactly the places this programme spends most — construction labour for the planning reform, care workers for the Fair Pay Agreement and the £34.5bn adult social care bill, clinical staff instead of agency premiums in the NHS. The cohort arriving under free movement is also disproportionately working-age, so it pays income tax, the Health Levy and Local Income Tax while drawing little pension or acute health spending in the early years. The cost of more people is real; so is the saving from more workers; and we have booked neither. A reader who thinks the net effect is negative can strike the 0.6% and the dividend falls to about £40bn, which moves the central case from essentially level with the official forecast to about £6bn above it.

HIGH SCENARIO — cumulative extra GDP of 6.5% by Year 5

Assumes: Single Market uplift reaching 3.4% of GDP by Year 5 — the upper half of the independent 2-4% range, trade and investment only; employer tax reform and planning delivering 2.2% against the central case's 1.9% on a strong hiring response to the Employer Levy cut; labour supply and free movement adding 0.9%; welfare delivering £12bn; no gilt repricing. Components sum to 6.5%, the figure in the heading, and at ~£11bn per 1% of GDP that is the £72bn carried in the table.

Summary Comparison

Key Sensitivities

Bond Market Management

The 2022 mini-budget demonstrated the cost of large unfunded announcements without independent validation. F³'s strategy is built so that the first thing markets see is a falling deficit:

• OBR pre-commitment — the full programme is submitted to independent OBR scrutiny before implementation and the results published in full, whatever they show
• Revenue first, cuts gated — Year 1 adds about £27bn to borrowing against the official path, of which £2bn is threshold-freeze receipts deliberately handed back; Years 2-4 close that gap and Year 5 returns to level; every tax-cutting step, including the Employer Levy core step-down, is pre-legislated but fires only on OBR certification in its published forecast, with NAO audit after the event
• A credible fiscal rule — the glide path in Chapter 2 is met or enforced by the gates in every scenario, including LOW. A rule the plan itself would break is not a rule; ours holds
• A specified emergency brake — if gilt yields rise more than 100bp above the OBR forecast within 12 months, two pre-named measures trigger automatically: the next pre-legislated step is suspended, and a temporary 1p surcharge on the flat rate applies until yields normalise. Markets know the circuit breaker’s exact wiring in advance
• No leveraged state investment — the British Future Fund will be capitalised only from certified headroom, saved North Sea receipts and in-kind shares (see Chapter 10), never from gilt issuance. The state will not borrow to buy equities

Water-acquisition event-risk is sequenced, not bundled: failed companies are taken over through court-supervised Special Administration as insolvencies arise, on their own timetable, at independently determined fair value — not as part of a single fiscal event.

One deliberate asymmetry in how we treat interest is worth stating plainly, because it runs in the reader's favour rather than ours. The interest line in our scenarios only ever adds cost: it carries the risk premium markets might charge if gilt yields rise, scenario by scenario. It does not claim the mirror-image saving — the lower debt-servicing bill that follows wherever we run below the official path, as the central case does in Year 4 and the High case does from Year 3 onward. The Low case never runs below it, so in that scenario there is no mirror-image saving to forgo. That saving is real, if modest at these margins. We do not book it, for two reasons. It depends on the precise path of the debt stock and the maturity of the gilt portfolio — only newly-issued and refinanced debt reprices, while the existing stock is locked at the rates it was issued at — and estimating it honestly would require a full debt-dynamics model rather than a single assumed figure; and inventing a number we could not stand behind is exactly the discipline this paper refuses to break. The effect is that our numbers carry the downside of interest without claiming its upside, so the true fiscal position is, if anything, modestly better than the scenarios show. We would rather understate the improvement than overstate it.

The Honest Summary

Year-one borrowing rises, because the spending commitments and the structural transition costs arrive before most of the growth dividend. It then falls through the remainder of the parliament. Against the latest ONS provisional outturn of £129.8bn for the year to March 2026, F³ borrows about £12bn more in Year 1; against the OBR's forecast of £115bn for 2026-27, about £27bn more. The structural cost of the completed reform arrives in Years 4-5, by which time the growth dividend, welfare savings and compounding pension reform are at full strength. Every commitment in this paper is sequenced behind that arithmetic, not in front of it.

Our long-term goal remains reducing national debt below 60% of GDP within a generation — through growth, not austerity.

Chapter 1: Tax Reform

One Simple Income Tax

We will replace today’s complicated income tax and National Insurance system — with its multiple bands, thresholds and rates — with a single, transparent tax:

• First £15,000 entirely tax-free — a single allowance that absorbs and replaces the thicket of reliefs below it
• 38% flat income tax on all earnings above £15,000 — glides toward 35% as headroom is certified, after debt reduction — pre-legislated and gated, not dated — each step conditional on the Office for Budget Responsibility certifying that the fiscal glide path is holding
• National Insurance abolished. For employees: outright, from day one — the largest single simplification in the system
• For employers: abolished and replaced by an 8.5% Employer Levy — a 7.5% core (half the old 15% rate, though not half the bill: the Levy falls on full payroll with no threshold, so on a £874.5bn base it yields about £65.6bn against employer NIC’s £108.7bn — the employer payroll cost falls by roughly 40%, not 50%) plus a 1-point Training Point paid into a pooled National Training Fund — phased via 12.25% in Year 1. We should be precise about which half of that is a promise. Replacing NIC with the Levy is sequenced for Year 2 and depends on nobody. Taking the core from 11.25% to 7.5% costs about £32.8bn a year and is gated on certified headroom, taken in increments as the money appears: on our central case that means little before Year 3 and about £29bn of the £52.4bn of gated giveaways fired by Year 5, taking the core to roughly 7.9% and the headline to about 8.9% rather than the 7.5% and 8.5% full implementation would deliver. The balance follows in the next parliament as headroom allows, and on our low case it does not happen at all. Every employer figure in this paper shows the position once the step has fired in full; until it does, the Levy sits above 7.5% and the saving is correspondingly smaller. We would rather an employer planned on that than on the headline. Small firms draw fully-funded apprentice training from the Fund; large firms that train draw their point back directly. The Training Point is not money the employer loses.

One honest sentence on the employer side: replacing NIC with a levy at half the rate is not the same as removing payroll taxation altogether, and we will not pretend otherwise. Unreplaced abolition leaves a £50bn-a-year structural hole that only heroic growth assumptions can fill — and every credible way of filling it lands on workers. So the core Employer Levy cuts the cost of employment by roughly £35bn a year — in full for every employer that trains — and carries a statutory, headroom-funded path to its own abolition: each future cut fires on OBR certification in its published forecast, written into the same statute as the glide path, exactly like Steps 2 to 4. The Training Point above the core is different in kind: an employer that trains pays nothing on it; an employer that does not funds the Guarantee for those who do (Chapter 6).

There is a quieter benefit here that we deliberately do not put a number on. Abolishing employee National Insurance does not merely remove a tax; it removes an entire parallel calculation from every payslip in the country. Today an employer runs two separate deductions on each employee’s pay — income tax and National Insurance — with different thresholds, different rules and different edge cases; scrapping the second leaves one honest deduction where there were two. That is a real and recurring simplification for the roughly 1.4 million employers who operate payroll, and a system that is correspondingly cheaper and easier for HMRC to run. We do not book either saving, and we are honest about why. The saving to business is genuine but diffuse, and sizing it would mean inventing a figure of exactly the kind this paper refuses; the saving to HMRC’s own running costs is real but modest, and in any case we are choosing to reinvest in HMRC’s capacity (Chapter 1’s tax-gap section), not to shrink it. So this goes in the same column as the interest we do not claim and the recoveries we do not bank: an unclaimed benefit that makes the true position a little better than the figures show. It is simply one more reason that a simpler tax system is a better one — not a saving we are spending in advance.

Why Cutting the Cost of Employment Matters Now

There is a forward-looking reason to move tax off employment, beyond the principle that you should not tax what you want more of. Artificial intelligence and automation are, for the first time, making it a genuine choice for many firms whether to employ a person or deploy a machine. In that world, a tax levied specifically on the act of employing a human is not neutral — it quietly tips the scales toward replacing the worker, faster than the underlying economics alone would. We are not trying to hold back automation; the productivity gains from AI are something this paper actively wants (Chapter 17). The point is narrower and more defensible: the tax system should not put its own thumb on the scale against human employment at the exact moment that choice is becoming live. Halving the employer charge removes that distortion.

There is a structural argument too, and it is the stronger one. A tax base built on employment is a base that shrinks if automation erodes headcount over the coming decades. Leaning the public finances heavily on payroll taxes — as the current National Insurance system does — is therefore not just unfair to workers; it is a bet on a base that the technology of the next twenty years may steadily undercut. A forward-looking tax system does the opposite: it shifts the weight onto bases that do not evaporate when a firm automates — profits, consumption, and the immobile assets this paper already favours. Moving off employment taxation is not only the humane choice. In an AI economy it is the fiscally prudent one.

A worker earning £35,000 today pays income tax and NICs across multiple bands and thresholds. Under F³ they pay one published combined rate — 42% including the Health Levy and Local Income Tax, 39% from Step 4 — on income above £15,000, and nothing at all below it.

Duties and Excise — The Other £75bn

Duties raise roughly £75bn a year and most manifestos pretend they do not exist. Our positions, briefly and completely:

• Fuel duty (£24bn): frozen in cash permanently, withering as the fleet electrifies — never replaced by pay-per-mile (Chapter 11)
• Vehicle Excise Duty (£8bn): abolished and replaced by the flat Road Charge (Chapter 11)
• Alcohol duty (£13bn): CPI-indexed, draught relief for pubs kept, the band structure simplified
• Tobacco duty (£8bn): the escalator stays — a declining base we are content to see decline
• Vaping duty: set at £3.00 per 10ml — still one flat rate, no nicotine tiers — with an above-inflation escalator thereafter. A statutory rule keeps vaping duty below half the equivalent cigarette duty, so switching stays the cheap option: priced for adults switching, not children starting. £100m of the proceeds funds Trading Standards enforcement against the illicit market. We book no revenue from this at all: the escalator is worth well under £100m a year by Year 5, and the honest counterpart is that the ~£8bn tobacco duty base erodes toward zero over a generation as the smoking age rises. Vaping duty does not replace it and we do not pretend it does
• Gambling duties (£4bn): the elevated online rates stay — consistent with Chapter 17’s approach to engineered addiction
• Air Passenger Duty and Insurance Premium Tax (£13bn): frozen and CPI-indexed; IPT — a stealth tax on prudence — reviewed once the glide path is met
• Customs duties (£5bn): retained in full — outside the EU Customs Union, Britain sets and keeps its own external tariffs and is not in the EU’s own-resources system (Chapter 4)

Local Government Finance

Council Tax is arbitrary, outdated and unfair — based on property valuations from 1991. A retired person in a large family home pays the same as a high-earning professional next door, and a £1.5 million house pays barely more than a £400,000 one because Band H is a ceiling. We replace it with a hybrid that takes the best of both worlds: a small income component that tracks ability to pay, and a small, modern property component that gives local services a base that does not vanish in a downturn:

• 2% Local Income Tax, applied to income above £15,000 — the same threshold as the national flat rate, collected through PAYE and self-assessment and distributed to councils by population and need
• 0.3% annual charge on residential property values, retained locally — valued every year by the same automated models the mortgage market already runs on Land Registry data, with a ±5% cap on annual movement so no bill ever jumps with a hot market
• Protection built in, not bolted on: households with income below £15,000 in a home below the national median value (~£290,000) are exempt outright — they pay nothing and accrue nothing. Below £15,000 in an above-median home, the charge is deferred: it rolls up at CPI only and settles from the property at sale or death, so no one is ever forced to pay from income they do not have
• The owner pays, never the tenant: for rented homes the charge falls on the landlord, so low-income renters are outside it entirely — and because the property element is retained locally, every council gains a direct fiscal stake in approving the new homes of Chapter 8

The full marginal rate on income above £15,000 — including the 2% Health Levy — is 42% throughout this parliament on our central case, falling to 39% only when certified headroom allows the headline rate to glide to 35% — which on our own scenario tables is after this parliament, not within it. Below £15,000: no income tax, no levy, no local income tax. We publish the combined figure because it is the number that actually governs your payslip; manifestos that quote their taxes one at a time are hoping you will not add them up. Together the two components raise about £45bn — around £20bn from the 2% rate on the same broad income base as the national income tax (employment, self-employment, pension and most investment income above the threshold, roughly £1.1 trillion in all), and £25bn from the 0.3% charge on a housing stock worth some £9 trillion, net of the exemption — against the £50bn Council Tax they replace: a net cost of about £5bn, and £5.4bn on the unrounded figures used in the scorecard. An earlier draft of this paper put that shortfall at £3bn, which did not follow from its own two components; the figure here is the arithmetic. The design answers the honest objection to each pure model. A pure local income tax is cyclical — earnings fall in a recession exactly when local services are needed most — and invites the avoidance this paper fights elsewhere: structure your income below a threshold and a seven-figure asset drops out of the base entirely. A pure property tax ignores income. The hybrid takes the stability of property and the fairness of income, at low rates on both — and the property side is fairer than the Council Tax it replaces in precisely the way that matters: under a proportional charge the £1.5 million house pays proportionally, while the struggling retiree next door is exempt or deferred. On valuation we make the honest concession in advance: Denmark's recent revaluation programme ran late and over budget, so we claim banded precision, not pound-level accuracy — published valuations, a simple appeal right, and the annual cap. Both yields are modelled figures we publish for an independent body to re-run, exactly as we treat every revenue estimate in this paper.

A Clean Slate: Council Debt and the End of the Postcode Lottery

Reforming how councils are funded is only half the problem. The other half is the debt they are already drowning in. A string of councils — Woking, Thurrock, Croydon, Birmingham — have issued section 114 notices, the local-government equivalent of declaring themselves unable to balance the books, and many more are one bad year from the same. Total local-authority debt runs comfortably beyond £100bn. No funding reform will hold if councils are still using the new settlement to service old borrowing they can never repay. So we deal with the debt directly — but we distinguish, as ever, between the different kinds of it, because a blanket bailout would reward recklessness and punish prudence in equal measure.

First, the funding reset itself does more than modernise a tax. By replacing Council Tax with a Local Income Tax collected nationally and distributed to councils by population and need — and by compensating councils centrally for the business rates we abolish — we end the postcode lottery at its root. Today a council’s spending power depends heavily on the property wealth and business base that happen to sit within its boundary, so the poorest areas, with the weakest tax bases and the greatest need, are structurally short-changed. Funding by need rather than by local property luck irons that out automatically: the equalisation that the current system attempts through a tangle of grants and top-ups is built into the design instead. We are honest that this is a more centrally-funded model of local government than Britain is used to, and that it trades some local fiscal autonomy for fairness and stability — a trade we make deliberately, and defend, because the autonomy the current system offers poorer councils is largely the freedom to be underfunded.

On the debt itself, the great majority is owed not to private lenders but to the Treasury, through the Public Works Loan Board. Debt the government is itself the creditor of can be restructured by direct decision rather than financial engineering — and we will:

• Restructure and, where necessary, write down the PWLB debt of councils in distress, distinguishing prudent borrowing from speculation. Debt that built homes, schools, depots and infrastructure is sound borrowing against real public assets, and we will re-profile it onto sustainable terms. Debt run up gambling on commercial property — the shopping centres and office blocks that sank Woking and Thurrock — is a different matter: those bets are written down, and where a council speculated recklessly the write-down comes with binding conditions and, in the worst cases, the loss of the borrowing freedoms that were abused. Because this debt is owed within government, restructuring it is an intra-government transfer: it does not add to the external deficit or the borrowing this paper is scored against, and we do not pretend it is a spending giveaway
• Challenge the private LOBO loans rather than legitimise them. Around £15bn of council debt sits not in PWLB loans but in so-called Lender Option, Borrower Option loans sold by banks in the 2000s — loans with low “teaser” rates that step up sharply, embedded derivatives councils were never equipped to price, and exit penalties that can double the cost. Deprived boroughs like Newham have paid tens of millions a year servicing them. We will not bundle this debt up and sell it on — securitising predatory loans to make them someone else’s problem is precisely the pre-2008 mistake, and it would lock in terms that should never have existed. Instead we back councils to challenge and exit these loans: a government-supported LOBO resolution programme that uses every legitimate lever — including the long-standing legal argument, rooted in the 1989 Hammersmith ruling, that such deals may have been beyond councils’ lawful powers — to renegotiate or unwind them penalty-free, as Newham was able to in 2019, and refinance the remainder into cheap Treasury lending. Where banks mis-sold, the banks that profited share the loss, exactly as we require of the bondholders of failed water companies (Chapter 15)

Taken together, this is a genuine clean slate: sound debt re-profiled, reckless debt written down with consequences, predatory debt challenged rather than dignified, and a funding system that stops the debt re-accumulating because councils are financed according to what they need rather than what their postcode happens to be worth. We will not put a false precision on the cost: the scale of any PWLB write-down is a matter for the settlement negotiated council by council, and the LOBO recoveries depend on cases yet to be run. What we can say plainly is that the PWLB element is money moving within government rather than new borrowing, and that the alternative — leaving councils to fail one by one, cutting services to the bone to service debts they cannot carry — is the more expensive path, just paid in a currency of closed libraries and uncollected bins rather than in the public accounts.

Business Taxation

Britain’s businesses face a triple burden: Corporation Tax, Business Rates, and employer NICs. The honest way to understand our reform is not as a Corporation Tax rise in isolation, but as a deliberate swap: we tax profit more heavily so that we can tax jobs, premises and investment less heavily. A company does not care about any single tax; it cares about its total cost base. So we cut two of the three burdens outright and lean harder on the third.

Consider a hotel in Surrey, Manchester or Edinburgh. Today it pays Corporation Tax, employer NICs, and Business Rates — and of the three, the rates bill is usually the most hated, because it falls on the building whether or not the hotel turns a profit. Most operators would gladly take a higher rate on the profit they actually make in exchange for losing a fixed charge that takes no account of whether they made any. That is the trade F³ offers: a 30% rate on profit, against zero rates on premises and a halved tax on every wage. For the businesses that employ people and occupy buildings — the overwhelming majority — the swap is a clear gain, as the worked examples in the business chapter show.

Within that swap, we cut two of the three burdens:

• Corporation Tax set at 30%, with the OECD Pillar Two global minimum closing the artificial profit-shifting that erodes the base — the brass-plate structures, royalty routing and transfer-pricing games that booked UK profit in near-zero jurisdictions. The £16bn yield is scored net of residual shifting and of permanent full expensing retained at the higher rate
• Business Rates abolished in tranches matched to the fiscal steps — retail, hospitality, leisure and schools at Step 1, every remaining sector by Step 3 — ending a tax that penalises physical investment and punishes success on the high street
• Permanent full expensing, written into the same clause of the same Finance Bill as the 30% rate — every pound of qualifying investment deducted in full, in year one, permanently

The pairing is the point, and it answers the obvious objection to a 30% headline rate. A headline rate taxes profit already being made in Britain; full expensing untaxes the decision to invest in making more. With every pound of qualifying investment deducted in full in year one, the effective marginal tax rate on a new investment financed from retained earnings is close to zero — among the most generous investment positions in the OECD — even as the return on existing profit is taxed at 30%. A company deciding whether to build its next line in Britain faces one of the friendliest regimes in the developed world; a company already booking British profit contributes at the new rate. That is ‘tax what cannot leave’ applied to the corporate base. And the two are written into the same clause of the same Finance Bill so no future chancellor can quietly keep the rate and drop the relief: they stand or fall together, and the £16bn yield is scored net of the expensing’s full cost.

What the Global Minimum Does — and What It Does Not

It is worth being precise, because the rule is often overstated. The OECD Pillar Two agreement, in force since 2024 across roughly 140 countries including the EU and UK, requires large multinational groups — those with global revenue above €750m — to pay an effective rate of at least 15% on profit in every country they operate in. Where a group books profit somewhere taxed below 15%, a top-up brings it to 15%; and if the low-tax country does not collect that top-up, another country where the group operates can. The floor is therefore enforced even against jurisdictions that would rather not enforce it.

The Acquisition Debt Neutrality Rule

There is a distinction this paper draws everywhere else and should draw here too: between borrowing to build something and borrowing to buy something that already exists. Interest on both is currently deductible against UK profits on the same terms, subject to a Corporate Interest Restriction that caps net deductions at 30% of EBITDA above a £2m allowance. The effect is that a fund can acquire a British company, load the acquisition debt onto the company it has just bought, and deduct the financing cost of the purchase from the profits the company was already making. Nothing new is built. The tax base shrinks anyway.

We keep normal deductions in full where borrowing does something. New factories, plant, equipment and infrastructure; working capital; genuine third-party project finance; and smaller businesses, where the £2m allowance means the rule never bites. What changes is narrower:

• Interest incurred principally to acquire existing shares — in particular debt pushed down onto a company following a leveraged buyout — is restricted to 20% of EBITDA rather than 30%
• Related-party interest routed to a low-tax jurisdiction is disallowed unless it is genuinely taxed there
• The £2m de minimis and the group-ratio alternative are retained, so the rule reaches large structured acquisitions and not ordinary corporate borrowing

We book £0.4bn, the bottom of a £0.4-0.9bn range, and we should say why the bottom. The original Corporate Interest Restriction raised something like £0.9-1.2bn, and a targeted tightening of one limb of it must raise less. Beyond that, two things cut the figure. Raising Corporation Tax to 30% makes every interest deduction more valuable than it is today, so the incentive to restructure around the rule is larger under this programme than under the current rate — we should expect behavioural response, not assume compliance. And Pillar Two already tops up profit shifted to low-tax jurisdictions, so the related-party limb partly duplicates a rule we are also relying on elsewhere. Booking the top of the range would be double-counting our own machinery.

What this does for Britain is real but bounded. It closes the artificial games — profit with no genuine activity behind it, parked in near-zero jurisdictions — because that profit now carries at least 15% wherever it lands. It does not, however, stop a country charging 15% while Britain charges 30%. Ireland is the obvious case: rather than lose the top-up to others, it raised its rate to 15% for large multinationals, keeping 12.5% only for smaller firms. So the floor did not end tax competition — it put a 15% floor under it.

We are honest about the consequence: a genuine 15-point gap with Ireland on real, mobile activity is a competitive disadvantage Pillar Two does nothing about, and a 30% rate must earn its keep against it. It does so on three grounds. First, profit shifting that is artificial — the bulk of the lost base — is now caught regardless of the headline gap. Second, what businesses actually locate for is rarely the rate alone: it is market access, talent, the rule of law and stability — which is precisely why Single Market re-entry, fast-track skilled visas and an independently gated fiscal framework matter more to a 30% Britain than a few points off the rate ever would. Third, the rate is gated like everything else: if the evidence shows 30% is costing more in lost activity than it raises, the structural surplus path that funds the rate cuts elsewhere in this paper applies here too. We set the rate where the revenue is needed and the base is now defensible — not where a race we have chosen not to run would pull it. One caveat we state plainly, because it is the most contestable number in this paper: the £16bn is scored on a partial behavioural response, not zero and not heroic, and it is precisely the figure we would expect the Treasury, the IFS and the OBR to re-run. We publish it as a reviewable central estimate, not a certainty, and the gating means that if it disappoints, a tax cut waits rather than a deficit opening.

On Relying on Any Single Tax

The sharper challenge is not whether 30% is too high — it is whether too much rides on Corporation Tax at all. We fund a great deal from it: the abolition of employee National Insurance, the £15,000 allowance, Business Rates abolition, and more. If Corporation Tax under-delivers — because profits soften, or because some shifting survives Pillar Two — does the whole programme wobble? It is a fair question, and we answer it the same way we answer every revenue risk in this paper: through the gates. No tax cut in this programme is unconditional. If Corporation Tax, or the Single Market dividend, or the welfare savings, or the Wealth Floor under-performs, the consequence is not a hole in the budget — it is that the next gated tax cut waits until the money is there. The deficit glide path is protected first; the giveaways come second. That is precisely why the rate cuts are staged and independently certified rather than promised up front. Over-reliance on a single source is a danger only for a plan whose spending is locked in regardless of revenue. Ours is not: the discipline that protects against optimistic growth protects equally against a disappointing Corporation Tax yield. We would rather a tax cut arrive a year late than a deficit arrive on schedule.

How Britain Compares — the United States

The fear that a 30% rate makes Britain uncompetitive against the United States rests on a misreading of American tax. The US federal corporate rate is 21%, but almost every state adds its own on top — averaging around six and a half points, and reaching a combined rate near 30% in states like New Jersey and California. America's true corporate rate is not 21%; it is a 25-to-30% range once the state is counted. F³'s 30% sits at the top of that band, not outside it.

Payroll is the same story. A US employer pays 7.65% in federal FICA on every salary — Social Security up to a cap, Medicare uncapped — and then federal and state unemployment taxes on top, which in a high-cost state push the real employer payroll cost into double digits. F³'s Employer Levy is a single 7.5% core with no separate unemployment tax bolted on. A British employer under F³ therefore faces a payroll-tax cost comparable to, or lower than, an American employer in a major state — and a corporate rate in the same range as New Jersey or California.

So the honest international picture is not a high-tax Britain undercut by a low-tax America. It is two economies in much the same band on headline rates — with F³ then handing its businesses things the US system does not: no tax at all on commercial premises, near-frictionless access to a market of 450 million on the doorstep, and a fiscal framework that does not lurch every budget. We compete on the things that compound, not on a race to the bottom on the rate.

Does 30% Drive Business Away? What the Evidence Shows

This is the right question to ask of any rate rise, and the evidence — including the evidence against parts of our own instinct — gives a clear answer. When Britain cut its corporation tax from 28% to 19% over the 2010s, two things happened, and they are usually confused. Company headquarters and booked profit did move to Britain: accountancy data at the time tracked roughly sixty multinationals considering UK relocation as a direct result, and the long run of corporate inversions away from Britain reversed. So the rate genuinely moves where profit is declared. But business investment did not rise — Britain had the lowest business investment in the G7 by 2019 despite having the G7’s lowest corporate rate, partly because the cuts were paid for by lengthening capital write-off periods, which quietly raised the effective tax on new investment even as the headline rate fell.

Two conclusions follow, and F³ is built on both. First, the corporate rate is mainly a tool for competing over where profit is booked — which is precisely why we pair a 30% rate with Pillar Two, so the booking game is closed and the rate can do its honest job. Second, investment responds to the cost of capital and to allowances far more than to the headline rate, which is why our growth strategy rests on investment incentives, Single Market access and stability rather than on a low number we know does not deliver investment on its own.

And 30% is not an outlier. Of the jurisdictions the OECD tracks, twenty-six levy headline rates at or above 30%. France sits at 36%. Germany — Britain’s closest competitor for serious industrial investment — reaches almost exactly 30% once municipal trade taxes are counted. Australia and several others sit at 30% flat. The economies Britain competes with for real operations, as opposed to brass-plate profit-routing, are clustered at or above the F³ rate, not below it. A 30% Britain with no commercial-property tax, near-frictionless access to its largest market, and a stable fiscal framework is more attractive to a business that actually builds things than a 19% Britain that taxed its premises, sat outside the single market, and changed its rules every year.

Abolishing Business Rates alone will transform the economics of retail, hospitality, manufacturing and logistics — sectors that have been hollowed out by a tax designed for a different era.

The Landowner Levy: The Rates Windfall, Clawed Back

There is an honest problem with abolishing Business Rates, and we would rather name it than have it named for us: decades of evidence say that much of the benefit of cutting a tax on premises ends up, over time, in the pockets of the people who own the premises, as rents rise to absorb the saving. Rates are legally paid by occupiers but economically shared with landlords. Abolish them outright and a meaningful slice of a £26bn tax cut aimed at shopkeepers, manufacturers and publicans quietly becomes a windfall for freeholders — including the offshore funds and tax-exempt vehicles that own much of Britain’s commercial property and pay little or no UK tax on the gain.

The Landowner Levy claws that windfall back at source. From Year 2, a 1% annual charge applies to the value of commercial land and premises above a £500,000 site threshold — paid by the freeholder, never the occupier, assessed on existing use so no one is taxed on hope value, with schools and charities exempt. The threshold takes the corner shop’s freehold, the small workshop and most owner-occupied small premises out of scope entirely; the charge falls where the windfall lands — on large commercial estates and institutional portfolios. It raises around £12bn a year when fully in: roughly half the cost of Rates abolition, recovered from the balance sheets that would otherwise have absorbed the gain.

This is the oldest idea in tax economics, endorsed by everyone from Adam Smith to Milton Friedman: land cannot move, cannot be hidden, and does not stop working when you tax it — a levy on landowners is the closest thing economics has to a tax without a growth cost, and it is precisely the ‘tax what cannot leave’ principle this paper is built on. Occupiers get the full benefit of abolition from day one; the Levy simply ensures the gift ends up where it was addressed. To the landlord who objects, we make the point plainly: this charge takes back only part of a windfall we are simultaneously handing you. No Rates abolition, no Levy. Taken together you are better off — and the businesses in your buildings are transformed.

Tidying the Bolt-On Taxes

Britain has roughly ninety taxes, and a cluster of the smaller ones exist only as patches over holes in the bigger ones. A serious reform of the main taxes should say what happens to the patches. Our principle is simple: where our reform fixes the hole, the patch can go; where the patch does a job our reform does not, it stays — repointed and honestly justified, not left lying around out of habit.

• The Digital Services Tax and the Diverted Profits Tax: fold them in as the global system bites. Both are workarounds for the same problem our Corporation Tax reform targets — multinationals booking UK economic activity as profit somewhere cheaper. The Digital Services Tax (a 2% charge on big-tech revenues) and the Diverted Profits Tax (the “Google tax”) were stopgaps for an international system that could not tax digital profit where value was created. As the OECD Pillar One and Pillar Two framework that underpins our Corporation Tax base takes hold, these unilateral patches should retire into it rather than sit alongside it. We are candid about why this is more than housekeeping: the Digital Services Tax is a perennial flashpoint in trade negotiations with the United States, which treats it as discriminatory against American firms. Folding it into the multilateral framework removes an irritant and signals that Britain taxes profit through the proper, internationally-agreed route — not that we are going soft on big tech, but that the grown-up mechanism has arrived. The honest caveat: we only retire the patch when the multilateral replacement is actually delivering the revenue, not before
• The Bank Surcharge and Bank Levy: abolish both — banks pay the standard rate like everyone else. We do not believe in sector-specific taxes, and bank taxation is where they have proliferated. Banks currently pay Corporation Tax plus a 3% surcharge — a combined 28% — and, separately, a Bank Levy on their balance sheets worth around £1.3bn a year. We abolish the surcharge: under our reform banks simply pay the standard 30% Corporation Tax, the same rate as every other company. Note what that does — it takes banks from 28% to 30%, a slightly higher rate than today, but through the normal corporate system rather than a special one, and it is already inside our Corporation Tax yield, which applies the 30% headline to the whole corporate base, banks included. So abolishing the surcharge costs the scorecard nothing: those profits are already taxed at 30% in our figures. We also abolish the Bank Levy, the balance-sheet charge — a genuine but modest cost of around £1.3bn a year, since the Levy sits outside Corporation Tax and is not captured in our headline yield. We score that £1.3bn as its own line in the scorecard rather than dress it up as free. The prize is a banking sector taxed like any other business — no surcharge, no levy, just the standard 30% — which is both simpler and more competitive against New York, Frankfurt and Singapore, none of which impose bank-specific taxes, and which keeps faith with our financial-services growth ambition (Chapter 15). Banks that pose genuine systemic risk are addressed where that risk actually lives — in capital and prudential regulation — not through a balance-sheet tax that competitors do not levy
• The Annual Tax on Enveloped Dwellings: keep it, repointed from avoidance to transparency. This annual charge falls on homes held inside a company “envelope” rather than owned by a person. It was built to kill a specific dodge: holding a house in a company so that, on sale, the owner could sell the company’s shares and escape Stamp Duty — while also hiding who really owned the property. Abolishing Stamp Duty (above) removes the first half of that motive, so part of this tax’s original rationale falls away with it. But the second half — stopping anonymous corporate ownership of British homes — matters as much as ever, and arguably more given our wider transparency agenda and non-dom reform (Chapter 14). So we retain the charge, but reframe it honestly for what it now does: a transparency tax on hidden property ownership, not an anti-Stamp-Duty backstop for a Stamp Duty that no longer exists
• The regulatory levies stay industry-funded — and that is the point. A cluster of charges (the Financial Conduct Authority and Prudential Regulation Authority levies, the Financial Services Compensation Scheme levy, the Pension Protection Fund levy, the Economic Crime Levy) are sometimes lumped in with “taxes,” but they are not taxes on the public — they are the cost of running financial regulation and the compensation schemes that protect consumers, paid by the industries that benefit from and pose the risk to those systems. We keep them funded that way deliberately: the alternative is to shift the cost of policing the City onto the general taxpayer, which would be both unfair and a quiet expansion of the state. Keeping the regulated sectors paying for their own supervision is the smaller-state position, not the larger one

Property Tax Reform

Stamp Duty is one of the most economically damaging taxes in Britain. It freezes the housing market, prevents people from moving for work, and distorts property values. We will:

• Abolish Stamp Duty on all property transactions
• Introduce Capital Gains Tax on property gains. The treatment differs by what the property is: the home you live in is handled through the deferred Capital Account below; second homes and buy-to-let are investments, and are taxed like any other investment — see the distinction set out beneath
• Movers pay nothing at the point of moving. Instead of taxing the gain on each sale, the gain is recorded in a personal Capital Account and carried forward — so the act of moving house is never taxed, and labour mobility is never penalised. This is the principle from our tax philosophy made real: we do not tax movement
• The accrued gain is settled once, at death, and is captured outside the estate — it is a registered charge that crystallises and is paid first, so it does not form part of the estate for Inheritance Tax. With the IHT threshold at £3m, the overwhelming majority of estates pay no IHT at all; for them this is simply a one-off settlement of the real housing gain, and nothing more
• The gain is real, not nominal. The original purchase price is uprated by CPI to the date of disposal, and tax falls only on the gain above inflation — we never tax the part of a price rise that is merely inflation. Acquisition and improvement costs, and the historic Stamp Duty paid on purchase, are all deductible: we abolish the transaction tax going forward and credit what was already paid
• It is taxed at the headline rate (38%, falling to 35% at Step 4), with no holding-period taper. The taper rewards genuinely productive patient capital — a business, a growth equity. A home is not that: its gain is mostly passive land-price appreciation, so there is no public interest in rewarding a longer hold, and applying the taper would simply reward land-banking and gut the revenue. The £15,000 annual exempt amount applies to the gain, as to any other
• Trusts cannot be used to escape it. Placing a property into trust is itself a disposal that crystallises the accrued gain, and property held in trust accrues and settles on the same basis — the same trust look-through the National Wealth Floor uses. Wealth cannot be wrapped out of sight of this charge

One distinction matters and we make it explicit. The deferred Capital Account — no tax on moving, settled once at death — applies to the home you actually live in, your primary residence, because that is not really an investment: it is where you live, and taxing someone for moving between homes serves no purpose. A second home or a buy-to-let is different. It is an investment held for return, and it is taxed like any other investment: Capital Gains Tax on the real gain at the point of sale, at the headline rate, with no deferral to death and no holding-period taper. There is no Capital Account shelter for investment property and no reward for holding it longer — the same principle that denies the taper to a primary residence applies with even greater force to property held purely for gain. This is what ties the housing and landlord chapters together: we stop punishing landlords through Section 24 and the Stamp Duty surcharge, but we tax the gain on investment property squarely, when it is realised, as the investment it is.

• The charge can be paid voluntarily at any time, and an outstanding balance is uprated by CPI only — never a real-terms penalty. Someone who sells and pockets cash can settle from the proceeds; someone asset-rich and cash-poor simply carries the balance to death

This creates a fairer system where tax falls on real gains rather than transactions — and, because no cash leaves anyone’s pocket at the point of moving, it is better for mobility than both the Stamp Duty it replaces and a conventional sale-by-sale gains tax. The gain does not escape at death the way it does under today’s system, where the uplift on death wipes it: here it is recorded from the first sale and settled once, in real terms, at the end. The mechanism is a personal Capital Account — administratively comparable to the student loan book the country already runs: a lifelong, indexed, per-person balance. We are honest that building it is a real undertaking, not a stroke of the pen. The line is scored on the real gains crystallising at death net of allowable costs and the historic Stamp Duty credit; at the headline rate on full real gains it is broadly self-funding against the transaction tax it replaces, and we flag it as a reviewable estimate an independent body should re-run.

Narrowing the Tax Gap — Honestly

No tax chapter is complete without confronting the money the system is owed but does not collect — the tax gap. It is large and it is growing: HMRC’s most recent estimate puts it at around £59bn for 2024-25, some 6.4% of all the tax that should be paid, and the figures for earlier years have repeatedly been revised upward as better data comes in. That is a serious sum, and closing even part of it would matter. But the way the tax gap is usually invoked in politics is precisely the dishonesty this paper exists to refuse, so we have to be careful about what we claim.

Start with what the gap actually is, because the popular story is wrong. It is not, for the most part, a pot of gold sitting in the accounts of clever rich people and multinationals waiting to be seized. HMRC’s own breakdown shows that small businesses account for the clear majority of the gap — well over half — and that the single largest cause, across the whole system, is not deliberate avoidance but failure to take reasonable care: ordinary error, running at around a third of the total. Deliberate avoidance of the kind that fills headlines is one of the smallest components, and has been shrinking for years. This matters enormously for policy, because it tells you that most of the tax gap will not be closed by a crackdown on a few villains. It will be closed, if at all, by making the system simpler to comply with and better resourced to enforce — which is far less exciting than a war on tax dodgers, and far more true.

So our approach follows the diagnosis, and much of it is already in this paper:

• A simpler system leaks less. If the biggest single cause of the gap is honest error, then simplification is a compliance measure, not just a convenience. The aligned income, capital-gains and inheritance rates of this chapter remove the boundaries that the avoidance industry games; the radically simplified small-business VAT of Chapter 3 removes the complexity in which small-company mistakes breed; the move toward digital, pre-populated tax records reduces the errors of manual filing. We narrow the gap most not by hunting mistakes after the fact, but by building a system in which fewer are made
• A properly resourced HMRC, costed from public data and subject to independent verification. Enforcement genuinely pays for itself — every pound spent on HMRC’s compliance work returns many times over in tax collected or protected. We will invest in that capacity: the skilled compliance and debt-management staff, and the modern data systems, that a tax authority hollowed out over years now lacks. We put the cost of that investment at up to around £1bn a year, and — this is the important part — we score it as a cost and bank none of the yield. The recoveries are real but they arrive with a lag and to an uncertain timetable, and this paper will not do what the government has done in staking billions of committed spending on a compliance windfall that the Office for Budget Responsibility itself doubts will fully arrive. Better-collected tax is upside to our deficit, not a war-chest we have already spent
• Measure the wealthy gap before claiming it. There is one part of the gap where the honest answer is that we do not know its size — and neither, officially, does the public. HMRC holds internal estimates of the tax lost through offshore non-compliance by wealthy individuals that it does not publish, and the National Audit Office has said the true figure is materially larger than the small sum HMRC discloses. We will require it to be published. You cannot close, or honestly cost, a gap you refuse to measure; transparency about the offshore and wealthy gap is the precondition for doing anything serious about it, and it costs nothing but candour

The contrast with the usual politics of the tax gap is the whole point. Governments of both stripes have reached for uncollected tax as the painless way to fund promises — the magic money tree that lets you avoid naming a loser. We do the opposite. We name the gap honestly, we describe what it actually is rather than what is convenient, we invest in closing it without pretending that investment pays for itself on a political timetable, and we bank nothing we have not collected. If the recoveries come, they make our numbers better than we have shown. That is the only honest direction for a surprise to run.

Chapter 2: Fiscal Responsibility

Borrowing Rules: A Falling Path, Enforced by Our Own Tax Cuts

Fiscal rules fail in Britain because they bind future governments and never the present one. Ours binds us — visibly, automatically, from Year 1. We will enshrine in law a deficit glide path (the deficit being this year’s borrowing — 4.2% of GDP today — not the £2.8 trillion accumulated debt, which stands at 94.1% of GDP and falls only as deficits do):

• The deficit never exceeds today's 4.2% of GDP after a single transition year. We have to be exact about that carve-out rather than hide it: on the central case Year 1 peaks at about 4.4% of GDP, above today's level, because the Health Levy's spending half and the structural transition costs land before the growth dividend does. The rule therefore binds from Year 2 onward, and Year 1 is written into the legislation as a declared transition year with a stated ceiling. A fiscal rule that its own author breaches in the first year is not a rule, so we would rather legislate the exception in public than discover it in office
• About 1.6% of GDP by Year 5 in the central case — 1.6% in the Low case and about 1.1% in the High case. These figures are far below the 3.8% an earlier draft of this paper stated, because that draft predated the rebasing onto the OBR's published path and the deficit endpoints fell sharply as a result
• Below 3% of GDP from Year 2 onward, not Year 7 as an earlier draft had it — on the corrected path the Year-7 test is redundant, and we have replaced it rather than claim credit for clearing a bar we pass five years early
• A 1% structural deficit by the early 2030s — the point at which debt falls decisively as a share of the economy
• Exceptions permitted only during formally declared recessions or national emergencies
• Any surplus in good years goes directly to debt repayment — not new spending

Enforcement is automatic: Steps 2, 3 and 4 of the tax programme (Chapter 1) are pre-legislated but gated — each proceeds only when the Office for Budget Responsibility certifies, in its published forecast, that the deficit is on or below this path — an arithmetic test, not a judgement — and the National Audit Office reports afterwards on whether the test was applied honestly. A rule that pauses our own tax cuts is harder than a cap that pauses nothing. If growth disappoints, the next step waits; the path holds either way.

Office for Fiscal Honesty

Three bodies do three different jobs here, and the distinction matters because it is what makes the gates real rather than rhetorical. The Office for Budget Responsibility certifies, before a step fires, that the headroom exists — a forecast judgement, which is its statutory function. The National Audit Office reports afterwards on whether the rule was actually followed — an audit of what happened, which is its. And a new Office for Fiscal Honesty does the thing neither currently does: scrutinise promises before they are made, and name the risks nobody has an incentive to name. The existing OBR will be strengthened and the Office for Fiscal Honesty established to:

• Independently audit all government spending promises before they are made
• Publish annual debt reduction progress reports
• Provide fully transparent public accounts, accessible to every citizen
• Publish an annual assessment of the take-up risk — the gap between what citizens are entitled to and what they currently claim — and the fiscal exposure if artificial intelligence closes it

One risk in particular the Office is charged with naming, because no one else has. Every fiscal forecast in Britain — ours included — rests on an unstated assumption that benefit take-up stays where it has historically sat, well below full entitlement. Somewhere between £16 billion and £23 billion of legally-owed support goes unclaimed each year, suppressed not by policy but by friction: the forms, the means tests, the appeals no one has the time or confidence to pursue. Artificial intelligence is dismantling that friction, and as it does, spending no forecast contains will crystallise as a legal obligation to pay. We do not book a number against this, because to do so would be to guess. But we refuse to hide it. It is a genuine and material risk to the public finances, and naming it honestly is itself the argument for two things this paper already commits to: simplifying entitlements so the system can afford to pay everyone in full, and making the more subjective assessments — where judgement, not data, decides — rigorous enough to withstand a surge in well-prepared claims. A benefit that survives only because people fail to claim it is not sustainable, and a subjective test that survives only because few challenge it is not robust.

Long-Term Goal

Reduce public debt below 60% of GDP within a generation — the threshold at which debt becomes a structural drag on growth rather than a manageable tool of economic management.

A More Productive State: The Civil Service

The civil service has grown sharply. Headcount fell to a low of around 416,000 at the time of the 2016 referendum; by 2025 it stood at roughly 550,000 — about 36% higher, and close to its early-2000s peak. A serious fiscal paper has to ask why, and answer honestly. The growth was not a sudden outbreak of waste. It has two identifiable causes: Brexit, which repatriated regulatory functions from Brussels and required new border and customs capacity, adding around 40,000 between 2016 and the pandemic; and Covid, which added a further 56,000 between early 2020 and 2022 to run furlough, testing and the vaccine rollout. We say this plainly because the populist story — that a bloated blob expanded out of greed — is not what the figures show, and a plan built on a false diagnosis will fail.

The honest diagnosis points to the honest remedy — and away from the one that keeps failing. Successive governments have announced arbitrary headcount targets: a return to 2019-20 levels, a return to the 2016 size, a 15% cut in running costs. They keep missing them, because a cap plucked from the air either pushes the work out to more expensive consultants and contractors or is quietly abandoned when services suffer. We will not pretend a number is a policy. Our approach is the reverse: get the size of the state right by getting its productivity right, and let a smaller headcount follow as a result rather than chasing it as a target.

• Reverse what is genuinely reversible. If the growth was driven by Brexit and Covid, then unwinding those drivers is the principled basis for reduction — not an arbitrary cap. The Covid surge was always meant to be temporary and should unwind further than it has. And our own European policy does something no rival plan can claim: by rejoining the Single Market (Chapter 4), we re-export some of the regulatory and customs function that Brexit forced Britain to rebuild in-house. We do not just want a leaner civil service; our policy actively removes one of the two reasons it grew
• Digitise and automate the routine. The technology this paper backs elsewhere (Chapter 17) applies first and most obviously to government itself: AI and modern digital services can take on the repetitive casework — processing, checking, form-handling — that absorbs a large share of administrative effort. The aim is fewer people doing low-value manual work, and the people who remain better paid and doing work that needs judgement. A more productive state is a better employer, not merely a cheaper one
• Cut the consultancy bill, not just the payroll. The headline civil service number can be a distraction: governments that slash permanent staff often spend more hiring the same skills back from consultancies at several times the cost. A genuine productivity drive builds capability in-house — particularly digital and commercial skills — so the state stops renting expensively what it should own. The right measure is the total cost of getting government’s work done, not the headcount line alone

We are honest about three things. First, headcount is not the same as the size of the state: fewer civil servants does not automatically mean lower spending. Second, on the cost per post we have corrected our own basis rather than defend it. An earlier draft used a conventional figure of around £45,000 per post and described it as covering salary, accommodation, pension and employer contributions. That is not arithmetically possible. The Civil Service employer pension contribution is 28.97% of salary across all bands for 2024-25 to 2026-27, and employer National Insurance adds 15% above the £5,000 secondary threshold, so on-costs alone are roughly 44% of salary before a desk is paid for. A £45,000 all-in figure would imply an average salary of about £31,800 with no accommodation at all, or £27,600 once estate and IT are included — neither of which is credible. Built properly, a post costs about £51,000 at a £32,000 salary, £56,000 at £35,000 and £60,000 at £38,000, each including roughly £6,000 of accommodation and support. On that basis 132,000 posts is worth about £6.8bn to £7.9bn a year, and the £8bn carried in our scorecard sits at the top of that range — it implies an average salary near £38,500 including accommodation. We flag it as being at the upper end rather than in the middle, and the figure to firm it up is the Cabinet Office's published median civil service salary. Third, there are functions where fewer people simply means worse service — at the front line, in inspection and in delivery — and we do not pretend otherwise.

Building Britain Affordably

There is a hidden tax on everything Britain tries to build, and it may be the single greatest brake on national renewal. We do not just build too little — we build at a cost that shames our peers. Independent analysis finds that, compared with seven other wealthy nations, Britain pays roughly twice as much to build new railways and around a tenth more per mile of road; HS2 has run at something like eight and a half times the cost of comparable European lines. At about £396m per mile, the first phase of HS2 is the most expensive above-ground railway ever built anywhere in the world — more than eight times the cost per mile of France’s Tours–Bordeaux high-speed line. This matters far beyond any single project, because the same planning system, the same land-acquisition laws and the same contracting habits sit behind every home, every grid connection, every reservoir and every data centre this paper depends on.

The deeper damage is not the overspend itself but what it does to ambition: when everything costs several times what it should, we simply build less of it. Crossrail 2 sits frozen and HS2 was cut to a stub, while France, Spain and Japan keep laying track. One transport researcher estimates that at Scandinavian construction costs Britain could build four Crossrails for less than a quarter of what Crossrail 2 alone was projected to cost. A country that cannot build affordably cannot grow, cannot house its people, and cannot power its industries — which is why this belongs in the chapter on fiscal responsibility, not buried in transport: every pound wasted gold-plating a project is a pound that builds nothing.

The causes are well understood, and honesty requires naming all of them rather than the one most convenient to blame:

• A consenting process that has become an industry in itself. The environmental statement for phase one of HS2 ran to some 50,000 pages; the planning application for the Lower Thames Crossing reportedly cost over £250m to assemble — a quarter of a billion pounds spent so that one arm of the state could ask another for permission, with no guarantee of a yes. We will standardise and time-limit the consenting process for nationally significant projects: a defined document set, fixed statutory deadlines, and decisions that stick
• Repeated, open-ended legal challenge. Judicial review is a vital constitutional safeguard and we will not abolish it. But the same project being challenged again and again, each round costing months and millions while teams idle, has become a veto in slow motion. We will reform the process for major infrastructure — limiting repeat challenges on points already decided and tightening the grounds and timetable — while preserving genuine recourse against unlawful decisions. We are clear-eyed that this touches civil-liberties territory, and the reform must be proportionate; the test is to stop abuse of process, not to silence legitimate objection
• Bespoke gold-plating instead of off-the-shelf design. Britain habitually redesigns from scratch what others buy off the shelf, and layers on requirements no comparable country imposes — the much-cited £100m HS2 bat shelter is a symbol of a wider habit. We will adopt standardised, repeatable designs and proven international specifications as the default, with bespoke engineering the exception that must be justified, not the norm
• Weak clienting and scope that never stops moving. Not all of this is the planning system’s fault, and we will not pretend otherwise: HS2 also suffered years of shifting specifications, optimistic budgeting and poor cost control — the employee who warned that costs were being understated was vindicated at tribunal in 2025. We will set scope and budget once, with a single accountable sponsor and disciplined change control, so that “just one more revision” stops being the most expensive sentence in British infrastructure

These reforms cost the Exchequer almost nothing — they are rules and discipline, not spending — yet they make every other capital commitment in this paper go further: the homes of Chapter 8, the grid and the cheaper energy of Chapter 10, the railways of Chapter 11, and the data-centre and laboratory build-out of Chapter 17. Bringing Britain’s build costs even partway back toward European norms would free billions, or equivalently let the same budget build two or three times as much. We are honest that no single reform is a silver bullet and that cultural change in how the state procures takes a parliament, not a Budget; but the prize — a country that can build again, and build affordably — is among the largest on offer anywhere in this programme.

Digital Infrastructure: Connectivity as a Utility

The same build-cost discipline applies to the infrastructure of the modern economy: fast, reliable internet and mobile coverage. In a country that does most of its work, learning and commerce online, a fast broadband line and a usable mobile signal are no longer luxuries — they are utilities, as essential as water and power, and we treat them that way. Britain’s problem is not the cities, which are well served; it is the persistent gap in rural and smaller-town coverage, the not-spots and the half-built 5G that leave whole communities and businesses throttled. This is not mainly a money problem — it is a planning and competition problem, the same friction that drives up every other build cost.

• Make mast and fibre rollout easy to build. The biggest brake on rural coverage is planning friction — the cost and delay of getting masts approved and fibre laid. We extend permitted-development rights for telecoms infrastructure, simplify mast-sharing and street-works rules, and apply the same standardised, time-limited consenting we apply to other infrastructure, so coverage reaches the places the market alone will not
• A genuine universal service obligation. We set a meaningful minimum standard for broadband speed and mobile coverage that every household and business is entitled to, backed by shared-rural-network commitments and targeted subsidy only for the genuinely uncommercial final percentage — the remote premises the market will never reach on its own. The state’s job is to set the obligation and clear the obstacles, not to build the network itself
• Use the spectrum we have well, and plan for what comes next. Spectrum is a finite national asset; we will license it to maximise coverage and competition rather than simply to raise the most at auction, with rural coverage obligations attached to licences, and a clear path for next-generation networks — including the satellite connectivity that can now reach the hardest-to-serve places quickly

Fair Taxes. Honest Spending. A Debt-Free Future.

Chapter 3: Small Business and Enterprise

Small businesses are the backbone of the British economy — employing more people than any other sector, anchoring high streets, and generating the innovation and dynamism that larger companies cannot. Yet the current tax and regulatory system treats small business creation as an afterthought and growth as something to be penalised.

The VAT Growth Cliff Edge — Fixed

The VAT registration threshold creates one of the most destructive growth disincentives in the British economy. Currently set at £90,000, it forces a binary choice: stay below the threshold and remain competitive, or cross it and immediately face 20% VAT on sales, significant administrative burden, and often a loss of price competitiveness against larger rivals.

The result is documented and damaging: businesses deliberately cap their turnover, decline contracts, and refuse to hire — not because they cannot grow, but because they cannot afford to. This is a government-designed brake on enterprise.

Single Market constraint: EU VAT rules cap the full VAT exemption threshold at €85,000 — approximately £72,000. Britain’s current £90,000 threshold is already above this level. F³ will negotiate to maintain the current threshold as an EEA accession condition, and introduce a two-tier structure that achieves the core goal of removing the growth cliff edge within EEA rules:

• Tier 1 — Full VAT exemption: maintain current threshold at approximately £90,000, negotiated as an EEA accession condition (Britain has the highest threshold in Europe and the strongest case for preserving it)
• Tier 2 — Expanded Flat Rate Scheme: businesses with turnover between £90,000 and £350,000 pay VAT at a simple flat rate of 8% on gross turnover — no input tax calculations, no complex accounting, one quarterly number
• Both tiers index-linked to CPI — ending the fiscal drag that repeatedly brings small businesses into the VAT system against the spirit of the policy

The Flat Rate Scheme already exists in the UK — F³ dramatically expands its coverage and raises the ceiling from £150,000 to £350,000. A business turning over £200,000 currently faces the full complexity of standard VAT. Under F³ it pays 8% on turnover in a single quarterly return. The compliance saving is substantial even though the liability is not zero. The core problem — the cliff edge that traps businesses — is eliminated through simplicity rather than full exemption.

The net revenue cost of this two-tier structure is approximately £0.8bn annually — lower than the original £150k full exemption proposal, reflecting the Flat Rate Scheme generating some revenue from businesses currently capping below £90k.

Simplified VAT Reporting

For businesses below the simplified scheme threshold:

• Quarterly VAT submissions only — no monthly filing
• Simplified flat-rate scheme with reduced administrative requirements
• Digital-first but not digital-only — no small business penalised for not having sophisticated accounting software

Entrepreneurs should spend their time serving customers, not filling in forms.

The Full Small Business Package

Taken together, the F³ programme delivers the most comprehensive package of small business support in a generation:

• Business Rates abolished entirely — the single tax most damaging to high streets, hospitality and physical retail
• National Insurance abolished — outright for employees; for employers an 8.5% levy: a 7.5% core at half the old rate, plus a Training Point that funds free apprentice training for small firms from a national pool — so the local plumber or hairdresser pays a wage, not a training bill, to take one on
• VAT two-tier structure — full exemption to £90,000, simple 8% flat rate to £350,000, eliminating the growth cliff edge
• Simplified VAT reporting — reducing compliance costs
• Faster planning approvals — making it easier to open premises, expand or change use
• Free town-centre parking — bringing customers back to the high streets where small businesses trade
• Single Market rejoining — restoring export access to 450 million European customers without the friction costs that currently make cross-border trade unviable for small operators

The combined effect of Business Rates abolition and the Employer Levy saves the average small employer a substantial sum annually — but we should be precise rather than slogan-like about the payroll half of it. 'Employer NIC substantially reduced' describes the core rate at full implementation, which our own central case does not reach within this parliament. Three things sit on top of the headline: the Training Point adds a point back (recoverable by firms that train), the 2% care contribution to the National Care Fund is a new charge on qualifying earnings, and the Employment Allowance is abolished, which matters most to the smallest employers. The worked examples in Part Eight show all four effects at Year 1 rates, at the Year-5 central rate, and at full implementation, so an employer can see which number applies when.

Employment Status: One Clear Deal, Honestly Enforced

Cheaper to hire is only half the story; the other half is the terms on which people work — and here Britain has made a genuine mess. Employment law recognises three statuses — employee, worker and the self-employed — while the tax system recognises only two. The gap between them is where a decade of confusion has bred. Firms have an incentive to label staff self-employed to avoid the employer tax on hiring; individuals sued to be recognised as ‘workers’; and the courts have spent years deciding case by case who is really what, from the Uber drivers found to be workers to the Deliveroo riders found to be genuinely self-employed. The result is uncertainty for everyone: a business cannot be sure what it owes, and a person cannot be sure what they are owed. Worst of all is the trap in the middle — the contractor taxed as an employee who receives none of an employee’s rights in return. Taxed like staff, treated like a stranger. That is not a defensible position; it is simply an unfinished one.

Our instinct throughout this paper is to simplify rather than to patch, and employment status is no exception. We support moving from three statuses to two: worker, with a clear set of rights, and the genuinely self-employed, who trade on their own account and carry their own risk. The test for which is which should be the one the courts have already converged on — not the label on the contract but the reality of the relationship, above all whether a person is genuinely independent or is substantially controlled by, and dependent on, a single business. A cleaner line means fewer tribunals, less uncertainty, and far less room for the deliberate misclassification that has become a business model in parts of the economy.

The deepest fix, though, is one we have already made elsewhere in this paper without anyone calling it employment reform. The reason misclassification pays is the tax gap between employment and self-employment: an employer that calls a worker self-employed sidesteps the employer’s National Insurance on their pay. Our tax reform closes much of that gap at source. With employee National Insurance abolished and the employer’s contribution replaced by a single, broad Employer Levy (Chapter 1), the tax advantage of dressing up employment as self-employment shrinks dramatically. Take away the tax reward for misclassification and a large part of the problem simply stops being worth the trouble — the honest structure and the cheaper one become the same thing. It is the principle of this whole paper applied to the labour market: tax the same activity the same way, and the games fall away.

On rights themselves, we favour a single, clear deal in place of today’s patchwork, in which some protections begin on the first day, some after six months, and some only after two years. We would set one qualifying period — six months — across the core protections, including statutory sick pay. That gives an employer a genuine, predictable probationary window in which to take a chance on someone without assuming the full weight of employment obligations from day one — the single biggest thing that makes a small firm hesitate before hiring. And it gives the worker a clear date on which they know their protections vest in full. One probation, one threshold, no guessing.

But a probationary window is a licence to part ways, not a licence to mistreat — and once it has passed, the protection on the other side must be real. Today it often is not, because compensation for unfair dismissal is capped at a level a large employer can treat as a modest cost of doing business. We would remove that cap. Tribunals should be able to award uncompensated losses in full where a dismissal is found genuinely unfair, as they already can in discrimination cases. We should be clear about the trade-off, because it is the honest objection to this proposal: yes, this raises the stakes of dismissing someone after their probation, and deliberately so. A protection an employer can cheaply buy its way out of is not a protection; it is a price list. The design is the balance — a genuinely free hand for six months, and genuine consequences after — and it asks employers to do the one thing the current system lets them dodge: get the decision right, or make the person whole.

None of this is a cost to the Exchequer — statutory sick pay is met by employers, not the state, and the rest is regulation rather than spending, so none of it appears in our scorecard. It is a cost, where it is a cost at all, to employers who dismiss unfairly or who built a model on misclassifying their staff. For the great majority of businesses — who hire honestly, treat people decently and part ways fairly when they must — the effect is not a new burden but a simpler, more predictable set of rules, sitting alongside the thousands of pounds a year our tax reforms have just saved them.

The State as Customer: Procurement as a Growth Lever

The state is the single largest customer in the British economy, spending roughly £385bn a year on goods, works and services — about a third of all public spending. How it spends that money is one of the most powerful and least-used growth levers in government’s hands. For a small or scaling company, a government contract is more than revenue: it is a credential, it is reliable cash flow paid within 30 days, and it is the reference that wins the next ten private contracts. Yet historically only around a fifth of public procurement has gone to small and medium-sized firms, and too much has flowed to a handful of large incumbents and outsourcing giants. We treat procurement not as a back-office function but as industrial strategy conducted through the chequebook the state already holds.

The legislative groundwork exists — the Procurement Act 2023 already places a duty on public bodies to consider barriers to SME participation, mandates 30-day payment down the supply chain, and lets authorities reserve smaller contracts for SMEs and UK suppliers. The failure has been one of will and execution, not statute. We make it bite:

• Hard, published SME spend targets with teeth. Every department sets and publishes a direct-SME spend target and reports against it annually; departments that miss must explain why and set out corrective action. Transparency is the enforcement mechanism — a target no one reports against is a target no one meets
• End late payment, the silent killer of small firms. Late payment closes dozens of UK businesses every day and turns small suppliers into involuntary lenders to large ones. We enforce 30-day payment rigorously through the public supply chain — including down to subcontractors, via project bank accounts on major contracts — and bar persistently late-paying prime contractors from future public work. Prompt payment is not a courtesy; it is the cheapest growth policy there is
• Break big contracts into lots small firms can actually win. Default to disaggregating large procurements into smaller lots wherever practical, publish a genuine forward pipeline so SMEs can prepare to bid, and strip out the disproportionate insurance, turnover and track-record requirements that exclude capable smaller suppliers before they start
• Procurement as a market for innovation. Government should be the first, demanding customer for British innovation — using small-business research contracts and challenge-led procurement to buy from young companies in the Industrial Strategy sectors, the same firms the British Future Fund (Chapter 15) invests in. A startup with a government reference customer and a Future Fund stake behind it is a startup that scales here rather than selling early abroad: the chequebook and the cheque-writer pulling in the same direction

The truth is that this is redirection, not new money: it costs little to the Exchequer because the £385bn is already being spent — the question is who wins it. The named loser is the comfortable incumbent that has relied on scale and complexity to keep smaller rivals out. And we are honest about the tension to manage: SME-friendly procurement must not become a cover for poorer value or weaker delivery, so value for money and capability remain the test — we are widening the field of credible bidders, not lowering the bar they must clear.

The Crown Commercial Dividend

There is money in how the state pays, not only in what it buys. Government already aims to pay small suppliers within five days and every valid invoice within thirty. That prompt payment stays free for small firms — it is a policy we support and would not sell back to them. But for large suppliers, the state's covenant is worth something, and at present it gives it away. A company paid on day one instead of day thirty has avoided a month of working-capital cost, and government can borrow more cheaply than almost any of its suppliers can. That gap is a saving nobody is currently collecting.

So we will offer large suppliers accelerated payment in exchange for a discount, calculated dynamically and taken at the supplier's option:

• Small and medium suppliers keep prompt payment at no cost — the five-day and thirty-day commitments are unchanged and unpriced
• Large suppliers may elect payment on day one to five in return for a discount set by the platform, reflecting the number of days accelerated and prevailing rates
• The supplier chooses which invoices to accelerate, invoice by invoice. Nothing is imposed, and declining costs a supplier nothing
• The same platform detects duplicate invoices automatically and pursues contractual service credits, delay damages and warranty recoveries — money already owed under signed contracts that the state routinely fails to claim

The arithmetic is deliberately modest. If £50-100bn of the £385bn participated at a 1% discount, and paying twenty-five days early costs government roughly 0.27% at current funding rates, the net saving is somewhere around £0.36-0.73bn a year. We book £0.3bn, below the bottom of that range, for two reasons a reader should be able to check. Public-sector dynamic discounting exists but only at far smaller scale, so take-up at this level is an assumption rather than an observation. And this is a saving on spending, not a tax receipt: it reduces what departments pay out, and it is scored that way rather than dressed up as revenue. Only cash discounts actually taken and recoveries actually received count. We have not booked a penny of the duplicate-invoice or service-credit recoveries, though we expect them to be real.

A Digital State: Identity Done Once, Done Securely

Ask anyone who has recently tried to prove who they are to the British state, or to a bank, employer or landlord acting on its behalf. They will have photocopied a passport, dug out utility bills, and quite possibly paid a solicitor a three-figure fee to certify that they are themselves. We do this over and over, to dozens of bodies that each hold their own partial, duplicated, error-prone record of us. It is slow, it is expensive, it is a gift to fraudsters, and almost every other developed democracy has stopped doing it. Britain is unusual not in worrying about identity systems, but in not having a modern one.

We will build a secure national digital identity — a single, verified, government-backed credential that a citizen can use, at their choice, to prove who they are, access public services, receive benefits and pensions, hold a driving licence, share a medical record with a clinician, and demonstrate the right to work or rent. Done well, the prize is large in three directions at once: convenience for citizens (the end of the repeated, costly identity rigmarole), efficiency for the state (Estonia, which runs exactly such a system, estimates it saves the equivalent of around 2% of GDP a year and the work of well over a thousand officials, by asking citizens for each piece of information only once), and integrity (a reliable right-to-work and right-to-rent check is central to the immigration system this paper sets out in Chapter 9, and digital benefit delivery sharply cuts the fraud and error that cost billions).

One Account for Debts Owed to the Public

The digital identity architecture above makes possible something government still does badly: knowing, in one place, what a person or company already owes across every public body. At present HMRC, the DVLA, local authorities, the courts and the student loans system each pursue their own debts in their own queues, and the state can be paying a grant out of one door while chasing the same person for an undisputed liability at another. We will consolidate that into a single view of uncontested public debt, and check it before making a non-essential payment.

• Before a tax refund, grant, supplier payment or other discretionary payment is made, the account is checked and net settlement is offered or applied
• No deductions from essential benefits, and hardship and vulnerability protections apply throughout
• No offset at all while a liability is under appeal — the debt must be uncontested
• Affordable payment plans remain available, and recovery effort concentrates on those demonstrably able but unwilling to pay
• For serial phoenix operators — directors who repeatedly abandon PAYE and VAT liabilities and start again — a tax security deposit may be required before operating another limited company, subject to appeal and risk assessment, so that an honest second attempt remains possible

We score this at zero, and the reason matters more than the number. Current government plans already assume around £11bn of additional debt collection by 2029-30 and about £250m from measures against phoenixism. Anything this reform achieves has to be measured against that baseline, not against a world in which nobody was collecting. Only an independently verified improvement beyond existing plans would ever be counted, and we would rather book nothing and deliver something than claim a share of money the Exchequer has already promised itself. What we do claim is the principle: this collects liabilities that already exist. It raises nobody's lawful tax bill by a penny.

But we are candid that a national identity system is one of the most sensitive things a government can build, and that Britain has rejected past attempts for good reasons. The dangers are real: a giant central database is a honeypot for hackers and hostile states; a rigidly digital system can lock vulnerable people out of the benefits and services they are entitled to; and an identity tool built for convenience can creep into an instrument of surveillance. A proposal that waved these away would not deserve to be trusted. India’s Aadhaar is instructive on both counts. It is, by most measures, one of the most successful state-capacity projects ever attempted — it brought over a billion people into a verifiable identity system, cut fraud and leakage sharply, and got benefits to hundreds of millions who had previously lost a slice of them to middlemen. It also produced real cases of exclusion, where people entitled to help were denied it because a fingerprint failed or a record did not match. We take both lessons seriously, while being honest about proportion and context: the exclusions, though real and damaging to those affected, were a small fraction of an enormous whole, and they bit hardest precisely because many of those people had no other way to prove who they were. Britain’s position is different — we already have near-universal documentation, so a citizen who fails a digital check has passports, national insurance records, NHS numbers and bank identities to fall back on, which makes that specific failure mode far less likely here. We are not importing a system wholesale; we are learning from the best of Aadhaar’s reach and the best of Estonia’s design, and building for British conditions. The design is the policy:

• No single central database. Following Estonia’s X-Road architecture, data stays where it already lives — with the NHS, DVLA, HMRC, DWP — and is connected only when needed, rather than piled into one giant store. There is no honeypot to steal, and no single point of failure; Estonia adopted precisely this design after a major cyber-attack exposed the fragility of centralised systems. The system is a secure exchange, not a super-database
• The citizen sees and controls access. Every time any body looks at a citizen’s data, it is logged, and the citizen can see who accessed what and why — with unauthorised access a criminal offence. The principle, which we adopt explicitly, is that your data is yours: the state and its agencies are accountable to you for every use of it, not the other way around
• A legal right to live without it. No one will ever be denied a benefit, a service or their rights because they cannot or will not use the digital credential. A properly staffed non-digital route is guaranteed in law, precisely to avoid the exclusion that has done real harm elsewhere. The credential is a convenience offered to citizens, not a condition imposed on them
• Hard statutory limits on what it can be used for. The purposes the identity may serve are defined in legislation and cannot be expanded by administrative drift. It is not, and may not become, an internal passport, a location tracker, or a tool for monitoring lawful behaviour. Function creep is the fear; a legal boundary, enforced by an independent commissioner, is the answer

It is also true that even the best design is not perfect: Estonia has had to fix genuine flaws in its cards, its logging of law-enforcement access is not complete, and no system is unhackable. Security here means well-engineered, independently audited and continuously maintained, not magic. But the choice is not between a digital identity and no identity system — it is between the expensive, insecure, fraud-prone patchwork we have now, and a modern one built on the right principles. Handled with the seriousness it demands, a digital identity is one of the largest single efficiency, security and convenience gains available to the British state. We think it is worth doing, and worth doing properly.

Chapter 4: Britain and Europe

Part One — The Case for the Single Market

What We Are Proposing — And What We Are Not

Let us be completely clear about what F³ proposes and what it does not, because this will be misrepresented.

This Is Not Reversing Brexit

Britain voted in 2016 to leave the European Union. F³ respects that decision entirely. We are not proposing to rejoin the EU, to restore EU citizenship, to accept the jurisdiction of the European Court of Justice over British domestic law, or to hand back the political independence that Brexit delivered.

What Brexit did not require — and what we are restoring — is the economic relationship that made British businesses competitive. Norway is not in the EU. Iceland is not in the EU. Liechtenstein is not in the EU. All three have been in the Single Market for thirty years. None of them have surrendered their political independence. None of them fly the EU flag or sit in the European Parliament. They simply trade freely with their largest neighbour.

Britain had both political independence and economic access before 2016. We are not asking the British people to choose between them. We are asking them to recognise that we can have both — as three other independent nations already do.

The Five Things Brexit Promised — And What F³ Delivers

Brexit was sold on five promises. Here is the honest scorecard — and what F³ actually delivers on each:

The Small Boats — A Brexit Problem With a Negotiated Fix

The numbers frame the problem better than any rhetoric. In 2018, the first year official figures were kept, 299 people crossed the Channel by small boat. By 2022 the figure was 45,774 — roughly a hundred-and-fifty-fold rise in four years — and it remained around 41,000 in 2025. We are careful about the claim we draw from this: Brexit did not create illegal migration, and the smuggling gangs, the closure of other routes, and global displacement all play their part. What Brexit did was remove Britain’s most effective legal tool for responding to it. Before 2021 Britain was part of the EU’s Dublin Regulation, the framework for returning an asylum seeker to the member state responsible for their claim — usually the first country they entered. We should be equally precise about what Dublin did and did not deliver: returns under it were limited in number, procedurally complex, and far from automatic — only a minority of cases ever resulted in a transfer. But it was the only legal returns route Britain had, and Brexit removed it overnight while the Trade and Cooperation Agreement secured nothing in its place. The result is a system that now returns only a small fraction of arrivals — around 4% since 2018 — which is why every workaround since has struggled, and why a negotiated returns agreement, not a gimmick, is the fix.

Every attempt since to solve the boats problem outside a returns framework — Rwanda, offshore processing, endless legal battles — has failed precisely because Britain has no legal mechanism to return people to the safe country they came from. A returns agreement with the EU is the missing piece, and Single Market accession is what makes the EU willing to negotiate one: it is the goodwill and the institutional relationship, not the EEA treaty itself, that opens the door. Non-EU states such as Norway participate in EU returns arrangements through separate association agreements — and a Britain inside the Single Market is far better placed to secure the same than one negotiating from outside, as the failure of every post-Brexit attempt has shown.

Free Movement — Honest and Positive

Single Market membership means free movement of EU citizens. We are honest about this. What we are also honest about is what free movement actually means in practice — and what it does not mean.

Free movement means EU citizens can come to Britain to work, study and live — as they could before 2021. It does not mean open borders to the world. It does not mean uncontrolled numbers. EU citizens who come to Britain work, pay taxes, and contribute to public services. The evidence consistently shows EU free movement has been net positive for Britain’s economy, public finances and public services.

The problems people associate with immigration — pressure on housing, GPs, school places — are not caused by free movement. They are caused by a failure to build enough houses, train enough doctors, and fund enough schools. Our platform addresses all three directly. The answer to those pressures is supply, not restriction.

And free movement runs both ways — which is the half of the argument too often left unsaid. Single Market membership does not just mean EU citizens can come here; it means the door swings open for us too. A British graduate could take a job in Amsterdam or Berlin without a work permit; a student could study in Paris or Bologna without a visa; a retiree could move to Spain or Portugal without a residence permit, and stay for good rather than counting down 90 days. Professional qualifications would be recognised across the continent again, so a British nurse, architect or engineer could practise in Lisbon without re-qualifying. The everyday irritations of a closed border would ease as well: surcharge-free mobile roaming restored by right rather than left to your provider’s goodwill, and simpler pet travel without repeating the vet paperwork on every trip. These are not abstractions for businesses — they are the freedoms a generation of Britons grew up with and lost, returned to every family with a child who wants to study abroad, a relative who wants to retire in the sun, or a career that could span a continent.

Trade Independence — The Honest Position

Critics will claim that Single Market membership means giving up independent trade policy. This is not accurate — but it requires a precise answer.

Norway, inside the Single Market, has concluded its own free trade agreements with India, Canada and many others — through the EFTA framework. Britain joining the EEA would mean coordinating future trade deals through EFTA alongside Norway, Iceland and Liechtenstein rather than purely unilaterally. Britain’s existing deals — including CPTPP — are not automatically disturbed, but nor are they untouched: EFTA accession requires the consent of all four members, participation in EFTA’s own agreements has to be negotiated with each third party, and some counterparties would have the right to be consulted. We assume negotiation, not continuity by default.

The trade-off is honest, and we will not dress it up: Britain coordinates future trade deals through EFTA rather than acting purely alone, in exchange for near-frictionless regulatory access to a market of 450 million people on its doorstep. We say near-frictionless deliberately, because outside the Customs Union some customs formality remains — more on that below. But the economic arithmetic is not close. The EU takes 42% of British exports; CPTPP countries take 8%, and those deals are worth a fraction of a percent of GDP. Deep single-market access is worth far more than the marginal freedom to negotiate independently — which is precisely why we keep that freedom rather than surrender it to a customs union for a benefit we can largely secure another way.

Most of the Customs Union Benefit, Without the Customs Union

Here is the part the old Leave-versus-Remain argument misses entirely, and it is genuinely good news. You do not need to be in the Customs Union to remove most of the friction people actually notice. The proof is already happening: the UK and EU have agreed in principle a sanitary and phytosanitary (SPS) — or veterinary — agreement, under which the great majority of movements of food, animals and plants cross the border without the certificates and checks that have clogged it since 2021. The government estimates it is worth up to roughly £5bn a year, and because it aligns Great Britain with the rules Northern Ireland already follows, it also eases the Irish Sea border. F³ would secure and build on exactly this kind of targeted agreement from inside the single market, where it is easier to reach.

We are equally honest about what such a deal does not do, because overclaiming here is how credibility dies. An SPS agreement removes the regulatory layer of friction — the agri-food checks — but it does not remove customs declarations or rules-of-origin paperwork. That residual customs friction is the one thing only a full customs union eliminates. So the real choice is narrow and specific: accept that remaining customs paperwork, or join a customs union and hand Brussels control of Britain’s external trade policy — no seat at its trade negotiations, forced to open our market to its FTA partners without reciprocal access, the Turkey trap. We judge that price far too high to pay for removing paperwork we can manage, especially when a veterinary agreement and single-market membership already capture the larger share of the prize. That is the F³ position in one line: take the friction-reduction that does not cost sovereignty, decline the bit that does.

And the residual paperwork is exactly where technology earns its keep. We are clear about the order of importance: single-market membership and a veterinary agreement do the heavy lifting on friction; the customs formality that remains, because we choose to stay outside the Customs Union, is a smaller and increasingly automatable problem. Customs is fundamentally an information process — what a good is, where it came from, what it is worth — and that is precisely the kind of task modern AI handles well. Used properly, and building on the digitisation already in HMRC’s own customs roadmap (Chapter 17), AI can read invoices and shipping documents and assemble declarations automatically, propose the right commodity codes from a product description, and risk-score consignments so that trusted, consistent traders flow through a green lane while scrutiny falls on the genuinely doubtful. The Customs Declaration Service already processes around 78 million declarations a year covering over £1trn of goods; the goal is to make each of those cheaper, faster and near-automatic for the business filing it. We are deliberately modest about what this achieves: AI does not abolish the customs border or remove a single legal requirement — only a customs union would do that, at the sovereignty cost we have just declined. What it does is shrink the administrative drag of the paperwork we keep, so that the cost of staying out of the Customs Union is smaller still. It is also, not incidentally, the same productivity logic we apply to the state itself (Chapter 2): software doing the routine processing that once needed armies of forms.

The Economic Case

The economic dividend from Single Market access is the single largest lever in this paper. The independent literature consistently puts rejoining the single market at 2–4% of GDP over the long run — on the order of £60–120bn of additional tax base at the top end — because the bulk of the Brexit cost was never tariffs but the non-tariff barriers the single market removes: regulatory divergence, services restrictions, and friction in goods. We are explicit that this is the single-market figure, not a customs-union one: outside the Customs Union, customs declarations and rules-of-origin checks remain, which is why we book the conservative bottom of the range and treat the full figure as a ceiling, not a forecast. Ireland, with fewer natural advantages, built its prosperity on exactly this foundation. Britain can do the same.

The Rule-Taker Question: An Honest Answer

Norway’s own 2024 EEA Review acknowledged the democratic deficit of adopting EU legislation without a vote. F³ does not pretend this is ideal. We accept it as a worthwhile trade-off — and we will say so openly rather than hiding it.

F³ makes this choice with open eyes. Single Market access is worth accepting regulatory alignment in most areas. Where it is not — defence procurement, certain security matters — explicit treaty carve-outs exist and will be used.

Chapter 5: Healthcare

Universal Healthcare Guaranteed

Healthcare remains available to every person in Britain, regardless of income. The NHS continues to provide:

• Emergency and urgent care
• GP services
• Hospital treatment
• Essential medicines

Free at the point of use. Always. For everyone.

Funding Reform

The Levy does two jobs, and we split it explicitly rather than let one pound do both. About half of it — £4bn in Year 1 rising to roughly £8bn by Year 5 — is additional, ring-fenced money on top of the existing base, and it is what pays for the improvements this chapter sets out: the dentistry contract rebuild, elective and diagnostic capacity, and the workforce to deliver them. It ramps rather than arriving at once, because the binding constraint in the first year is trained people, not money. Together with the £2bn mental-health expansion scored separately, that is the health half. The other half replaces general-taxation funding of the NHS, releasing it to reduce the deficit — and we say so plainly, because the alternative is to claim the same £20bn twice: once as new money for health and once as headroom in the scorecard. Both halves appear as separate lines in the fiscal table, so the reader can check that the pound is spent once. What the Levy also buys, in either half, is a secure home for a visible portion of health funding, independent of the annual spending-review scramble and the political cycle.

Let us be plain about what this is: a new, additional 2p tax on income above £15,000. It is the largest tax rise in this manifesto, it raises about £20bn a year, and it appears — in full — in every household example we publish. We will not insult the reader by calling a tax rise a 'replacement'. Nor will we claim it is the only new charge here: the 2% individual care contribution on earnings above £150,000 and the Landowner Levy on commercial freeholds are also new taxes, smaller and narrower, and both are in the scorecard. An earlier draft of this paper called the Health Levy 'the one genuinely new tax'. That stopped being true as the programme grew, and we have withdrawn the claim rather than defend it on a technicality.

One point we state openly rather than bury, because it is a real choice and pensioners deserve a straight answer. The Health Levy applies to income above £15,000 whatever its source — employment, self-employment, or pension. This is a deliberate departure from National Insurance, which was never charged on pension income or on those above state-pension age. We make the change for one reason: the NHS is used most heavily, by some margin, by older people, and a levy ring-fenced for the health service cannot coherently exempt the group that relies on it most. The £15,000 floor protects every pensioner of modest means — anyone on the state pension alone, or close to it, pays nothing. Only pensioners with substantial additional private or occupational income contribute, and they contribute at the same 2% as a worker on the same income. We think that is fair, and we would rather defend it in the open than pretend pension income sits outside the system while working income carries the whole load.

NHS Dentistry — An Urgent Repair Job

NHS dentistry has effectively collapsed as a public service. The 2006 dental contract pays dentists a fixed number of Units of Dental Activity regardless of complexity — making NHS practice economically unviable. The result: millions cannot access an NHS dentist. Adults are pulling their own teeth. Children are being hospitalised for preventable decay.

Free STEM degrees improve the dental pipeline over 10-15 years. Free EU movement restores access to European dental professionals in the short term. But the immediate problem is contract design, not workforce numbers.

• Replace the 2006 UDA contract with a capitation and quality model — dentists paid for registered patients and outcomes
• Minimum NHS commitment for all practising dentists receiving public subsidy
• Dental therapists and hygienists empowered to carry out a wider range of routine treatments

Dentistry and the Australian Model

Dentistry is the one place the Australian split already operates in Britain — and it shows what the model does and does not promise. Most adult dentistry is already co-paid through NHS bands; most Australians pay privately for dental care, which sits largely outside Medicare. So the dentistry rebuild is not about importing the Australian funding split — it is already more privately funded than the Australian norm. It is about making the public commitment that remains actually deliverable:

• Children, emergency, and clinically urgent dental care sit in the NHS guarantee — not the co-payment tier. A child hospitalised for decay is a public-health failure, never a billing event
• The capitation contract makes NHS list-dentistry viable again, so the guarantee has practitioners willing to deliver it — the binding constraint is contract design, not the funding model
• Banded co-payments for routine adult treatment continue, with the existing exemptions (under-18s, pregnancy, low income) preserved in full — the means-tested floor the Australian system lacks

The principle holds across the whole health chapter: the Australian model adds a private tier on top of a guaranteed public core — it never charges children, emergencies, or those who cannot pay. Dentistry is the proof that the line can be drawn and held.

Private Sector Partnership

Following the principles that make Australia’s Medicare system the envy of the developed world:

• Means-tested support for private hospital insurance — ensuring lower-income patients are never disadvantaged
• A 2% surcharge on the total income of individuals earning above £150,000 who decline private hospital cover — modelled on Australia’s Medicare Levy Surcharge, and this time modelled on it properly. Australia charges the surcharge on the whole of a liable person’s income, not merely the slice above the threshold, and that distinction is the entire mechanism: a marginal charge on the excess raises almost nothing near the threshold and gives nobody a reason to insure. Ours works the same way, with a shade-in between £150,000 and £187,500 — 10p in the pound across that band, reaching the full 2% at £187,500 — which is the Australian Taxation Office’s own formula, so no one meets a cliff edge. We set the threshold at £150,000 for a reason anyone can check against our own tables. It is the income at which this programme starts paying people back. Below roughly £121,000 a household is worse off under F³, and it is only at about £150,000 that the gain from abolishing the 60% taper trap and employee National Insurance first covers the cost of a private premium. We are not prepared to levy a surcharge on people our own reforms have made worse off. Above £150,000 the tax cut pays for the cover; below it there is nothing to pay with. The threshold is CPI-indexed. One further leg makes the market work, because guaranteed issue without it collapses: a late-joiner loading, under which premiums rise for each year cover is deferred beyond a set age, exactly as Australia’s Lifetime Health Cover loading does. It costs the Exchequer nothing and it is what actually stops people waiting until they are ill. Now the part that cuts against us. The surcharge raises roughly £2bn in the first year and less thereafter — by design, because everyone who responds by taking cover stops paying it. It is a revenue line that shrinks as it succeeds, and we book it that way. What we do not book is the NHS saving, and we want to be exact about why. There are about 800,000 taxpayers above £150,000. Even if every one of them went private, the activity they would take out of the NHS is worth on the order of £1bn a year; a realistic behavioural response is a fraction of that. The much larger savings this kind of shift is usually credited with would require uptake across millions of people, which in Australia is driven by a premium rebate costing several billion a year — quite plausibly more than it saves. So we score the surcharge as revenue and score the health saving at zero. If a saving materialises it is upside, and we treat it precisely as we treat the growth dividend: named, explained, and not banked
• Abolish business rates across the NHS estate. It is absurd that NHS hospitals, GP surgeries, clinics and ambulance stations pay business rates at all — one arm of the state taxing another, roughly £1bn a year cycled out of frontline healthcare and into local-authority coffers, the same circular waste we identify in the cost of building (Chapter 2). It is more absurd still that while NHS hospitals pay the full bill, around a quarter of private hospitals registered as charities receive up to 80% relief. We end it: no business rates anywhere on the NHS estate, freeing around £1bn a year for patient care. Because this is money moving between arms of government rather than new spending, it is fiscally neutral overall — local government is compensated in full through the funding settlement (Chapter 2) — but for the NHS it is roughly £1bn of real frontline resource returned
• Relieve the NHS by fixing social care, not by loading care onto it. A large share of the NHS's worst pressure — beds occupied by patients who are medically fit to leave but have nowhere safe to go — is really a social-care failure showing up on a hospital ward. Because we pre-fund social care separately, through the National Care Fund and the reinstated Dilnot cap (Chapter 20), rather than leaving it to compete for the same money, we attack delayed discharges at their source. We should be honest about the timing: the Fund does not relieve today's care bill inside this parliament, so the effect on delayed discharges builds over years rather than arriving with the legislation. Properly funded care is still one of the most effective things any government can do to free NHS capacity, and the Health Levy is not asked to carry the cost of care on top of health — the two are funded separately and deliberately
• Make private cover fair, or the surcharge is not. A surcharge that asks higher earners to take private cover only works if they can actually buy it — and today an insurer can refuse someone outright, or write out the very condition they most need covered, because of their medical history. It cannot be right to penalise someone for not buying a product they are not allowed to buy. So we legislate, in the original spirit of America’s Affordable Care Act, three protections together: guaranteed issue (an insurer must offer cover to anyone who applies), a ban on pre-existing-condition exclusions and the broad get-out clauses that ride alongside them (the policy must actually cover the conditions people have), and community rating (premiums may not be loaded on the basis of personal health status). We are honest about the trade-off community rating involves: it means the young and healthy pay somewhat more so that the older and sicker can be covered at all. We think that is the right trade — it is, in miniature, the same principle of pooled risk that underpins the NHS itself. And here the design locks together. The Affordable Care Act’s architects learned that these protections cannot stand alone: if healthy people can wait until they are ill to buy in, the risk pool curdles and premiums spiral — the “death spiral” that wrecked several markets that tried guaranteed issue without a counterweight. The ACA needed a mandate to keep healthy people in the pool. Our 2% surcharge is precisely that counterweight: it gives healthy higher earners a standing reason to hold cover rather than buy only when sick, which is what makes a fair, exclusion-free private market sustainable. The surcharge and the protections are not two separate ideas — they are two halves of one mechanism, each of which needs the other to work
• Greater use of independent providers and private facilities to reduce waiting lists

This is not privatisation. It is intelligent use of all available capacity — exactly as Australia has done for 40 years.

The Expected Outcomes

• Shorter waiting lists
• Better patient outcomes
• More NHS capacity for those who need it most
• Reduced pressure on public healthcare as the private sector absorbs a greater share — an outcome we expect, deliberately do not put a figure on, and do not book

Chapter 6: Education and Families

School Calendar Reform

Britain’s six-week summer holiday was designed for an agricultural economy. It no longer reflects modern family life, costs working parents thousands in childcare, and contributes to educational regression — particularly for disadvantaged children. The aim is simple: a school calendar designed for modern working families, with the year’s holiday weeks distributed more evenly rather than concentrated in one long summer block.

• Summer break reduced to approximately 4 weeks
• Holidays distributed more evenly throughout the year
• Schools encouraged and funded to run holiday clubs and enrichment activities

This directly improves productivity by keeping more parents — disproportionately women — in work at full hours throughout the year.

University Tuition Fees

University tuition fees have been frozen, cut and raised at political whim for two decades — creating uncertainty for universities, students and the sector as a whole. The result is institutions that cannot plan long-term and students who cannot understand what they are committing to.

• Tuition fees index-linked to inflation (CPI) from the point of this paper’s implementation
• No further political interference in fee levels — universities plan on a stable, predictable basis
• Student loan repayment terms reviewed to ensure graduates from lower-income backgrounds are not disproportionately penalised

Index-linking ends the cycle of political manipulation while giving universities the financial certainty to invest in teaching and research.

Keeping Our Universities World-Class — and Solvent

Index-linking stops the fee eroding further, but it does not repair the hole already dug. The real-terms value of the tuition fee has fallen by around a quarter since 2012, so universities now lose money on every British undergraduate they teach and have plugged the gap with international-student fees. That model is breaking: with recruitment volatile, the Office for Students reports that close to half of English universities are now running deficits, and a number are within sight of insolvency. Meanwhile the research that underpins our global standing runs at a structural loss — public and charitable grants cover only around 70 to 80 pence of every pound of research actually done, a shortfall of several billion a year. This is not a footnote. Our universities are, with the City and the life-sciences sector, one of the handful of things at which Britain is genuinely world-class, and they are being hollowed out by a funding model that no longer adds up.

The reality is that one of this paper’s own commitments adds to the pressure. Rejoining the Single Market restores home-fee status to EU students, which is right in principle and welcome to them — but it means EU students who currently pay full international fees will pay the capped home rate instead, and those studying the strategic subjects qualify for free tuition on equal terms. That is a real reduction in one stream of university income, and we will not pretend a policy we support elsewhere is costless here. It strengthens the case for putting university funding on a sounder footing rather than leaving institutions to chase ever more overseas students to stay afloat.

So we make a deliberate choice, and we cost it. A new University Research and Strategic-Teaching Sustainability Fund commits around £3bn a year, and — unlike much else in this paper, which is gated to fire only as the OBR allows — it is a standing priority from Year 1. It is worth stating what that means for the arithmetic: it is a genuine cost that raises the central deficit by about £3bn and narrows the fiscal headroom the package reaches by Year 5. We make that trade openly, because a world-class research base starved into decline is a false economy that costs far more in lost growth, lost life-sciences leadership and lost talent than the £3bn it takes to sustain it.

The Fund is weighted deliberately toward two things only universities do: the research base, and the high-cost strategic teaching — medicine, engineering, the sciences — that costs more to deliver than any capped fee can cover. It is not a general subsidy to the sector to carry on as it is. In return we expect reform: honest published data on course outcomes and graduate earnings, restraint on the grade inflation that a volume-chasing funding model produced, and governance that plans for sustainability rather than gambling on ever-rising recruitment. And we are clear about what this is not: it is emphatically not a licence to drift toward the American model, where a handful of universities sit on vast endowments and charge tuition to match. We want institutions funded well enough that they need not behave like businesses — neither starved into corporatised volume-chasing, nor hoarding wealth like a hedge fund with a campus attached.

One route among several. This investment sits deliberately alongside, not above, the vocational path. The 18-to-22 Guarantee and the Training Point (Chapter 6) fund an apprenticeship or further-education place for every young person who wants one, and they do so through a redesigned, economy-wide training levy that reaches every employer — fixing a broken Apprenticeship Levy rather than adding new Exchequer cost. The two routes are funded differently because they are short of different things: the vocational route needed a better mechanism, which it now has; the university route faces a genuine funding hole that no mechanism can close without money. Parity of esteem does not mean identical funding plumbing — it means neither route is left to fail, and a young person’s choice between a lab and a workshop is a choice between two well-founded futures, not between a favoured path and a neglected one. University is one route among several, and we fund it as such: properly, but not preferentially.

Paying for It: No One Sorted by Money

There is one route left to account for, and it is the one where the equity risk is sharpest. Three of our pathways see money flow to the young person: the strategic free degree, the vocational apprenticeship, and military training all pay their way. The exception is the student who wants to study a non-strategic subject — history, law, the humanities and much of the social sciences — direct at university, and who meets the fee through a loan. We have to be honest about the danger that creates. If three routes are funded and one is not, a poorer student who is debt-averse — and the evidence is clear that debt-aversion is sharply concentrated among students from low-income homes — may be steered away from the course they actually want and toward a funded route they do not, simply because they cannot comfortably pay. A system that sorted young people into pathways by the size of their parents’ bank balance would be a betrayal of everything this paper stands for. So we close the gap.

Two things make this affordable rather than ruinous. First, the three funded routes take a substantial share of each cohort, so the number of young people left needing means-tested support for a fee- or cost-bearing course is smaller than the headline suggests. Second, with far more eighteen-year-olds in funded training, education or service rather than idle, youth unemployment and the benefit spending attached to it fall — an offset we count honestly against the cost rather than ignoring. The measures:

• Restore means-tested maintenance grants, for academic and vocational study alike. The barrier that actually stops poorer students is not tuition debt; it is the cost of living while studying — you cannot pay rent with a tuition loan, and the abolition of maintenance grants in 2016 loaded the heaviest debt precisely onto the poorest. We restore proper, non-repayable maintenance grants for students from lower-income households, and — crucially — they are route-neutral: available to someone taking a college plumbing course or a nursing diploma just as much as a degree. The grant follows the student’s chosen direction, not a single approved path
• A genuinely progressive graduate contribution, not a debt. For fees on the non-strategic academic route, the income-contingent loan is reformed to work as it should: you contribute a share of what you earn above a real threshold, for a limited period, after which the balance is written off. It is a graduate contribution proportionate to the benefit received, not a debt that frightens people away from the classroom. We will set the threshold and write-off period so that graduates from lower-income backgrounds are not left paying longest for the smallest gain
• No one forced down a path by money, and no one sorted by a rejection. If a young person wants the vocational route and does not win an apprenticeship place, the answer is another way into that trade — a funded or grant-supported college course — not a shrug toward an unwanted degree. Support meets the destination the young person is actually aiming for. The point is not that university is a backstop for the disappointed; it is that whichever direction a young person chooses, academic or vocational, the means to pursue it does not depend on family wealth

We put the net cost of the maintenance-grant commitment at around £2bn a year after the youth-unemployment saving, and we carry it in our figures rather than waving it through: it is part of the £35bn of static policy cost the central case absorbs by Year 5, and it is among the most justified two billion in this paper. A country that has built three funded routes into adulthood, and then made the fourth open to the poorest on the same terms as the richest, is a country that has genuinely stopped sorting its young people by money — which is the whole point.

Education Is Not Taxed — Independent Schools

F³ holds a principle without exception: education is never taxed. Parents who choose independent education have already paid for the state system through their Local Income Tax — they take nothing from it and relieve it of a place. Taxing them again for educating their own child is double payment dressed as fairness, and it is precisely the social engineering through the tax system this paper rejects everywhere else.

• No VAT on school fees — core school and university education is VAT-exempt right across Europe, required by EU law under the VAT Directive, and was so in Britain until 2025. We restore that exemption, ending a tax that makes Britain a genuine outlier among developed nations and bringing us back into line with EEA norms
• Business rates abolished for schools in the first tranche — at Step 1, alongside the high street — because a school is not a warehouse to be taxed on its floor space

And there is a hard-headed economic case beneath the principle. British independent and boarding schools are a soft-power asset in the same class as our universities and the World Service: they educate the children of global elites who go home with a lifelong connection to Britain, they earn billions in export income, and they anchor a brand no competitor can copy. You do not tax a strategic export — you promote it. Britain spent years taxing its own advantage; F³ stops.

Free Degrees for Strategic Subjects

Britain faces critical shortages in medicine, nursing, engineering, computer science and life sciences — shortages that cost the NHS and the economy billions annually. The state already invests £230,000 per medical graduate, yet 52% of foundation year 2 doctors were unemployed in August 2025 while patients waited months for care. We train doctors at enormous public expense, then fail to employ them. This is a systemic failure of workforce planning, not a funding argument against free medical education.

F³ will introduce free tuition — no fees, no student loan — for degrees in the following strategic subjects:

• Medicine and dentistry
• Nursing, midwifery and allied health professions
• Engineering — all disciplines
• Computer science and AI
• Mathematics and physics
• Life sciences — biology, biochemistry, pharmacology

In exchange, all graduates in these subjects — British, EU or international — undertake an identical five-year service commitment in Britain. This is not indentured labour. It is a fair return on a significant public investment, applied without discrimination regardless of nationality. Equal treatment is a Single Market principle — and equal obligations is the same logic applied consistently.

The sliding scale repayment for those who leave early applies equally to all:

• Leave before completing year 1 of service: repay 100% of tuition
• Leave after year 1: repay 80%
• Leave after year 2: repay 60%
• Leave after year 3: repay 40%
• Leave after year 4: repay 20%
• Complete all 5 years: nothing owed

There is no nationality-based discount. A French medical student who trains free in Britain and immediately returns to Paris has received exactly the same public subsidy as a British student doing the same. The obligation is identical. This is also the strongest legal defence against any Single Market challenge — non-discrimination is a core EEA principle, and we apply it in both directions.

Single Market note: EU/EEA students are entitled to home fee status under EEA membership — meaning they qualify for the free degree programme on identical terms to British students. The cost rises from approximately £1.5bn to approximately £2bn annually once EU student uptake is factored in. This is offset by reduced recruitment of expensive overseas professionals, lower NHS agency spend, and the long-term economic return on a stronger STEM pipeline across Britain and Europe.

The 18–22 Guarantee: No One Leaves School Into Nothing

At eighteen, every young person in Britain holds a funded place — or has turned one down. The Guarantee is statutory: an apprenticeship or a further-education place for every 18-to-22-year-old, with four apprenticeship streams built so the offer covers everyone:

• Professional — degree-level apprenticeships in law, accounting, engineering, digital and health: earn, qualify, carry no debt
• Vocational — construction, manufacturing, energy, care and the trades this paper’s housing and infrastructure chapters will demand in volume
• Supported — apprenticeships designed for young people with special educational needs and disabilities (SEND): physical disabilities, learning disabilities and autism. Adjusted pace, job coaching, and a funding uplift that makes employers whole. Capability is the test; the format flexes. For this stream alone the entitlement runs to age 30, not 22: SEND young people often reach the starting line later, and a system that closes the door at 22 fails exactly the people it should hold it open for. And it is properly funded — the support that makes these placements work, the coaching, the workplace adjustments, the longer ramp, is fully subsidised rather than left as a cost the employer must swallow or the young person must go without. A placement that needs more to succeed gets more; that is the point of the stream
• Military — an expanded armed-forces apprenticeship intake across engineering, logistics, cyber and medicine, aligned with the 3% defence commitment: a trade and a service record by twenty-one

The alternative is a funded further-education place. What is not an option is nothing. The cliff edge at eighteen — where school ends and no institution catches those who do not go to university — is where NEET numbers, welfare entry and a lifetime of worklessness begin. The welfare chapter’s youth employment guarantee catches those who fall; this Guarantee is built so far fewer do.

Paid For By The Training Point

The Employer Levy is set at 8.5%: a 7.5% core plus a 1-point Training Point. The Training Point from every employer flows into a single National Training Fund — not millions of tiny per-firm pots, which is the design error that would otherwise leave a hairdresser with a £240 balance and no way to use it. The Fund pools the contributions of the whole economy and pays for the training itself. How an employer relates to it depends on size:

This is not a new layer on top of what exists — it replaces a tax already in place. Britain has had an Apprenticeship Levy since 2017: a 0.5% charge on employer pay bills above £3m, ring-fenced for training but use-it-or-lose-it, with unspent funds clawed back to the Treasury after two years. It has not worked. More than £3bn of it has expired unspent, because it was built for large employers running long, formal apprenticeships and left everyone else out: a firm below the £3m threshold paid nothing in but also drew little out, and a firm above it too often forfeited what it could not spend in time. The government’s own rebrand to a “Growth and Skills Levy” keeps the same £3m threshold and the same clawback — and from 2026 actually tightens it, halving the spending window to twelve months. We do something cleaner: we abolish the Apprenticeship Levy outright and fold its purpose into the Training Point. The Training Point fixes its two structural flaws at a stroke. It is economy-wide rather than gated at £3m, so the incentive to train reaches every employer, not just the largest. And because the money pools into one national fund that pays providers directly — rather than sitting in per-firm accounts on a countdown to clawback — nothing is forfeited for being too small to use or too slow to spend. To be clear for the scorecard: the old Levy’s receipts are not banked anywhere in this paper. It is gone, replaced pound-for-purpose by the Training Point, which is scored as fiscally neutral. We are not collecting both.

• Large employers that run their own apprenticeships draw their Training Point back from the Fund directly, pound for pound, to cover the cost — so a big firm that trains pays an effective 7.5%, and a big firm that does not leaves its point in the Fund for those who do
• Small employers do not self-fund and do not need a large balance of their own. They take on an apprentice and the National Training Fund pays the training and assessment in full for every apprentice under 25 — the apprentice costs the small firm a wage and a workplace, never a training bill. This mirrors the direction the current system is already moving, where small firms taking young apprentices have their training fully funded
• Firms too small to host a full apprenticeship can instead draw on Fund-supported group training schemes shared across a trade or locality, so the corner shop and the single-site garage are not shut out

So the plumber and the hairdresser are not asked to build a fund large enough to train from — an impossible ask on a small payroll. They are asked to give an apprentice a job, and the pooled Fund does the rest. The Training Point is the mechanism that fills the Fund; the entitlement it buys is free apprentice training for the small employer and a direct rebate for the large one. It never disappears into general revenue, and it is scored as fiscally neutral for exactly that reason.

One honest subtlety, because it is a fair challenge: if apprenticeship take-up surges — which is the aim — could the Fund pay out more in training than the Training Point raises? In principle yes, and we treat that as a feature with a funded ceiling, not a hidden hole. The Point is calibrated to fund the realistic take-up the 18-to-22 cohort can absorb; if demand runs ahead of that, the additional cost is met from the same place every other commitment in this paper is — scored openly, against the glide path, and gated if the money is not there. A surge in apprenticeships that strained the Fund would be one of the better problems a government could have, and we would fund it as a deliberate investment rather than pretend it away. What we will not do is the usual trick of scoring a training scheme as free by quietly assuming nobody uses it.

How the Earn-Back Is Policed

Because the Fund pays training providers directly rather than handing employers cash, the small-firm route is hard to abuse by design — there is no rebate cheque to inflate, only an accredited course delivered to a registered apprentice. The large-firm direct rebate is where reclassifying ordinary wages as ‘training’ could tempt, so that mechanism is built to make abuse expensive and easy to catch. Every apprenticeship claimed must be registered on an Apprentice Training Record: the apprentice’s National Insurance number, start date, the accredited qualification being worked toward, and the registered training provider. HMRC matches those National Insurance numbers automatically against the employer’s existing PAYE records. Two things follow. First, the age band does most of the work for free: the Guarantee covers 18-to-22-year-olds, so a claim for an existing mid-career employee fails on the date of birth before anyone even looks at it. Second, where gaming is found — a long-standing employee re-badged as a trainee, a qualification that does not exist, a provider that is not accredited — the penalty is deliberately punitive: the full amount repaid for every affected employee, plus a multiple of the sum wrongly reclaimed, plus director-level liability for systematic abuse. The honest firm gets its training funded or its point rebated with a single registration; the gaming firm pays several times what it tried to save.

Parental Leave — Flexible and Shareable

Chapter 9 makes the demographic case plainly: Britain’s birth rate has fallen well below replacement, and a country that needs more children cannot be silent on the support new parents receive. But the barrier to family life is often not only money — it is rigidity. The current system assumes the mother stays home, hands fathers a meagre fortnight on a use-it-or-lose-it basis, and forces an all-or-nothing choice between work and care. F³ reforms the structure rather than the cost: we keep statutory pay at its current levels — this is not an unfunded giveaway — and instead make the entitlement genuinely flexible and genuinely shareable, because that is what lets modern families actually use it.

• Full shareability: the total leave entitlement belongs to the family, to divide between two parents however they choose — all to one, split evenly, or any combination. We end the assumption that one parent is the carer and the other the earner
• Flexible by the day, not the block: leave can be taken in separate periods rather than one continuous stretch, so parents can overlap at the hardest moments, alternate around work, or phase a return gradually
• A genuine, non-transferable slice reserved for the second parent — paid at the same statutory rate, but ring-fenced so it is lost if not used, the one design that evidence shows actually gets fathers to take leave and normalises shared care
• Self-employed parents brought fully within the system — the entitlement follows the parent, not the employment contract, so a freelancer or company founder is no longer locked out of leave for their own child
• The right to return part-time: a parent may return to the same role on reduced hours for a defined period, dropping to one-and-a-bit household incomes exactly the situation the transferable allowance above is designed to cushion

The pieces are deliberately built to interlock. The demographic case explains why family formation matters; flexible, shareable leave removes the rigidity that deters it; the transferable allowance cushions the income drop when a parent reduces hours; and keeping pay at current levels means the whole package is a structural reform the public finances can carry, not a cheque the glide path cannot honour. Pro-family by design, not by spending.

Minimum Wage and Tax Thresholds

Two of the most important numbers in working people’s financial lives — the minimum wage and the income tax threshold — are currently set by political decision and changed at political whim. We will remove that uncertainty:

• National Minimum Wage index-linked to CPI inflation annually — ending the practice of below-inflation awards that erode real pay
• The £15,000 income tax-free threshold index-linked to CPI — so fiscal drag does not silently pull more workers into taxation over time
• One allowance replaces many: the £15,000 income threshold subsumes the personal allowance, the blind person’s allowance, and the £1,000 trading and property allowances — a single, generous, universal sum instead of a means-tested patchwork. Gift Aid is retained, because it funds charities rather than individuals; the £500 dividend allowance is retained, CPI-indexed, purely as a no-filing rule for small shareholders
• A new £5,000 tax-free savings allowance, CPI-indexed and outside any wrapper, replaces the old £1,000 personal savings allowance — protecting the interest on a meaningful cash buffer for every saver, with no account to open and no provider to choose. This sits alongside the £15,000 income threshold; ordinary savers pay no tax on their interest, and only those with very large cash holdings pay the flat rate on the excess
• VAT thresholds index-linked to CPI — so small businesses are not gradually drawn back into the VAT net by inflation
• The £15,000 CGT annual exempt amount index-linked to CPI — the same anti-fiscal-drag protection applied to capital as to income, and the same £15,000 that starts the income tax scale
• The £10 million National Wealth Floor scope threshold index-linked to CPI — the floor is for the extraordinarily wealthy, and stays that way
• The £5,000 tax-free savings allowance index-linked to CPI — held stable in real terms, never eroded by stealth, and never quietly cut without a vote. It is an allowance, not a wrapper: there is no ISA account to open, because the ISA regime is abolished and its job is done by the universal allowances instead
• A fully transferable income tax-free allowance between spouses and civil partners — one partner may transfer their entire unused £15,000 to the other, sheltering up to £30,000 of income from the national flat rate in a single-earner household. This applies where the couple has a dependent child under 18, or a dependent child of any age with a disability, and is available from the tax year in which a child is born — so it covers the parental-leave period from the start, when a household first drops to one-and-a-bit incomes, rather than making the family wait for the next April. Single parents keep their own full £15,000 and the family-support provision in Chapter 6
• The transfer is of the income-tax allowance only. The 2% Local Income Tax keeps its own £15,000 threshold per person and is unaffected: the earning spouse remains liable for local income tax on income above their own £15,000, exactly as every other taxpayer is. A transferred allowance lowers the household's national income-tax bill; it never exempts anyone from funding their local services
• Road Charge tiers index-linked to CPI — the flat rate stays flat in real terms
• The Pension Credit Winter Heating Element linked to the average domestic gas price rather than CPI — uniquely among our thresholds, because it exists to cover a specific volatile cost and should track that cost up and down, with no floor and an annually-reviewed cap
• All principal tax thresholds reviewed and uprated annually by the Office for Fiscal Honesty

Index-linking these thresholds removes a persistent tool of stealth taxation — governments have repeatedly frozen thresholds to raise revenue without a visible tax rise. Under F³, that practice ends.

Supporting Families

We will replace Child Benefit — a cash payment that does not guarantee children are better fed, clothed or supported — with direct provision:

• Free breakfasts for all school-age children — universal, with no means test. A fed child learns; the stigma of a means-tested queue is itself a harm, and testing a daily meal costs more to administer than it saves
• Free lunches for all school-age children — universal, on the same principle
• Uniform vouchers — means-tested at a household income of £80,000 (CPI-indexed). Periodic and higher-value, these are easy to target at the families who need them
• Holiday support payments — means-tested at the same £80,000 household threshold, replacing the patchwork of holiday-hunger schemes with one clear entitlement
• Enhanced support for children under five

This ensures support reaches children directly, reduces stigma, and delivers better nutritional and educational outcomes than cash transfers.

Childcare: One Entitlement, Funded at Cost

Britain runs three childcare systems at once — Tax-Free Childcare, the free-hours entitlements, and the childcare element of Universal Credit — each with its own eligibility rules, its own application process, its own cliff edges, and its own way of failing the parents it was designed for. Take-up of Tax-Free Childcare has never come close to forecast because working parents cannot navigate it; the free hours are ‘free’ at a funding rate that loses providers money on every funded place, so nurseries cross-subsidise from unfunded hours, restrict funded places, or close altogether. The result is the worst of all worlds: among the highest childcare costs in the developed world, a shrinking provider base, and a benefits maze that punishes exactly the second earners the economy needs back.

We replace all three with one entitlement: a single right to funded childcare for working families, one application, with providers paid directly by the state at a rate set by an independently audited assessment of actual delivery cost — reviewed annually, published openly, varied by region and age band. If the audited cost of a place is what a place costs, that is what we pay. No provider should lose money on a government promise, and no parent should discover that a ‘free’ place does not exist at the advertised price.

Delivery leans on the estate Britain already owns: schools. The school day and the school building end mid-afternoon; working parents do not. Wraparound and holiday provision expands through school premises — extending the school-based family provision this chapter already builds for meals, uniforms and holiday support — so capacity grows fastest where buildings, safeguarding and staffing structures already exist, rather than waiting for a private market that is currently contracting.

We score this honestly at £3bn a year in Year 1, rising to £5bn as the audited funding rate and the expanded entitlement phase in — over and above the spending the three abolished schemes release. It is one of the largest new spending commitments in this paper, and we make it without apology, because it is a growth policy before it is a family policy: childcare is the single biggest lever on the labour supply of prime-age parents, and every serious study of why mothers’ employment in Britain lags its peers lands on the same answer — the cost and chaos of childcare. A tax plan built on making work pay cannot leave standing a childcare system built on making work impossible. The transferable allowance elsewhere in this chapter supports families who choose a parent at home; this entitlement supports those who choose to work. The state’s job is to make both choices possible — not to price one of them out of existence.

Chapter 7: Special Educational Needs Reform

The Problem

Britain’s SEND system is broken — not from lack of spending, but from lack of capacity and a system built around bureaucracy rather than provision:

• SEND spending has risen sharply year after year
• EHCP numbers continue to grow, overwhelming the system
• Tribunal cases have exploded as families fight for basic support
• Parents spend years navigating legal battles that should never be necessary
• Councils face enormous SEND deficits and potential bankruptcy
• Too many children are sent to expensive private special schools because local places simply do not exist

The crisis is not a lack of money. It is a lack of places, a lack of specialist staff, and a bureaucratic process that forces families to fight the system rather than helping their children. Parents should not need lawyers to secure support, and councils should not be forced into financial crisis because specialist provision does not exist.

Reform 1: Diagnosis-Led Funding

For lower and moderate needs — ADHD, dyslexia, mild autism, anxiety, speech and language difficulties — funding will follow diagnosis automatically. No lengthy EHCP battle. No appeals. No waiting years for a decision that should take weeks. We should be straight about the tension between this and the take-up risk named in Chapter 2, because a reader will spot it. Making funding automatic on diagnosis does make the entitlement easier to obtain, and diagnosis rates are rising. Two things keep it affordable. The money follows the child to the school as a funding allocation, not to the household as a cash benefit, so the incentive to seek a diagnosis for its own sake is far weaker. And the gate is a clinical judgement made by someone other than the claimant, which is exactly the kind of assessment we argue should be retained and properly resourced rather than automated away. The honest position is that this is a bet on diagnosis being cheaper to fund early than to litigate late, and we monitor it as a named risk rather than assume it away.

• Schools receive dedicated funding directly upon diagnosis
• Trained SENCOs in every school with genuine authority and accountability
• Earlier intervention, at the point children actually need it

Reform 2: Reserve EHCPs for Complex Cases

Education, Health and Care Plans will be reserved for the children who genuinely need the full weight of multi-agency co-ordination:

• Severe autism, Down syndrome, profound learning disabilities
• Significant physical disabilities requiring specialist environments
• Complex cases requiring co-ordinated health, education and social care

This reduces administrative burden, allows specialists to focus on the highest-need children, and ends the situation where families with moderate-need children spend years in a system designed for something else entirely.

Reform 3: Specialist Units in Mainstream Schools

Every large secondary school will be encouraged and funded to develop dedicated specialist provision on site:

• Autism resource bases
• SEND resource centres
• Speech and language hubs
• Sensory support spaces

This keeps children close to home, close to their peers, and within mainstream education wherever possible — while providing the specialist support they need. It is better for children, better for families, and far cheaper than out-of-area independent placements.

Reform 4: National SEND Capital Fund

We will establish a £5 billion capital programme over five years — £1 billion per year — to build:

• New special schools in areas of acute shortage
• Dedicated autism schools
• SEND units attached to mainstream schools
• Therapy and assessment facilities

This will be funded through government capital borrowing — not day-to-day spending. Borrowing to build schools is investment, not consumption. Just as governments borrow to build roads, hospitals and railways, investing in specialist educational capacity generates long-term fiscal returns.

The arithmetic is compelling. An independent special school placement currently costs councils £60,000–£150,000+ per pupil annually. A maintained special school place costs approximately £35,000. A new 200-place school saves around £13 million per year once full. The capital programme pays for itself within a decade through reduced placement costs alone.

Reform 5: Parent-Directed SEN Budget

Families will receive greater direct control over their child’s support spending, usable for:

• Specialist tutoring and teaching support
• Therapy — occupational, speech and language, psychological
• Specialist software and assistive technology
• Equipment and sensory tools

With proper safeguards against misuse and clear accountability for outcomes.

Reform 6: Transport That Doesn’t Stop at 16

There is a cliff edge in the current system that quietly ends some young people’s education, and almost no one outside the families affected knows it exists. A child with special educational needs who is placed by the council — not by parental choice — in a school more than three miles away is entitled to free home-to-school transport up to the age of 16. On their sixteenth birthday, in the eyes of the law, that entitlement simply stops. The young person is still in education or training, as the law now requires them to be until 18; the placement is still one the council chose, not the family; the disability that makes independent travel impossible has not gone away. But the free transport has. What happens next is a postcode lottery: two-thirds of councils have withdrawn free post-16 transport, and where a council still helps, it typically charges the family a contribution of several hundred to over a thousand pounds a year. The result is predictable and documented — families forced to give up work to drive their disabled young person to college, and in the worst cases, young people dropping out of education altogether because they cannot get there.

This is indefensible, and the fix is narrow and cheap. Where a placement post-16 is the council’s decision rather than the family’s, and where the young person genuinely cannot travel independently because of their special educational needs or disability, free transport should continue — not to 16, but to 25, matching the age to which this paper already extends SEND support. This is not free transport for every sixth-former, nor for a family that chose a school across the county when a nearer one would serve. It is the removal of an arbitrary age cliff for the specific group who were placed where they are by the state and cannot make the journey alone. We cost it at approximately —£150m a year: much of the provision already exists but is charged to families, so the true additional cost is the contributions we would absorb plus the young people currently deterred from coming forward. It is a small sum to stop a disabled teenager’s education ending at the school gate for want of a bus.

The Coming Surge: When Claiming Becomes Easy

Every part of the British state that pays out money on application shares an unspoken assumption, one that appears in no legislation and no forecast but underpins the cost of all of it: that most people who could claim will not. The gap between what citizens are entitled to and what they actually receive is not small. Somewhere between £16 billion and £23 billion of legally-owed support goes unclaimed every year — £2.5 billion in Pension Credit alone, with over 900,000 eligible families not claiming; more than £2 billion in Attendance Allowance; £3.3 billion in council tax support. Take-up of Pension Credit is not merely low at 62 percent, it is falling. The Treasury does not describe this unclaimed money as a saving, but it depends on it. Every spending forecast is built on the quiet assumption that the maze does its work — that the forms, the means tests, the evidence requirements and the appeals process will continue to deter a large fraction of the people they are meant to serve.

Artificial intelligence is about to remove the maze. An assistant that can read the eligibility rules, complete the forms, assemble the evidence, draft the appeal, cite the relevant case law and chase the correspondence turns a daunting bureaucratic ordeal into a task of minutes. This is, in every individual case, a good thing: people receiving what Parliament decided they are owed is not an abuse of the system, it is the system working as designed for the first time. But it detonates the hidden assumption. If take-up across the major entitlements moves from around 60 percent toward 90 or 95 percent — and there is no reason to think it will not, once claiming is effortless — the cost is not marginal. It is potentially ten to twenty billion pounds a year of spending that no forecast contains, that no government has budgeted for, and that the state is legally obliged to pay. We are the first to say plainly: this is one of the largest unpriced risks to the public finances, and neither established party is planning for it.

The administrative shock arrives alongside the fiscal one. The Department for Work and Pensions, HMRC, local authorities and the tribunal service are all staffed and structured around today's volume of applications, challenges and correspondence. A step-change in demand — and worse, a step-change in well-constructed, hard-to-refuse applications and appeals, each of which must still be assessed by a human — would overwhelm systems that already run months behind. A tribunal service that cannot cope with today's appeal rate does not survive a doubling of it.

The wrong responses are the obvious ones. To celebrate is to ignore that the system was never funded for full take-up. To resist — to tell citizens not to use AI to claim — is both futile and indefensible; no government can tell people not to understand their own rights. The honest path is harder: if an entitlement would break the public finances at full take-up, that is not an argument for keeping take-up low. It is proof that the entitlement is badly designed — too complex, too conditional, too dependent on friction to control its own cost. The coming surge is not merely a threat. It is the forcing function that finally makes the case for the simpler, more automatic system this paper argues for throughout.

So we act before the surge, not after it, on three principles. First, automatic payment wherever the state already knows. For entitlements the government can determine from data it already holds — Pension Credit against income and pension records, the passported benefits that flow from a Universal Credit award — the citizen should not have to claim at all; the state should simply pay. This is fairer, cheaper to administer, and it removes the appeals problem at its source: there can be no wave of challenges against decisions that were never adversarial. Second, simplification so that full take-up is affordable by design. A benefit that is sustainable only because a third of the people entitled to it never claim is living on borrowed time; we would rather design a smaller number of clear, adequate entitlements the state can afford to pay in full. Third, retained human assessment where judgement is genuine. For entitlements that turn on real, contested judgement — the disability assessments at the heart of Personal Independence Payment, where a person's circumstances cannot be read off a database — the answer is not to pay automatically, which would invite error and fraud, but to match the citizen's tools with the state's: if claimants arrive with AI-assembled evidence, assessors need AI-assisted triage to process it fairly and at speed. The state cannot meet a modernised public with a bureaucracy frozen in the last decade.

The Fiscal Position

This is one of the few areas in this paper where modest investment demonstrably reduces long-term public spending:

The net annual revenue cost of approximately £750m is more than offset over time by the returns from the capital programme. As new state special school places replace expensive independent placements, the savings grow year on year.

Chapter 8: Housing and Planning

Planning Reform

• Presumption in favour of development — the default answer to planning applications is yes, not no
• Statutory decision deadlines with escalating consequences when councils miss them — a fee-refund ladder that rises to full refund, and deemed consent only as a final backstop for total inaction
• Simplified planning system with fewer grounds for obstruction
• Planning objections must include a constructive alternative — not simply opposition

On deadlines specifically, the consequence is graduated so that it bites without being reckless. A council that misses the statutory determination target refunds a quarter of the planning fee if the decision is more than two weeks late, half at four weeks, and the whole fee at eight weeks. Only if an application is more than twelve weeks overdue, with no decision and no agreed extension, is it deemed granted. This is a penalty against poor service, not a raid on council budgets — a council that decides on time keeps its fee in full and pays nothing — and the deemed-grant backstop exists so that total inaction has a hard stop, while a genuinely contentious application can still be refused on its merits within the window.

Developer Charges: Simpler to Build, Honest About Safety

Building a home in Britain attracts a thicket of separate developer charges, and the thicket itself is part of why we build too little. Two of them — the Community Infrastructure Levy and Section 106 agreements — are meant to fund the roads, schools, GP surgeries and affordable homes that new development requires. The principle is right: development should pay for the infrastructure it needs. But the execution is slow, unpredictable and negotiated case by case, which adds cost and delay and gives blockers another lever. We fold both into a single, simpler, transparent infrastructure charge — a published rate rather than a bespoke negotiation on every site — so that developers know the cost up front and councils still get the contribution. One commitment makes that last clause mean something: the published rate is set so that it raises at least what Section 106 and the Levy currently deliver in value, measured nationally and reviewed each Parliament. A charge set below that would not be simplification, it would be a cut to affordable housing wearing simplification’s clothes, and we would rather bind ourselves against it in advance. This is a faster route to the same end, consistent with the pro-building thrust of this chapter; it is not a cut to infrastructure or affordable-housing funding, and we say so plainly.

A second pair of charges exists for an entirely different reason, and here we are honest about a duplication. The Residential Property Developer Tax (a profits tax on large housebuilders) and the Building Safety Levy (a charge on new residential development) were both created after Grenfell to do the same job: make developers, rather than leaseholders or taxpayers, pay to remove dangerous cladding. Two instruments aimed at one purpose is exactly the kind of bolt-on duplication this paper sets out to clean up. We consolidate them into a single building-safety remediation charge. Crucially, we treat it as what it is — temporary. It exists to fix a specific scandal, and it sunsets when that job is genuinely done: when the buildings are made safe, the charge ends, rather than quietly becoming a permanent tax on building the homes the country needs. Leaseholders, who were never responsible, do not pay; developers who profited from the system that allowed the crisis do.

VAT That Punishes Improvement

The tax system contains a perverse signal. Build a new home and the construction is zero-rated for VAT; improve, extend or retrofit the home you already have and the work carries the full 20%. We penalise precisely what an ageing housing stock and a net-zero transition most need — insulation, heat pumps, the extension that lets a growing family stay put rather than move. And a 20% wedge on domestic building work is the single biggest driver of the cash economy: it is the direct financial incentive for the off-books job, the trader who takes a deposit and vanishes, and the homeowner left with no contract and no protection when it goes wrong.

The scale of this is not marginal. A 2025 survey by Resi, the HomeOwners Alliance and YouGov found that roughly a third of homeowners — about 5.3 million people — pay cash specifically to avoid VAT on improvements, and that around 4 million more are deterred from improving their homes at all by the VAT cost. The Federation of Master Builders estimates that cutting the rate to 5% would add at least £15bn to the economy.

We remove the distortion in both directions: a single 5% rate on all residential building work, new build and improvement alike. New construction rises from zero to 5%; renovation, repair and retrofit fall from 20% to 5%. The treatment is finally neutral — the tax no longer cares whether you build new or improve what stands. The net cost to the Exchequer is modest, around £1bn a year once the 5% now charged on new build is set against the reduction on improvement, and we book that figure in full. But we expect the true cost to be lower, and quite possibly to vanish, because the static number assumes only that today's declared work pays a lower rate. It cannot capture the two effects that matter most: the cash-in-hand work that comes into the formal economy once the incentive to hide it collapses, and the millions of deterred homeowners who finally proceed. We bank none of that behavioural gain — but we are confident it is large.

Licensing the Trades

In much of Europe you need a licence to lay bricks or wire a plug. In Britain, anyone with a van and a Facebook page can strip the back off someone's house, take a deposit, and disappear. There is no minimum competence standard, no public register, and no straightforward recourse when the work is dangerous or the money is gone. This fails the public, and it fails good builders most of all, because they are undercut on price by firms that carry none of the cost of doing the job properly — no insurance, no training, no accountability.

We introduce a licensing scheme for the building trades: a competence and insurance standard required to trade, a public register homeowners can check, and removal from it for firms that endanger the public or defraud their customers. The scheme is funded by a modest licence fee, not the Exchequer. This is not red tape for its own sake — it is the same principle we already apply to gas engineers and electricians, extended to the people rebuilding the nation's homes. It pairs naturally with the VAT reform above: as work moves out of the cash economy and into the formal one, a licensing regime ensures the newly-visible market meets a real standard rather than simply becoming visible. Good firms have nothing to fear and everything to gain.

Policy Interaction: Immigration and Housing Supply

The interaction is manageable but requires honest monitoring. F³ sets an explicit housing supply target: 300,000 new homes per year by year 3, rising to 350,000 by year 5. If housing starts fall materially below this trajectory in years 1-2, the immigration openness will be reviewed — not reversed, but the pace of new route expansion moderated until supply catches up. The Land Banking Tax is the primary supply accelerant: by making land banking economically painful from day one, it creates immediate financial incentive to build. The planning presumption in favour of development removes the bureaucratic barrier. Together these should deliver supply within 12-18 months — ahead of immigration volumes building to their full level.

There is a second honest condition, and it is measured in people, not permissions. Three hundred thousand homes a year is a workforce commitment before it is a planning commitment: the construction industry needs tens of thousands of additional recruits a year to meet existing demand alone — before this programme, before the SMR build-out, before grid, geothermal and data centres compete for the same trades. Planning reform without people pours faster permissions into the same bottleneck. So the housing target is backed by a construction workforce plan: construction trades sit first in the queue for the Training Point and the 18-to-22 Guarantee, so an employer that trains bricklayers, electricians and groundworkers earns its Levy point back from day one; and a time-limited construction shortage visa route operates alongside, tapering as domestic training delivers — because the honest sequencing is to train British workers and borrow experienced ones while the training completes. If the workforce does not materialise, the homes will not either, and a plan unwilling to say that is not a plan.

Land Banking Tax

Land banking — holding developable land with planning permission without building on it — is costing Britain tens of thousands of homes every year. Britain’s eight biggest housebuilders alone sit on nearly 500,000 unbuilt plots with a combined estimated value of £200bn. The total national picture, including smaller developers and strategic land holders, represents £300–400bn of developable land lying idle.

We will introduce an annual Land Banking Tax of 2% on the value of land that holds planning permission but remains unbuilt after two years:

• Applied to all landowners — housebuilders, developers, investment funds, and private holders
• Calculated on the assessed development value of the land, not agricultural value
• Exemptions for sites with active construction underway or where development is genuinely blocked by infrastructure constraints
• Revenue ring-fenced in the first parliament for affordable housing and planning capacity

At 2% on an estimated £300bn taxable base, the levy raises approximately £6bn annually before behavioural change. As developers accelerate building to avoid the charge, revenue falls — but housing supply rises, house prices moderate, and economic growth follows. By year three the tax may raise only £2–3bn, but Britain will have gained tens of thousands of additional homes.

Ending Social Engineering in Housing

The current planning system imposes mandatory affordable housing quotas — typically 30-40% of all new developments, rising to 50% under recent Labour reforms — as a condition of planning permission. The intention is admirable. The outcome is counterproductive.

Developers respond to unviable affordable housing requirements in two ways: they either abandon the scheme entirely — currently 8,500 planned affordable homes are at risk because housing associations have withdrawn from Section 106 agreements — or they submit viability assessments arguing the site cannot bear the cost, triggering years of negotiation, legal challenge and delay. Meanwhile, nothing gets built.

The fundamental economic truth is simple: housing becomes affordable when there is enough of it. Cities with high supply — Tokyo, Minneapolis, Auckland since its planning reforms — have moderate and falling house prices. Cities that restrict supply while mandating affordable percentages have expensive housing and less of it.

F³ will remove mandatory affordable housing quotas as a condition of planning permission:

• No central government-imposed affordable housing percentage requirement on private developments
• Developers build what the market needs — more supply of all types moderates prices across the board
• Local authorities retain the ability to use their own land and resources to build social housing directly — this is the appropriate mechanism for delivering genuinely affordable homes
• The Land Banking Tax and planning reform together will unlock far more homes than any quota system has delivered

This is not an argument against social housing — it is an argument that social housing should be built by the public sector on public land, not extracted as a levy from private developers in a way that makes development unviable. The proceeds of the Land Banking Tax, ring-fenced in the first parliament for affordable housing, provide the public sector funding to do exactly that.

Property Tax Reform

Replacing Stamp Duty with Capital Gains Tax (see Chapter 1) directly improves housing market mobility — allowing people to move for work, family or downsizing without punitive transaction costs.

The Private Rental Sector: Helping Tenants by Fixing Supply

Britain has spent a decade taxing its private landlords as though renting were a social ill to be stamped out. It is not. Millions of people rent — the young, the mobile, those saving for a deposit, those who simply do not want to own — and they are served by a private rental sector that the last Labour government actively built up in the early 2000s. Then policy reversed: a 3% Stamp Duty surcharge on additional homes, the stripping-back of capital gains reliefs, and above all Section 24 — the rule that stops landlords deducting their mortgage interest as a business cost. The intent was to discourage landlords. The effect was to do exactly that: landlords sold up, rental supply shrank, and rents rose. The policy designed to help renters made renting more expensive.

Section 24 deserves singling out, because it is genuinely anomalous. Every business in Britain deducts its financing costs before it is taxed on profit — except a landlord, who since Section 24 is taxed on rental income before deducting the mortgage interest paid to the bank, receiving only a restricted 20% credit instead. In a period of high interest rates this can tax a landlord at an effective rate well above 100% of their real profit — they are taxed on money they never keep. No other business is treated this way, and there is no principled reason a landlord should be.

F³ does not introduce a special regime for landlords. It simply stops punishing them, and the relief falls out naturally from the universal tax system the rest of this paper builds:

• Abolishing Stamp Duty on all property purchases removes the additional-homes surcharge entirely — there is no longer a penalty rate for buying a property to let
• Capital Gains Tax on a let property is charged on the real gain — net of inflation, acquisition costs and capital improvements — at the point of sale, at the headline rate. It is taxed as the investment it is: no holding-period taper (that is reserved for productive business capital, not passive property), but no longer inflated by taxing paper gains that are merely inflation, as the current nominal-gain system does
• Section 24 is repealed: rental businesses deduct their genuine financing costs like every other business, and are taxed on real profit — at the single flat rate, with no special surcharge and no special relief

None of this is a charter for bad landlords. Decent standards, security of tenure and protection from unfair eviction remain — a healthy rental sector and well-protected tenants are not in tension. And renting is not the enemy of owning. F³ wants more of both: more first-time buyers and a healthy rental market, which is why the real answer to housing is the one this chapter leads with — build more. Renting and owning only compete when homes are scarce; the enemy is not the landlord or the owner-occupier, it is the planning system that stops Britain building enough for either. Fix supply, and a private rental sector becomes what it should be — a service to the people who use it, not a political punchbag.

Chapter 9: Immigration

Britain’s immigration policy has been pulled in contradictory directions for a generation — too restrictive for the talent and skills the economy needs, too permissive in the areas where public confidence has been lost. F³ will reverse both failures simultaneously: make it significantly easier for high-value immigration, and significantly firmer on illegal entry and asylum abuse.

Why Britain Needs People — The Honest Starting Point

Begin with the demographics, because everything else follows from them. Britain’s birth rate has fallen to around 1.4 children per woman — well below the 2.1 needed to hold the population steady. The working-age population that pays the tax which funds pensions, the NHS and social care is shrinking relative to the number of retired people who draw on them. A country in that position has only three options: accept managed decline, raise the birth rate overnight (no government has ever managed it), or bring in people who work and contribute. There is no fourth door.

So the honest question is not whether Britain needs immigration — it does, and the maths is not close — but which immigration. The country needs the doctor, the engineer, the care worker, the founder, the graduate who stays and builds: people who pay in more than they take out and who fill the gaps an ageing workforce leaves. It does not need, and the public will not accept, uncontrolled illegal entry or an asylum system gamed by people with no claim. The failure of the last decade was to muddle the two — to wave at a single net-migration number while losing control of its composition. F³ separates them cleanly: more of the migration that builds the country, far less of the migration that strains it.

Free Movement: An Honest Position

Free movement of EU citizens is a fundamental rule of the Single Market. Norway has it. Iceland has it. Liechtenstein has it. There is no credible path to Single Market membership that does not include it. Any politician who claims otherwise is not being straight with the public.

F³ makes this case openly. Free movement from the EU — of workers, students, professionals and families — has been broadly economically beneficial to Britain. The problems attributed to immigration are largely problems of public service capacity, housing supply and wage pressure in specific sectors — all of which our platform addresses directly. The solution is to build more houses, fund public services properly, and raise wages through productivity growth — not to restrict the movement of European citizens who contribute significantly to Britain’s economy and culture.

With free movement restored, seasonal worker schemes for EU nationals become unnecessary. The Single Market itself provides the flexible labour access that British agriculture, hospitality and other sectors need without bureaucratic visa processes.

Attracting Global Talent: An Ambitious Open Door

Beyond EU free movement, Britain should be the most attractive destination in the world for high-performing individuals. Labour has spent its time in office trying to reduce immigration routes despite the obvious economic benefits. This is economically illiterate and nationally self-defeating.

F³ will actively recruit the world’s best minds, most ambitious entrepreneurs and most productive professionals. We will make the visa system for high-value migrants faster, simpler and more generous than any comparable country:

• High Earner Fast Track — for talent from outside the EEA (EEA citizens already enjoy free movement and need no visa): any non-EEA individual with a confirmed salary or contract above £120,000 receives an expedited visa within 14 days, no caps, minimal bureaucracy. The point is to add the rest of the world to the European talent pool free movement already restores
• Global Talent Visa — dramatically expanded and simplified for scientists, researchers, engineers, artists, musicians, architects and other exceptional talent
• Entrepreneur Visa — any individual bringing verifiable capital of £250,000+ or founding a company with credible backing receives immediate residency
• Student visas — actively promoted, graduate routes maintained and extended, international students treated as the economic and soft power asset they are

The Trump Opportunity: Recruiting America’s Intellectual Refugees

The United States is conducting a remarkable experiment: systematically alienating the global talent on which its innovation economy depends. Federal research funding has been frozen or cut. Universities are under political attack. Scientists, academics, technologists, artists and financiers are actively looking to leave.

Americans now make up the fastest-growing group of foreign applicants for UK jobs — up 2.4 percentage points in a single year. A Nature survey found 75% of US scientists were considering leaving for Europe or Canada. Nobel laureates are relocating. Dozens of researchers are moving to France following high-profile recruitment campaigns.

Britain is watching this opportunity with one hand tied behind its back — bureaucratic visa processes, hostile rhetoric about immigration, and a government that has cut rather than expanded talent routes. F³ will change this decisively.

• A dedicated US Talent Welcome Programme — streamlined visa processing for American scientists, researchers, technologists, life scientists, financiers and artists seeking to relocate
• Research asylum — formal fast-track routes for researchers whose work is being defunded or suppressed in the US, modelled on France’s Safe Place for Science initiative
• University partnerships — UK universities actively funded and encouraged to recruit displaced US faculty
• Tax incentives for the first three years — non-domiciled status maintained for new arrivals, making Britain financially competitive with the US for relocated talent

Non-EU Skills Migration

Outside the EU free movement framework, Britain maintains a points-based assessment for non-EU immigration:

• Skills-based system — genuine shortage occupations prioritised, with regular independent review of the shortage occupation list
• No salary floor that prices out genuinely needed roles in care, education and public services — shortage occupation designation takes precedence
• Dependant rights maintained — isolating workers from their families is not an immigration policy, it is cruelty that drives talent elsewhere

Asylum: A System That Has Lost Public Confidence

Britain’s asylum system has lost public confidence — genuinely exploited at one end, while the most desperate refugees wait far too long for protection at the other. Restoring trust means being honest about why: the legal framework, and the way it has come to be interpreted, no longer fits the world it operates in. We set out below what that means and what we would do about it.

The result is a system that has lost public confidence and is genuinely being exploited — while genuine refugees in the most desperate circumstances wait longer for protection. That is good for nobody.

The Returns Opportunity

One of the least-discussed consequences of Brexit is that Britain lost its ability to return asylum seekers to the EU country they arrived from. Until 2021 the Dublin III Regulation allowed returns to France and other EU states; Brexit ended that, and the Trade and Cooperation Agreement secured no replacement. The EU has since moved on: from June 2026 Dublin III is itself replaced by the EU’s new Pact on Migration and Asylum. So the prize is not literally to ‘rejoin Dublin’ — that system is being superseded — but to negotiate a returns agreement with the EU equivalent to the one Brexit removed, of the kind non-member states already hold.

Single Market accession creates the opportunity to fix this — not because the EEA treaty includes asylum cooperation (it does not), but because the goodwill and institutional relationship that come with membership are what make the EU willing to negotiate a returns agreement at all. We will pursue a returns framework with the EU alongside accession — restoring our legal ability to return people who cross from France, and removing the primary incentive for small boat crossings.

We should be candid about the counterfactual, because honesty is the point of this document. Full EU membership would put Britain back inside the EU’s asylum framework, and it is fair to say rejoining would, in that narrow sense, address the small-boats problem more directly than anything short of it — though even inside that framework, as the pre-2021 record shows, returns were never a complete or automatic solution. F³ is not advocating that — we have set out elsewhere why full membership is the wrong destination for Britain, and Single Market accession the right one. We raise the point only to be straight: the small-boats crisis is, at root, a problem Brexit made far harder to manage — Britain left the returns system and secured nothing in its place. We do not propose to undo Brexit to fix it — we propose to negotiate back the one piece that was lost.

A Law Built for a Different Age

The 1951 Refugee Convention was written for the world that produced it: post-war Europe, people fleeing identifiable persecution across land borders, in an age before mass air travel, organised people-smuggling, and economic migration at today’s scale. Its founding principle — that those genuinely fleeing persecution deserve protection — is timeless and right, and F³ holds to it without reservation. But the 1951 text, and still more the expansive way courts have interpreted it over seventy years, was not designed for dinghies crossing the Channel or for asylum claims used as a route around ordinary immigration control. The problem is not the principle. It is that the drafting and its interpretation belong to a world that no longer exists.

The proof that this is about law and interpretation, not an immovable moral fact, is that comparable countries — bound by exactly the same Refugee Convention and the same European Convention on Human Rights — accept far fewer claims than Britain and remove far more people. Denmark, the Netherlands and others do it through fast inadmissibility decisions, a firmly applied first-safe-country principle, temporary rather than automatic permanent status, tight family-reunion rules, and a narrower domestic interpretation of the same treaty text. The Conventions do not compel Britain’s outcomes; Britain’s own statute and case law do. A great deal can be tightened at home without leaving anything.

Reform With Allies, Not Exit Alone

So F³ will tighten the system within the Conventions, and press for their modernisation alongside the many states that now want the same:

• Legislate a firm first-safe-country inadmissibility rule and a faster, simpler process, using the room the Conventions already allow — the lever Denmark and others already pull
• Pursue reform with like-minded partners: in May 2025 nine European states, led by Italy and Denmark, jointly called for a new interpretation of the European Convention on Human Rights to allow states to expel foreign nationals who commit serious crimes, and by December 2025 a broader group of European governments had aligned behind the same case — so Britain would be working with the current, not against it
• Maintain safe and legal routes for genuine refugees — expanded, well-funded and properly administered, because a credible system is firm on abuse precisely so it can be generous to need
• Remove failed applicants promptly to home countries or designated safe third countries, underpinned by the returns agreement above

Deporting Serious Foreign Criminals

There is a category the returns debate above does not reach, and it is the one the public finds hardest to accept: foreign nationals who commit the gravest crimes on British soil — serious violence, terrorism, the sexual abuse of children — and who, having served their sentence, cannot be removed. We are clear that this is different in kind from the asylum question. It is not about people whose only fault is an unprocessed claim; it is about convicted serious criminals whose continued presence the country has every right to refuse. Two things currently frustrate their removal, and we address both honestly.

The first is a quirk of domestic law: a provision of the Immigration Act 1971 protects Commonwealth citizens resident in Britain since before 1973 from deportation. That protection exists for an honourable reason — it is the same shield that should have prevented the Windrush injustice, where people who came lawfully to rebuild post-war Britain were wrongly threatened with removal because they could not document a status they plainly held. We will not touch that protection for the law-abiding, who are its overwhelming beneficiaries and to whom Britain owes a debt. But a protection designed to secure the innocent was never intended to shield the perpetrators of the most serious crimes, and we will legislate a narrow carve-out so that it cannot: conviction for the gravest offences removes the historical bar to deportation, while leaving it wholly intact for everyone else. The principle is simple — no immigration protection, however venerable its origins, should be a sanctuary for those who rape children or murder.

The second obstacle is harder and more honest: even with the legal bar removed, a person can only be deported if somewhere will receive them, and home countries sometimes refuse to take back their own nationals — occasionally after the individual has renounced citizenship or destroyed documents precisely to frustrate removal. No domestic law can compel a foreign government to cooperate. So we act on two fronts:

• Make returns a first-order diplomatic priority, with leverage. Britain gives significant development aid, grants large numbers of visas, and offers trade access to many of the countries that decline to take back their nationals. We will make cooperation on returns an explicit condition of that relationship — aid, visa allocations and trade facilitation weighed against a country’s willingness to readmit its own citizens — rather than treating the returns question as separate from the wider bilateral deal, as Britain too often has
• A lawful third-country backstop for serious criminals only. Where a home country genuinely will not receive a serious foreign criminal, the alternative to indefinite management in Britain is removal to a safe third country under a formal agreement. We are precise about the limits of this, because Britain has an expensive lesson to learn from. The Rwanda scheme failed — struck down by our own courts and abandoned at a cost of hundreds of millions — because it tried to offshore asylum seekers to a country the courts did not accept as safe, without adequate legal safeguards. We propose something narrower and lawful: a backstop confined to convicted serious criminals who have served their sentences and cannot be returned home, operated only under genuine safe-country agreements with proper legal process and protection against onward refoulement. The distinction from Rwanda is not cosmetic — it is the difference between offshoring the vulnerable to dodge our obligations, and removing dangerous criminals to a safe country when their own will not have them

We are candid that even this will not resolve every case immediately: some individuals will, for a period, have to be managed in Britain under supervision and restriction while agreements are built. We do not pretend otherwise. But a country that sets out to remove its most dangerous foreign offenders — through honest legal reform, hard diplomatic leverage and a lawful backstop — will succeed in far more cases than one that treats each failure as inevitable. And it can do so, as with everything else in this chapter, without abandoning the rule of law that makes Britain worth defending.

Why We Do Not Leave the Conventions

Some now argue Britain should simply leave the European Convention on Human Rights and the Refugee Convention altogether. We understand the frustration beneath that argument, and we do not pretend the question is illegitimate — for a party that has ruled out the European single market, leaving the ECHR is a more available choice, and the debate is a real one. But for F³ it is the wrong choice, because it collides directly with the rest of this programme. The single market accession that anchors our growth strategy rests on the European Economic Area, whose founding agreement is built on a commitment to ‘peace, democracy and human rights’ — the ECHR — and every member has ratified it. Leaving the ECHR would also breach the Good Friday Agreement, which writes the Convention into the law of Northern Ireland, and the existing UK-EU Trade and Cooperation Agreement, which the EU has said it would suspend in part if we left. Tearing up the Conventions to control the border, while simultaneously seeking to rejoin the European market that is built upon them, is not a strategy — it is a contradiction.

More legal routes in. Firm borders against illegal entry. Treaties modernised with allies, not abandoned alone. A system the public can trust.

Chapter 10: Energy Security and Net Zero

The Honest Transition

Britain is committed to net zero — but honesty requires acknowledging that the transition takes time, and that intermittent renewables alone cannot power a modern economy. The path to a clean energy future runs through a pragmatic bridge, not ideological purity.

Natural gas will remain part of Britain’s energy mix during the transition. Expanding domestic gas production is not a retreat from net zero — it is a rational response to energy security, reducing dependence on imports, keeping bills lower during the transition, and ensuring continuity of supply as we build the clean infrastructure of the future. We will expand domestic natural gas production as a transitional measure, with a clear sunset horizon tied to nuclear and storage capacity coming online.

The North Sea: Don’t Leave It in the Ground — and Don’t Spend the Proceeds

Britain has roughly ten to fifteen billion barrels of oil and gas equivalent still recoverable beneath its waters within the limits of our net zero commitments, of which current producers are lined up to extract only about four billion. We do not propose to leave the rest in the ground. With 85% of homes still heated by gas and around 23 million petrol and diesel vehicles still on the road, the choice during the transition is not between domestic oil and gas and none — it is between our own production and imported liquefied natural gas, which is typically more carbon-intensive once shipping is counted. Producing what we still need, at home, is both the more secure and the cleaner option. The question is how we tax it — and what we do with the money.

We should be honest about scale before we begin: this is no longer a fiscal windfall. North Sea receipts peaked at £9.9bn in 2022-23 and the OBR forecasts them falling from around £2.7bn in 2025-26 to roughly £0.3bn by 2030 as fields mature and the temporary windfall levy expires. The era in which oil could have built Britain a sovereign fund of hundreds of billions — as it built Norway’s, now worth over £1.5 trillion — is behind us, squandered across the 1980s and the rounds of rate cuts that followed. What remains is smaller, but it is not nothing, and the principle is what matters: a finite national asset should be priced properly and saved, not given away and spent. We propose to get the last chapter right even though we got the first one wrong.

Two Regimes: Stability for Existing Fields, Incentive for New Ones

The UK’s problem is not the headline rate. At 78% — 30% ring-fence corporation tax plus the Energy Profits Levy and supplementary charge — Britain already taxes its oil and gas as heavily as Norway. The difference is that Norway’s 78% is permanent and predictable, while ours is a windfall levy that has been raised and extended four times since 2022, deterring the very investment we need: production is forecast to fall around 40% by 2030 without reform, with roughly a thousand jobs a month already going. We will not repeat that mistake. We separate the resource into two clearly distinct regimes, taxing where behaviour cannot change and incentivising where it can.

• Existing production — a permanent, stable 78%. We abolish the variable Energy Profits Levy and fold its effect into a single, fixed, permanent headline rate of 78% on fields already producing. The capital here is sunk — the wells are drilled, the platforms built — so a high rate does not deter investment that has already happened; it simply collects the public’s fair share without the chaos that is currently suppressing receipts. Crucially, the rate stops moving. No more annual surprises, no escalator, no ministerial whim. Predictability is the single thing Norway had that we did not, and it is free to give.
• New production — a permanent, profit-based regime at a materially lower headline rate of 43%, with a generous first-year investment allowance. Here the investment decision is live, so the lever must encourage drilling rather than punish it. We set the rate at 43% — simply the level we judge high enough to take a fair public share yet low enough to unlock the investment the current 78% is killing. It sits in the range the industry itself has proposed (around 40%) and is coupled with full expensing of development capital, so the tax rewards reinvestment rather than merely cutting the headline. A price-triggered element can lift the effective rate when oil and gas prices are exceptionally high, capturing genuine windfalls without the permanent deterrent. This is the difference between a regime that maximises drilling and one that maximises the fund; we choose to grow the asset base, and tax the profits steadily, rather than tax a shrinking base into disappearance.

The Norway Rule: New Receipts Are Saved, Not Spent

Every pound of tax raised from new fields is paid directly into the British Future Fund (Chapter 15) and never into day-to-day spending. This is the Norwegian principle applied honestly: oil is a one-time inheritance, not income, and a country that spends it on current consumption has eaten its savings. The existing surplus-only capitalisation rule of the Future Fund is unaffected — this is a second, ring-fenced revenue stream into the same patient national pot, not a relaxation of the rule that the Fund never borrows. We should be clear about one consequence, in keeping with this paper’s method: receipts from new fields that would, under today’s rules, have flowed to the Treasury’s general account are instead saved. We forgo that day-to-day revenue deliberately. Our fiscal framework does not bank a penny of it — the deficit numbers in this paper assume the OBR’s existing, declining North Sea forecast and no more. That is the prudent treatment: a long-run plan should never lean on volatile hydrocarbon receipts, and ours does not. The Fund is the upside, held ring-fenced and separately reported, not spent on day-to-day services — and we should be precise that a ring-fenced public fund is still inside the public accounts, not outside them, for the country’s future rather than this parliament’s.

This is not a contradiction of net zero, and we will not pretend the tension does not exist. New licensing is bounded by the same sunset horizon as the rest of this chapter: it substitutes for dirtier imports during a defined transition window, not in perpetuity, and it winds down as nuclear, geothermal and storage capacity come online. The low-carbon programme set out in the rest of this chapter — the Rolls-Royce SMRs, thorium, Cornish geothermal and lithium — is the destination; the North Sea is the bridge that funds part of the journey and keeps the lights on while we build it. We reject both the fantasy that Britain can stop producing tomorrow without importing the difference, and the fantasy that the North Sea is a long-term answer. It is a declining asset to be extracted responsibly, taxed fairly, and banked for good.

Two honest caveats remain, which we flag rather than bury. First, the volumes from new licensing are genuinely uncertain — we present no central revenue estimate for the Fund because any figure would be speculative; what we commit to is the mechanism, not a number. Second, a 78% rate on existing fields against a 43% rate on new ones creates a visible gap between neighbouring producers; we defend it on the clear principle that tax should fall on sunk capital and incentive should fall on capital not yet committed, but it is a deliberate distinction, openly made, not an accident.

Ending the Hydrogen Distraction

Significant public money has been directed toward green hydrogen as a heating and transport solution. The evidence does not support this. Green hydrogen is enormously energy-inefficient to produce, expensive to store, and requires entirely new distribution infrastructure. Heat pumps and direct electrification are demonstrably more efficient pathways. We will end government investment in hydrogen as a domestic heating solution and redirect that funding to proven technologies that will actually deliver lower bills and lower emissions.

Nuclear: The Backbone of Clean Baseload

Britain has already taken the right first steps. Rolls-Royce has signed a contract to build the UK’s first three Small Modular Reactors at Wylfa in North Wales, backed by £2.5bn government funding and £599m from the National Wealth Fund. Rolls-Royce estimates the programme will contribute up to $73bn (around £58bn) to the UK economy over its lifetime, with 90% of manufacturing in UK factories. We will back this programme fully and accelerate it.

• Full backing for the Rolls-Royce SMR programme at Wylfa — first units targeting grid connection in the mid-2030s
• Fast-track identification of further SMR sites beyond Wylfa — the programme should not stop at three units
• Support for Sizewell C and new large-scale nuclear sites as identified by Great British Energy-Nuclear
• Long-term procurement frameworks giving the British nuclear supply chain the certainty to invest in capacity

Thorium Molten Salt Reactors: The Next Generation

Beyond conventional nuclear lies a technology that could transform Britain’s long-term energy position. Thorium-based Molten Salt Reactors offer significant potential advantages over conventional uranium reactors — and Britain should be at the forefront of their development.

• Thorium is 3-4 times more abundant than uranium and Britain has domestic deposits
• MSRs operate at atmospheric pressure — eliminating the meltdown risk inherent in pressurised water reactors
• Produces dramatically less long-lived radioactive waste — hazardous for hundreds rather than tens of thousands of years
• Cannot easily be weaponised — a significant non-proliferation advantage

Copenhagen Atomics, a Danish company in active conversation with the UK’s National Nuclear Laboratory, has developed a working prototype and claims 300 units delivering 12GW of capacity could be deployed by 2035 under an as-a-service model — at no upfront capital cost to British taxpayers, saving an estimated £8bn annually versus current energy baseline costs. We reproduce that claim because it is the vendor’s and it is on the record, and we decline to adopt it. No thorium molten-salt reactor has yet been licensed or operated commercially anywhere; 12GW by 2035 would require licensing, supply chain and deployment at a pace the UK has not achieved for any nuclear technology in fifty years; and an as-a-service model does not remove capital cost so much as move it to a counterparty whose own financing has to come from somewhere. We therefore book none of the £8bn. Thorium appears in this programme as a research and licensing commitment — funding the regulatory pathway so that if the technology works Britain is not five years behind — not as a source of capacity or savings in any scenario in this paper.

We will establish a dedicated Thorium MSR development programme — inviting international developers to partner with UK institutions, fast-tracking regulatory assessment, and positioning Britain as the leading Western nation in next-generation nuclear technology. China is already ahead in this race. Britain should not cede that ground.

Geothermal: Britain’s Untapped Strategic Asset

Britain is sitting on an energy and critical minerals asset it has barely begun to exploit. The UK’s first deep geothermal power plant went live at United Downs in Cornwall in February 2026 — and what it revealed should fundamentally reshape our energy strategy.

• The British Geological Survey estimates over 200GW of thermal resource at temperatures suitable for electricity generation lies beneath Britain’s surface
• Geothermal provides 24/7 baseload power — entirely weather-independent, unlike wind and solar
• The Cornwall site simultaneously produces battery-grade lithium carbonate at among the highest concentrations found anywhere in the world
• Scaling lithium production to 18,000 tonnes annually would supply 65% of Britain’s EV battery needs from domestic sources

Geothermal is not just an energy story — it is a critical minerals story. Domestic lithium production reduces supply chain dependence, supports British EV manufacturing, and strengthens national economic security. We will establish a national deep geothermal development programme with fast-track planning consent and long-term revenue certainty for developers.

Fixing the Wind Power Scandal

The problem has two parts. First, location. The majority of UK wind capacity sits in Scotland — far from the major demand centres of England. The transmission network cannot move all that power south when the wind blows strongly. The result: Scottish wind farms are paid to turn off, and gas plants in England are paid to fire up to replace them. Total curtailment payments exceeded £2.5bn in 2025 alone. As Octopus Energy’s Greg Jackson has observed, some energy companies own both a constrained Scottish wind farm and a flexible gas plant in England — allowing them to be paid twice during the same constraint event. Consumers pay both times.

Second, the Contract for Difference structure itself. CfDs guarantee revenue regardless of where power is generated or whether the grid can use it. There is no financial incentive for developers to site projects near demand, near storage, or in ways that reduce grid pressure. The subsidy system is location-blind in a country where location is the entire problem.

We will reform CfDs for all new contracts:

• Locational pricing — strike prices reflect proximity to demand centres and available grid capacity
• Storage co-location incentives — developers pairing generation with battery or pumped hydro storage receive premium contract terms
• Local energy pricing — communities near generation assets pay lower bills, creating genuine social licence for new development and reducing long-distance transmission costs
• End structural curtailment payments — new CfDs include grid risk sharing so developers have a direct financial interest in siting projects where power can actually reach consumers

Energy Storage: The Missing Link

Intermittent renewables become reliable when paired with adequate storage. Britain has massively underinvested here. We will accelerate:

• Grid-scale battery storage co-located with generation and at key grid nodes
• Pumped hydro — Britain’s geography is well suited and planning barriers will be removed
• Vehicle-to-grid technology — turning the growing EV fleet into a distributed national storage asset

Taking the Levies Off the Bill: Cheaper Electricity Than Gas

There is a hidden tax on your electricity bill, and it is working against everything this chapter is trying to do. A set of legacy green levies — chiefly the Renewables Obligation (around £89 on a typical bill) and Feed-in Tariffs — are loaded onto electricity, not gas. The perverse result is that the cleaner fuel is made artificially more expensive than the dirtier one. A household thinking of switching from a gas boiler to a heat pump, or running an electric car, is financially penalised for going green — the exact opposite of the outcome we want. Martin Lewis has called this “policy perversion”, and he is right.

We make a clean distinction the current debate usually blurs, between subsidies that are merely sitting on bills and mechanisms that actually lower them:

• Move the legacy subsidies off bills — precisely, and costed from public data and subject to independent verification. The Renewables Obligation and the old Feed-in Tariff generation subsidy are closed schemes: they pay for renewable capacity already built and buy no new generation. They are pure legacy cost, and there is no reason their runoff should sit on electricity bills making power dearer than gas. Scope matters, because it drives the cost, and there are three groups to be clear about. Households: most of the work is already done and sits in the baseline we inherit — the current government has already moved the bulk of the domestic Renewables Obligation onto general taxation; completing the remaining domestic portion is a small, incremental step. Energy-intensive industry: the steel, chemicals, glass and ceramics producers whose competitiveness is most exposed are already 100% exempt from the Renewables Obligation under the existing relief and the 2024 “British Industry Supercharger” — so the strategic-industry case is largely already handled, and our task there is to protect and extend that exemption so it reliably covers strategic energy-intensive activity, including qualifying data centres. The group that falls through the gap is everyone in between: the small and commercial businesses — offices, shops, warehouses, hospitality, smaller manufacturers — too small to qualify for the energy-intensive exemption but still carrying the full levy. For them we move 50% of the non-exempt business Renewables Obligation onto general taxation. The truth is that this is a partial fix, not a complete one: it halves the distortion rather than removing it, a deliberate balance between competitiveness and fiscal prudence, with the remainder falling away as the scheme tapers anyway. And we cost it: a derived estimate of £1.5–2.5bn a year, declining, scored in the table and flagged for confirmation against the published non-domestic split — not waved through as free. The whole cost is in any case a wasting one: the Renewables Obligation runs off as its 20-year contracts expire through to 2037, falling from roughly £8bn now toward £6bn by the end of the decade as the oldest projects drop away
• Keep — and reform — the mechanisms that make electricity cheaper. Contracts for Difference and the Capacity Market are different in kind. A Contract for Difference gives a generator a fixed price decoupled from gas, which is precisely what protects bills from gas-price spikes: without CfD-backed wind, wholesale electricity would be materially more expensive, not less. So we do not scrap CfDs — we keep them and fix their one real flaw, location-blindness, through the locational pricing and grid-risk-sharing reforms set out above in Fixing the Wind Power Scandal. The Capacity Market, which keeps the lights on, stays. The test is simple and honest: a charge that lowers or stabilises the price of power earns its place on the system; a legacy subsidy that merely sits on the bill does not
• Keep the sensible half of Feed-in Tariffs — pay fairly for power fed back to the grid. The old scheme bundled two different things: a subsidy for generating, which is closed and rightly running off, and a payment for the surplus a household exports to the grid, which is simply good sense — excess clean power should not be wasted. That export function already exists as the Smart Export Guarantee, which obliges larger suppliers to buy exported power, but it is weak: suppliers set their own rates and some offer next to nothing. We strengthen it with a mandated floor price, set a few percent below the prevailing wholesale price — illustratively around 5%. That is high enough to pay households and businesses the genuine value of the power they export, rather than the token rates offered today, while leaving the supplier a small margin to cover costs. Crucially, because the floor is anchored to wholesale value rather than the retail tariff, it is never recovered from other households’ bills — the supplier pays roughly what the power is worth to them, so there is nothing to socialise onto everyone else. A fair price for your surplus, no added cost to your neighbour

On the cost, and precise about its shape. Moving these legacy levies off bills does not make the cost vanish — it moves to the Exchequer. The household portion is small and largely already done; the energy-intensive producers are already exempt; what we add is a costed, partial shift of the non-exempt business levy, estimated at £1.5–2.5bn a year and declining. None of it is a permanent new line: it is a slice of a wasting, closed subsidy that runs off through the 2030s, so the cost shrinks every year rather than compounding. We score it in the table, we flag it as an estimate pending the published split, and we do not inflate it into a false headline figure. We judge the commitment worth making, because cheaper electricity than gas is the precondition for the electrification — heat pumps, electric vehicles, electric industry — that the rest of this chapter depends on. For households the prize is a meaningful cut to the electricity bill and an end to the absurdity of taxing the clean fuel to subsidise the dirty one. For energy-intensive industry the protection of the existing exemption keeps electricity-intensive manufacturing competitive, and it sits alongside the cheaper baseload from the nuclear and geothermal build-out set out earlier — the deeper fixes to industrial power costs being grid connection, network charges and the wholesale price, which this chapter addresses directly.

Carbon Pricing: Make It Count, and Stop Giving It Away

Britain prices carbon through the UK Emissions Trading Scheme — a cap-and-trade market, separate from the bill levies above, under which around a thousand businesses in power, heavy industry and aviation must buy allowances for the carbon they emit. We are clear about what it is, because the honest case here is not the populist one. The Scheme is not a stealth tax dressed up as green policy; it is a genuine and well-designed decarbonisation tool. Its purpose is not to raise money but to cut emissions: the cap shrinks every year — from roughly 44 million allowances in 2027 toward 24 million by 2030 — so the Treasury expects it to raise less over time, not more, as firms decarbonise. Carbon pricing of this kind has been one of the single largest contributors to the near-halving of Britain’s emissions since 1990. We support it, and we would not pretend otherwise to score a cheap point against business.

But there are two things wrong with how Britain runs it, and both are worth fixing. The first is that, almost alone in Europe, the United Kingdom recycles none of the revenue back into the green transition. The Scheme raised about £2.6bn in 2024-25 and roughly £17.8bn over 2021–25, and every penny disappears into the general pot. HM Treasury does not earmark any of it for climate action — while the revised EU rules require member states to direct at least half of their carbon revenue into exactly that. So British businesses pay a carbon price whose proceeds do nothing visible to help them, or anyone else, actually decarbonise. That is a fair grievance, and it corrodes the consent the policy depends on.

• Recycle the carbon revenue into the transition. We will hypothecate a meaningful share of UK ETS auction revenue — bringing Britain into line with the European norm of at least half — into the decarbonisation that the price is meant to drive: grid upgrades, industrial decarbonisation and carbon capture, and help for the energy-intensive and smaller firms most exposed to the carbon price. This is a redirection of money the Exchequer already collects, not new spending: it is broadly neutral for the deficit, and we treat it as such in our figures. What changes is not how much is raised but what it does — a carbon price that visibly funds the transition is one business and the public will keep supporting

The second problem is the one that is about to get more expensive. Because the UK ETS is a small, low-liquidity market, its carbon price has been volatile and has often sat below the EU’s. From January 2026 the EU’s Carbon Border Adjustment Mechanism charges imports from countries with a lower carbon price the difference at the EU border. On current pricing that could mean UK exporters paying around £800 million a year straight into EU coffers simply to trade with our largest market — the government’s own estimate of the exposure — and, perversely, even Britain’s 100% clean wind, solar and nuclear power being charged a carbon price at the EU border. Paying a tax to Brussels, on clean British energy, for the privilege of exporting: it is hard to imagine a worse outcome, and it follows directly from standing apart.

• Link the UK and EU Emissions Trading Schemes. The two governments agreed in May 2025 to work towards linking their carbon markets, and we will complete it. Linkage ends the Carbon Border Adjustment charge between us — British exporters stop paying the differential into EU coffers — deepens a thin and volatile UK market into a far larger and more stable one, gives a stronger and steadier investment signal for low-carbon capital, and is expected to raise more revenue for the British Exchequer, not less. It is the same logic as the rest of our European strategy (Chapter 4): on the things where standing apart simply costs us money for no sovereignty worth the name, we realign with our largest market. Carbon is one of the clearest examples

Taken together these two moves turn the carbon price from a grievance into an asset: linked to Europe so it is deeper, steadier and no longer a border tax against us, and recycled at home so the money it raises visibly funds the transition it exists to drive. It is also true that neither move is a revenue windfall to be banked — linkage’s yield depends on where the joint carbon price settles, and recycling redirects existing money rather than creating new money — so our fiscal framework leans on neither. The case is not fiscal. It is that a carbon price worth having is one that is stable, fair to British exporters, and seen to do the job it was created for.

The Carbon Border: What We Keep, and What We Drop

There is a second carbon border to be honest about, because Britain is not only being charged at the EU’s frontier — it is building a frontier of its own. From January 2027 the UK’s own Carbon Border Adjustment Mechanism, now law under the Finance Act 2026, will charge higher-carbon imports of steel, cement, aluminium, fertiliser and hydrogen the difference between their carbon price and ours. It is worth being clear-eyed about what this kind of measure is, and is not. A carbon border is not a growth policy and not a revenue-raiser; independent analysis suggests both the EU and UK versions would raise little, and that a fully successful one would raise nothing at all, because it would simply prompt other countries to price carbon at the same level. Nor is it costless at home: while it protects the carbon-intensive producers who already pay our carbon price, it raises input costs for the far larger set of downstream firms — the car makers, the builders, the manufacturers — who buy their steel and cement. We should not pretend otherwise.

So why have a carbon border at all? Because it is the unavoidable complement to having a carbon price. If Britain makes its own industry pay for the carbon it emits — which the Emissions Trading Scheme does, deliberately — then either we adjust for that at the border or we simply watch carbon-intensive production, and the jobs and emissions with it, relocate to countries that do not price carbon at all. That is not decarbonisation; it is exporting our industry and importing the same carbon by another route. A carbon border is what stops a domestic carbon price from quietly de-industrialising the country. It is protective, not dynamic — the price of having a carbon price worth the name.

Which brings us to the honest simplification our European strategy makes possible. Once the UK and EU carbon markets are linked, Britain does not need a standalone carbon border of its own at all. The precedent is clear: Norway, Iceland and Liechtenstein sit inside the EU’s carbon market and are exempt from its border charge, while sheltering behind the EU’s common external border against the rest of the world. Britain would occupy exactly that position. Linkage makes trade in both directions across the Channel border-free — no charge on our exports to the EU, none on theirs to us — while the EU’s external carbon border, which we would stand behind, continues to protect against high-carbon imports from elsewhere. That lets us do something this paper always prefers where it can: remove a whole layer of duplicated domestic bureaucracy. The separate UK mechanism, with its own HMRC administration, its own returns and its own compliance burden on 1 January 2027, can be folded into the common European border rather than run in parallel with it. British industry keeps its protection against third-country carbon leakage; British firms trading with Europe lose a border charge entirely; and the country runs one carbon border instead of two. It is the same principle as the rest of our approach to Europe: where standing alone means duplicated cost for no sovereignty worth the name, we align with our largest market — and here, alignment lets us scrap a domestic scheme rather than build one.

We accept that this depends on the linkage actually being completed, and until it is, the UK mechanism stands and does its job. But the direction is clear, and it is the honest one: a carbon border is a necessary shield for as long as we price carbon alone, and a redundant duplication the moment we stop being alone. We would rather say that plainly than defend two overlapping bureaucracies for the sake of appearances.

The Economic Dividend

Lower energy costs feed directly into business competitiveness, household spending power, and industrial viability. Energy-intensive industries — steel, chemicals, ceramics, glass — can survive and grow in Britain rather than relocating to countries with cheaper power. Domestic lithium production supports the EV and battery industries. Nuclear and geothermal provide the long-term price stability that serious industrial investment requires.

Chapter 11: Transport

Drivers are not the enemy. Commuters are not a problem to be solved. Transport policy should serve the people who use it — not punish them for needing to get somewhere.

Roads

• Motorway speed limit increased to 80mph — recognising that modern vehicles are categorically safer than those of 1965 when the current limit was set
• 20mph limits restricted to residential streets and school zones — not imposed blanket across entire towns and arterial routes

The Road Charge

We will abolish Vehicle Excise Duty and the planned pay-per-mile scheme, and replace both with a single flat Road Charge — one published rate per vehicle class, CPI-indexed, with every pound of revenue allocated in public:

• Cars and vans: £450 a year — petrol, diesel, hybrid or electric, identical
• Motorcycles: £100 a year
• E-bikes: £25 a year, with registration — a number plate, not a revenue line
• Pedal cycles: nothing, ever — no registration, no charge. The line is the motor
• HGVs: the existing weight-based schedule stays — a 44-tonne lorry does exponentially more road damage than a hatchback, and flat-rating it would be a subsidy
• Exemptions preserved: disabled drivers and vehicles over 40 years old

No CO2 bands, no first-year supercar rates, no list-price supplements, no mileage reporting. A Range Rover already pays more VAT at purchase and more fuel duty at the pump than a Fiesta — the state collects on price and on use elsewhere. The access charge has one job, and it does it in one line.

The arithmetic, in full, because a flat charge invites the question. There are 41.7 million licensed vehicles in the UK, of which roughly 38 million are cars and light vans. Net of disabled and historic-vehicle exemptions — about 2.5 million — that leaves 35.5 million chargeable. Take off the 1.3 per cent the Department for Transport estimates evades vehicle tax, and a £450 charge on cars and vans plus £100 on motorcycles yields about £15.9bn. We then take a further 2 per cent off for behavioural response — vehicles taken off the road, declared SORN, or not replaced when scrapped — which is the sort of reaction a charge of this size should be expected to provoke, and which we would rather assume than be surprised by. That gives £15.6bn, and £15.6bn is the figure carried in the scorecard. Against it, VED is abolished: the OBR forecast it would raise £9.1bn in 2025-26. The net new revenue is therefore about £6.5bn, and that is what pays for the commitments attached to it.

The larger is the road itself. The 2026 ALARM survey puts the carriageway repair backlog in England and Wales at £18.62bn. We commit £4bn a year to clearing it, falling to a maintenance level once cleared — the survey's own finding is that clearing the backlog reduces annual maintenance costs by about £1bn, so the programme pays for part of itself. We should be clear about one constraint money does not solve: the Asphalt Industry Alliance judges the work itself would take twelve years, not four, because contractor and materials capacity binds before funding does. Ours is a funding commitment; the delivery profile depends on building that capacity, and we say so rather than promise a completion date we cannot control. The second commitment is free parking — and it is narrower than the phrase usually implies, so we should define it before defending it. It means hospitals, town centres and railway stations. It does not mean free parking everywhere; airport car parks, private operators, city-centre congestion schemes and residents' permit systems are all untouched.

The rationale in each of the three cases is growth or coherence, not generosity. Town centres charge for parking while out-of-town retail parks give it away free — a disadvantage created by planning policy, not by the market, and imposed on exactly the high streets this paper is already trying to rescue by abolishing Business Rates. Hospital charges are a tax on illness levied on the sick and the people visiting them, and serve no policy purpose we can identify. And charging people to park at a railway station is a tax on modal shift inside a programme whose whole transport chapter is trying to get people onto trains. Three narrow, arguable cases; not a giveaway to drivers in general.

The compensation figure has been corrected downward, and the correction runs against us in presentation and for us in the arithmetic, so it is worth showing. The RAC Foundation puts English councils' parking income at £2.163bn in 2024-25, and an earlier draft of this paper compensated councils for all of it. That was wrong. The £2.163bn is total parking income and includes penalty charge notices and residents' permits, neither of which this policy stops — Leicestershire's on-street account, to take one published example, records £842,000 of PCNs and £217,000 of permits against no fee income at all. Councils keep that money because they keep those charges. What they actually lose is the fee income from the three categories above, which we put at roughly £1.3-1.5bn; we compensate at £1.4bn. On these figures the road package is a net saving of about £1.1bn in Year 1, rising to about £4.1bn by Year 5 as the repair programme falls back to maintenance. We book that, and we should say why we treat it differently from the health saving in Chapter 5, which we declined to book at all. This number rests on no behavioural assumption about how anybody responds. It is a fixed charge on a known vehicle stock, set against a published OBR forecast for the tax it replaces — arithmetic, not a projection about how people will behave. We have still taken 2% off the gross for the extra evasion and off-road notifications that doubling a charge will provoke.

Where Every Pound Goes

The Road Charge raises £15.6bn a year on the figures above. The allocation is published and audited annually:

• £1.4bn — free parking at town centres, hospitals and railway stations, compensating councils, NHS trusts and station operators for the fee income they lose (not for penalty notices or residents' permits, which continue)
• £4bn a year — the Local Roads Renewal programme, against the £18.62bn carriageway repair backlog the 2026 ALARM survey identifies, falling to a maintenance level once cleared
• £1.2bn — local transport infrastructure
• £9bn — the general revenue VED already provides, plus cover for the fuel duty freeze below

And there is a loser to name, because a flat charge creates one. £450 is roughly double the current average of about £220 across the fleet, and more than double the £195 standard rate a modern car pays. The honest household arithmetic: the standard rate rises from £195 to £450 — £255 a year more. Against it: £300-600 of parking charges gone for those who regularly park in a town centre, at a hospital or at a station, and the average £460 pothole repair bill engineered away. For most driving households this is a net gain in year one. For a rural driver with a driveway who rarely pays to park, it is a straight increase of a couple of hundred pounds a year with no offset. That is a real cost to a real group, and it is the price of not metering anybody's mileage. Multi-car rural households, who gain least from the Rail Pass, are the clearest case, and we name them rather than hoping nobody adds it up.

Fuel Duty: Frozen, Withering, Never Per-Mile

Fuel duty is frozen in cash terms permanently — no escalator, no staged returns. As the fleet electrifies it withers on schedule: a deliberate, published, fifteen-year tax cut for motorists, part-funded inside the Road Charge allocation above. And we make the commitment Labour would not: there will be no pay-per-mile charging in Britain — not by tracker, not by odometer return, not by stealth. They reach £450 a year by making electric drivers report their mileage; we reach it with one flat line and no forms.

E-Bikes: Registered, Charged, Off the Pavement

Riding on the pavement has been illegal since 1835 — for every cycle, electric or not. The law is not the gap; enforcement is, because an unregistered machine with no plate faces no consequence. The £25 Road Charge tier closes that gap:

• Every e-bike registered, with a visible identifier — making theft recoverable, hit-and-runs traceable and enforcement possible
• Pavement riding on a motorised cycle: a £100 fixed penalty, actually enforced — resourced within our police restoration programme
• Derestricted and throttle-only machines are already motorcycles in law: unregistered ones are seized, full stop
• Delivery platforms made liable for their riders' machines being legal and registered — the illegal e-moto epidemic is a business model, and the businesses will answer for it
• Battery safety standards for imports — ending the fire risk in flats and hallways
• Speed pedelecs (28mph assist) sit in the motorcycle tier, as the law already classifies them

Ordinary pedal cycling stays exactly as it is: free, unregistered, encouraged. The line is the motor — and the principle is the one running through this whole chapter: use the road, share the cost, follow the rules everyone else follows.

Rail: A Transformation, Not a Tweak

The problem with UK rail is not simply that fares are high. It is that the pricing system is incomprehensibly complex, peak fares are punishing, and walk-up tickets bear no relationship to the cost of provision. We will reform all three.

The F³ National Travel Pass

This is a growth and labour-mobility measure that happens to cut travel costs — not a travel subsidy that might help growth. It belongs in the same family as planning reform, Single Market re-entry and tax simplification: policies that widen the market in which people can work, hire, learn and trade. Britain’s productivity problem is partly a geography problem — workers priced out of the jobs they could reach, firms unable to recruit beyond a narrow radius, regions economically severed from one another by the cost of getting between them. A flat, affordable pass covering both rail and bus attacks that directly:

The obvious question is capacity, and it is worth answering rather than ducking. Germany launched the €49 Deutschlandticket nationally in a single step in May 2023, without staggering, and the demand response was overwhelmingly modal shift and additional off-peak journeys by people already using public transport — not a wave of new peak commuters. That is what the evidence would predict here too: a season-ticket holder does not start commuting because the ticket is cheaper, they already commute. So we launch nationally rather than in phases. What the Pass changes is the cost of journeys people are largely already making, plus the discretionary off-peak travel that fills empty seats — which is where the spare capacity actually is. Where a specific corridor is genuinely at its peak limit, the answer is rolling stock and paths on that corridor, funded from the capital side, not a slower rollout that denies the benefit to everyone else while the problem is fixed for a few.

• Labour mobility: a worker can take a better job two towns away without a £4,000 season ticket deciding it for them
• Wider hiring pools: employers recruit from a larger catchment, easing the skills shortages that hold back growth
• Access to education, training and apprenticeships: the 18–22 Guarantee and the free STEM pipeline are only as good as a young person’s ability to physically reach them
• Regional integration: cheaper movement between cities knits regional economies together, spreading the agglomeration gains today concentrated in London

The timing suits the reform already underway. With Great British Railways bringing track, stations and train operations into a single public body, a flat national rail fare becomes far simpler to deliver than under the old fragmented franchises — there is no longer a tangle of competing operators to compensate or revenue to carve up. Buses are harder, because outside London most are privately run and deregulated: delivering a true national pass means bringing bus operators into the scheme through the franchising and partnership powers now expanding under devolution, and we are honest that this is a delivery task, not a stroke of the pen. Inspired by Germany’s transformative Deutschlandticket — which attracted 11 million subscribers within months of launch — the British version:

• £75 per month or £800 per year for a national pass covering all standard-class rail travel and local bus services
• £45 per month or £480 per year for a regional pass covering rail and bus within a defined zone
• Concessionary rates for under-25s and lower-income over-65s
• Available to individuals and as an employer benefit

On the economics, because the rest of this paper demands it. A pass at this price is deliberately set below what heavy users currently pay, so it does require net public subsidy — we estimate around £3.5bn a year once the Commuter Mobility Levy is netted off, in the same range as Germany’s Deutschlandticket. We treat that as a growth investment, not a giveaway: economic infrastructure priced like a road network, paid for because the mobility it unlocks — workers reaching better jobs, firms recruiting from wider pools, young people reaching training — is what generates the return. We keep the price low on purpose. Set it too high and the occasional user, whose flexible purchase cross-subsidises the heavy commuter, simply does not buy it, and the model unravels. The subsidy is the point, and we cost it openly rather than pretending a payroll levy alone makes it free.

A London commuter currently spending £4,000 a year on season tickets would save over £3,000. A family making occasional trips to visit relatives would gain flexibility without the lottery of dynamic pricing. A business offering the pass as a staff benefit would see measurable improvements in recruitment, retention and punctuality.

Peak Fare Cap

Walk-up and peak fares will be capped at a fixed multiple of the equivalent off-peak fare — ending the practice of charging commuters who have no choice but to travel at peak times whatever the market will bear.

Ticketing Simplification

The complexity of UK rail ticketing is itself a hidden tax. We will mandate a simplified national fare structure — no more than five fare types on any route — and require all operators to sell the cheapest available fare for any given journey proactively, without passengers needing to know which obscure ticket type to request.

The Commuter Mobility Levy

Large employers are the primary beneficiaries of an efficient commuter network. A business with hundreds of employees in a city centre depends on public transport to function. Yet the cost of that infrastructure falls entirely on passengers and the public purse.

We will introduce a Commuter Mobility Levy on companies with 250 or more employees or a payroll above a specified threshold:

• A modest levy of approximately 0.3% of payroll above the threshold
• Revenue ring-fenced exclusively for: reduced commuter rail fares, peak fare caps, station improvements, park-and-ride expansion, and local transport links to employment hubs
• Applied regionally — revenue raised in the North funds Northern rail, revenue raised in London funds London commuter routes
• Far smaller than the Business Rates we are abolishing — net effect on large employers remains strongly positive

At 0.3% of eligible payroll, the levy raises approximately £1.2bn annually. That is a contribution from the large employers who most depend on an efficient commuter network — not the whole cost. It offsets part of the pass subsidy costed in the fiscal framework; the balance is the explicit growth investment described above. Large employers gain a more reliable, punctual workforce, and we present the levy as what it is — a partial, fair contribution — rather than pretending it makes the pass free.

Rail Reliability

• Stronger performance contracts with operators — with meaningful financial penalties for persistent delay
• Long-term investment in capacity, electrification and station infrastructure
• Open-access competition on suitable routes — the single biggest driver of lower fares where it has been tried

Connectivity: What We Build, and How We Decide

Everything above is about the trains that already run — what they cost to board and whether they arrive on time. But Britain also needs new connections, and here we have to be honest about a national embarrassment: we are extraordinarily bad at building them. The problem is not a shortage of proposed railways. Governments have announced them for decades — Northern Powerhouse Rail, Crossrail 2, the eastern leg of HS2 — and the drawing boards groan with schemes that were promised, half-started, rescoped and quietly shelved. The problem is that we cannot build what we propose at a price the country can bear, and so, over and over, we build less than we said, later than we said, for more than we said. HS2 is the monument to this failure: the most expensive railway per mile ever built anywhere on earth, cut back to a stub while France and Spain kept laying track at a fraction of the cost (Chapter 2). We will not add another glossy line to that graveyard of announcements. We will do the less glamorous thing that actually delivers connectivity: fix how Britain chooses and builds, first.

That means two disciplines, in order. The first is the build-cost reform this paper sets out in full in Chapter 2 — the standardised consenting, the off-the-shelf design, the single accountable sponsor, the scope that stops moving. Until Britain can build a mile of railway for something like what its neighbours pay, no amount of ambition on the map is worth anything; it will simply become the next overrun. Bringing our costs even partway back toward European norms is what turns connectivity from a fantasy into a budget line. The second discipline is honest selection. We will choose what to build by connectivity and economic return per pound — published, checkable business cases judged on the growth and access they actually deliver — not by which project carries the most political glamour or lands in the most marginal seats. A railway is an investment, and it should have to prove its return like one.

Apply that test honestly and it points somewhere specific, and long neglected. For a generation, British rail investment has been overwhelmingly radial and capital-centric — faster ways into and out of London — while the connections that would do most to rebalance the economy have languished. The single clearest gap is east-west connectivity across the North: the corridor linking Liverpool, Manchester, Leeds, Sheffield and Hull, where journeys between great neighbouring cities are still slow, infrequent and unreliable in a way that would be unthinkable between comparable cities in Germany or the Netherlands. Knitting the northern cities together does more for national rebalancing, and has a stronger agglomeration case, than another marginal minute shaved off the journey to London. We name this as the priority of principle — the place the connectivity-per-pound test most clearly points — and we are deliberately disciplined about what we do not do next:

• We name the priority, not a costed route. We will not repeat the central error we criticise — announcing a specific line with a specific price tag before the engineering, the design and the honest cost work exist to stand behind it. The east-west northern corridor is where the case is strongest and where a reforming government should concentrate; the specific alignment, phasing and budget are for proper appraisal under the disciplines above, not for a manifesto to invent. Naming a number we cannot yet defend is how the graveyard of announcements got so full
• Upgrade and electrify before we reach for the megaproject. The instinct to solve every connectivity problem with a grand new line is part of the disease. A great deal of the gain — more frequent, faster, electrified services on existing and improved alignments — can be delivered sooner and far more cheaply than a new-build high-speed railway, and we will exhaust the cheaper, faster options before committing to the expensive ones. Incremental capacity that arrives this decade beats a spectacular line that arrives, over budget, in the 2040s
• Deliver it under the same discipline as everything else. Whatever is built proceeds under the single-sponsor, fixed-scope, standardised-design regime of Chapter 2, with the business case published and the cost held to account. Connectivity earns its place in this paper not because railways are romantic, but because — built affordably and chosen honestly — they are among the highest-return investments a state can make. Built the way Britain has built them lately, they are among the most ruinous. The difference is entirely in the discipline

This is the unglamorous truth about rail connectivity, and we would rather tell it than sell you a map. Britain does not need another politician holding up a coloured line and a groundbreaking date. It needs a state that can build a railway for what a railway should cost, choose the right ones for honest reasons, and finish them. Do that, and the connections this country has promised itself for thirty years finally become affordable. Fail to do it, and no map, however bold, will ever leave the drawing board.

Chapter 12: Defence and National Security

The Commitment

The UK currently spends 2.4% of GDP on defence — £60bn annually, rising to £73.5bn by 2028-29. NATO’s emerging expectation is 3.5% of GDP. The gap between commitment and requirement is real, growing, and cannot be filled by tinkering at the margins of existing budgets.

We will increase defence spending to 3% of GDP within this parliament and set a statutory, dated path to NATO’s 3.5% core-defence commitment by 2035 — matching the seriousness of the threat environment with the seriousness of our response.

Honest accounting: 3% of GDP is roughly £15bn a year above current plans by the end of the parliament. Most of that is resource spending — pay, training, munitions, operations — and it appears in our fiscal scorecard as a ramped cost of £3bn in Year 1 rising to £10bn by Year 5, reflecting the first steps of the climb toward 3.5%. The capital share — ships, aircraft, infrastructure — is financed through the Defence Bond. We do not pretend salaries can be paid with bonds, and we will not park the largest spending commitment in this paper outside its own fiscal framework.

That last point is not a throwaway. When the government published its own Defence Investment Plan in mid-2026, it left roughly a third of the promised uplift — getting on for £5bn — formally “to be funded at a future Budget”: a commitment with no source, deferred to a decision not yet taken. We regard that as precisely the gap between announcement and delivery this paper exists to close. Every pound of our defence commitment is costed and sourced here — resource in the scorecard, capital on the Bond — not promised now and funded later, or never.

We are also honest about a number we decline to simply adopt. Defence chiefs have put the shortfall in the government’s plans at around £28bn over four years, and much commentary treats that figure as the measure of what Britain must find. We are not persuaded it is the right question. That number prices the existing programme of record — the traditional platforms and force design largely inherited from a previous era — and asks how to pay for it in full. But the lesson of the section above is that a pound spent on mass drones, counter-drone and cyber can buy far more deterrence than a pound spent on another exquisite platform delivered a decade late. The goal is not to fund yesterday’s shopping list to the last item; it is to buy the right capability mix. It is also true that this cuts both ways: some capabilities — the nuclear deterrent, integrated air defence, the submarine fleet — are irreducibly expensive and cannot be replaced by cheap and many, so we do not claim a modern mix is simply cheaper. But we do claim it is different, and that measuring adequacy against the old list is how a country spends heavily and still buys the wrong war.

The F³ Defence Bond

We will fund a significant portion of the defence spending increase through a public Defence Bond programme — modelled on the success of Green Savings Bonds and the wartime bond model that has served democracies in every major conflict:

• Retail Defence Bonds available to every British citizen and institution
• Competitive fixed-rate returns over 3, 5 and 10 year terms
• Revenue ring-fenced exclusively for capital defence investment — ships, aircraft, vehicles, cyber infrastructure
• Not counted against day-to-day borrowing rules — defence capital investment is treated as national infrastructure, not consumption

Defence bonds do three things simultaneously: they fund the capability Britain needs, they give British savers a patriotic and financially sound investment, and they build public engagement with and support for defence investment. At a time when the transatlantic security guarantee is less certain than at any point since 1945, that public buy-in matters.

Buy British: A Defence Industrial Strategy

Britain has world-class defence industries in aerospace, shipbuilding, cyber, electronics and AI. Too often, procurement decisions default to overseas suppliers when British alternatives exist or could be developed. We will end this.

Article 346 of the Treaty on the Functioning of the European Union explicitly allows member states to exempt defence procurement from Single Market rules where essential security interests are at stake. France routinely uses this to buy French nuclear submarines. Belgium awarded a 20-year strategic partnership for light weapons to FN Herstal without competitive tendering. Germany buys German armoured vehicles. Britain will use the same legal tools every major European defence nation already deploys.

• A strong preference for British design and systems integration on all security-sensitive defence contracts — invoked through Article 346 exemptions where legally appropriate
• British content requirements for non-sensitive components where this can be justified on industrial policy grounds
• Long-term procurement frameworks that give British defence industry the certainty to invest in capacity and skills
• Defence as an industrial policy — every pound spent on British defence capability circulates in the British economy, supports British jobs, and builds sovereign technological capability

We will not apply a blanket Buy British rule that would be legally unsustainable — but we will apply the Article 346 security exemption as broadly and assertively as the treaty allows, and more assertively than any recent British government has done. The legal tools exist. The political will has been absent. F³ will supply it.

Cyber and Technology

• Significant increase in investment in cyber defence and offensive capability
• Britain’s AI and technology sector enlisted as a national security asset — with appropriate frameworks for responsible use
• Sovereign satellite and communications capability
• Electronic warfare and counter-drone systems — the warfare of the near future, not the last war

What We Buy: A Capability Mix for the War of Now

Deciding how much to spend is the easy part; spending it on the right things is what actually buys security. The war in Ukraine has rewritten the rules of what matters on a modern battlefield, and any honest defence plan has to say where the money goes — not just how much of it there is. Our guiding principle is simple: buy for the war of now and the war of next, not the war of last time. That means a deliberate shift in the balance of investment, without abandoning the capabilities that still underwrite deterrence.

• Mass, attritable, autonomous systems first. The single clearest lesson of Ukraine is that cheap drones produced in their thousands now do work that once required platforms costing a thousand times as much, and that the side which can design, adapt and manufacture them fastest wins the tempo. We prioritise uncrewed air, sea and ground systems, loitering munitions, and above all the counter-drone and electronic-warfare systems to defeat them — built for scale and continuous iteration, not gold-plated perfection. This is the fastest-growing line in our defence investment, deliberately
• Cyber and space as first-order domains. Offensive and defensive cyber capability, resilient sovereign satellite communications and surveillance, and the electronic-warfare edge that decides whether anything else on the battlefield works. These are no longer support functions; they are where a peer conflict is won or lost first, and we fund them as such (building on the cyber commitments above)
• A conventional backbone kept strong, not hollowed. None of this means abandoning the exquisite few. Integrated air and missile defence, the shipbuilding that sustains the surface fleet and the sovereign submarine capability, combat air, and the continuous-at-sea nuclear deterrent remain the foundation of credible deterrence, and the Defence Bond funds their capital renewal. The shift is one of balance — more mass and more autonomy alongside the platforms, not instead of them
• Munitions stockpiles and the industrial base to refill them. Ukraine exposed how thin European magazines had become. We rebuild deep stockpiles of missiles and large-calibre ammunition and — more importantly — the standing industrial capacity to produce them at wartime rates, because a stockpile without a production line is a one-shot weapon. This is where defence spending and industrial strategy (Chapter 4) become the same policy

The thread running through all four is that modern defence capability and a sovereign industrial base are inseparable: what we can build, we can sustain, adapt and scale under pressure; what we can only buy, we can lose access to at the worst moment. That is why the procurement reforms below, and the European and Ukrainian industrial partnerships that follow, are not separate from the capability question — they are how it is answered.

Fixing Procurement: How We Buy, Not Just What

None of this works if it is run the way British defence procurement has been run for a generation. Consider Ajax. A £6.3bn contract for armoured reconnaissance vehicles, ordered in 2010, was declared ready for service in late 2025 — eight years late — only to be paused within weeks because the vehicles injured their own crews, with soldiers suffering hearing damage and nausea. The warning signs had been raised inside the programme as early as 2014 and formally in 2018, yet did not reach the responsible senior officer until 2020. And in the years the money was sunk, the battlefield moved on: cheap drones now do the reconnaissance Ajax was built for, more safely and at a fraction of the cost. Ajax is not a freak. It is what weak governance, blurred accountability, and a client too thin on engineering expertise to challenge its suppliers reliably produce — and the fault is not confined to defence. The pandemic’s protective-equipment scandal, with billions written off on unusable kit bought through fast-tracked lanes, was the same disease in a different department. A government that intends to spend three per cent of GDP on defence, and hundreds of billions across the state, cannot keep buying like this.

So the money in this paper comes with reform of how it is spent. Our aim is a procurement system that is fast where it must be, accountable when it fails, and technically literate enough not to be sold yesterday’s capability at tomorrow’s price. Four changes matter most:

• One named owner per programme, accountable for its life. Too many major contracts are run by a rotating cast with diffuse responsibility, so that when things go wrong — as with Ajax’s years of unreported warnings — no one is answerable. Every major programme gets a single senior responsible owner, in post for long enough to be held to account for delivery, with the reporting lines that carry bad news upward quickly rather than burying it
• Rebuild the client’s own technical expertise. The state has hollowed out its in-house engineering and commercial skill to the point where it cannot always tell a genuine advance from a repackaged commonplace, or hold a contractor to a specification. We will rebuild that capability inside government, because a buyer who does not understand what it is buying will always overpay for underperformance. This is the client-side counterpart to the industrial base we sustain on the supply side
• Fund what you approve. It is a peculiarly British absurdity to announce a capability with fanfare and then starve it of the money to happen — a pattern visible even in the government’s own recent plan, where headline programmes were endorsed and then left barely funded. Our whole method — every promise priced — forbids this. A capability we commit to is a capability we fund; one we cannot fund, we do not pretend to have
• Buy for iteration, not gold-plated perfection. The lesson of Ukraine is that continuously improved, good-enough kit fielded fast beats the exquisite system delivered a decade late. We will favour shorter, modular contracts with room to adapt, spiral development over decade-long fixed specifications, and a genuine route for small and innovative firms to sell into defence — the same open-the-field approach we bring to civilian procurement (Chapter 3)

It is worth being clear about why government keeps signing contracts that seem to tolerate overspending and slipping deadlines, because the answer is not simply incompetence and the fix is not simply “insist on hard deadlines.” Four structural forces are at work, and each has a remedy. The first is that fixed-price, fixed-date contracts only hold when the specification is stable — and government specifications are notoriously unstable, because ministers and priorities change mid-project, at which point the contractor is contractually entitled to more time and money. The discipline this demands is on the buyer’s side: lock the requirement before signing, and stop changing it. The second is the conspiracy of optimism — to win approval, bidders and officials alike have every incentive to lowball cost and timescale, because the honest number might get the project cancelled; so the “overrun” is really the gap between a fantasy that secured sign-off and the reality that always arrives. The remedy is independent, published cost assurance before approval, and consequences for the systematic optimism that has become a professional habit. The third is that genuine risk transfer is expensive: a truly ironclad fixed price costs more, because the supplier prices in every hazard, so government repeatedly chooses the cheaper-looking deal that leaves the taxpayer holding the risk — and then pays when it materialises. We will buy risk transfer where it is worth it and be honest that it costs more up front to cost less in the end. The fourth is the absence of a walk-away: for the largest programmes there is often only a handful of credible suppliers, sometimes one, and a buyer with no alternative has no leverage to enforce tough terms. Sustaining competition and a capable domestic base (above) is not only industrial strategy — it is what gives the state the power to hold any single supplier to account.

The through-line of all four is that tough contract terms are worth little unless the state has done the unglamorous work that makes them enforceable: a stable specification, an honest cost estimate, a real alternative to walk to, and the in-house expertise to know when it is being gamed. That is the reform. It applies across government, not only in defence — the same disciplines will govern the major civilian programmes and the digital-state build set out in Chapter 4.

We are honest about one real tension this runs into. Our commitment to buy British where we can (above) can cost more per unit than buying off an overseas shelf, and we will not pretend otherwise. But the Ajax saga shows the other side of that ledger: a domestic industrial base allowed to wither — no major armoured-vehicle order for over a decade — loses the very skills that make delivery reliable, so that buying abroad and buying badly at home become two versions of the same failure. The answer is not to abandon British industry to save money, nor to shield it from the discipline of delivering. It is to sustain a capable sovereign base and hold it to demanding standards — paying a measured premium for security of supply and skilled jobs, while refusing to pay for lateness, failure or gold-plating dressed up as sovereignty.

European Defence Cooperation

Rejoining the Single Market is not only an economic decision — it is the single most consequential thing Britain can do for its defence industry, and the timing could hardly be sharper. In 2025 the EU created SAFE (Security Action for Europe), a £130bn-plus joint procurement instrument, alongside a wider European defence-financing push worth hundreds of billions over the decade. The rules matter: SAFE requires most of a contract’s value to go to suppliers inside the EU, the EEA or Ukraine, and it treats EEA-EFTA states on fully equal terms with member states — able to bid, to sell, and to be bought from without restriction. A mere “partner country” is limited; an EEA member is not. Britain signed a Security and Defence Partnership with the EU in 2025, but the deeper participation stalled — which already has a price, from British production lines being locked out of jointly-funded orders to Franco-British missiles being sourced from their French plant rather than their British one. EEA membership resolves this at the root: it puts British defence firms inside the tent of the largest coordinated rearmament in Europe since the Cold War, on equal terms, by right. This is a direct dividend of the Single Market choice this paper makes in Chapter 4, and one too rarely counted.

And we will build a deliberate defence-industrial partnership with Ukraine. Ukraine is now the most battle-tested defence innovator on earth: it has learned, under fire and at extraordinary speed, how to design, iterate and mass-produce the drones and counter-drone systems that are reshaping warfare faster than any peacetime procurement system could. Europe has understood this — the EU-Ukraine Drone Alliance and the EUDIS Tech Alliance already link dozens of European and Ukrainian firms to scale exactly these technologies. Britain should be at the centre of that, not outside it. We will pursue joint development and production with Ukrainian industry in drones, counter-drone and electronic warfare; embed British engineers where the learning is happening; and channel the knowledge back into the British supply chain and the companies the British Future Fund backs (Chapter 15). Helping Ukraine build its defensive-industrial capacity and learning from it are the same act. Alongside this we will deepen bilateral cooperation with France, Germany, Poland and the Nordic and Baltic states, the natural partners in a Europe that must now provide more of its own defence as US engagement narrows.

Chapter 13: Pensions and Retirement

Britain’s pension system has two distinct problems: a state pension that is becoming unaffordable in its current form, and a private pension system that leaves too many workers — particularly the self-employed, part-time and lower-paid — with inadequate retirement savings. We will address both.

State Pension Reform

First, How the State Pension Is Funded — Because Abolishing NI Raises the Question

If employee National Insurance is abolished, what funds the state pension? The question is natural and the answer is reassuring, because it rests on a fact most people have never been told: National Insurance has not funded pensions in any real sense for decades. The UK state pension is pay-as-you-go — today's contributions pay today's pensioners, immediately, with nothing invested and no personal pot. The so-called National Insurance Fund is an accounting device that holds only a small working balance and is topped up from general taxation whenever it falls short. In substance, National Insurance was already a tax with a different name; the link between what you paid and what you receive was severed long before F³.

So folding employee National Insurance into the single income tax changes the label on the money, not the money itself. Pensions continue to be paid from general revenue, pay-as-you-go, without a day's interruption — exactly as they are now, and exactly as they are in the many developed countries that fund state pensions from general taxation rather than a separate contribution. Nothing about abolishing NI defunds the pension, because NI was never a fund.

What must be preserved is not the tax but the entitlement record — the qualifying-years count that determines who receives the full pension. That record continues unbroken on the same HMRC and PAYE systems that already track everyone's earnings: a year worked is a year credited, precisely as today, and the credits for those who cannot work — carers, the disabled, the unemployed — are retained in full (Chapter 20). You will build, keep and prove your state-pension rights exactly as before. The only thing that disappears is a separate, regressive payroll tax that stopped functioning as insurance generations ago.

The Pensioner Guarantee

Every pensioner deserves certainty and dignity in retirement. We replace the Triple Lock with the Pensioner Guarantee — a guaranteed, inflation-proof state pension for life, written into law.

• The State Pension will always rise with inflation. No pensioner will ever see their purchasing power eroded — that protection is guaranteed in law, not left to the discretion of the government of the day
• Pensioners continue to share in national prosperity through a five-yearly earnings review — an independent report to Parliament on where the pension stands against average earnings, with any correction taken openly against the fiscal position
• We are straight about what changes. The Guarantee floors the pension to inflation every year; what it gives up is the Triple Lock’s “highest of three” ratchet, under which the pension rose by earnings or 2.5% in years when those beat inflation. Over time that ratchet lifts the pension faster than prices — which is precisely why it is unaffordable, and why no honest costing can keep it. The Guarantee protects living standards in full; it does not promise to raise them faster than the economy can pay for, year after year. That is the difference, and we state it. We abolished one automatic lock; we will not legislate another

The Pensioner Guarantee replaces the Triple Lock with a system that is fair to today's pensioners and affordable for future generations — and we should be careful with the numbers here, because the well-known ones do not mean what they are usually taken to mean. The OBR's Fiscal Risks and Sustainability Report — restated in its July 2026 edition — puts the cost of the Triple Lock at £15.5bn a year by 2029-30 relative to earnings uprating, and £22.9bn relative to CPI uprating, against an original 2012 estimate of £5.2bn. Those are accumulated figures: they measure the gap between where the pension now is and where it would have been had a different rule applied since 2012. Switching prospectively does not recover that gap. It only stops it widening. The Resolution Foundation, working from the OBR's numbers, finds that moving to a smoothed earnings link from next year would save around £650m by 2029-30 rather than anything resembling £15bn. Our own switch is to CPI, which is less generous than earnings in most years and therefore saves more, but the mechanism is the same: the saving is the year-by-year differential, compounding from zero. That is why our scorecard shows £2bn in Year 1 rising to £10bn by Year 5 and not £22.9bn at any point. Anyone promising to save fifteen or twenty billion pounds by ending the Triple Lock in this parliament is quoting a number that describes the past, not the saving available in the future. The OBR's own framing makes the point for us: it describes the £15.5bn as the cost of having moved from an earnings link in 2012 to the Triple Lock — an accumulated gap, not a fund waiting to be reclaimed. Where the OBR does quantify a prospective saving it does so in the long run and as a share of GDP: uprating with earnings rather than the Triple Lock would reduce spending pressure by roughly 2% of GDP over its fifty-year horizon, against state pension spending otherwise rising from about 5% of GDP to around 9% by the mid-2070s. That is the scale of the prize, and it is a generational one rather than a parliamentary one.

Pensioner Benefits

The universal Winter Fuel Payment is abolished — not means-tested, abolished. Means-testing it would only swap one unfairness for another, creating exactly the kind of cliff edge this paper rejects, where the pensioner a pound over the threshold loses everything. Instead we do two cleaner things at once. For every pensioner, the universal benefit becomes the abolition of VAT on domestic energy (Chapter 1) — a permanent cut to the heating bill of every household in the country, pensioner or not. And for the poorest pensioners, a Winter Heating Element is built into Pension Credit, paid automatically each winter and linked to the average domestic gas price. This is the honest design: the help tracks the thing it exists to cover. In an expensive winter it rises automatically; in a cheap-gas year it falls, and if gas is cheap enough it falls to nothing — because at that point the heating bill is low, the energy-VAT abolition is already doing the work, and a payment that kept flowing regardless would be the very with-no-regard-to-need spending this paper abolished at the universal level. We deliberately set no floor: help should scale with the problem, not exist for its own sake. We do set a cap, reviewed annually against the same gas-price data Ofgem uses for the price cap — not to abandon pensioners in a crisis but to keep the commitment budgetable, because an uncapped payment indexed to a volatile global commodity is exactly the open-ended liability the fiscal glide path exists to prevent. The asymmetry is intentional: no floor, because cheap gas needs no subsidy; a cap, because a price spike must not blow a hole in the public finances. It rides on Pension Credit's existing taper, so it adds no new cliff. Free bus passes are similarly retargeted at pensioners on Pension Credit. As one in four pensioners is now a millionaire, a universal payment funded by working households was always a transfer from the struggling young to the comfortable old. We end the universal version and protect the pensioners who actually need it — by name, without a cliff.

And the other side of the ledger — what pensioners gain under F³: the £86,000 Dilnot cap on lifetime care costs, reinstated (Chapter 20) — the largest pro-pensioner commitment on offer from any quarter; a housing gain never taxed for moving and settled only once, in real terms, at death (Chapter 1); and a state pension guaranteed in law never to fall in real terms. We ask wealthier pensioners to give up universal perks. We give every pensioner protection from the catastrophic care costs no government has ever actually delivered.

Private Pensions: A System That Works — With Reforms

The existing private pension framework — tax relief on contributions, tax-free growth, and auto-enrolment — is broadly sound. We will keep what works and fix what does not.

Abolish Salary Sacrifice — For All Benefits

Salary sacrifice schemes — for pensions, cars, bikes and other benefits — were designed as a minor flexibility tool. They have grown into a major and incoherent tax avoidance mechanism. In 2024, £32bn of pension contributions alone used salary sacrifice arrangements, costing the exchequer billions in lost revenue that falls disproportionately on those who cannot access such schemes.

Under F³, employee National Insurance is abolished and employer NIC is halved. Salary sacrifice exists primarily to avoid NICs — so most of its rationale disappears with the reform itself. We will abolish all salary sacrifice arrangements cleanly and completely, replacing them with a simpler, fairer system where pension contributions receive straightforward tax relief at the contributor’s marginal rate.

Honest scoring: because the NIC saving that made salary sacrifice valuable is largely extinguished by our own reform, abolition yields roughly £0.5bn a year from boundary effects — not the £5bn a static reading against today’s system would claim, which would double-count revenue already moved by the NIC changes. The case for abolition now rests where it always should have: simplicity, and equal treatment between the employee of a sophisticated employer and everyone else.

Mandatory Employer Contributions

Auto-enrolment has been transformative but employer contributions remain too low and too discretionary. We will move to a mandatory employer contribution model — drawing on international best practice:

Australia’s Superannuation Guarantee requires employers to contribute 12% of ordinary earnings — a rate that has been legislatively increased year on year and is now settled at that level. Denmark and the Netherlands have similarly strong mandatory employer contribution frameworks that deliver retirement incomes their citizens can actually live on.

• Employer pension and care contributions raised from a 3% statutory minimum to 5% — 3% to the pension as now, and 2% ring-fenced by law to the National Care Fund (Chapter 20) — phased in over three years, with a five-year schedule for employers with fewer than 50 staff
• Auto-enrolment maintained and strengthened — with opt-out rather than opt-in as the default for all workers including part-time
• Extended to the self-employed through a simplified flat-rate contribution mechanism

We state the split plainly, and we state what changed. An earlier draft of this paper set a 10% employer minimum — 8% pension plus 2% care — and left the cost unpriced. That was a large promise resting on nothing, plausibly £25-45bn a year of new obligation on employers, and we have withdrawn it. The employer minimum rises from 3% to 5%: the pension element stays at 3%, and the additional 2 points are ring-fenced for care. With typical employee contributions of around 5%, total retirement-and-care saving reaches about 10% of qualifying pay. That is below Australia's 12% employer-only Superannuation Guarantee and we no longer claim otherwise. What it buys that Australia's system does not is the care side: a capped, pooled guarantee against catastrophic care costs, priced at £4.2bn a year of forgone Corporation Tax and carried in the scorecard.

Simplified Annual Allowance

With the ISA regime gone, the pension becomes Britain's single long-term savings vehicle — and a more generous one than any ISA. Where the ISA sheltered £20,000 a year of already-taxed money, the pension shelters £60,000, gives relief at your marginal rate on the way in, compounds tax-free, and carries no lifetime cap. For shorter-term saving, the £5,000 tax-free interest allowance and the £15,000 capital-gains allowance mean ordinary savers pay no tax on realistic returns without any wrapper at all. The job the ISA did is now done by a larger pension and allowances generous enough that the wrapper became redundant. The current pension relief system, though, is needlessly complex and creates perverse incentives:

• Annual contribution limit set at £60,000 — sufficient for serious retirement saving without being a tax shelter for the very wealthy
• Abolish the tapered annual allowance entirely — it is complicated, creates cliff edges, and discourages senior professionals from working
• No lifetime cap — the previous lifetime allowance was administratively burdensome and discouraged pension saving among those who most need long-term certainty
• One early access to the tax-free lump sum. The pension already lets you take 25% tax-free at retirement; under F³ you may take that 25% once, at any age, rather than only from 57 — turning the pension into the vehicle for major life capital, such as a first-home deposit, as well as retirement. It is once per lifetime, capped, and available only on funds held at least five years, so it cannot be used to wash income through the pension for instant relief. Taking it early naturally leaves a smaller pot to compound to retirement — a genuine choice with a real trade-off — but it gives every saver the flexibility the old wrapper-maze never did, inside a single account

Simple rules. Generous limits. No taper. Consistent treatment regardless of income level.

Chapter 14: Inheritance and Wealth

Ordinary family homes and working estates should not face large inheritance tax bills built over a lifetime of work. The state should not be a primary beneficiary of a family’s life savings.

Inheritance Tax Reform

• No inheritance tax on estates below £3 million — protecting the overwhelming majority of families including most working farms
• Estates taxed at the flat income-tax rate — 38%, falling to 35% when the income rate does at Step 4 — levied only on the excess above £3 million

The £3 million threshold is deliberately set to protect genuine family wealth without requiring complex carve-outs. The average working farm in England — 203 acres at £11,000 per acre plus farmhouse and buildings — sits at approximately £2.5–3 million. The threshold already protects most genuine working farms without needing Agricultural Property Relief or specific exemptions. Larger landholdings above £3 million, often held as investment assets rather than working farms, contribute proportionately at the flat rate — 38%, then 35%. One rate for income, the same rate for estates: when the statutory gate cuts it for the living at Step 4, it cuts it for legacies too. Both start and finish below today’s 40%.

Business Relief: Deferral, Not Exemption

Business Relief — the open-ended exemption for business assets — is abolished and replaced by the principle the National Wealth Floor uses: no exemption, no forced sale. Inheritance tax attributable to genuinely illiquid business assets is payable over ten years, interest-bearing at the gilt rate plus 2%, secured against the asset — so no family firm is ever broken up to pay a death bill, and no portfolio is ever assembled purely to dodge one. The £3m threshold already shields small firms entirely. The industry that marketed AIM portfolios as inheritance-tax products loses its product; the Growth Exchange gains a market held for growth instead of wrappers. And this base-broadening is what lets our +£4bn scoring stay conservative.

This is simpler, fairer and more honest than a system of complex reliefs that in practice benefit wealthy landowners and are lobbied for by those with the resources to exploit them.

Non-Domiciled Residents and Overseas Assets

The previous government’s decision to apply inheritance tax to the worldwide assets of long-term UK residents — regardless of where those assets are held — was a significant deterrent to wealthy individuals choosing to live, invest and spend in Britain.

Britain’s status as a global financial centre depends on attracting internationally mobile talent, entrepreneurs and investors. A policy that effectively taxes their entire global estate on death — including assets in countries where they also pay tax — is duplicative, punitive and self-defeating.

• Reverse the extension of IHT to overseas assets of non-domiciled long-term residents
• Restore the previous regime under which UK assets of non-doms are subject to IHT but overseas assets are not
• Maintain the £3 million threshold for UK-sited assets regardless of domicile status

This sends a clear signal that Britain welcomes internationally mobile wealth and talent — consistent with our broader platform of openness, growth and global ambition.

The National Wealth Floor

Everything else in this paper builds the environment in which wealth is created: courts that work, the Single Market restored, a free STEM pipeline, fast-track visas for talent, falling energy costs, CGT tapering to 18% for holdings of five years or more, no stamp duty on share trading, no taper traps. Wealth creation and accumulation are good — this platform exists to produce more of both. But when individuals hold more than entire nations while structuring their personal tax toward zero, the compact between wealth and the society that enables it has failed. The Floor repairs it: not a punishment for success, but a floor under contribution.

How It Works

The Floor is a hybrid, and the hybrid is the whole point. Wealth decides who is in scope; income decides what they must contribute. That combination defeats both failure modes — the French wealth tax that collapses under annual valuation, and the naive income rule that lets a billionaire with a modest salary slip through.

• The trigger is wealth: households with worldwide net wealth above £10 million (CPI-indexed) are in scope. Being in scope does not mean an annual wealth charge — it means one question is asked each year
• The test is income: an in-scope individual must pay UK tax equal to at least 25% of their economic income for the year. If the tax they already pay — income tax, capital gains tax, dividend tax — meets or exceeds that, nothing happens, and most will be unaffected. Only where actual tax falls below the floor is a top-up collected to reach it
• Economic income is defined to catch what structuring hides — the reason the genuinely rich so often pay so little. It counts realised income and gains in the ordinary way, distributions routed through personal companies and trusts, and high-value benefits in kind. It also reaches the ‘buy, borrow, die’ route — but not by pretending that a loan is income, because it is not. A loan is liquidity matched by a liability; it leaves net worth unchanged, and any rule that called it income would not survive first contact with an accountant. Instead, borrowing secured on an appreciated asset above the reporting threshold triggers a deemed part-disposal: a slice of the asset proportionate to the sum borrowed is treated as realised, the gain on that slice alone is taxed at the borrower’s marginal rate, and the base cost is adjusted so that the same gain is never taxed a second time when the asset is finally sold. What is taxed is the gain the borrower is actually living on, not the loan. It is a timing rule, not a new tax base — it brings forward tax that the current system allows to be deferred indefinitely and then wiped out at death. You cannot escape the Floor by taking your living as a loan against a share portfolio; equally, you do not pay tax on money you have merely borrowed
• Why the wealth gate at all? Because the Floor targets the genuinely wealthy, not a high earner having one good year. The £10 million test means a £400,000-salaried professional with no assets is never caught, while the £50 million asset-holder living tax-efficiently always is. Wealth says who is in scope; income says how much they must contribute
• Roughly 22,000 people are in scope. Below £10 million the Floor can never apply — by statute

Wealth, Exactly Defined — For Scope

Wealth no longer sets the charge; it sets who is in scope. That makes the whole exercise far lighter than a wealth tax — there is no annual pricing of every asset to levy a percentage on it, only a test of whether a household is clearly above £10 million. The detailed mechanism for that test (objective triggers and banded disclosure) is set out in the next section; what follows here is simply what counts as net wealth when the question is asked. Net wealth for scope means worldwide net assets, measured comprehensively so the line cannot be gamed:

• Cash, deposits and money-market holdings
• Listed securities and funds, at market value
• Unlisted and private company shares — at a provisional value (the most recent arm’s-length transaction, or a published assets-and-earnings formula), with a true-up at eventual sale so any under- or over-payment is corrected with interest. Valuation disputes are settled by reality, not by tribunal
• All real property, UK and overseas, including the main residence — at £10 million there is no sympathy case
• Pension wealth: defined-contribution pots at value, defined-benefit rights at transfer value
• Trust interests, looked through to settlors and beneficiaries; assets held in personal investment companies attributed pro-rata — the two classic escape routes, closed at the design stage
• Loans owed to the individual, crypto-assets at market value, and intellectual-property and royalty rights
• Art, collectibles, vehicles, vessels and aircraft — only individual items worth £250,000 or more count, so ordinary possessions are never in scope; a serious collection is valued as a collection
• Less all debts and liabilities: this is a net measure. Gifts to minor children are attributed to the parent; spouses are assessed individually, with anti-fragmentation rules

There are no exemptions from the scope test — not for business assets, not for pensions, not for homes — because a gap in the scope definition is how the wealthy would duck under the £10 million line. France exempted ‘business assets’ and watched everything become one. But the burden is lighter than a wealth tax: wealth now only decides scope, not the charge, so what an in-scope person pays is driven by their income, not by a forensic annual revaluation of their art and homes.

How Scope Is Established — Without Valuing Everyone

The obvious objection is administrative: does this mean HMRC must value every asset of every wealthy household every year? No — and the design is built specifically to avoid that, because annual universal valuation is exactly what sinks wealth taxes. Scope is established by objective triggers, banded disclosure, and challenge powers, not by a national audit.

Entry is flagged by data HMRC and other registries already hold. A household is asked to confirm its position only when it trips a visible high-wealth indicator — not before:

• UK property or listed equity holdings above £5 million (visible to the Land Registry and to HMRC)
• A disposal of shares, a business or other assets above £5 million in the year (HMRC already sees disposals)
• Dividend, capital-gains or partnership income above £1 million (already tax-reported)
• Loans of more than £2 million secured against assets — the reporting trigger for the deemed part-disposal that catches the ‘borrow against wealth’ route
• Control of trusts or companies holding substantial UK assets (the classic wrapper), or lifestyle spending above a high threshold as an anti-avoidance backstop
• Trust look-through: where an individual settled a trust, can benefit from it, or controls it, the trust’s assets and distributions are attributed to them for both the scope test and economic income — so wrapping wealth in a trust moves it out of sight of neither. Genuinely independent trusts with no retained benefit or control are treated on their own terms

These triggers sit below the £10 million line on purpose: they are detection flags, not the threshold. Tripping one does not mean you are in the Floor — it means HMRC asks one question, and you either demonstrate you are below £10 million and hear no more, or you enter the regime. The threshold remains £10 million; the triggers simply decide who is asked.

Disclosure is then proportionate to wealth, so no one argues over whether a fortune is £11.2m or £11.8m:

• £10m–£25m: self-certified net-worth position, with full HMRC challenge rights
• £25m–£100m: a formal asset statement every three years, not annually
• £100m+: annual enhanced disclosure

And valuation uses practical rules, not perfection: listed securities at market value on 5 April; property by Land Registry plus a professional valuation every three to five years, not yearly; private businesses by existing HMRC valuation methods with averaging allowed; art and collectibles only above a high per-item threshold; pension wealth counted for the scope test but never force-saleable to pay. The principle is that you do not need a perfect number for every asset — you need enough to identify, fairly and defensibly, who is clearly ultra-wealthy and may be paying very little tax. That is a far lighter task than a wealth tax, and it is why the Floor is administrable where annual wealth taxes are not.

Liquidity, Not Carve-Outs

• Contributions attributable to illiquid assets may be deferred until a liquidity event, accruing interest at the gilt rate plus 2%, secured by a charge over the asset. No founder is ever forced to sell their company to pay
• Payment in kind: contributors may settle in shares, transferred at market value to the British Future Fund — the state that built the environment takes its stake in the success it enabled
• In-kind receipts are revenue, not borrowing — the Future Fund’s surplus-only capitalisation rule is unaffected

The Exit Lock

Norway learned in 2022 that a wealth contribution without an exit charge is a relocation incentive — and fixed it only after the departure lounge had emptied. We adopt the fix before launch, not after:

• Leaving the UK triggers a deemed disposal: CGT on unrealised gains at departure, with a deferral election for moves within the EEA — the design settled European case law requires, keeping the Floor fully Single Market-compatible
• The floor itself trails after departure, on wealth accumulated while UK-resident — at a taper, not a cliff. A departing contributor who was fully within the Floor pays the full floor in their first year away, then four-fifths, three-fifths, two-fifths and one-fifth across the following four years, reaching nil after five. Leaving is a gradual exit from the obligation, not an overnight escape from it
• Outside the EEA, the exit charge is drafted to survive Britain’s double-tax treaties — because several of them would otherwise override a domestic exit charge on emigration. Most UK treaties follow the OECD model, which leaves gains on emigration to the departing state in defined cases, but some allocate taxing rights in ways that would blunt the charge. We will legislate the exit charge as a tax on gains accrued during UK residence, crystallised at departure — the form treaty practice already recognises (Canada, Australia and the United States all operate departure or expatriation charges alongside their treaty networks) — and, where a specific treaty would still frustrate it, that treaty will be renegotiated or notified for the limited change required, on the same timetable as accession. We name this rather than discover it in court: an exit charge that any competent adviser could treaty-shop around would not be worth legislating
• You are free to leave. Your dues are not

The Ten-Year Runway

The Floor applies from the tenth year of UK residence. Someone arriving under our talent programmes gets a decade before the floor reaches them — consistent with the restored non-dom architecture, and a materially better offer to arriving wealth than the regime Britain operates today. We are not ambivalent: we want the world’s wealth creators here. The Floor is designed so that coming remains rational and staying remains fair.

Three clarifications, because clever advisers will ask. First, the ten years are cumulative across a lifetime, not consecutive — leaving in year nine and returning later does not restart the clock. Second, the exit charge — deemed-disposal CGT on gains accrued while resident — applies to every leaver from day one, runway or not: everyone settles their capital gains on the way out. Third, and this closes the obvious gambit, the runway itself tapers rather than ending in a cliff. We considered letting the year-nine leaver walk owing no Credit at all — the simplest design — but it hands the wealthiest a clean exit precisely one year before the obligation begins, and rewards the adviser who books a departure in month one hundred and seven. So entry into the Floor phases in across the final years of the runway: someone who leaves in year nine pays a proportionate share reflecting how close they came to full residence, not zero. The honest principle is symmetry — the Floor fades in as residence lengthens and fades out as departure recedes, with no single date on which a fortune can step across the line untouched. Britain charges for the harvest, even where it never charged for the field — and it does not leave a one-day gate open in the fence.

Honest Revenue

The yield is the top-up paid by in-scope individuals whose effective tax rate falls below 25% of economic income — not a percentage of wealth. Those who already pay their share contribute nothing extra; the revenue comes from those structuring below the floor, brought up to it. We score the National Wealth Floor at £5bn a year from Year 2 (£3bn in our LOW scenario, £8bn in HIGH) — deliberately a fraction of what campaign groups claim for wealth taxes, because this manifesto’s credibility rests on not doing fantasy yields. This figure is the one most affected by the redesign and should be re-derived by independent costing under the economic-income basis; we publish it as a conservative placeholder, not a precise forecast. Year 1 is registration and scope infrastructure; revenue begins at Step 2.

And the name is the mechanism: it is called a Credit because it works as one. Those who pay, pay nothing more. Those who structure to zero meet, for the first time in British history, a floor.

Chapter 15: Financial Reform and Capital Markets

The Scale of the Problem

The numbers tell an unambiguous story of decline. In 2024, 88 companies delisted or transferred their primary listing from London — the largest exodus since the financial crisis. Proceeds raised from London IPOs fell 64% in the first half of 2025. AstraZeneca and Wise are among the companies exploring US listings. Deal numbers fell roughly 80% between 2022 and 2025.

The causes are structural and largely self-inflicted: stamp duty on share trading that no other major financial centre imposes, pension fund domestic equity allocations that have collapsed from over 50% in the 1990s to below 5% today, listing rules that have been too rigid for founder-led technology companies, and a decade of political and economic uncertainty that has depressed valuations relative to the US.

Abolish Stamp Duty on Share Trading

Stamp duty on shares is, as City leaders have called it, a self-inflicted wound. No other major financial centre taxes equity investment this way. It suppresses trading volumes, depresses share valuations, drives institutional investors into derivatives to avoid the charge — reducing transparency and compounding the impression of poor liquidity — and actively discourages companies from listing in London.

The economic case for abolition is overwhelming. Independent modelling shows abolition would:

• Permanently increase UK GDP by 0.2–0.7%
• Trigger a one-off 4% uplift in UK equity valuations — worth approximately £100bn in additional wealth
• Leave the average defined contribution pension pot more than £6,000 larger over a career
• Be revenue-positive in the medium term — the growth effects generate more tax than the £3.3bn annual stamp duty take

We will abolish stamp duty on UK share transactions outright — a charge that raises around £3-4bn a year but does so by suppressing the trading, listing and liquidity that make London a global financial centre. There is no carve-out and no exception: every ordinary share transaction in Britain becomes free of transaction tax. In its place, on buybacks only, we introduce a 1% charge on net repurchases — shares a company buys back and cancels, less new equity it issues in the same year. This is deliberately not a surviving fragment of stamp duty. It is a separate, named charge on a distinct act, and structuring it that way keeps the abolition clean rather than leaving an asterisk on it. The model is American: the United States has levied a 1% excise on net repurchases since 2022, and Canada operates a 2% version, so the rate we have chosen is the more modest of the two established precedents. The case is twofold. Ordinary trading moves a share from one owner to another and deepens the market, which is what we want more of; a buyback permanently removes the security from circulation, shrinking the very transaction base the abolition is meant to widen. And a company with surplus cash chooses between a dividend and a buyback — we prefer the dividend, because it puts money into savers' and pension funds' hands rather than concentrating ownership and flattering earnings per share. Netting new issuance matters: a firm raising equity to invest and simultaneously tidying its share count is not the target, and the American design excludes it for that reason.

On the costing we make one exception to a rule we apply everywhere else, and it needs stating rather than hiding in an annex. Throughout this paper we decline to book behavioural gains: vaping duty is nil-booked, Channel 4 raises nothing, the health saving from better social care is left unclaimed, and the Road Charge carries a 2% haircut. Here we do net part of the behavioural recovery. The static cost of abolition is £3-4bn; we book £1.2bn, which is £1.5bn for the abolition less the £0.3bn buyback charge, and the £1.5bn is net of partial volume and valuation recovery rather than the full static figure. The reason is specific to this measure. The entire case for abolishing stamp duty on shares IS behavioural — deeper markets, more listings, higher valuations. If the behavioural effect is not real, the policy should not be done at all. Booking the full static cost would therefore mean pricing a measure on the assumption that its own rationale is false, which is a different kind of dishonesty from optimism. We think roughly half the static cost is recovered over the medium term and we book the position on that basis. A reader who disagrees can restate the line at the full £3.2bn: it costs £2bn, and it moves the central case from essentially level with the official forecast to about £1bn above it. We would rather show that arithmetic than bury the assumption.

Reviving the IPO Market

London’s listing rules have been reformed — but not enough and not fast enough. Companies, particularly in technology, have consistently chosen New York over London because of valuation gaps, investor depth and structural flexibility. We will go further:

• Dual class share structures — formally embrace and simplify the rules allowing founder-led companies to maintain control post-IPO, as New York has done for decades
• Retail investor participation — make it structurally easier for retail investors to participate in IPOs at the same terms as institutions, deepening the domestic investor base
• A British Growth Exchange — a dedicated market for high-growth companies, designed around the four things AIM never had. Full design below
• Secondary listings — actively market London as the premier venue for international companies seeking a secondary listing, with streamlined dual-listing rules
• FTSE index reform — review the criteria that determine FTSE inclusion to ensure the index better reflects the growth companies of the future, not just the incumbents of the past

Pension Funds and UK Equity

The collapse of pension fund allocation to UK equities is perhaps the single biggest structural driver of London’s decline. When domestic institutions owned 50% of UK shares, London had deep, liquid markets and strong valuations. At below 5%, the market is structurally disadvantaged.

Mandating pension fund allocations to UK equity is too blunt and potentially harmful to savers' returns. The primary lever is stamp duty abolition — which immediately reduces the cost disadvantage of UK equity and triggers the £100bn valuation uplift that makes UK equities structurally more attractive without regulatory compulsion. Beyond that:

• Stamp duty abolition on share trading is the primary lever — the single most impactful measure for UK equity markets, revenue-positive in the medium term, and creating the structural demand that a dedicated savings account was intended to achieve. The 1% buyback charge pulls the same way: it favours returning cash as dividends over shrinking the share count, and it leaves everyday liquidity untouched
• The ISA regime is abolished outright — cash, stocks and shares, Lifetime, Junior and Help-to-Buy alike — and nothing replaces it as a wrapper. This is a change of position from an earlier draft of this paper, which kept a shares-and-funds ISA and abolished only the cash version; that draft sat inconsistently alongside the abolition set out in Chapter 1 and in the scorecard, and the abolition is the policy. Savers lose the account and keep the benefit: the £5,000 tax-free savings allowance shelters cash interest and the £15,000 capital-gains allowance shelters investment gains, both universal, both with nothing to open. Existing balances are protected in full and continue to grow tax-free; what ends is new subscriptions. The job the wrapper was meant to do — channel long-term capital into productive assets — is done here by abolishing stamp duty on shares, which changes the return on UK equity itself rather than the container it is held in

The £5,000 Savings Allowance: The Rules, In Full

An allowance this simple still needs its edges defined, and a paper that complains about complexity elsewhere should not leave the detail to be guessed at. The rules are these.

• It is an individual allowance and it applies to children as well as adults, because children have always had their own personal allowance and we are not creating a separate regime for them. The existing settlements rule is retained: interest on money provided by a parent, above a small de minimis, remains taxable on the parent. Without that rule a family of four would shelter £20,000 of interest — around £533,000 of cash at present rates — and the allowance would become an inheritance-planning device rather than a savings one
• Interest on joint accounts is split equally between the holders, which is the existing default and needs no new rule. A married couple therefore shelters £10,000 of interest between them
• It covers all savings income as already defined in statute: bank and building society interest, corporate bond and gilt coupons, peer-to-peer lending interest, and the interest element of purchased life annuities. We are not carving out asset classes, because carve-outs are how the ISA maze began. One consequence should be stated rather than discovered: gilt capital gains are already exempt from Capital Gains Tax, so gilts held by individuals become very lightly taxed. We think that is acceptable for sovereign debt and we say so rather than pretend not to have noticed
• It covers overseas interest, with foreign tax credit relief as now. UK residents are taxed on worldwide income, and exempting offshore interest would open exactly the arbitrage our capital gains chapter exists to close
• It is NOT transferable between spouses. The £15,000 income tax allowance is transferable between parents; this one is not. The joint-account rule already lets a couple use both allowances, and transferability here would add cost and complexity for no gain. One transfer in the system is enough
• It is additional to the £15,000 income threshold and is not tapered away as income rises. A saver with earnings can therefore shelter £15,000 of income and £5,000 of interest — £20,000 in total. We should be candid that this is more generous than the present system, in which the £5,000 starting rate for savings tapers to nothing as other income rises above £17,570 and so reaches only lower-income savers. Universalising it is the single largest driver of this measure's cost, and it is a deliberate simplification rather than a like-for-like replacement of the £1,000 personal savings allowance
• Interest above £5,000 is charged at the full combined rate — the national flat rate plus the 2% Health Levy plus the 2% Local Income Tax, so 42% falling to 39% when the headline rate glides to 35%. Savings income is not privileged relative to earned income above the allowance. Excluding it from the levies would recreate precisely the distinction between types of income that this design exists to remove

One caveat on the cost, and it is a real one. The £4.8bn in the scorecard was set so that the savings package as a whole lands at the cautious end of what the evidence supports. That makes it a fiscal target rather than a costing, and we would rather say so than let it pass as derived. It has not been rebuilt from the rules above, and four of them push the cost up — children, the breadth of qualifying income, overseas interest and the absence of a taper — while one pushes it down, since interest above the allowance carries both levies. Presented as a range rather than a point: on a low case the allowance costs about £3bn, on a central case about £4.8bn, and on a high case about £6bn, which is the ceiling set by the fact that only ~1.9 million people pay any tax on savings interest at all, on roughly £6bn of tax. The package nets as follows, and we set it out explicitly because otherwise it can look like double counting: abolishing the ISA regime recovers relief on new subscriptions, building from £1bn in Year 1 to £5bn by Year 5; abolishing the £1,000 Personal Savings Allowance raises £0.8bn; and the new £5,000 allowance costs £4.8bn. The £0.8bn is the offset against the £4.8bn gross, not a separate receipt. Net, the package raises about £1bn a year by Year 5. Settling the range needs HMRC's distribution of savings income by amount, which is not published at the necessary detail, so the central figure is marked for re-derivation and subject to independent verification. We would rather publish firm rules and a declared range than a confident point estimate and no rules.

• Solvency and investment rules — review the regulatory framework that has pushed pension funds into bonds and away from equities, removing the regulatory bias against long-term UK equity investment
• Sovereign wealth participation — the British Future Fund (below): headroom-funded, never gilt-funded, a patient cornerstone for UK equities
• Pension fund reporting — funds above a certain size required to explain publicly why they hold less than 10% in UK equities, creating accountability without mandating allocation

Single Market note: F³ originally proposed a British ISA restricted to UK-listed equities. This has been removed as incompatible with EEA free movement of capital rules — and, as it turned out, made redundant by a better idea. Rather than steer capital with a wrapper, F³ abolishes the ISA regime entirely and gives everyone a £15,000 annual capital-gains allowance and a £5,000 tax-free savings allowance, with no account to open. That achieves the same structural goal — more capital into productive assets — more powerfully, more simply, and without legal conflict. The contrast with the current approach is deliberate: where today's reform cuts the cash ISA limit and threatens a punitive charge on cash held in share accounts, F³ uses the carrot — generous universal allowances — not the stick.

The British Growth Exchange: A Stock Exchange, Not a Tax Wrapper

AIM’s decline — from 1,700 companies to under 700 — is not a branding problem. Institutions will not hold small-caps locked out of the index series; abundant private capital lets companies stay unlisted longer; and much of what remained was held for inheritance-tax relief rather than growth: an estate-planning product with a ticker attached. The Growth Exchange is built on the four things AIM never had:

• Index inclusion from day one — BGX constituents enter a FTSE Growth index family with automatic graduation to the FTSE 250 and FTSE 100 as they scale. Tracker demand follows the index; AIM’s exclusion was its liquidity death sentence
• An anchor investor — the British Future Fund may cornerstone BGX floats and warehouses the shares the state receives through National Wealth Floor payment-in-kind: a patient demand floor no junior market has ever had
• The pension channel — BGX holdings count toward the 10% UK-equity explain-or-comply reporting rule for large schemes
• One rulebook — a single proportionate FCA regime replaces the Nomad system and its £400,000-a-year quote-maintenance overhead

How It Fits the Tax System

• EIS, SEIS and VCT continue — and a BGX listing counts as 'unquoted' for their purposes, exactly as AIM does today, so floating never triggers a relief clawback. The ladder runs SEIS, to EIS, to BGX, to the main market — with no tax cliff at any rung
• CGT holding periods carry through the IPO — a founder’s or early backer’s taper clock does not reset at listing. Five patient years means 18%, uncapped: Business Asset Disposal Relief without the £1m ceiling, available to every long-term shareholder
• No stamp duty on trading — as on every ordinary share transaction in Britain under F³; a separate 1% charge applies only where a company buys back and cancels its own stock
• And no inheritance-tax wrapper: Business Relief becomes a ten-year deferral, not an exemption (Chapter 14). The Growth Exchange will be held for growth, because growth is all it offers

The British Future Fund: An Investment Trust the Nation Owns

In structure, the Fund is an investment trust with a single shareholder — the Treasury: closed-end, permanent capital, an independent board on staggered statutory terms, professional managers on published mandates, quarterly NAV, audited by the National Audit Office, holdings register public. The governance models are Temasek and Norway’s oil fund — not a quango, and not a minister’s chequebook.

• Capital from four sources only: certified fiscal headroom released under the glide path once debt targets are met — headroom, not surplus, since the plan does not reach surplus within this parliament — shares received through National Wealth Floor payment-in-kind, tax receipts from new North Sea oil and gas production (Chapter 10 — a finite national asset saved rather than spent, on the Norwegian principle), and its own reinvested returns. Never gilt issuance — the state does not borrow to buy equities
• Mandate: long-term total return for the nation, with authority to cornerstone British Growth Exchange floats — capped at 10% of any company, votes cast on a published passive policy. A patient shareholder, never a controlling one
• Anti-raid protection: the mandate sits in primary legislation; no distributions for the first decade, and thereafter only under a published spending rule once debt is on its statutory path
• Honest scale, honestly distinguished: the British Business Bank already deploys an economic capacity of around £25.6bn — that existing balance sheet is the foundation the Fund builds on, not a number we are inventing. What is genuinely new is the additional patient capital the Fund accretes on top: an acorn by design, perhaps £3–8bn by Year 5 from in-kind receipts and released headroom, growing if and as genuine surpluses emerge in the 2030s. We separate the two deliberately, so no one mistakes the inherited base for new money or the new money for the whole
• Not a new quango — the British Business Bank, reconstituted. We are clear-eyed that Britain already has a state development bank: the British Business Bank, whose capacity was raised to £25.6bn in 2025 and which now makes direct equity investments in strategic scale-ups. We do not propose a rival to it. The British Future Fund is that institution given the one thing the analysts who study it say it lacks — genuine autonomy and a commercial mandate. The diagnosis is shared: as the Bank’s own leadership puts it, Britain risks “remaining strong at company creation but weak at retention and scale.” But a development bank bound by civil-service pay scales and Whitehall decision cycles cannot move at the speed of the markets it invests in, nor attract the calibre of managers the best funds demand. We reconstitute it as an arm’s-length investment trust — Temasek-style governance, professional managers on published mandates, freed from the pay and process constraints — so that public capital is deployed with the velocity and credibility a sovereign investor needs. Same money, same mission; a structure built to actually deliver it

What it is not: leveraged, redeemable, or available for bailouts. An investment trust cannot be run on — and this one cannot be raided.

Invent Here, Own Here: The Commercialisation We Keep Losing

The Future Fund and the Growth Exchange are answers to a failure this country has grown shamefully used to, and it is worth naming it plainly, because it may be the most expensive habit in British economic life. We are brilliant at inventing things and hopeless at owning what we invent. Graphene was isolated at Manchester in 2004 and won a Nobel Prize two years later; the patents and the manufacturing base that grew from it are now overwhelmingly held overseas, with China alone accounting for the lion’s share. ARM, the crown jewel of British computing, whose chip designs sit in almost every smartphone on earth, was sold to SoftBank in 2016 and listed in New York rather than London. DeepMind, the most important artificial-intelligence company Britain has produced, was sold to Google before it was a decade old. The pattern is always the same: British science makes the breakthrough, and then the value — the jobs, the tax base, the strategic control, the compounding wealth of the decades that follow — is captured somewhere else. We hold the ideas long enough to be proud of them and not a moment longer. As the leadership of our own state development bank puts it, Britain is strong at company creation but weak at retention and scale.

This is not an accident of national character, and it is not mainly a failure of invention or even of early-stage funding, where Britain does comparatively well: a British start-up raises around three-quarters of its early money at home. It is a failure that strikes at a precise point in a company’s life — the scale-up stage, the stretch between a promising young company and a global one. That is where British capital thins out and all but disappears. By the scale-up rounds, the domestic share of funding falls to less than a third, and the founder who wants to keep growing has little choice but to take money from American or Asian funds — who, reasonably enough, expect the company, its headquarters, its listing and eventually its ownership to migrate toward them. The numbers are stark: British firms raised some £16bn in venture funding in a recent year against roughly £65bn in Silicon Valley, and the proportion of seed-backed companies surviving to their next major round has collapsed. The cause sits in plain sight. Britain’s pension funds hold around £3 trillion, among the deepest pools of capital in the world, and they invest strikingly little of it in British growth companies — sending it abroad, and with it the returns that ought to accrue to British savers and the businesses that ought to have been built here.

We accept that leaving the European single market made this harder, not easier. A company scaling in Britain today does so with more friction in reaching its nearest large market, and with some of the European investment channels that once fed it now narrowed — which is one more reason this paper rejoins that market (Chapter 4). But Brexit is an aggravating factor, not the origin; the retention gap predates it and would outlast it if left unaddressed. Other countries have faced the same pull and refused to be passive about it. Germany pairs targeted state investment in strategic technology with a foreign-investment law that requires scrutiny of acquisitions above a quarter of a strategic company’s shares. France has mobilised state and institutional capital behind its own champions in artificial intelligence and deep technology. The United States wins less through clever policy than through the sheer depth of its domestic capital and the pension money that flows into it. China directs sovereign funds explicitly at its industrial strategy. Britain has stood almost alone in treating the sale and departure of its best companies as a compliment rather than a loss.

Much of our answer is already in this chapter — it is what the Future Fund and the Growth Exchange are for, and it is why we are so insistent about turning some of that £3 trillion of pension capital toward British growth. But retention needs one thing more than capital, and we add it here:

• A national-interest test for strategic intellectual property. Where a company built on genuinely strategic, often publicly-funded British innovation — the frontier science, the sovereign-capability technologies — is to be sold to an overseas buyer, that sale should face a national-interest assessment, on the model Germany has operated for years. We are precise, because this cuts against instincts we otherwise hold: this is not blanket protectionism, and it is not a licence to block ordinary foreign investment, which Britain needs and welcomes. It is a narrow test for the crown jewels — the handful of companies whose technology is genuinely strategic and often built on public research — asking whether the country should retain a stake, a licence, or the manufacturing before the whole thing departs. We are equally honest that a screen has a cost: it can, at the margin, make Britain a slightly less frictionless place to sell a company, and some investment is deterred by any such test. We judge that a proportionate price for no longer handing away our most important technologies by default
• Commercialisation built into public research from the start. When the state funds the science — through the universities, the research councils, the sustainability fund of Chapter 6 — it should fund the path to a British company, not just the discovery that a foreign one commercialises. That means spin-out terms that let founders and universities build here rather than sell early, procurement that makes government the first customer for home-grown innovation (Chapter 3), and the patient scale-up capital above waiting when they grow. The public paid for the invention; the public should share in the ownership

Nowhere is the choice clearer, or more current, than nuclear fusion. Britain is, genuinely, among the world leaders: the national fusion programme at Culham, the STEP prototype-powerplant effort, and decades of accumulated expertise put us at the front of a field that could eventually reshape energy entirely. We are deliberately honest about the timescale — commercial fusion power is not years away but likely decades, and this paper does not cost a single kilowatt-hour of it or lean on it in any energy projection. That is precisely the point. Fusion is not, yet, an energy policy. It is the perfect test of whether Britain has learned anything: a world-leading, publicly-funded, strategically vital technology, arriving slowly enough that we can see the decision coming. We can do what we did with graphene — make the breakthroughs, publish the papers, and watch other countries build the industry and hold the patents. Or we can decide, this time, to coordinate the public and private effort behind our lead, keep the intellectual property and the supply chain anchored here, and own a share of the industry we are helping to invent. A fusion strategy worth the name is not a promise of limitless power by a certain date; it is a commitment that if the prize is real, this time Britain will still own a piece of it.

The Taxpayer Upside Rule

There is a pattern in British industrial policy that this chapter has already complained about twice, and it is worth naming as a rule rather than a grievance. The state routinely absorbs genuine commercial risk — grants for gigafactories, guarantees for reactors, subsidies for vaccine plants and chip fabs — and then retains none of the upside when the bet comes good. The losses are socialised and the gains are not. We have applied the opposite principle in two places already: NHS data, where the public takes a royalty or an equity share where its data demonstrably enables a product (Chapter 5), and the British Future Fund, which holds assets rather than handing them over. The Taxpayer Upside Rule generalises it.

Any bespoke grant, guarantee or subsidy to a commercial company above £5m must carry one of the following, chosen to fit the sector and published with the award:

• A capped royalty on revenues attributable to the supported activity
• Non-voting warrants, so the public holds an option on success without directing the company
• A revenue share triggered once the project reaches an agreed threshold
• Preferential public-sector pricing or guaranteed early access
• Repayment of the support once an agreed profitability threshold is crossed

The scope is deliberately narrow, because a rule that catches everything catches nothing. It applies to bespoke, negotiated support for individual commercial companies — the large energy, pharmaceutical, technology and industrial awards that are decided one at a time in a minister's office. It does not apply to ordinary R&D tax relief, to universities, to charities, or to standardised small-business support, where the administrative cost would exceed any return and where the public benefit is the point.

Two limits we impose on ourselves. The public takes an option on success, not control: warrants are non-voting, and nothing here gives a department a seat in a boardroom or a veto over a commercial decision. And we book nothing. No anticipated equity gains, no assumed asset-sale proceeds, no notional portfolio value enters the fiscal framework. Only dividends, royalties and repayments actually received in cash become revenue, and they are scored in the year they arrive. The rule is therefore in the scorecard at zero, with the reason on the line. We would rather add a discipline that costs nothing and might return something than claim a number for money the state has not yet been paid.

Capital Gains Tax: A Fair and Coherent Framework

Britain’s CGT system is a patchwork of rates, reliefs, exemptions and cliff edges that has been repeatedly tinkered with for political rather than economic reasons. The current system is neither fair nor growth-promoting. We will replace it with a coherent framework built around three principles: alignment with income tax, protection from inflation, and reward for long-term patient capital.

Rate Alignment

CGT is aligned with the prevailing flat income tax rate — 38%, falling to 35% at Step 4 — so the top of the holding-period taper always equals the income rate of the day. The distortion between income and capital gains rates creates perverse incentives — to disguise income as capital, to structure remuneration artificially, and to defer sales for tax rather than economic reasons. Alignment ends them: the shortest-held gains are taxed exactly as income, and only genuine long-term holding earns the taper below.

Dividends follow the same principle, with one deliberate adjustment. Dividend income is taxed at the flat rate — 38%, then 35% — but with a fixed credit for the corporation tax already paid at 30%, so the same profit is not fully taxed twice. The combined burden on distributed company profit therefore lands close to the rate on labour income, removing the incentive to dress salary as dividend that a naked rate gap would create. The £500 dividend allowance is retained and CPI-indexed, sparing small shareholders from filing over trivial sums.

Indexation Relief

Gains attributable to inflation will not be taxed — only real gains. If an asset bought for £100,000 is sold for £120,000 after inflation has risen 15%, the taxable gain is £5,000 not £20,000. This is the crucial protection for genuine risk-takers and long-term investors — you pay tax only on wealth genuinely created, not on monetary illusion. Indexation will be calculated using CPI from the date of acquisition.

Holding Period Taper

Patient capital — investment held for years or decades — creates more genuine economic value than short-term trading. The tax system should reflect this. We will introduce a holding period taper on the effective CGT rate after indexation:

An entrepreneur who builds a business over 15 years and sells it pays 10% CGT on real gains — after inflation is stripped out. A short-term trader pays the full income rate of the day. This is the right incentive structure: it rewards genuine wealth creation and long-term risk-taking while ensuring the state benefits proportionately from shorter-term gains.

The obvious objection is that a low terminal rate reopens the very arbitrage the alignment was built to close: convert income into capital, hold, and pay less than a salary pays. We take that objection seriously enough to have moved the number. An earlier draft set the terminal rate at 10% after ten years, which left a thirty-two point gap to the 42% combined rate on earnings — a two-thirds discount, and indefensible as a permanent feature. The terminal rate is 18% after five years, and the gap is 24 points rather than 32. Three things then close what remains. The taper runs on real gains only, after indexation, so inflation is stripped out before the rate applies — and because indexation already answers the 'a decade's gain taxed as a year's' complaint, the taper is doing a second and narrower job, which is why it does not need to fall as far as 10% to be justified. The clock cannot be reset by reorganisation. And the National Wealth Floor sits above the whole structure for those who arrange to have almost no taxable income at all. We also chose 18% rather than a rounder number deliberately: it is the rate the United Kingdom itself settled on for capital gains in April 2008, when taper relief was abolished, so it is neither invented nor untested.

Annual Exempt Amount

The annual CGT exempt amount has been salami-sliced from £12,300 to £3,000 — a stealth tax increase that has swept ordinary investors and small savers into CGT liability. We will restore it to £15,000 and index-link it to CPI. The figure is deliberately the same £15,000 that begins the income tax scale and the same £15,000 that is transferable between parents: one number does three jobs, and simplification is the argument for it as much as generosity is. An earlier draft set it at £20,000; £15,000 raises about £0.7bn more, on roughly 350,000-400,000 CGT payers, and we would rather have the coherence and the £0.7bn than the larger headline.

The Holding-Period Taper — Bands Defined

• Under 2 years: the full income rate — 38%, then 35% at Step 4. Speculation pays exactly what earned income pays
• 2 to 5 years: 25%
• 5 to 10 years: 17.5%
• 10 years and beyond: 10% — uncapped, for every shareholder, with no employment condition attached
• Clocks follow the shares: holding periods carry through IPOs, reconstructions and share-for-share exchanges — and through being fired. The taper rewards patient ownership, not job titles

Worked Example: The Founder Forced Out

Seven years in, two funding rounds, a £70m company — and a 20% founder forced out and forced to sell. Proceeds of £14m against a near-nil base cost. Seven years puts the gain in the terminal band at 18%: tax of roughly £2.52m, against about £3.3m under today's BADR-plus-24% system. Note what the schedule no longer does: an earlier draft rewarded reaching year ten with a 10% rate, so this founder would have paid £1.4m had the coup come three years later. That was a £1.1m swing turning on the timing of someone else's boardroom decision, and we have removed it — the taper now terminates at five years, so patience is rewarded up to the point where it is genuinely earned and not beyond. Two design choices matter here. First, the taper attaches to the shares, not the job: today's reliefs carry employment conditions that a boardroom coup can poison; ours cannot be weaponised by the people removing you. Second, the National Wealth Floor does not quietly claw the taper back.

Business Asset Disposal Relief

The existing Business Asset Disposal Relief — which charges 18% on the first £1m of qualifying business gains — will be replaced by the holding period taper, which is more generous for long-term entrepreneurs and simpler to administer. The artificial £1m lifetime cap disappears; what matters is how long you built the business, not an arbitrary ceiling.

Financial Regulation

London’s regulatory framework — the FCA, the PRA, the Bank of England — is broadly sound. The instinct to regulate proportionately and principles-based rather than rules-based has served Britain well. We will maintain this approach and resist the temptation to follow the EU into ever-more-prescriptive financial regulation.

• FCA mandate explicitly includes competitiveness and growth as primary objectives alongside consumer protection — as recently legislated but insufficiently operationalised
• Fintech and digital assets — maintain Britain’s position as the leading global centre for responsible fintech innovation with a clear, predictable regulatory sandbox approach
• Green finance — support London’s position as the world’s leading green finance centre without mandating disclosures that add cost without adding information value

Chapter 16: Arts, Culture and Tourism

Film and Television

The UK’s film and television tax relief framework — now the Audio Visual Expenditure Credit at 34% for film and high-end TV — has made Britain one of the world’s leading production destinations. Pinewood, Shepperton, Crown Works: the infrastructure is world-class. We will maintain and strengthen this framework.

• Maintain the AVEC at current rates — providing long-term certainty for inward investment decisions
• Extend enhanced relief for independent British films — supporting domestic storytelling, not just international productions using British facilities
• Fast-track planning for studio expansion — removing barriers to new studio development outside London and the South East
• Restore VAT-free shopping for international visitors — reversing a post-Brexit own goal that has cost London and major cities billions in lost high-value tourist spending

Music

Britain produces world-class music — from classical to grime, from the Beatles to Adele. Yet the music industry receives far less fiscal support than film and television, despite comparable cultural and economic impact. We will address this imbalance.

• Introduce a Music Production Tax Credit — equivalent to the AVEC for recorded music production, incentivising artists to record in Britain
• Music venue relief — extending business rate abolition with enhanced support for grassroots music venues, the pipeline through which all major artists develop
• Export support for British music — targeted support for breaking British artists in international markets
• Protect and strengthen music education in schools — the pipeline from which British musical talent flows

Channel 4 and the British Future Fund

Channel 4 was created as a publisher-broadcaster — commissioning content from independent producers rather than making it in-house. That model remains valuable, and we intend to keep it.

Earlier drafts of this paper proposed privatising Channel 4 with conditions. We have changed our position. Ownership will instead transfer to the British Future Fund, which is where this paper sends public assets generally — so the publisher-broadcaster model is preserved, the independent production sector keeps its commissioner, and the privatisation fight is not fought for a prize that turns out to be small. We should be plain about the fiscal effect: it is nil, and we book nothing. Channel 4 receives no licence fee, is commercially funded and runs close to break-even, so there is no revenue stream to capture and no sale proceeds to spend. Nor can the Fund borrow against it and call the proceeds income — that is the borrowing-is-income error this paper rejects everywhere else, and we are not going to make it here.

BBC World Service and Soft Power

The BBC World Service is one of Britain’s most powerful soft power assets — trusted by hundreds of millions of people in countries where free media does not exist. At a time of growing global information warfare, cutting it is a strategic own goal.

• Restore and expand BBC World Service foreign language broadcasting — with particular focus on Russia, China, the Middle East and Africa
• Fund the expansion through the Foreign, Commonwealth and Development Office budget — treating it as the strategic communications asset it is, not a cultural nicety
• Invest in digital distribution — ensuring World Service content reaches audiences through platforms that work in restricted media environments

Tourism

• Restore VAT-free shopping for international tourists — already noted above but worth emphasising: this costs little and generates significant high-value visitor spending
• Support hospitality and cultural industries — removing regulatory burdens that price out small venues and independent operators
• Revitalise town centres through free parking, business rate abolition and planning reform — the same policies that support housing and growth also support culture

Chapter 17: Technology and Artificial Intelligence

The Opportunity

Consultancy estimates put the potential prize as high as £550bn added to UK GDP by 2035 — a headline figure we treat with appropriate caution, since such projections are inherently uncertain and tend to assume flawless adoption. But the direction is not in doubt, and the foundations are real: British AI firms raised £4.7bn in investment in 2025 alone, and we have DeepMind, Arm, a world-class university research base, the English language, the deepest financial markets in Europe, and a fintech and life sciences ecosystem that is genuinely world-leading.

What Britain lacks is not talent, not ideas, and not capital. It lacks two things: the scale-up infrastructure to turn world-class research into world-class companies, and — increasingly — the power to run it. This second constraint is now the binding one. Nearly £10bn of data centres were approved in 2025 but less than £1bn were actually built; OpenAI’s Stargate UK, intended as the centrepiece of the North East AI Growth Zone, was paused partly over energy costs. Around 140 proposed data centre projects are now queuing for roughly 50GW of grid capacity — more than Britain’s entire peak electricity demand of about 45GW — with some connection waits stretching to fifteen years. The lesson is blunt: in an AI economy, power is the new planning permission. A serious AI strategy is therefore, before anything else, an energy and grid strategy.

The F³ AI and Tech Platform

• Power first: treat AI compute as a national strategic priority and build the electricity to run it. The 6GW of AI-capable data centre capacity the government expects by 2030 would absorb the output of roughly four large nuclear reactors. We do not pretend that capacity appears by wishing for it: it is powered by the nuclear, SMR, geothermal and gas-bridge build-out in Chapter 10, and by fixing the grid bottlenecks set out there. The honest corollary, which we state plainly: prioritising compute strains the Clean Power 2030 timeline and competes for grid capacity with housing, heat and transport electrification — new housing in west London has already been stalled by data centres taking the available connection. We make that choice deliberately, and we carry it by building generation rather than by rationing it away
• Grid connection, not just planning consent, is the real gate. Data centres are already designated Critical National Infrastructure and planning is rarely the true blocker — the multi-year connection queue is. We will give strategically important compute — AI Growth Zones and sovereign-capability projects — priority access to grid connections, paired with a hard condition: large data centres must fund their own high-voltage connections and must be demand-flexible, able to curtail or shift load when the system is stressed. This is not a giveaway. National Grid trials have already cut data centre demand by a third within seconds without disrupting critical work; a flexible data centre is a grid asset that spreads fixed network costs over more hours, rather than a free rider that pushes up everyone else’s bills through higher network charges
• Back it with British capital, not just hope. Expand the AI Growth Zones programme beyond Wales to create clusters in every major British city — and, crucially, point our own investment machinery at the scale-up gap this chapter identifies. The British Future Fund (Chapter 15) will take patient, minority equity stakes in sovereign compute capacity and in scaling British AI firms, on its standard commercial mandate — capped, passive, and headroom-funded, never borrowed; the British Growth Exchange (Chapter 15) gives those firms a domestic listing route so the next DeepMind floats in London, not New York. The state co-invests as a patient shareholder in strategic capability; it does not nationalise the cloud. The principle is no longer fringe: in 2025 even the United States took a roughly 10% equity stake in Intel on national-security grounds. We would do it more cleanly than that — through an arm’s-length fund on a commercial mandate, as a minority holder taking forward stakes in capability, not as political pressure on a chosen firm or a rescue of a failing one
• Sovereign AI capability — ensuring Britain retains independent AI development capacity rather than becoming wholly dependent on US or Chinese models, on the same strategic-autonomy logic as our defence procurement (Chapter 12). What we will not do is pretend to fund what the private sector funds. The overwhelming majority of data centre capital is private — the £31bn UK-US technology partnership and hyperscaler investment dwarf anything the state would put in — so our scorecard prices only the genuinely public commitments here (the British Future Fund’s compute allocation, drawn from its existing headroom-funded capital, and the cost of grid-priority administration). The rest — the private build this policy unlocks — is enabling, not spending, and we count it as such. We claim credit for the conditions we create, not for capital we do not provide

Regulation: An Honest Position

The EU AI Act is the world’s first comprehensive AI regulatory framework — binding from August 2026, with fines up to €35m or 7% of global turnover. It is important to be honest about what this means for Britain.

The EU AI Act already applies to British companies. Any UK business that develops, deploys or sells AI systems affecting users in the EU must comply — regardless of whether Britain is in the Single Market. This is not a future concern. It is current reality, and Single Market membership changes the practical position only at the margins.

What Single Market membership does change is this: currently Britain is subject to EU AI regulation extraterritorially, with zero input into how those rules are written. EEA membership provides informal consultation rights during drafting and expert participation in the process — not a vote, but not nothing. For a country of Britain’s size and technological weight, that informal influence is meaningful.

F³ will maintain a principles-based domestic AI framework for purely domestic applications — lighter-touch than the EU AI Act where British law permits it. But we will not pretend that British AI companies can ignore EU rules while selling into European markets. The honest position is:

• Domestic applications — principles-based, outcomes-focused regulation with clear red lines but no unnecessary burden on low-risk innovation
• EU-facing applications — compliance with the AI Act is unavoidable and already required; Single Market membership gives Britain informal input into future revisions
• Sector-specific high-stakes frameworks — healthcare, criminal justice, financial services — developed with industry, not imposed upon it
• International standards leadership — Britain actively shaping global AI governance norms through the Council of Europe Framework Convention on AI, which the UK has already signed

The strongest argument for our position is not that Britain escapes EU AI regulation — it does not. It is that Single Market membership, combined with a lighter domestic framework and active engagement in international standard-setting, gives Britain more influence over the global direction of AI governance than the current arrangement of rule-taking from outside with no voice whatsoever.

AI in Public Services

The productivity gains from AI in public services are potentially enormous — and largely unrealised. We will drive adoption across:

• NHS — AI-assisted diagnosis, administrative automation, drug discovery and clinical trial acceleration
• HMRC — AI-driven tax compliance and fraud detection, reducing the tax gap and cutting administrative burden on honest taxpayers
• Planning — AI-assisted planning decisions, reducing the months-long delays that currently block development
• Criminal justice — AI tools for case management, reducing the backlog that is denying justice to victims and defendants alike

Skills and Talent

Britain cannot lead in AI without the people to build it. We will:

• Expand computer science teaching from primary level — coding and data literacy as core curriculum alongside reading and maths
• Create an AI skills visa — a fast-track route for the world’s top AI talent to work in Britain, with no cap on numbers
• Fund postgraduate AI research places at British universities — keeping our best graduates here rather than watching them leave for the US
• Target 10 million workers upskilled in AI by 2030 — delivering the existing government commitment with proper resourcing

Life Sciences and the NHS Data Advantage

Britain’s life sciences sector is already world-class — and it sits on an asset no other large country possesses. The NHS is a single-payer system with cradle-to-grave, population-wide, linked health records stretching back decades. For training diagnostic AI, running real-world studies, and discovering which treatments work in whom, that is among the most valuable health datasets on earth. We have barely begun to use it. The independent Sudlow Review (2024) concluded bluntly that Britain is squandering this resource: researchers can wait months or years for access, delaying advances in cancer, dementia and heart disease. Our approach has two stages, in order: fix access first, then capture fair value on top. Neither happens without public trust, so we start there.

Trust first, and we are honest about why. This has been tried before and it failed: the care.data programme collapsed in 2014 because the public rightly sensed it was being asked to hand over medical records with too little transparency and too little say. We will not repeat that. The model we back is the one that has since earned cautious trust and is already being built — Secure Data Environments, where de-identified data never leaves the NHS’s control. Researchers come to the data in a secure setting; only aggregated, non-identifying analysis comes out. The principle is simple and absolute: access, never transfer. Patient records are not sold, not exported, and not handed to anyone. What is monetised is insight, not identity.

Stage One: Fix Access

The binding constraint is not privacy law — it is delay. The fix is largely designed already; what is missing is the will to finish and resource it.

• Complete and fully fund the Health Data Research Service — the national front door recommended by Sudlow, backed by £600m from government and the Wellcome Trust — and set a hard service standard: a decision on a compliant research-access request in weeks, not years
• Standardise the regional Secure Data Environments onto one set of governance, approval and accreditation rules, so a researcher does not face a different process in every region — the single biggest practical brake the sector reports
• Formally designate NHS health data as critical national infrastructure, as Sudlow urged — funded, governed and protected with the seriousness we give power and transport, and turning the NHS into the best clinical-trials platform in the world

Stage Two: Capture Fair Value for the Public

At present, where commercial value is created using NHS data, the public too often captures little of it. That is the asymmetry we correct — not by selling anything, but by ensuring the nation shares in what its data helps create. The data-stays-home rule above is unchanged; this is about the terms of access, not the nature of it.

• Fair commercial access terms: companies pay to use the secure-environment infrastructure on transparent, published terms, with academic and NHS-led research kept low-cost or free — the public asset is priced fairly, not given away
• A public stake where the data is material: where NHS data demonstrably enables a commercial product, the public takes a return — a royalty, an equity share routed to the British Future Fund (Chapter 15), or preferential UK pricing and early access to the resulting treatment. The same principle as the rest of this paper: where the state materially enables success, it takes a stake in it
• Returns ring-fenced back into the NHS and research: value captured from NHS data is reinvested in the service that generated it and in further research — visibly, so the public can see that sharing their data funds their care

We are clear about the limits. We attach no revenue figure to this: the value of health data is real but genuinely hard to quantify in advance, and we will not invent a number to make the policy look richer than it is. The dependency runs the other way — the prize is not a windfall but faster cures, a stronger sector, and a public that finally shares in value it currently gives away. And the constraint is permanent: the moment value-capture is seen to compromise privacy or the data-stays-home rule, public trust collapses and the whole asset is lost — as care.data proved. Value-capture is therefore always subordinate to trust, never the reverse.

Rewarding Discovery: R&D and the Patent Box

There is a strategic-autonomy case here, of the same kind this paper makes on energy, defence and compute. China now accounts for around a fifth of the world’s drugs in development — nearly double the combined share of Britain, France, Germany, Italy and Spain — and roughly a third of global pharmaceutical out-licensing by value now involves Chinese-discovered molecules, the product of a decade of deliberate regulatory and industrial policy. A Britain that merely buys in others’ discoveries is a Britain dependent on them; the goal is for Britain to be a source of new medicine, not only a customer for it. We are not naive about this: licensing and collaboration have genuine benefits, and we do not pursue techno-nationalism for its own sake. But originating discovery here is both an economic prize and a question of resilience, and the tax system is the most direct lever we hold.

• A strengthened Patent Box: profits earned from patents developed in Britain are taxed at a reduced rate. This is the single most targeted tool we have, because it rewards exactly what we are short of — not discovery alone, but discovery commercialised and kept here. It is the direct counterpart to our headline Corporation Tax at 30%: the full rate on ordinary profit, a deliberately lower rate on the fruits of British R&D. We name the honest objection: a Patent Box can be expensive deadweight, rewarding IP that would have stayed in Britain anyway rather than changing decisions at the margin — a point the Institute for Fiscal Studies and others have rightly pressed. Our answer is design. The relief is tied to a strict “nexus” test — the lower rate applies only in proportion to the R&D genuinely carried out in Britain, not to IP merely parked here for the tax break — and we would have the OBR and HMRC publish its actual additionality, so that if it is buying behaviour we already had, we can say so and change it. A targeted relief we are willing to measure honestly is worth more than a generous one we refuse to examine
• A stable, generous R&D tax credit: not headline-grabbing rate changes every fiscal event, but a predictable regime firms can plan a decade of investment around. Predictability is itself an incentive — the same lesson we drew on the North Sea. Constant tinkering with R&D relief has done real damage to confidence; we would stop
• Full expensing of R&D capital, consistent with the investment-led logic applied throughout this paper’s tax chapters — the cost of the lab, the equipment and the kit written off in full, in the year it is spent

The Golden Triangle: From Discovery to Company

Britain’s problem has never been the quality of its science. The Oxford–Cambridge–London “golden triangle” is one of the densest concentrations of world-class life-sciences research anywhere on earth. Our problem is the step from discovery to company: too many British breakthroughs become American or, increasingly, Chinese businesses. We have been brilliant at the Nobel Prize and poor at the IPO. Three things change that.

• Fix university spinout terms. For too long, British universities took founder-crushing equity stakes in the companies built on their research — often far more than US peers — leaving founders too little to attract the next round of investment. We back the move to standardised, founder-friendly spinout terms across the sector, so a discovery in a Cambridge lab has a fighting chance of becoming a Cambridge company rather than dying in negotiation or fleeing abroad
• Build the physical cluster. The binding constraint in the golden triangle is now mundane and fixable: a chronic shortage of laboratory and incubator space, and the housing and transport to support the people who work in it. We treat life-sciences lab capacity as strategic infrastructure — fast-tracked alongside the compute and energy infrastructure elsewhere in this paper — and back the expansion of the science parks and campuses the cluster has outgrown
• Keep the scale-ups British. The British Future Fund and British Growth Exchange (Chapter 15) are pointed squarely at this gap: patient capital so a promising biotech does not have to sell to a US acquirer at phase one to survive, and a domestic listing route so it can float in London. Paired with the patient-capital crossover funding that lets a company cross the “valley of death” between discovery and clinical proof, this is how a British discovery becomes a British company that stays British

The Wider Sector

Around the data advantage sits the rest of the offer: a fast, predictable MHRA — backing the regulator to be the quickest credible approval route in the world, not the slowest; the British Future Fund and British Growth Exchange (Chapter 15) pointed at scaling British biotech so our best companies float and grow here rather than sell early to a US acquirer; free university tuition for life-sciences disciplines with a UK service obligation (Chapter 6); and the manufacturing capacity to make here what we discover here. Fintech, Britain’s other world-leading innovation sector, is backed on the same logic — the FCA sandbox model that made London a global fintech capital, maintained and extended, with the same scale-up finance behind it.

Social Media, Cohesion and Democracy

Social media is the most significant driver of social division in Britain today. The evidence is now substantial and largely beyond dispute: algorithmic amplification of outrage is documented and deliberate, the mental health impact on young people is beyond reasonable doubt, and foreign state actors — as the opening of this paper illustrates — routinely use social media infrastructure to amplify division for geopolitical ends.

The policy challenge is framing. Any government intervention risks being characterised as censorship. F³ rejects both the authoritarian instinct to regulate speech and the libertarian instinct to regulate nothing. We target the mechanism of harm — the algorithm — not the content.

We are not regulating what people say. We are regulating the commercial exploitation of human psychology for profit, and the deliberate amplification of content designed to divide us. The algorithm is not neutral — it is a product, and like every product it must meet basic standards.

• Algorithm transparency — platforms above £500m UK revenue must publish how their recommendation systems work and what content they amplify. Not censorship. Sunlight.
• Opt-in algorithmic curation — the default for all users is chronological feed. If you want the algorithm, you opt in. This one change would fundamentally transform the information environment.
• Under-18 protections — no algorithmic recommendation, age-verified accounts, default safe settings, no targeted advertising. The evidence on harm to young people is strong enough to justify firm action.
• Foreign state interference liability — platforms that knowingly host coordinated inauthentic behaviour by state actors face significant fines. Makes it commercially rational to police this seriously.
• Digital literacy curriculum — mandatory critical media literacy from secondary school. Understanding how algorithms work, how to identify disinformation, how to evaluate sources.

Platform Liability: Money and Amplification Carry Responsibility

Platforms have hidden for thirty years behind a single claim: we are just the pipes. The claim was always selective — pipes do not sell advertising against the sewage — and under F³ it ends where it was always false. Liability follows the platform’s own conduct, in three tiers:

Tier 1 — Paid and Promoted: Full Joint Liability

Where a platform is paid to distribute content — advertising, sponsored posts, boosted reach — it is jointly liable for that content as a publisher: defamation, fraud, scams, incitement, all of it. A platform that takes a scammer’s money and delivers the scammer’s victims is not an intermediary; it is a partner. Victims of fraud originating in paid placement are entitled to restitution from the platform, recoverable in turn from the advertiser — and platforms contribute to authorised-push-payment fraud reimbursement where first contact came through their paid or promoted surfaces.

Tier 2 — Amplified: Amplification Is Publication

Where a platform’s recommender system pushes content beyond its organic reach — chooses it, ranks it, injects it into the feeds of people who never asked for it — the platform is jointly liable for what it amplified. This is the editorial act: a front page assembled by machine is still a front page. The synergy with this chapter’s algorithm rules is deliberate: a platform operating chronological, neutral, opt-in feeds makes no editorial choice and keeps full hosting protection. Amplify and you own it; merely carry it and you don’t. The liability rule enforces the algorithm policy through incentives, not inspectors.

Tier 3 — Hosted: Immunity With Teeth

For organic, unamplified content, conditional immunity remains — because the alternative is collateral censorship. No moderator can adjudicate truth or honest opinion at scale, so blanket liability would mean every contested review, every allegation and every criticism deleted on receipt of a lawyer’s letter — and only the giants could carry the risk, entrenching exactly the companies this chapter exists to discipline. But the conditions acquire teeth:

• 48-hour takedown on a valid notice of unlawful content — defamation with particulars, fraud, incitement — with daily penalties for default
• Identity disclosure: on a court order, platforms must produce verified account information so victims can sue the actual author. Monetised accounts are identity-verified at onboarding — anonymity for speech, not for income
• Senior-manager liability for systemic failure — a named executive answers for the system, not for each post
• Repeat-offender duties: accounts and advertisers with adjudicated violations lose monetisation and amplification before they lose the account

Single Market note: under EEA membership Britain adopts the Digital Services Act, which grants hosting immunity to neutral intermediaries. Tier 2 is therefore drafted inside the DSA’s own 'active role' doctrine — a recommender that selects and promotes is not a neutral host, and paid placement never was. We expect the argument; we have written it down in advance.

F³ explicitly does not: regulate what individuals say (within existing law), remove content on political grounds, create government content moderation bodies, or give ministers power over platform decisions. Liability in this chapter attaches to platform conduct — money and amplification — never to the lawful speech of individuals.

Chapter 18: Criminal Justice

A System in Crisis

The scale of the criminal justice backlog is extraordinary and largely invisible to public debate. There are currently 379,000 outstanding cases in magistrates' courts and 80,000 in the Crown Court — with 21,000 cases open for over a year. Some defendants wait until 2027 or 2028 for trial. Victims live for years in limbo before seeing justice.

This is not an accident. Criminal legal aid rates were effectively cut by 42% in real terms between the early 2000s and recent years. Court buildings are deteriorating. Technology fails routinely. Essential staff have left and not been replaced. The profession of criminal defence solicitor — the bedrock of access to justice — is in rapid decline, leaving entire regions of England and Wales as legal aid deserts where affordable legal representation simply does not exist.

F³ will fund a comprehensive criminal justice recovery programme — treating it as the public safety infrastructure it is.

Funding the Recovery

• Immediate 20% real-terms increase in criminal legal aid rates — stopping the haemorrhage of solicitors leaving the profession
• £2bn capital investment in court buildings and technology over five years — £0.4bn a year, ending the scandal of cases collapsing because of crumbling infrastructure and failing IT
• Sitting days funded to the maximum the judiciary can supply — and we should be honest that this is a smaller lever than it sounds. The Crown Court sat 110,647 days in 2025-26, an all-time high and 35% above the 82,000 of 2019-20, against a maximum judicial capacity the Lady Chief Justice has put at about 113,000. The headroom is roughly 2,000 days, or 2%, and it costs tens of millions rather than hundreds. Anyone promising to clear this backlog by sitting more days is promising something the estate cannot deliver
• Expand the magistracy and judicial appointments — because the binding constraint is people, not money. The Ministry of Justice's own position is that court capacity is not solely a matter of judicial sitting days but of the finite supply of advocates, defence and prosecution lawyers, legal advisers and court staff. That is why the legal aid uplift above is the central capacity measure in this chapter and not merely a fairness measure: paying advocates properly is how the sitting days actually get used
• AI-assisted case management — technology to reduce the administrative burden on judges, clerks and legal professionals

Sentencing: Punishment and Purpose

Public anger at sentencing is real and legitimate. When violent offenders receive sentences that bear no relationship to the harm they have caused, confidence in the justice system collapses — and with it, the deterrent effect that sentencing is meant to provide.

• Tougher sentencing for violent crime — mandatory minimum sentences for serious violence, knife crime with intent to harm, and repeat offenders
• Remote prisons for dangerous and violent offenders — purpose-built facilities in remote locations where the environment itself reflects the seriousness of the crime and proximity to criminal networks is eliminated
• White collar crime — heavy financial penalties rather than custodial sentences as the primary response; fines set as a multiple of the financial gain, not a fixed sum that becomes a cost of doing business
• Confiscation of assets — proceeds of crime pursued aggressively and comprehensively, with no statute of limitations on financial investigation

Rehabilitation: Investing in Second Chances

The evidence on what reduces reoffending is clear and has been ignored for decades. Prison without rehabilitation produces more crime. The majority of those in the criminal justice system come from a narrow band of disadvantaged backgrounds — care leavers, victims of abuse, those who grew up in poverty without stable family structures.

The devil finds work for idle hands. Young people without purpose, without activity, without hope are the recruiting pool for gangs and violence. Preventing crime is cheaper than prosecuting it, imprisoning it, and living with its consequences.

• Mandatory education and skills programmes in all prisons — every sentence is an opportunity to equip someone for a life without crime
• Mental health and addiction treatment as sentence requirements — addressing the underlying conditions that drive a significant proportion of offending
• Dedicated rehabilitation programmes for care leavers and those from abusive backgrounds — recognising that these individuals are often victims before they are offenders
• Expand the network of sports facilities, youth clubs and community activities — free or heavily subsidised, run through schools and sports centres, available evenings and weekends
• Fund youth workers and mentors in high-crime communities — not as a soft alternative to policing but as a complement to it

Policing

• Restore police numbers to 2010 levels as a minimum — neighbourhood policing that knows its community is the most effective crime prevention tool that exists
• Cut police bureaucracy — officers should police, not fill in forms; administrative burden reduction to free up frontline time
• Knife crime — serious enforcement combined with serious prevention; stop and search used intelligently and lawfully, not as a substitute for community engagement

Drugs Policy

The UK’s drugs policy debate is often presented as a binary choice between full legalisation and zero tolerance. F³ rejects both extremes in favour of a coherent evidence-based position.

Medical cannabis was legalised in Britain in 2018 but remains largely inaccessible through the NHS due to complex regulation and cautious guidance. This is indefensible. The evidence for medical cannabis in treating chronic pain, epilepsy, multiple sclerosis, chemotherapy-induced nausea and a range of other conditions is robust. Independent analysis shows wider NHS access could add £13.3bn to the UK economy over a decade, reduce hospital admissions by 28% for eligible patients, and help thousands of people with long-term conditions return to work. The Alfie Dingley case — a child whose NHS care costs fell by £130,000 annually after starting cannabis treatment, now seizure-free — illustrates the human case as powerfully as any economic model.

• Expand NHS access to medical cannabis for all conditions with robust clinical evidence — not just the three currently approved
• Streamline the regulatory framework for prescription — removing the barriers that force patients into expensive private clinics
• Support domestic production of medical cannabis — reducing import dependence and building a British industry

On broader decriminalisation: F³ does not support it. The county lines drug gangs destroying communities, the knife crime epidemic driven by drug territory disputes, and the devastation of addiction in families across Britain are not arguments for making drug acquisition easier. They are arguments for better treatment, better enforcement of serious drug supply offences, and the rehabilitation programmes described elsewhere in this chapter.

One Age of Adulthood

Britain currently decides that a person is old enough to have sex at 16, to join the armed forces at 16, to drive at 17, and to vote, drink, marry, gamble or serve on a jury at 18 — while holding them fully criminally responsible from the age of 10, the lowest threshold in western Europe. No principle produces that list. It is an accumulation of separate decisions taken decades apart, and the result is a country that tells a 16-year-old they are adult enough for one thing and a child for another, in the same week.

F³ will set a single age of adulthood at 18 in England and Wales, and legislate a separate, higher age of criminal responsibility at 12. The distinction matters and we make it deliberately: 18 governs what society permits you to do, and 12 governs when society holds you accountable. Those are different questions, and collapsing them is what produced the current muddle. Where a matter is devolved we legislate for England and Wales and leave Scotland and Northern Ireland to reach their own view; where a matter is reserved, the single age applies across the United Kingdom.

• Every adult privilege at 18 — alcohol, driving, gambling, tattooing, jury service and armed forces enlistment, aligned on a single threshold. Marriage in England and Wales is already 18 and stays there. The franchise for UK general elections is reserved and is set at 18
• The age of criminal responsibility in England and Wales raised from 10 to 12 — the lowest in western Europe becomes merely low, and England and Wales stop being an outlier against Scotland, which moved to 12 in 2019
• The age of consent in England and Wales raised from 16 to 18, with a statutory close-in-age exemption. The exemption is the substance of the reform, not a footnote: where both parties are 16 or over and within three years of each other, no offence is committed and the law is unchanged from today. What changes is that a 16- or 17-year-old gains protection against a materially older partner, and the existing offences below 16 and the position-of-trust provisions are untouched. The target is adults exploiting minors. Teenagers are not the target and must not become the casualty
• The participation age in England already runs to 18, and the 18–22 Guarantee (Chapter 6) funds a place for every young person at that age. The rest of the framework is brought into line with the threshold that already exists, rather than the other way round

Two qualifications, and one objection we do not accept. The objection first: it is usually said that ending enlistment at 16 costs the forces recruits. It does not, on this programme, because this programme already recruits them at 18. The 18–22 Guarantee's Military strand (Chapter 6) is an expanded armed-forces apprenticeship intake across engineering, logistics, cyber and medicine, aligned with the 3% defence commitment — a trade and a service record, entered at eighteen. And the participation age in England already runs to 18: a sixteen-year-old must be in education or training, and enlistment counts as the training. The forces are the standing exception to a threshold Parliament set years ago, not a casualty of a new one. What genuinely changes is the pipeline — junior entry closes, and the services compete for eighteen-year-olds alongside employers rather than reaching them two years earlier. The Guarantee's Military strand exists precisely to supply that, and we would expect the services to use it. The first real qualification is drafting: the close-in-age exemption on the age of consent has to be got right, because a badly drafted version either criminalises teenagers for conduct with each other or leaves the protective purpose unachieved. It should be settled with the Law Commission before a Bill, not in Committee. The second is that this is not a UK-wide reform and we do not present it as one.

On devolution we take the straightforward view. Marriage age, the age of criminal responsibility, the age of consent and the franchise for devolved elections are matters for the Scottish Parliament and the Northern Ireland Assembly, and it is not for a Westminster paper to tell them what to conclude. We legislate for England and Wales, we set the single age across reserved matters, and we leave the devolved administrations to decide their own. If that produces a United Kingdom with more than one age of adulthood for a period, that is what devolution means and we are not going to complain about it. We should also be plain that this cuts against the current direction of travel on the franchise, where the settled expectation is votes at 16. We take the other view, and the reason is consistency rather than any judgement about the political maturity of sixteen-year-olds: a country cannot coherently argue that someone is too young to buy a beer or sign a tenancy, and simultaneously old enough to choose a government. If the age of adulthood is 18, it is 18 for the ballot box too. If Parliament prefers 16, then the honest consequence is 16 for the rest of the list as well — what we reject is the present arrangement, in which the number changes according to which conversation is being had.

Chapter 19: Water — Clean Rivers, Whoever Owns the Pipes

What Actually Causes the Spills

Start with the physical reality, because most of the debate ignores it. The majority of Britain’s sewers are Victorian combined sewers: they carry sewage and rainwater in the same pipe. When it rains hard, the pipe fills, and rather than back up into homes the surplus is discharged into rivers through a combined sewer overflow. These overflows are not a malfunction — they were designed in, more than a century ago, as a relief valve. The scandal is not that they exist; it is that chronic underinvestment in separating sewers, building storage and upgrading treatment has left them discharging far more often than they should, and that for years nobody measured or enforced it.

The private water companies in England and Wales have a genuinely disgraceful record on this. Since privatisation in 1989 they have paid out around £85 billion in dividends while loading £60 billion of debt onto businesses that were sold debt-free — consumers pay the interest through their bills. Thames Water paid £7.2 billion to shareholders while dumping raw sewage, leaking 630 million litres a day, and bringing itself to the brink of collapse. That is financial engineering of a natural monopoly, and it is indefensible.

But Ownership Is Not the Variable

Here is the inconvenient fact for both tribes. Northern Ireland’s water has never been privatised — it has been in public ownership throughout — and it has the same sewage spills, the same ageing infrastructure, the same pollution problems. Scottish Water, also publicly owned, performs better than England on some measures but still discharges sewage into rivers and seas. Meanwhile, several privately operated systems in Europe run clean. If public ownership guaranteed clean rivers, Northern Ireland would have them. It does not. The reason is simple: NI Water is starved of capital because it competes with schools and hospitals for a fixed Stormont budget and loses, and its regulation is no fiercer than England’s. Public or private, the rivers are dirty for the same three reasons — too little investment, weak regulation, and prices or budgets that never covered the true cost of the network. F³ fixes those, and refuses to pretend that changing the nameplate on the owner is a substitute for any of them.

Fix the Machine, Not the Nameplate

Because water is a natural monopoly with captive customers, the discipline that works in a competitive market — switching supplier — does not exist. So the discipline has to come from regulation and from binding investment obligations, whoever owns the pipes. That is where F³ puts its effort:

• A hard, legally binding pollution-reduction schedule: a falling statutory cap on sewage discharges, company by company, year by year, with automatic penalties for breach — not the discretionary, rarely-used enforcement of the past. Fines are a deterrent, not a revenue stream: every penny is ring-fenced into river restoration, never general spending, so the state never comes to depend on pollution continuing. The goal is to collect nothing, because the rivers are clean
• A real regulator with teeth: Ofwat and the Environment Agency merged or re-tooled into a single body with the monitoring capacity, the data, and the statutory duty to prosecute, funded by a levy on the industry rather than by a stretched public budget
• Mandatory reinvestment: a ring-fence preventing dividends or distributions while a company is failing its pollution or leakage targets — you do not get to pay shareholders until the rivers are clean
• Universal real-time spill monitoring: every overflow metered and published live, so discharges are visible to the public the moment they happen and cannot be hidden as they were for years
• A funded programme to separate combined sewers and build storage in the worst-affected catchments — the physical fix that actually reduces overflows, financed over the decades the engineering genuinely takes

Policy Interaction: Failed-Company Takeovers and Bond Markets

The risk is real: a disorderly haircut on water company bonds could raise the risk premium on all UK infrastructure debt — partially offsetting the fiscal emergency brake credibility built elsewhere in this programme. F³ mitigates this through three commitments: (1) Special Administration process is court-supervised and transparent — no ministerial discretion on compensation; (2) fair value is determined by independent valuation, not government fiat; (3) the sequencing is explicit — companies approaching insolvency voluntarily first, profitable companies only after the insolvency route is exhausted. Infrastructure investors will distinguish between an orderly, legally supervised acquisition of an insolvent utility and political expropriation of profitable assets. We are doing the former, not the latter.

Public Ownership Where Companies Have Failed

The government’s own estimate of £100bn for full renationalisation is based on Regulatory Capital Value — 35 years of inflation-adjusted investment regardless of its quality or the debt loaded onto customers. But our approach is not to buy the whole sector at that figure. It is to take over only the companies that fail, through insolvency, at their true worth after debt and liabilities — which for the worst offenders is a fraction of RCV, and for Thames Water may be close to zero. The blanket £100bn number prices a policy we are not pursuing. A near-zero purchase price is not, however, a near-zero fiscal exposure, and we will not let the first number stand for the second. Acquiring a failed water company means inheriting its environmental liabilities, its statutory investment programme, its pension obligations and its operational risk — and the reason the equity is worthless is precisely that those obligations exceed what the business can carry. The acquisition is cheap; the undertaking is not. We do not book a purchase cost, and we do not pretend the absence of one is a saving. The scale of the inherited obligation is company-specific and would have to be established in each insolvency, which is why no figure for it appears in our scorecard.

Several companies — Thames Water foremost among them — are already functionally insolvent. Their bondholders and shareholders have already been told to expect a significant haircut. Special Administration, the formal insolvency process for regulated utilities, is the mechanism through which the government can acquire these companies at or near their actual worth — which in Thames Water’s case may be close to zero after accounting for their debt and environmental liabilities.

• Companies that fail — Thames Water foremost — pass into public ownership through Special Administration, the existing insolvency process for regulated utilities. Not as an ideological programme of blanket renationalisation, but because a company that has bankrupted itself has forfeited the right to run a monopoly. The taxpayer acquires it at its true worth — which after debt and environmental liabilities may be close to zero — never at the inflated Regulatory Capital Value that would reward decades of underinvestment
• Compensation paid at fair value — not at the inflated RCV figure that rewards decades of underinvestment and excessive debt loading
• Debt taken on only where backed by genuine asset value — bondholders who financed dividend extraction rather than infrastructure investment cannot expect full recovery
• A new National Water Authority to manage the public estate — not a return to the pre-1989 monolithic structure but a regionally organised public body with genuine accountability
• Ring-fenced investment programme — every penny of profit reinvested in infrastructure, leakage reduction and environmental standards

Where a company passes into public hands this way, the dividends it used to extract — running at roughly £2bn a year across the sector at its peak — are redirected into infrastructure and bills instead. But the test is performance, not ownership: a private operator hitting its pollution and investment targets keeps operating; a public body that fails them faces the same binding schedule and the same penalties. We are not promising to nationalise clean rivers into existence. We are promising to make every operator, public or private, deliver them — and to take over the ones that collapse.

The Other Half of the Problem: Farming

Water companies are the headline villain, but they are not the whole story, and a manifesto that pretended otherwise would be dishonest. Agriculture is one of the largest sources of river pollution in Britain — in many rural catchments the largest. Slurry from livestock, nitrogen and phosphate from fertiliser, and soil washed off bare fields all run into rivers, feeding the algal blooms that strip the oxygen and kill the fish. The River Wye, among others, has been devastated largely by intensive poultry units, not sewage works. Renationalising a water company does nothing about any of this.

• Enforceable rules on slurry storage and fertiliser application, with the Environment Agency resourced to actually inspect and prosecute — the rules largely exist on paper today but are barely enforced
• Buffer strips of uncropped land along watercourses, and cover crops on bare winter fields, as a condition of agricultural support payments — paying farmers for the clean-water outcomes the public wants rather than simply for owning land
• Catchment-level limits on intensive livestock density where a river is already failing, so a single overloaded catchment cannot be pushed past recovery

This is the same F³ principle applied to farming as to water companies: target the behaviour that causes the harm, fund the outcomes you want, and enforce the rules that already exist — rather than fighting a symbolic battle while the rivers stay dirty.

Waste Crime: The Flytipping Epidemic

The same enforcement collapse that left sewage unmeasured has let waste crime explode. This is no longer a stray mattress on a country lane: it is industrial-scale flytipping — lorry-loads of construction and commercial waste dumped in fields, laybys and on the edges of towns by organised operators who undercut legitimate disposal, pocket the difference, and vanish. Local councils in England now clear over a million flytipping incidents a year, at a cost running into tens of millions, and landowners hit with private dumps face the cleanup bill themselves. Prosecutions, meanwhile, are derisory — a tiny fraction of incidents end in a meaningful penalty, so for an organised gang the maths is simple: the profit dwarfs the risk.

• Treat organised, commercial-scale flytipping as the serious crime it is — vehicle seizure and crushing, proceeds-of-crime confiscation, and custodial sentences for repeat operators, not the token fines currently priced in as a cost of doing business
• Mandatory, digitised waste-tracking from source to disposal, so a load of building waste can be followed and whoever dumped it identified — closing the paper-trail gap the gangs exploit
• Make the duty of care real: businesses that hand waste to an unlicensed carrier share liability for where it ends up, so the cheap no-questions-asked operator loses his market
• Reimburse private landowners for clearing waste dumped on their land where they report it promptly — they are victims of a crime, not its cause, and should not carry the cost of the state’s enforcement failure

As with rivers, the rules already largely exist. What has been missing is the will and the capacity to enforce them, and a penalty severe enough that dumping stops paying. F³ supplies both.

Chapter 20: Social Care — A Pre-Funded Future

The Scale of the Crisis

Total spending on adult social care in England reached £34.5bn in 2024-25 — and the Health Foundation estimates an additional £8.3bn will be needed by 2032 just to keep pace with demographic demand. The sector employs about 1.6 million people with roughly 111,000 vacancies — down from the 152,000 an earlier draft cited, on Skills for Care’s 2025 workforce report, and a fall driven substantially by international recruitment that the current visa regime is now restricting. Local authorities are overwhelmed, care homes are closing, and hundreds of thousands of elderly people are not receiving the care they need.

The fundamental problem is structural: social care is funded year-to-year from local authority budgets, creating chronic underfunding and a system that cannot plan beyond the next spending review. Eighty percent of people aged 65 will need some social care before they die — but the amount is unknowable in advance. It is an insurance problem that is currently funded as a consumption problem.

The Singapore Model

Singapore solved this problem through a mandatory pre-funded long-term care insurance system. CareShield Life provides every citizen with lifetime long-term care coverage. Premiums are paid during working years, risk-pooled across cohorts, and fully payable from MediSave — Singapore’s mandatory healthcare savings account. Payouts start at around S$689 per month, rising in the years before a claim, and are then paid for life once care needs begin. The scheme is self-funding and sustainable.

The principle is elegant: you fund your future care during your working years, when the cost is small and spread over time, rather than facing catastrophic bills in old age. Those who die without needing care contribute to a pool that covers those who need extensive care. The risk is pooled across a whole generation.

The National Care Guarantee

F³ will create a National Care Fund: a single pooled insurance fund for long-term care, not a savings account in each worker's name. What a worker buys into it is not a balance but a guarantee, and the guarantee has two parts:

• Care will never cost you more than £86,000 in your lifetime — the Dilnot cap, legislated in 2014 and abandoned, reinstated and honoured
• You will never have to sell your home to pay for it

We designed this as individual accounts first, and the arithmetic sent us back. Two per cent of median earnings over a full forty-year career accumulates roughly £45,000 — about half the £86,000 cap. A career interrupted by caring, illness or unemployment reaches far less, and reaches least for the people most likely to need care. Meanwhile care costs are not spread evenly: roughly a quarter of people need no long-term care at all, while a minority need years of it. That is the shape of an insurance risk, not a savings target. A pooled fund covers the whole of that distribution. Forty million individual pots would leave most people short at precisely the moment the money was called for, and would invite a question the design could never answer well — where has my money gone? The honest answer under pooling is that there is no individual money to go anywhere, and that is the mechanism rather than a flaw in it.

• The Fund pays the bill; there is no personal balance, and therefore no balance that can run out
• Contributions are pooled across the whole working population and across generations
• Those who never need care fund those who need a great deal — that is the point of it, not a defect
• There is no residual to return at death, and we do not imply a property right the design cannot deliver
• Entitlement follows care need and is never conditional on what any individual has paid in

Who Pays In

The Fund is financed by two contributions, one on employers and one on the highest earners.

• Employers pay 2% of qualifying earnings — the £6,240 to £50,270 band used for auto-enrolment — into the Fund. In practice the employer minimum rises from 3% to 5%, with the two additional points ring-fenced by law for care rather than added to the pension pot. This raises approximately £14.0bn a year.
• Individuals pay 2% of earnings above £150,000. This raises approximately £1.5bn a year. That figure is an estimate and we mark it as one: it assumes roughly two-thirds of the £115bn of income above £150,000 is earned rather than investment income, and it should be re-derived independently rather than relied on as it stands.
• £150,000 is the same threshold as the private hospital cover surcharge in Chapter 5 — one threshold, one justification, one number to remember. We looked at £100,000 and rejected it: at that level the £120,000 household in our own worked examples tips back into a net loss of about £251, where at £150,000 it stays ahead by about £149. We are not willing to publish a design that quietly reverses one of our own published examples.
• The upper tier falls on individuals rather than employers deliberately. The benefit is flat — the same £86,000 cap protects everyone, whatever they earn — so a flat employer charge on high earners would subsidise the top; and loading it onto employers would make senior hires more expensive at exactly the moment the rest of this paper is trying to make hiring cheaper.

Together the two tiers put roughly £15.5bn a year into the Fund. That is smaller than the 10% employer pension-and-care mandate this design replaces — a mandate we previously left unpriced and which would plausibly have cost employers £25-45bn a year. We are replacing a large unpriced promise with a smaller priced one. The headline is less impressive and the paper is better for it.

The Numbers

Why pooling works where individual accounts do not:

• 2% of median earnings, a full 40-year career, 5% real return: approximately £45,000 accumulated
• Against a lifetime cap of £86,000 — so a complete working life funds about half of the promise
• Average residential care: approximately £50,000 a year; average stay 2.5 years, about £125,000 in total
• Roughly a quarter of people need no long-term care at all; a minority need several years of it
• Pooled, the ~£15.5bn a year the Fund receives carries the guarantee across that distribution. Held individually, most pots would fall short exactly when they were called on

The Dilnot Cap

The £86,000 lifetime cap legislated in 2014 and then abandoned is reinstated, and it is the Fund's core promise rather than a supplement to it. Above the cap, costs are met. Below it, the means-tested floor that exists today is retained and strengthened, so people with few assets are not asked for £86,000 they do not have. Pooling also disposes of a problem the account design had to patch: people who could never build a balance — those out of work through caring, disability or long-term illness — needed a whole apparatus of Treasury-funded credits to be made whole. Under a pooled fund they need nothing, because there was never an individual balance to be short of. The guarantee is universal by construction.

Workforce

• Free movement from the EU restores the labour supply that Brexit removed — care workers from across Europe can return
• Care worker pay raised through the Social Care Fair Pay Agreement and the care contribution paid on their behalf
• A Social Care Apprenticeship pathway from entry level to specialist and management roles

The Fiscal Position

The Fund's money is not the Exchequer's, and we should be precise about that. The roughly £15.5bn a year flowing into the National Care Fund is not government revenue, is not available for anything else, and is scored nowhere in the deficit. What the Exchequer does lose is Corporation Tax. Employer contributions are deductible against profits, so £14bn of new employer contributions at a 30% rate costs about £4.2bn a year in forgone Corporation Tax. That £4.2bn is in the scorecard as a cost. It is the single largest reason the fiscal position in Chapter 2 is worse than in earlier drafts of this paper, and we would rather book £4.2bn we can defend than carry a £25-45bn mandate we never priced at all. Two further points, both uncomfortable. The Fund does not relieve today's care spending inside this parliament: adult social care costs local authorities about £34.5bn a year and continues to be met from general taxation throughout the window, so nothing in this chapter reduces the deficit in the next five years. And the transition remains genuinely unsolved. Today's pensioners and those close to retirement will never have contributed, yet the cap protects them from the day it is legislated — a liability falling on general taxation for decades against a Fund seeded at about £5bn over the first parliament. Pooling makes the steady state work. It does not make the transition free, and we have not found a way to make it so. We state that as an open problem rather than assert a solution we do not have.

Chapter 21: Agriculture, Food Security and Rural Britain

Leaving CAP Was Right. What Replaced It Is Not.

The Common Agricultural Policy was poor policy — paying farmers based on how much land they farmed regardless of productivity or environmental outcomes. Britain was right to leave it. The problem is what replaced it.

ELMS has been implemented as a cliff edge. Farms receiving £160,000 in CAP payments in 2020 received £62,000 in 2024 and will receive just £7,200 in 2025. Beef and sheep farmers — the custodians of Britain’s most iconic landscapes — have been the hardest hit, with many facing insolvency.

The F³ Agricultural Framework

• Public goods payment — farmers paid for environmental stewardship, flood prevention, biodiversity enhancement and landscape management. Simple, predictable, not a labyrinthine application process
• Food security baseline — minimum domestic production capacity maintained as a strategic asset. Britain should produce at least 70% of its temperate food
• Productivity investment — grants and low-interest loans for technology adoption, precision agriculture and energy efficiency

Single Market and Food Costs

Rejoining the Single Market is the single most impactful measure for reducing food costs. Trade friction since Brexit has added approximately 6% to UK food import costs. A single-market footing plus a veterinary (SPS) agreement removes most of the agri-food checks that drive those costs, reducing them substantially. Britain produces 17% of its fruit and 55% of its vegetables domestically — highly import-dependent. Restoring near-frictionless agri-food trade brings back competitive pricing for these staples.

Food Standards Are Non-Negotiable

Trade deals with the United States and other countries must not come at the cost of British food standards. Chlorinated chicken, hormone-treated beef and products from intensive farming systems that would be illegal in Britain will not be permitted to undercut British producers. This is not protectionism — it is the application of consistent standards regardless of where food is produced.

Rural Communities

• Rural broadband and mobile coverage as statutory infrastructure obligations
• Rural transport — community transport funding and flexible bus routes
• Planning flexibility for rural dwellings and farm diversification

Chapter 22: Welfare Reform

The Scale of the Problem

Total working-age and children’s welfare spending is £145bn annually. The full breakdown reveals the scale of the challenge:

Of these, the State Pension is addressed through the Pensioner Guarantee (Chapter 13), which replaces the Triple Lock with an inflation-proofed pension protected in law. Child Benefit is replaced by school-based provision. The F³ welfare reform programme focuses on the three areas where spending has risen most sharply and where the case for structural reform is strongest: disability and incapacity benefits, Universal Credit conditionality, and housing benefit.

Disability and incapacity benefit spending has risen from £36bn in 2019-20 to £77bn today and is projected to approach £100bn by the end of the decade without reform. The number of working-age people receiving disability benefits has risen from 2.1 million to 3.4 million in five years — an increase that cannot be explained by a sudden deterioration in national health.

The causes are complex and demand honest acknowledgment:

• Mental health claims — particularly anxiety and depression — have risen dramatically, especially among young people post-pandemic. More people now receive maximum PIP for generalised anxiety than were born deaf. This reflects a genuine mental health crisis that needs treatment, not simply payment
• Assessment failures — the current system simultaneously denies support to people who desperately need it while paying others who do not
• Perverse incentives — a system designed around inactivity rather than supported employment traps people in dependency
• No regular reassessment — only 6% of PIP assessments are currently face-to-face; many awards run indefinitely without review

The result serves nobody well. Genuinely disabled people wait years for support while fighting bureaucratic battles. People who could work with the right support are written off. Taxpayers fund a system that is neither compassionate nor effective. F³ rejects the false choice between cruelty and complacency.

The F³ Principles

Our welfare reform is built on three principles that are simultaneously non-negotiable:

• Genuine need receives genuine support — no person with a serious disability or health condition should be worse off
• The system exists to enable people, not trap them — supported employment should always be the goal where it is achievable
• Public money must be spent honestly — a system that cannot distinguish between genuine need and exploited entitlement fails everyone, including those it is meant to help

PIP Reform — Personal Independence Payment

PIP is the largest single working-age disability benefit at £25bn annually, rising to a projected £37bn by 2029-30 without reform. It was designed to help with the extra costs of disability — mobility aids, adapted equipment, care support. It has drifted far beyond that original purpose.

F³ will reform PIP assessment comprehensively:

• Face-to-face assessments increased to 50% of all PIP assessments — the current 6% face-to-face rate is indefensible for a benefit of this scale
• This is funded, not assumed: returning to majority face-to-face assessment costs roughly £0.4–0.5bn a year in assessor capacity, and the two-yearly work-capability reassessments add to that. The net savings in the scorecard are stated after these delivery costs — the cost of doing the assessing is inside the line, not hidden beside it. We should be equally plain about what the savings figure is and is not. It is our own estimate, subject to independent verification, and no published costing exists for this mechanism. The comparators that do exist are for a different lever: a minimum four-point score in one daily living activity, which the OBR put at £3.9bn by 2029-30 while calling its own estimate highly uncertain and noting that previous reforms have often saved less than expected. DWP's impact assessment reached a similar total by a different route — about £3.5bn from PIP plus roughly £500m in knock-on Carer's benefit savings. The wider Green Paper package including the Universal Credit health element was costed at £5bn, and independent analysis put PIP narrowing plus UC health at £6.4bn or more. Two further facts belong here rather than in a footnote. Even on the government's own reforming path, working-age disability spending still rises — from about £23bn in 2024-25 to £31bn by 2029-30 — so this slows growth rather than cutting the bill. And the four-point reform was withdrawn in July 2025 after a Commons rebellion, so any programme relying on savings of this kind should treat parliamentary deliverability as a live risk and not an afterthought
• Assessors must hold relevant clinical qualifications — a generalist nurse should not be assessing complex neurological or psychiatric conditions
• Awards time-limited by default — 2-3 year reviews for most conditions, with indefinite awards reserved for genuinely permanent and severe conditions
• Ad-hoc reviews triggered by credible lifestyle evidence — activities demonstrably inconsistent with claimed limitations trigger reassessment
• Fraud treated as serious criminality — deliberate misrepresentation prosecuted, not just administratively corrected
• Conditions primarily treatable through the NHS — anxiety, depression, mild ADHD — addressed through NHS provision and employment support, not permanent PIP awards

Universal Credit Reform

Universal Credit at £60bn is the backbone of the working-age welfare system. Its structure — combining housing, childcare, disability and income support in one payment tapering at 55p per pound earned — creates significant work disincentives for many recipients. F³ will reform the conditionality and taper:

• Reduce the UC taper rate from 55% to 45% — allowing people to keep more of what they earn as they move back into work
• Strengthen work search requirements — those capable of work who decline reasonable offers face meaningful sanctions, applied consistently and fairly
• Mandatory work capability reassessments every two years for all claimants on health-related elements — regular review rather than indefinite award
• Assessors trained and resourced for fluctuating and mental-health conditions, with the NHS mental-health expansion (a stated precondition of this reform) carrying the clinical load a benefits assessment cannot. The point of more face-to-face contact is accuracy in both directions — awards refused that should never have been made, and awards granted that the broken remote process wrongly denied
• In-work progression requirements — those in part-time work actively supported and expected to increase hours where possible

Carer's Allowance — Ending the Cliff Edge

Unpaid carers save the state an estimated £160bn a year by caring for relatives the NHS and social care would otherwise have to. In return, Carer's Allowance pays around £86 a week — and traps them with one of the cruellest rules in the entire benefits system. Earn a single pound over the weekly limit and the whole payment vanishes for that week: a cliff edge, not a taper. The result was a national scandal — between 2015 and 2025, thousands of carers drifted over the limit after small pay rises they did not realise breached it, and were pursued for debts running into thousands of pounds, with penalties on top. A benefit meant to support the most selfless people in the country became a trap that punished them for working an extra hour. F³ ends it:

• Replace the cliff edge with a taper. Above the earnings limit, Carer's Allowance withdraws gradually — so an extra hour of work always leaves a carer better off, and no one is ever pursued for thousands because a pay rise tipped them a pound over a line
• Carers of school-age children can work school hours without penalty. The earnings limit for carers of children is set so that part-time work during school hours never costs them their allowance — recognising that caring and a modest job are not mutually exclusive, and that work during the hours a child is in school is exactly the flexibility a carer can offer
• A more generous limit for carers of non-school dependents — modestly higher than today's, in recognition that round-the-clock caring for an adult or a child out of education leaves less room for paid work, but that some work should still always pay
• Recognise caring intensity. Following the direction Scotland has already taken, payments reflect the hours of care provided — 20, 35 and 50-plus hours — rather than a single flat rate that treats light and round-the-clock caring identically
• Allow caring hours for more than one person to be combined toward the qualifying threshold — so someone caring for two relatives part-time is no longer locked out for failing to reach 35 hours for either alone
• Write off the historic overpayment debts caused by the DWP's own admittedly unclear guidance, and end the practice of pursuing carers for the department's failure to design a workable rule

The cost is modest — Carer's Allowance is a ~£4bn line, and a taper plus higher limits adds a fraction of that — and it is among the most defensible spending in this paper. A country that leans on £160bn of unpaid care has no business punishing the people who provide it for daring to also hold down a job. We score the change conservatively and treat it, openly, as money well spent.

Housing Benefit Reform

Housing benefit at £37bn is significantly driven by high private rents and the failure to build enough homes. The most effective housing benefit reform is a planning reform that increases supply and moderates rents — which our platform delivers through land banking tax, presumption in favour of development, and removal of viability-killing affordable housing quotas.

In parallel:

• Local Housing Allowance rates reviewed and uprated to reflect actual local rents — the current freeze has pushed claimants into housing poverty while doing nothing to reduce rents
• Benefit cap maintained but uprated with inflation — protecting the principle while preventing real-terms erosion
• Stronger incentives for local authorities to build social housing directly — using Land Banking Tax proceeds as the funding mechanism

Assessment Reform

The current assessment system is broken in both directions — too harsh on some, too permissive on others. F³ will reform it comprehensively:

• Face-to-face assessments increased to 50% of all PIP and Work Capability Assessments
• Assessors must be qualified healthcare professionals with relevant expertise
• Awards time-limited by default — most awards set for 2-3 years with mandatory review
• Ad-hoc reviews triggered by substantive evidence — reported activity or third-party information that materially contradicts a claimed limitation, assessed by a qualified professional. On publicly available social-media material we take a deliberately narrow position, because both the blanket and the permissive versions are wrong. There is no proactive monitoring of claimants: it is never a trigger, and no one goes looking. But where an investigation has already been opened on independent grounds, publicly available material may be considered as one source of evidence among others, weighed by a professional who understands that a photograph shows a moment and not a capacity. Refusing to look at relevant evidence in an active fraud investigation would be indefensible; trawling disabled people’s accounts to generate suspicion in the first place would be worse
• Fraud penalties significantly strengthened — deliberate misrepresentation treated as a serious criminal matter

NEETs and Young People

The rise of young people who are Not in Education, Employment or Training is one of the most serious long-term challenges facing Britain. It is not a simple story of laziness or entitlement — it reflects failures of the education system, the mental health crisis among young people, the lack of affordable housing near jobs, and the absence of routes into work that do not require a university degree.

F³ addresses the root causes through our education, housing, mental health and criminal justice reforms — but also directly:

• A national youth employment guarantee — every person under 25 who has been out of education and employment for more than 6 months receives a guaranteed offer: employment, apprenticeship, training or community service
• Benefit conditionality for under-25s — those who decline all reasonable offers without good cause face a graduated reduction in benefits. This is not punitive; it is the same expectation we have of any adult in a functioning society
• Apprenticeship expansion — make it as easy for an employer to take on an apprentice as to hire a graduate, with comparable status and career prospects
• Mental health support as a condition of benefit receipt — those whose NEET status is driven by mental health conditions receive treatment as a priority, with benefits maintained during engagement with treatment

Policy Interaction: Welfare Reform and NHS Capacity

The NHS mental health backlog is already severe. Removing PIP from conditions treatable through NHS without expanding NHS capacity first would trap people between two failing systems — no benefit, no treatment. F³ will therefore: (1) expand NHS talking therapies and IAPT capacity in year 1 as a condition of welfare reform proceeding; (2) set minimum NHS mental health waiting time standards before PIP eligibility tightening takes effect; (3) guarantee a supported employment offer to every person whose PIP is reduced. Welfare reform delivers its fiscal savings only if the NHS can absorb the demand — and that requires explicit investment sequencing, not a simultaneous cut.

One further interaction deserves stating plainly, because the biggest welfare-to-work lever in this paper sits outside this chapter entirely. Millions of working-age people are economically inactive through ill health — the defining labour-market problem of the post-Covid decade — and no assessment reform returns someone to work faster than treating the condition that took them out of it. The NHS elective recovery and the mental-health expansion in Chapter 5 are therefore labour-market policy as much as health policy: every month cut from a waiting list is measured not only in relieved pain but in people back at work, paying the taxes this framework depends on. We fund NHS capacity first and tighten eligibility second not merely because it is fair, but because it is the order in which the arithmetic works.

The Fiscal Position

We score this the way the OBR would score it — costs first, savings only where precedent supports them:

• PIP reassessment and eligibility reform: +£4–6bn a year by Year 5 (gross) — in line with, not beyond, what the OBR scored for far milder packages. This is PIP alone; the Universal Credit and work-capability side of the same reform carries the balance to the £8bn combined figure that appears in the scorecard
• UC taper cut from 55% to 45%: −£3bn a year, scored as a cost. We believe employment effects will claw much of it back; we do not book that belief as money
• Year-1 NHS mental-health expansion — the precondition for this chapter: −£2bn a year, scored
• Housing benefit reduction through increased supply: +£2–3bn by end of parliament
• NEET youth employment guarantee: funded by the ring-fenced Training Point, scored neutral in the main table

Net position: £4–6bn a year by Year 5 from PIP, and £8bn once Universal Credit and work-capability assessment reform is included — roughly half the headline other parties would claim from the same reforms, because we have scored the cost of doing it properly. The fiscal framework uses these numbers, not the press-release version.

What F³ Means For You: Real Income Examples

Tax policy is abstract until it lands in your pay packet. Here is what F³ means for five salary levels — against the current system, with nothing excluded.

These tables are complete, with one assumption stated up front. The private hospital cover surcharge in Chapter 5 is 2% of total income for those above £150,000 who decline qualifying cover. Only the £300,000 household in these examples is above that threshold, and its table assumes the individual holds qualifying cover, so no surcharge appears. If they decline it, the surcharge is 2% of the whole £300,000 — £6,000 a year — which takes their Years 1-3 gain from £7,314 to £1,314, and their post-glide gain from £15,864 to £9,864. Against that, holding cover has a premium cost of its own, which is a private expense we do not put in a tax table. Neither version is hidden: the surcharge is a choice, and both outcomes are printed here. The F³ deduction is the full combined rate on income above £15,000 — 38% income tax + 2% Local Income Tax + 2% Health Levy = 42% in Years 1–3, falling to 39% once certified headroom allows the headline rate to glide to 35%. On the timing, we should be blunt rather than let a column heading imply something we do not mean: on our central case the flat rate does NOT fall below 38% within this parliament. The gated giveaways that fire are the Employer Levy step-down, and the rate glide begins in the next parliament as headroom allows. Every 'after the glide' figure below is therefore the eventual position, not a five-year promise, and a reader should treat the Years 1-3 column as the one that describes this parliament. Employee National Insurance is abolished. The first £15,000 carries no tax, no levy and no local charge. We also show Council Tax and the 0.3% property charge that replaces it, because the Local Income Tax is only half of that swap and a payslip-only comparison would flatter us. One charge is still excluded and named here so the claim stays true: the 2% surcharge on total income above £150,000 for those who decline private hospital cover (Chapter 5) — which at these income levels applies to none of the five households below. If you can find a deduction we have left out, write to us.

Full-Time Minimum Wage Worker (£26,437)

Gain: +£251 a year in Years 1–3, rising to +£594 once the rate glides to 35%. Note what is doing the work, and what is not: on the payslip alone this household is £921 worse off, because a £15,000 threshold at 42% does not fully replace a £12,570 allowance plus abolished employee NIC at this income. The entire gain comes from abolishing Council Tax. It is a real gain, but it is thin, and we would rather show that than dress it up.

£40,000 — Typical Skilled Worker

Cost: −£1,647 a year (−£32 a week) in Years 1–3, narrowing to −£897 (−£17 a week) once the rate glides to 35%. This household is a loser under this programme, on the payslip and after Council Tax abolition, and we name it. See the offsets table below before judging the net position.

£80,000 — Senior Professional

Cost: −£3,027 a year (−£58 a week) in Years 1–3, narrowing to −£1,077 (−£21 a week) once the rate glides to 35%. This is the most exposed point on the whole income scale and we say so. A commuting two-child household recovers part of it through the offsets below, but not all of it.

£120,000 — Higher Earner (Currently Caught by the Taper Trap)

Gain: +£1,030 in Years 1–3, +£4,180 once the rate glides to 35%. This household gains while the £80,000 household loses, for one reason: the 60% marginal-rate trap between £100,000 and £125,140 is abolished entirely. That is the single largest structural improvement in the schedule, and it lands here.

£300,000 — Very High Earner

Gain: +£7,314 in Years 1–3, +£15,864 once the rate glides to 35% — on the assumption that this household holds qualifying private hospital cover. Declining it triggers the 2% surcharge on the whole £300,000, which is £6,000 a year and reduces those gains to +£1,314 and +£9,864 respectively — after the 2% care contribution on earnings above £150,000, which costs this household £3,000 a year and which we show on the face of the table rather than in a footnote. We print the gain because our opponents will. The defence is the ladder below — every rung still rises — the uncapped 0.3% property charge on high-value homes where Band H was a ceiling, and the abolition of the allowance-withdrawal games that made today's published 45% a fiction.

The Effective-Rate Ladder

Progressivity is measured by effective rates — the share of total income actually paid. Here is the ladder, levy-inclusive:

The ladder rises at every rung — the definition of a progressive system — though we will not claim the largest fall is at the bottom, because it is not: 1.0 point at minimum wage against 2.4 at £300,000. A named and narrowing transitional cost in the middle, and no 60% marginal-rate spike that today makes an extra pound earned at £110,000 worth less than an extra pound earned at £300,000.

What the Middle Gets Back

The transitional cost between roughly £28,000 and £100,000 is real and named. So are the offsets — line items with prices, not vague promises of better services:

A two-child household at £80,000 with one rail commuter recovers the full £3,027 transitional cost from the first line alone. On the limit of this: a household that neither commutes by rail nor has school-age children — the childless driver on £80,000 — draws less from the offset list, and for them the Health Levy is felt more directly. They still gain from the abolished employee National Insurance, the energy VAT cut, free parking, and a 0.3% property charge that is typically half their old Council Tax bill, but the offset is smaller. We do not claim every household is made whole by the offsets; we claim most are, we name the ones who are not, and we point to the reason the cost exists at all — a National Health Service funded honestly, by a levy that everyone above the threshold pays and can see. The transitional cost narrows at Step 4 for them as for everyone else.

The honest distribution: households below roughly £28,000 gain from day one, and so does the £100,000–£125,140 band whose 60% taper trap we abolish outright. Between about £28,000 and £100,000 there is a named transitional cost — the visible price of the Health Levy — narrowing at Step 4 and offset for most households by the items above. Above £110,000, abolishing the taper produces gains we publish rather than bury. If the country judges the middle’s share too high, the honest lever is a Health Levy threshold — a choice we cost at roughly £8bn, openly, rather than pretending it is free.

On distributional analysis by decile we have deliberately published nothing. An earlier draft of this paper carried a chart showing a net effect by income decile — from about +£600 in the lowest to about −£2,200 in the highest. Those figures looked precise and were not: they came from no household microsimulation, and a chart of that kind cannot be produced honestly without one. We have removed it rather than dress an estimate as a result. What we publish instead is the five worked households in this chapter, each shown line by line and each checkable, and the effective-rate ladder that follows. A reader who wants a decile analysis is entitled to one, and we would rather they got it from a model that exists than from a chart of ours that does not.

Winners and Those Who Pay More

Most manifestos only talk about winners. We believe honest politics requires acknowledging who pays more under our proposals — because honest government begins before election day. Here is our assessment across all 22 chapters of this paper:

We make no apology for this redistribution. Those who gain most are working families, growing businesses and people who use public services. Those who pay more are the wealthy, the land-banking, and those who benefit from universal entitlements they do not need.

Winners and Losers: Business

The same honesty applies to business. F³ pulls four levers at once — it abolishes transaction taxes (Business Rates, Stamp Duty), it cuts the cost of employing people (employee NIC gone, employer NIC substantially reduced), it raises the headline Corporation Tax rate to 30% while enforcing it through the global minimum-tax floor, and it recovers, through a 1% Landowner Levy on large commercial sites, the share of the Rates windfall that would otherwise reach freeholders as higher rents. The effect is deliberate and uneven: it rewards businesses that are physical, domestic and employment-heavy, and it falls hardest on those whose UK profit is large, mobile, and currently booked elsewhere. Here is the picture by size.

Small Businesses

The clearest winners in the entire programme. A high-street firm pays no Business Rates, half the old employer NIC, no Stamp Duty when it moves premises, and trades into a high street with free parking and returning footfall — while its modest profits sit far below the level at which the 30% rate bites hard.

Medium-Sized Businesses

Net winners, but the balance is finer. The rates and hiring savings are large for an employment-heavy scale-up, and a domestic profit base means the 30% rate is paid in full rather than dodged — which is the point. The losers here are those who leaned on reliefs the programme withdraws.

Large Domestic Businesses

A mixed picture that tilts on one question: where is your profit booked? A large employer with genuinely UK operations gains enormously from rates abolition and the hiring-cost cut, and pays the 30% rate it was largely paying anyway. A large firm that has engineered its UK tax bill down toward zero is a clear loser — by design.

Multinationals

The band where F³ asks the most. The trade is explicit: a far more competitive operating environment — near-frictionless EU access, no rates, cheaper hiring, stable rules — in exchange for paying tax on profit genuinely earned in Britain rather than routing it through lower-tax jurisdictions. For multinationals with real UK substance this is a good deal; for those whose UK presence is largely a tax address, it is not.

It bears stating plainly, because it is the heart of the business case: for the great majority of firms, the savings outweigh the rise. The worked examples make it concrete — the manufacturer pays roughly £9m more in Corporation Tax but saves some £40m in Business Rates and £15m in employer NIC. Only a business with very large UK profit and a very light UK footprint — the offshore-booking multinational — ends up paying materially more overall. Everyone who actually builds, hires and occupies premises in Britain comes out ahead, even with a higher headline rate, because we cut the two taxes that fell on them regardless of whether they made a penny of profit.

The pattern is consistent with everything else in this paper: F³ taxes what is booked and rewards what is built. A business that employs people, occupies premises and makes its profit in Britain is a winner several times over. A business whose UK profit is large but whose UK tax is small is asked, at last, to pay it. We think that is the right trade — and we say so to their faces, before the election, not after.

Four Worked Examples

Illustrative figures, rounded, to show the direction and rough scale in each band. Real outcomes vary with property values, payroll and profit, but the pattern is robust.

The Corner Shop — a small independent retailer

A high-street shop: £400,000 turnover, eight part-time staff on a £140,000 payroll, £35,000 profit, premises with a £28,000 rateable value paying roughly £14,000 a year in Business Rates.

• Business Rates: £14,000 → £0. A direct saving of £14,000
• Employer payroll charge: £15,000 gross today (15% above the £5,000 secondary threshold on each of eight part-time staff), but the Employment Allowance takes £10,500 off it, so the shop actually pays about £4,500. F³ abolishes the Employment Allowance, so the Levy is paid without relief: at the Year 1 core of 11.25% plus the 1-point Training Point the shop pays about £17,150 on a £140,000 payroll, falling to roughly £12,500 at the Year-5 central rate of about 7.9% core and £11,900 if the core ever reaches 7.5%. The Training Point comes back as funded apprentice training
• Corporation Tax on £35,000 profit rises modestly with the 30% rate, costing a few hundred pounds more
• Care contribution: 2% of qualifying earnings (£6,240 to £50,270) into the National Care Fund — about £1,800 a year for this shop, and a new cost that did not exist before

Net effect: on Year 1 rates the shop is worse off on payroll and better off overall only because of Business Rates. Business Rates abolition is worth £14,000; against it, losing the Employment Allowance and paying the Year 1 Levy costs about £12,650 more than today's £4,500 NIC bill, the care contribution adds £1,800, and Corporation Tax adds a few hundred. That is roughly break-even in Year 1, improving to a gain of about £4,000 a year by Year 5 as the Levy core steps down, and about £4,600 if the core ever reaches 7.5%. An earlier draft of this example claimed £7,700 by comparing today's NIC with the fully implemented 7.5% Levy and omitting the care contribution altogether. Small employers with low pay per head are the group for whom losing the Employment Allowance bites hardest, and we would rather show that than average it away

The Regional Chain — a medium-sized employer

A 12-site hospitality group: £18m turnover, 320 staff on a £9m payroll, £1.2m profit, paying about £900,000 a year in Business Rates across its sites.

• Business Rates: £900,000 → £0
• Employer payroll charge: about £1.11m today (15% above the £5,000 secondary threshold on each of 320 staff), replaced by the Levy — about £1.10m in Year 1 at the 12.25% headline, roughly £804,000 by Year 5 at the central 8.93%, and £765,000 only if the core reaches 7.5% beyond this parliament. So the payroll saving is close to nil in Year 1 and around £306,000 by Year 5, not the £435,000 that full implementation would deliver
• Corporation Tax: £1.2m profit at 30% rather than 25% costs about £60,000 more
• Stamp Duty: nothing now payable when it acquires or relocates a site
• Care contribution: 2% of qualifying earnings into the National Care Fund — about £140,000 a year, a new cost across a 320-strong payroll

Net effect: better off by roughly £700,000 in Year 1 and £1.0m a year by Year 5, before counting Single Market supply-chain savings. Business Rates abolition does the heavy lifting at £900,000; the payroll saving is small in Year 1 and builds as the Levy steps down; the care contribution takes £140,000 and Corporation Tax £60,000. An earlier draft put this at £1.3m by applying the fully implemented 7.5% Levy from the start and omitting the care contribution

The Manufacturer — a large domestic business

A FTSE 250 manufacturer with UK operations: £1.4bn turnover, 6,000 UK staff on a £240m payroll, £180m UK profit, large plants and warehouses paying around £40m a year in Business Rates. Its profit is genuinely booked in Britain.

• Business Rates: roughly £40m → £0 — a very large saving on a heavy physical estate
• Landowner Levy: as freeholder of its own plants it pays the new 1% charge on site value above £500k — several million a year on a large industrial estate, a fraction of the £40m Rates saving
• Employer payroll charge: about £31.5m today (15% above the £5,000 secondary threshold across 6,000 staff), replaced by the Levy — about £29.4m in Year 1 at the 12.25% headline, roughly £21.4m by Year 5 at the central 8.93%, and £20.4m only if the core reaches 7.5% later. The payroll saving is therefore about £2.1m in Year 1 and £10.1m by Year 5, not the £13.5m full implementation implies
• Corporation Tax: £180m at 30% rather than 25% costs about £9m more — paid, because the profit is real and UK-based
• Care contribution: 2% of qualifying earnings into the National Care Fund — about £4.1m a year on a 6,000-strong payroll, the largest single new cost in this example

Net effect: better off by roughly £26m in Year 1 and £34m a year by Year 5 — still a very large gain, but built mostly on Business Rates rather than payroll. Rates abolition is worth about £40m; the payroll saving builds from £2.1m to £10.1m as the Levy steps down; against that, the care contribution costs £4.1m, Corporation Tax £9m and the Landowner Levy several million. A genuinely domestic large employer was always going to pay close to the headline rate and gains most from abolishing the taxes that fall on physical presence

The Tech Multinational — UK profit booked offshore

A global technology group: roughly £2bn of UK sales, a modest UK headcount, and a UK tax bill engineered down toward a few tens of millions through royalty and licensing payments to a low-tax jurisdiction. The contrast with the manufacturer is the whole point.

• Business Rates and employer NIC savings are real but small relative to its UK sales, because its UK physical and human footprint is deliberately light
• Single Market access is a genuine, large gain — near-frictionless sales across 450 million customers
• Corporation Tax at 30% enforced through Pillar Two, plus tighter rules on profit-shifting, mean the UK profit it has long routed offshore is increasingly taxed where the sales actually happen — potentially a nine-figure swing
• Its globally-mobile principals resident in Britain meet the National Wealth Floor; its platform arms meet the new content-liability tiers

Net effect: the operating environment improves markedly, but the tax bill rises sharply — by design. For a multinational with real UK substance this is a trade worth making; for one whose UK presence is largely a tax address, F³ is the end of a long free ride. Verdict: the band that pays for the others — and the one we are least apologetic about.

Closing Statement

Answering the Critics

Every serious political programme faces serious political attacks. This section sets out the strongest challenges from across the political spectrum — and our answers to them. We publish this because honest politics requires engaging with the best version of the opposition’s argument, not the weakest.

The Left Critique

The arithmetic does not support this — and we publish it levy-inclusive so the reader can check. A minimum-wage worker gains £251 a year — thin, and entirely from abolishing Council Tax, because on the payslip alone that household is £921 worse off. We publish that rather than the flattering version. Employee National Insurance is abolished and the first £15,000 is untaxed. The gain at £300,000 is larger in cash — £7,314, after the 2% care contribution on earnings above £150,000 has taken £3,000 of it — and we print it. And we will not claim the ladder falls furthest at the bottom, because it does not: the effective rate falls 1.0 point at minimum wage against 2.4 points at £300,000. A single rate applied above a single threshold does that, arithmetically, and no amount of framing changes it. And the payslip is no longer the whole story: the 0.3% property charge is uncapped where Council Tax's Band H was a ceiling, so the £1.5 million house now contributes proportionally; the Wealth Floor binds above £10 million; and the housing gain settles at death. The top's total contribution rises. The bottom's falls. And every rung of the ladder still rises with income.

Effective rates, counting every tax and charge including Council Tax and the property charge that replaces it: 20.4% at £26,437 (down from 21.4% today); 27.8% at £40,000 (up from 23.6%); 35.4% at £80,000 (up from 31.7%); 38.0% at £120,000 (down from 38.9%); 41.9% at £300,000 (down from 44.3%). A flat rate above a generous threshold produces a rising ladder — a progressive system by definition — without the 60% taper trap today's 'progressive' bands hide in the middle.

Effective tax rates under F³ versus the current system, counting every tax and charge including Council Tax and the property charge that replaces it — lower at the bottom and the top, a named and narrowing cost in the middle, and a clean rising ladder at every rung.

And what we do not do is hide the middle. Between about £28,000 and £100,000 there is a transitional cost of roughly £30 to £58 a week in Years 1–3 before offsets — the visible price of a 2p ring-fenced Health Levy we name rather than disguise, and one most commuting households recover through the offsets we itemise. Trickle-down is when you cut taxes at the top and ask the bottom to wait. We cut them at the bottom first and ask the comfortable to read the bill.

The Red Wall Critique

The Pensioner Guarantee does not cut pensions — it uprates them with inflation every year, a real-terms guarantee written into law, with a five-yearly earnings review reported to Parliament. The Triple Lock's ratchet has added £22.9bn a year to state pension spending by 2029-30 relative to CPI uprating on the OBR's figures, restated in July 2026, three times the original 2012 estimate — a transfer from working households, many of them struggling, to pensioners, one in four of whom is a millionaire. We should be precise about what our reform recovers: not that accumulated £22.9bn, which is already in the pension and stays there, but the future differential, worth about £10bn a year by the end of the parliament. That is the regressive policy, not our reform.

Child Benefit is replaced, not removed. Free school breakfasts and lunches — universal for every child — plus uniform and holiday support for households under £80,000 deliver more direct benefit to children than cash that may or may not be spent on them. A lower- or middle-income family with two school-age children receives more under our provision model than under the current cash payment; higher-income households keep the universal meals their children eat every day.

And the charge that we shift the burden from employers to workers is arithmetically false. Employer NIC is replaced, not removed: an 8.5% Employer Levy remains on every payroll — 7.5% core plus the earn-back Training Point — with the employer pension-and-care minimum rising from 3% to 5% on top of it. The biggest winners from Step 1 are the lowest-paid workers; the biggest new obligations fall on the largest employers.

On immigration — we are opening doors for talent while closing routes for illegal entry. A system that welcomes a doctor from India, an engineer from America and a researcher from Germany while firmly returning those who arrive illegally is not hostile to working families. It is what a functioning immigration system looks like.

The Brexit Populist Critique

Britain voted to leave the European Union. We are not proposing to rejoin it. We are proposing to rejoin the Single Market — the economic arrangement, not the political union. Norway has been in the Single Market since 1994. Norway is not in the EU. Norway has its own parliament, its own laws, its own foreign policy and its own supreme court. Nobody seriously argues Norway has surrendered its independence.

On free movement — yes, we are honest that Single Market membership includes free movement of EU citizens, as it did before 2021. What changed in 2021? Net migration hit record highs. The boats crisis exploded. EU doctors, nurses and workers left. Free movement of EU citizens was not the cause of Britain’s immigration problems. Its removal demonstrably did not solve them.

On the numbers — because this is the real question, not free movement as such. Britain’s net migration is already high; the failure is in its composition, not its absence. Our birth rate has fallen well below replacement, the working-age population that funds pensions and the NHS is shrinking relative to the retired, and the country genuinely needs people. What it needs is the right people: skilled workers who pay in more than they draw out, the doctors and engineers and founders who build the tax base. So F³ does the thing the slogans never manage — it raises skilled, contributory migration while bearing down hard on the two categories the public actually objects to: illegal entry and unmanaged asylum. More of the migration that pays for the state; far less of the migration that strains it. That is not a number you wave at; it is a composition you engineer.

On stopping the boats — our policy is the only one that actually works. A returns agreement requires EU cooperation, and that cooperation is far easier to secure from inside the Single Market than outside it. Every alternative — Rwanda, offshore processing, legal challenges — has failed. We are offering a solution, not a slogan.

On ‘four taxes dressed up as one’ — the combined rate on your payslip above £15,000 is 42%: 38% income tax, 2% local income tax, 2% Health Levy. We print that number in every household example in this document. Below £15,000: nothing. The one charge that sits off the payslip — the 0.3% property charge that replaces Council Tax — is billed to owners once a year and printed in the same examples. One payslip line, no NIC tables, no Council Tax bands, no taper. We will take a published 42% over today’s disguised 62% taper zone every single time — and so will anyone who has ever tried to read their own payslip.

On trade independence — Britain keeps CPTPP. On the rest we should be careful with the verb: Britain would seek to preserve its existing trade agreements, and we do not present that as automatic. EFTA's own position is that accession to the EEA is not automatic and that entry into EFTA's existing agreements must be negotiated with the consent of the other parties, so continuity is a negotiating objective rather than a legal inheritance, and it should be assessed agreement by agreement. Future deals are negotiated through the EFTA framework alongside Norway, Iceland and Liechtenstein. The EU takes 42% of British exports. CPTPP takes 8%. The arithmetic of where Britain's trade interests lie is not ambiguous.

Three More Attacks — Answered Before They Are Made

An honest platform red-teams itself. Here are the three strongest attacks we expect that the sections above do not cover — and our answers.

“A pay cut for nurses to fund a pay rise for bankers”

The poster writes itself: a senior nurse on £80,000 pays about £58 a week more in Years 1–3 before offsets, while a banker on £300,000 gains £7,314. Our answer is printed in our own tables. The gains at the top come from abolishing the taper and allowance-withdrawal distortions, and they are smaller than they look: the 2% care contribution on earnings above £150,000 takes £3,000 of that banker's gain, which is why the figure is £7,314 and not the £10,314 the payslip alone would show. And we will not claim the ladder falls furthest at the bottom, because counting every tax and charge it does not: the top falls from 44.3% to 41.9%, the bottom from 21.4% to 20.4%. The middle's cost is the Health Levy — half of it ring-fenced for named NHS improvements, half replacing general funding — named either way, and the offsets — Rail Pass worth up to £3,200 for a full-time intercity commuter, which is a maximum rather than a typical figure, school provision worth ~£850 a child, energy VAT, a property charge typically half the old Council Tax bill — are real money against it. From Step 4 the £80,000 cost narrows to £1,077 a year — £21 a week. The banker's £2 million house pays the uncapped 0.3% where Band H once shielded it. And above £10 million, the National Wealth Floor is designed to ensure — for the first time — that those at the very top cannot structure their way to a lower share of their income than the nurse on the poster. And if the country judges the middle's contribution too high, the honest lever is a Health Levy threshold — a choice we cost at roughly £8bn, openly, rather than pretending it is free.

“Seven takings from pensioners — and nothing given”

The list is real and we own it: triple lock, the universal winter fuel payment, universal bus passes, IHT at the flat-tax rate above £3m, the insurance surcharge for high earners. Now the other column, which the attack always omits: the £86,000 Dilnot cap on lifetime care costs, reinstated — the single largest pro-pensioner commitment on offer from any quarter, protecting every pensioner from the catastrophic costs that today consume entire family homes; a housing gain that is never taxed for moving and settled only once, in real terms, at death — so the equity that funds later-life care is never eroded by a transaction tax; and a state pension guaranteed in law never to fall in real terms, with a five-yearly earnings review. We ask wealthier pensioners to give up universal perks. We give every pensioner the protection no government — including the ones that promised it — has ever actually delivered.

And there is a principle beneath the trade, not just a ledger. The pensioner perks were always a strange kind of policy: a benefit you qualified for by age, not by need. F³ replaces benefits-by-demographic with benefits-for-everyone. VAT comes off energy — for every household, including every pensioner. Trains and parking get cheaper — for everyone, including every pensioner. Disability support is protected — for the pensioners who actually need it. The question is whether the state should hand a universal payment to a millionaire because of his date of birth, or take that money and make energy, transport and care cheaper for the whole country. We think helping everyone beats subsidising one age group regardless of need — and most pensioners, asked plainly, think so too.

“The EU hasn’t agreed to any of this”

Correct — and the architecture assumes nothing. Steps 1 and 2 are unilateral in the sense that matters here: employee NIC abolition, the replacement of employer NIC by the Employer Levy, the Business Rates phase-out, planning, courts, energy, schools and welfare reform all proceed on Parliament's timetable alone, with nobody's permission required. One thing inside Step 2 is not a matter of timetable and we should not let the word carry it: the Levy's core rate step-down from 11.25% to 7.5% is gated on certified headroom, so it is unilateral as to permission and conditional as to money. Single Market accession is Step 3, on its own diplomatic track, with the consent of the EU and the existing EFTA states required and acknowledged — and the EEA fee of roughly £3bn a year costed openly in our scorecard. The LOW scenario in our fiscal framework is precisely the world in which Brussels says no or says slowly: no Single Market dividend, no Step 4, no Levy step-down and no EEA fee either. In that world borrowing runs above the OBR's published path for four years and ends about £1bn above it. Our plan is not contingent on Brussels' goodwill. It is improved by it.

Part Two — Getting There: Compatibility, Sequencing and the City

Part One made the case for rejoining the Single Market. This second part turns to delivery: whether the rest of this paper’s programme is compatible with membership, what happens if Brussels is slow or says no, and how the City is protected in the process.

Single Market Compatibility: What F³ Has Checked

A credible Single Market rejoining argument requires honest acknowledgment of which F³ policies need adjustment under EEA rules. We have conducted a full compatibility audit across all 22 chapters. Here are the findings:

The vast majority of F³ policies are fully compatible with Single Market membership. Three required adjustment: VAT threshold redesigned as a two-tier flat rate scheme, the British ISA removed (the universal £15k capital-gains allowance achieves the same goal more powerfully), and STEM free degrees extended to EEA students on equal terms. This is what serious policy development looks like.

Sequencing: If Brussels Says No — Or Says Slowly

Nothing in Steps 1 and 2 requires anyone’s consent but Parliament’s. Employee NIC abolition, the Employer Levy, the Business Rates phase-out, planning reform, the courts programme, energy, schools and welfare reform all proceed unilaterally, on our timetable. Single Market accession is Step 3, on its own diplomatic track — and we acknowledge plainly that it requires the consent of the EU and of the existing EFTA states, whose concerns about admitting an economy ten times Norway’s size we will address directly rather than wish away.

The LOW scenario in our fiscal framework is the protracted-or-failed-negotiation case: no Single Market dividend is booked, the EEA fee never starts, Step 4 does not fire and the Employer Levy core never steps down. Borrowing runs above the OBR's published path throughout and ends about £1bn above it. We should be careful not to over-claim here, because an earlier draft of this paper did: the Low case ends closest to the official path of our three scenarios, not furthest below it, and after our savings and family-allowance re-costings it ends slightly above. Never firing the giveaways protects it from their cost, but it does not compensate for the growth it never gets. Our plan is not contingent on Brussels' goodwill. It is improved by it.

The City: Equivalence, Autonomy and the Trade

Honesty about financial services cuts both ways. The 'equivalence' Britain was promised never arrived: Brussels granted only a time-limited decision on clearing — because EU firms need London’s clearing houses — and withheld the rest as leverage. Waiting for equivalence was a strategy of hoping the counterparty would be generous. It was not.

Meanwhile, regulatory autonomy delivered real wins we do not pretend away: Solvency UK released insurer capital for investment, the bonus cap went, the listing rules were rebuilt for founders, and the share-trading obligation was scrapped. The exodus never matched the forecasts — a few thousand roles moved, not seventy-five thousand — though EU share trading left for Amsterdam and some derivatives business for New York.

EEA membership replaces the equivalence mirage with the thing the City actually asked for in 2016: passporting — the right to sell from London into every EEA market without a subsidiary in Dublin or Frankfurt. The cost is re-adopting the EEA-relevant financial rulebook and constraining parts of Solvency UK. We name the trade rather than hiding it — and we make it, because passporting into the world’s largest capital market is worth more than autonomy over rules the EU’s own Listing Act is now copying from us. But the trade is negotiated, not swallowed whole: the EEA Agreement provides for adaptations and derogations, and the EFTA pillar gives non-EU members a distinct voice. Two specific carve-outs are named accession objectives — transition arrangements for capital already deployed under Solvency UK, and removal of the bankers’ bonus cap. On remuneration we have the stronger argument: the cap is contested even inside the EU as a matter of proportionality and member-state competence, several EEA states apply it only grudgingly, and Britain has already shown the sky does not fall without it. We will press that case rather than concede it by default.

It is worth saying, too, about what passporting will and will not do — because the City of 2027 is not the City of 2016. In the years since Brexit, firms did not wait for a deal: they built subsidiaries in Dublin, Paris, Amsterdam and Frankfurt, moved staff and capital to them, and shifted whole business lines — euro-denominated share trading chief among them — out of London for good. Regained passporting will not simply reverse that. Some of what left has put down roots and will stay where it is, and a part of the City has made its peace with a more autonomous, divergent future and is in no hurry to re-adopt an EU rulebook it no longer has a vote on. We do not pretend, then, that this is the City clamouring to be rescued. What we offer is narrower and more honest: the removal of a real and continuing disadvantage — the loss of frictionless access to the world’s largest capital market, replaced by an equivalence regime that never materialised — for those firms, activities and future growth that are still London’s to keep or to lose. London’s pre-eminence was damaged by Brexit but never broken; the case here is for stopping the slow bleed and competing from a position of restored access, not for turning back the clock.

Annex: The Scorecard, Line by Line

This is the whole of it. Eighty-one measures, grouped into the nine blocks used in Chapter 2, each with the note that explains how it is scored. Convention throughout: a minus sign improves the deficit; a plus sign is a cost. Block nets and the net revenue effect are computed in the fiscal model from unrounded figures, not from the rounded lines printed here, so a column of displayed measures can differ from its stated block net by up to about £0.1bn. Where the source supports more precision than £0.1bn — the small relief abolitions in the ninth block, which HMRC publishes to the nearest £5m — we print it, so that block adds exactly. Nothing is dropped between the model and this page; what varies is only how many decimals each source can honestly carry.

Where a measure carries no note it is because the measure is self-explanatory or the reasoning sits in the relevant chapter. Two lines are deliberately scored at zero rather than as savings — HMRC compliance yield and the health saving from private cover — and one, the Training Point, is ring-fenced and nets to nothing by design. Those are choices, not omissions, and they are argued where they arise.

Annex: Sources and Evidence

The claim on the cover is that every figure comes from public data. This annex is where that claim is tested. Every externally-derived number in the fiscal model appears below with its reference code, its value, its year and its source, in the form a reader needs to go and check it. Where a figure is contested, we say so; where we have declined to adopt a source, we say that too.

Three conventions apply throughout. Figures are £bn a year unless stated. Where a source gives a range we take the cautious end. And where a number rests on a behavioural assumption rather than a published statistic, it is not in this table at all — it is argued in the chapter that uses it, and in most cases it is scored at zero.

Tax base, receipts and reliefs

What we have deliberately not sourced, because we do not book it

Four numbers that other papers would bank appear nowhere in the table above, because we do not claim them. HMRC compliance yield — real, but a behavioural projection rather than a statistic. The NHS saving from private-cover uptake — the mechanism reaches perhaps £1bn against the £3–9bn such shifts are usually credited with, so we score it at zero. The productivity gain from a simpler tax code — real, unmeasurable, unbooked. The output effect of clearing the road-maintenance backlog — same. Each is named in its chapter and excluded from the arithmetic.

One further exclusion is worth stating plainly because it cuts against us. The **thorium molten-salt claim of 12GW by 2035* is the vendor’s, is reproduced as the vendor’s, and is adopted nowhere: no such reactor has been licensed or operated commercially anywhere in the world, and we book none of the £8bn saving attached to it.*

Future Forward Foundation

Building Prosperity. Securing the Future.

Future Forward Foundation · A Policy Paper 1