What we could put back · if we stopped paying margin to private contractors for NHS services.
A calculator. Audited contractor margins applied to the outsourced spend in your district, your trust, or the whole country. The number isn't a campaign slogan — it's a band, with sources after every line, and you can see the working.
"A director of a company must act in the way he considers, in good faith, would be most likely to promote the success of the company for the benefit of its members as a whole."
— Companies Act 2006, section 172(1) · the duty applies to every director of every limited company contracted to the NHS
A director who voluntarily reduces their NHS-contract margin — without a commercial reason — is arguably breaching that duty. This is not malice. It is statute. The Better Business Act campaign has been pushing to add stakeholder duties since 2020; it has not passed.
Until then, every £1 of NHS spend going to a for-profit entity faces a structural extraction pressure that does not exist when the same service runs inside an NHS trust. That's the premise. The rest of this page is the receipt.
verified inputs
The numbers that survived adversarial verification.
Each row was sourced from primary audited filings or government press releases, then independently challenged 3 times. Only what passed 2/3 votes is below. Killed claims are listed openly in the methodology section so we don't repeat them.
Source
Period
Revenue
Margin
Verdict
Practice Plus Group · NHS 111, OOH, surgical hubs, prison health
FY24
£599.3m
9.1% EBITDA
verified
Practice Plus Group dividend to shareholders (FY24)
FY24
£34.1m
—
verified
Spire Healthcare · NHS-funded elective + private
2024
£1,511.2m
9.1% operating
verified
Care UK · NHS-funded residential
FY23
£490.5m
6.1%
verified
NHS clinical activity outsourced (doubled since 2016/17)
NHS PFI unitary charges (1.7% of DHSC cash budget)
2016/17
£2.0bn
—
verified
NHS Property Services · OCS cleaning insourcing
Dec 2018 →
1,692 staff
£41.5m / 5yr projected
verified
GMB FOI · NHS non-clinical contracts (under-half response rate)
—
£1.83bn floor
—
verified
What we do NOT publish: the "£6.7bn extracted · £10m / week · £558m / year" campaign headline was killed in 3-vote adversarial verification because the methodology proxies company-wide margins onto NHS-specific revenue (the framework is fair, the output is not defensible). The "105 NHS PFI / £12bn capital" stat was also killed. We use audited margins only.
the calculator
Run it for any scope · district, ICB, or whole country.
Defaults below are pre-loaded with verified national 2023/24 figures. Adjust any row for your district. Floor = audited margin × verified spend. Ceiling adds plausible categories that need fresh 2024 sourcing (we flag them).
£m
£m
£m
£m
£m
£m
Floor reinjection (verified categories only)—
Ceiling reinjection (verified + plausible)—
10-year cumulative · floor → ceiling—
As % of NHS England budget (~£200bn baseline)—
10-year cumulative reinjection · ceiling scenario
staged unwind · contracts roll off as they expire
action
Write to your MP · auto-filled from your calculator inputs.
Show your working — or this is just another campaign slogan.
What's verified
Every figure flagged v was sourced from a primary audited filing, NAO report, DHSC press release, or peer-reviewed paper. Each claim was challenged 3 times by independent verifier agents; only 2/3 confirmations survived.
Rows flagged est use industry-standard figures or older verified figures with reasonable inflation. Before publication of any specific district paper, these need re-sourcing from 2024/25 NAO, HoC Library briefings, or contractor statutory accounts.
The biggest open questions: total NHS budget 2024/25 breakdown · current PFI annual + portfolio · FM contractor NHS-segment margins · IT contractor NHS margins · consultancy 2024 spend.
Killed claims · the numbers we refuse to publish
Claim
Why killed
"£6.7bn extracted from NHS · £10m / week · £558m / year (2012-2024)" — We Own It campaign headline
Methodology proxies company-wide margins (5.18% from Moody's BvD) onto £130.7bn of NHS contract value. Framework is fair, output is not audited NHS-segment profit. refuted 1-2
"105 NHS PFI projects, £12bn capital value" — widely cited
Did not survive primary-source check. Current portfolio needs fresh sourcing. refuted 0-3
"Practice Plus Group owned by Bridgepoint"
Widely reported, did not survive primary-source check on FY24 statutory accounts. Re-verify before asserting. refuted 1-2
"Spire hospital segment 18% EBITDA, 2024"
Not present in audited filing. Group operating margin of 9.1% DOES hold. refuted 0-3
Honest caveats
TUPE protects staff but it also protects the cost base. Insourcing doesn't automatically cut payroll — it captures the margin and the markup above payroll.
PFI termination is brutal. Treasury calculations typically pay the SPV the NPV of remaining unitary charges. The realistic PFI play is wait for contract expiry (most end 2030-2042).
In-house capability gaps are real. Some specialist categories (neurosurgery, complex genomics, advanced imaging AI) genuinely cannot be insourced because the NHS has lost the skill.
Quality is not automatic. Peer-reviewed evidence (PMC5267843) links cleaning outsourcing to higher MRSA rates — but quality drops happen on both sides of the boundary. Honest analysis requires a counter-narrative section.
The Streeting agency savings are partly offset by bank-shift substitution at higher unit cost in some trusts (e.g. Imperial College Healthcare bank shifts averaging £5,509 vs £2,116 agency in 2025/26 sample).
how it actually happens
The implementation plan for one ICB.
Same template fits Mid & South Essex, Hampshire & Isle of Wight, Lancashire & South Cumbria, or any struggling district with broad outsourcing and public procurement data.
Phase 0 · Discovery (months 1-6)
Full procurement audit · every contract over £100k, vendor, expiry, break clause. Publish vendor margins from latest Companies House filings on the ICB website. Map TUPE register: how many staff transfer in by category. Capital ask: kitchens, cleaning equipment, fleet, IT licences direct.
Phase 1 · Low-friction (months 6-18)
Cleaning · catering · non-emergency patient transport · interpreting · bank-staffing scale-up to displace commercial agency. Welsh insourcing toolkit categories. Expected reinjection: 3-5% of contracted spend, growing through years 2-3.
Phase 2 · Specialist + clinical (months 18-36)
Community services contracts as they expire · mental health bed contracts where capacity can be built · lab / pathology JVs renegotiated. Expected reinjection: ~£40-80m / year cumulative by end of year 3 in an average ICB.
Phase 3 · The hard problems (months 36-60)
PFI renegotiation OR planned hand-back at expiry (patience > termination penalties). IT on natural renewal. Consultancy starved: cap external consultancy at 0.5% of trust spend · build in-house transformation team.
Capital ask for one district: historical NHSPS comparator suggests ~£8-10m one-off setup to bring 1,692 staff in-house. For a full district touching all categories, plan £25-50m one-off investment with payback inside 3-4 years on the verified categories alone.
questions
Honest answers.
Why not just publish the £10m / week number every campaigner uses?
Because it was killed in adversarial verification. The framework (margin × contract value) is sound. The specific output (£6.7bn, £10m/week, £558m/year) proxies company-wide margins onto NHS-only revenue — that's not the same as audited NHS-segment profit. If we publish that number, the first analyst from the Big 4 to read it will use it to discredit everything else we say. We use audited margins only.
Are you saying private healthcare workers are the problem?
No. Cleaners, agency nurses, IT contractors all do the same work as in-house staff and largely deserve the same pay. The argument is that the margin layer above their wage — directors' duty under s.172 to extract profit for shareholders — is structurally taking money out of NHS care. Bringing services in-house captures that margin layer, not the wages.
Won't TUPE just preserve the same cost base?
Partly yes — that's the honest answer. TUPE moves staff in on existing terms, so day-1 payroll is the same. What you capture is the margin and overhead above payroll. NHS Property Services' 2018 cleaning insourcing projected £41.5m over 5 years on this basis. Realised data is mixed (~£2m in year one of one sample). The principle is proven; the magnitude is debated.
Why can't we just terminate PFI?
Treasury's standard PFI termination clauses typically pay the SPV the net present value of all remaining unitary charges. So termination costs nearly as much as completion. The realistic play is to plan for contract expiry (most end 2030-2042) and not renew. That's why the calculator's PFI recovery slider defaults low — we don't want to overclaim what's recoverable in the near term.
What about the quality argument? Won't insourcing make things worse?
Sometimes. Honest answer: peer-reviewed evidence (PMC5267843) links outsourced cleaning to higher MRSA rates in some trusts. But quality drops happen on both sides — the GMB, RCN, and UNISON have documented two-tier T&Cs (e.g. ISS Mediclean staff on statutory-only sick pay vs NHS 6 months full + 6 months half) that hurt continuity and morale. We need both sides of the evidence base in any publication.
Has any country actually done this?
Wales has a formal insourcing toolkit anchored in the Wellbeing of Future Generations Act, prioritising FM, catering, ICT and social care. Madrid cancelled its hospital PFI in 2014. New Zealand merged its 20 District Health Boards into Te Whatu Ora in 2022. None of these is a perfect comparator but all show insourcing is a real, legal, policy-tested option.
Who built this?
Simon Gant · part of the GroundLevel sovereign-tools estate (groundlevel, groundlevel-sdk, etc.). Source is on GitHub, MIT licensed. Methodology is auditable. Open an issue if you spot a stale source.