London already levies exactly the instrument Peter Murray OBE says the development industry needs: a fixed charge, published years in advance, priced per square metre, payable by every builder regardless of size. The Mayor sets it personally. It raised a fraction of the money that matters, and by law it cannot be spent on affordable housing. The uncertainty the urbanist describes is real, and the small builders it removed are real, but the charge he proposes already exists in the one form that cannot touch the problem.
That distinction is the whole argument, and it is usually skipped. When the urbanist set out his policy platform for the May 2028 mayoral election in a conversation recorded on 14 August, he built the case around a scene rather than a statistic.
What does a small builder actually ask a bank for?
The scene is a credit conversation, and it is the most precisely constructed thing he says.
“So if you’re a small builder, you say to your bank manager, can you give me half a million pounds to buy a site? And the bank manager says, well, how long do you need the money for? Oh, I don’t know. It might take one year, might take two years to get planning permission. And then the bank manager, well, how much do you need, really? And he said, well, we don’t know because. The planning authority only tells us towards the end of the project how much they’re going to charge us in what here we call Section 106, which is quasi taxation on projects.”
Read that as a lender rather than as a planner. A bank is not being asked to accept risk, which it does daily. It is being asked to size a facility against a liability with no known value and no known date. Term and quantum are the two variables a development loan is built from, and the borrower can supply neither. The loan is not declined because the scheme is bad. It is declined because it cannot be priced.
“And so the bank manager says, well, you know, I can’t lend you money like that.”
The consequence he attaches to that is a collapse in who builds Britain’s homes.
“Forty years ago, 40 per cent of homes in England were delivered by small builders, small developers. They’ve been wiped out by the bureaucracy and the uncertainty.”
The figure needs correcting in two small ways, and the correction strengthens it. The peak year is 1988, which is 38 years ago rather than 40, and the share is 39%, falling to about 10% by 2020, according to the House of Lords Built Environment Committee. The same collapse shows up in company counts rather than percentages: Savills research puts at least 12,000 active housebuilders in that 1988 peak against 2,423 recorded in the third quarter of 2024, with average output per builder rising from 20 homes a year in 1988 to 84 in 2023. Fewer firms, each building more, consolidation driven by fixed compliance costs rising faster than scheme size.
The mechanism is not mysterious, and it is not our inference. The National Audit Office found that viability assessments, the appraisals that determine how much a developer actually pays, can be manipulated, are difficult for planning authorities to challenge, and reward scale directly: “larger developers have the resources to employ consultants and legal experts to find ways to negotiate contributions down.” A negotiated obligation is a contest between professional teams. A builder doing nine units does not field one.
The delay half of the parable is harder to pin down, and it is worth being honest about why. There is no official median. The Ministry of Housing, Communities and Local Government publishes speed of decision as threshold percentages, not as an average duration, and its most recent quarterly release records 91% of major applications decided within 13 weeks or within an agreed extension, against 19% inside the statutory 13 weeks with no extension at all. The gap between those two numbers is the extension, and the extension is where the year goes. On the builders’ own account, 51% of small housebuilders reported waiting more than a year for permission, in a Home Builders Federation survey. That is self-reported evidence, not an administrative record, and should be read as such.
Two further pieces of evidence complicate the parable, and both belong beside it.
The builders themselves do not rank the problem the way the scene implies. The Federation of Master Builders’ 2025 survey of small housebuilders found 67% saying sites they were interested in were unviable because of the Section 106 or CIL obligations they anticipated, with around 31% naming the cost of those obligations as an impediment. That is the size of the charge, not the lateness of learning it. Brian Berry, the FMB’s chief executive, framed the consequence when the survey was published in December 2025: “This decline is not just bad for local house builders, it is bad for consumer choice, it is bad for the range of design being made available, and it is bad in terms of delivery.” A charge known at the outset but still too large does not bring those sites back.
The lending behaviour in the parable may also not be the lending behaviour in the market. The closest study to the question, research for the Royal Town Planning Institute by LSE London and the Bartlett in 2018, interviewed developers and financiers and found that “banks generally will not lend on development schemes until after planning permission is secured, so planning risk is not an issue for them.” On that account a lender never prices the uncertainty, because it waits until the obligation is settled and lends against a known figure. The same research did find the idea worth testing for precisely the tier of builder at issue, noting that small and medium-sized firms “could benefit most from a zoning-type system, if it meant that lenders were more willing to provide finance”. It is eight years old and predates this episode. It is also the nearest thing to evidence there is, and it does not straightforwardly support the scene.
Has fifty years of development taxation actually failed?
His verdict is unambiguous.
“And they have never in 50 years managed to find a particularly successful way of doing it.”
As a description of the ambition to capture land value through a single national tax, that holds. As a description of the system now operating, it does not. Section 106 and the Community Infrastructure Levy together accounted for developer contributions worth around £5.5 billion agreed in 2022 to 2023, on the Ministry’s own analysis, and Section 106 alone delivered around 27,700 affordable homes, 44% of all new affordable housing, in 2023 to 2024. A system moving those sums is open to the charge that it is opaque, slow and regressive towards small firms, which is the urbanist’s actual case and a strong one. It is not open to the charge that it failed.
The accurate framing is narrower and more damning: repeatedly attempted, never settled.
- Community Land Act 1975: Labour, in force from 1975. Abolished, repeal completed 1983. Repealed via the Local Government, Planning and Land Act 1980, completed by a 1983 order. Secondary planning-history sources attribute this to the unworkability of councils buying and reselling development land; no government post-mortem was located.
- Development Land Tax 1976: Labour, in force from 1976. Abolished 1985. Repealed by Finance Act 1985, section 93, for disposals from 19 March 1985. Hansard records the abolition costing the Exchequer £50m, a straightforward reversal rather than a stated technical failure.
- Section 106 planning obligations: Conservative, in force from 1990. Still operating, universal. Negotiated site by site. Delivered 44% of new affordable homes in 2023 to 2024. This is the instrument Peter proposes to replace.
- Planning Gain Supplement: Labour, proposed, never enacted. Abandoned before enactment, 2007. Recommended by the Barker Review in 2004, dropped in the October 2007 Pre-Budget Report in favour of a tariff, which became CIL.
- Community Infrastructure Levy: Labour, in force from 6 April 2010. Still operating, adopted by 162 of 308 planning authorities (52%) as at November 2024. A fixed, published rate per square metre. Transparent and poolable, but cannot be spent on affordable housing.
- Mayoral CIL: Mayor of London, in force from 1 April 2019 (current schedule). Still operating, London-wide. £25 to £80 per square metre by borough band, up to £185 for offices in central London. Restricted by regulation to transport, and applied to Crossrail.
- Infrastructure Levy: Conservative, never commenced. Abandoned before enactment, July 2024. Legislated in 2023 to replace CIL and largely absorb Section 106. No regulations were ever made; the Ministry confirmed in July 2024 it would not proceed.
Two rows in that table are not tombstones, and they are the two that decide whether the proposal is new.
What would a fixed charge actually change?
A fixed, knowable, published development charge is not a gap in the English system. It is CIL, and it has been running for sixteen years.
CIL is set by charging schedule, expressed in pounds per square metre, published before an application is made, and payable without negotiation. A builder can read the rate off a document, multiply it by floorspace and put the answer in a lending appraisal before buying the site. It does precisely what the parable asks for.
London goes further. The Mayor is himself a charging authority under Part 11 of the Planning Act 2008, and the current Mayoral CIL charging schedule was approved on 4 February 2019 and took effect on 1 April 2019. It charges £80 per square metre across eight central boroughs, £60 across nineteen more and £25 across the outer seven, indexed annually to a construction cost index. Not only fixed, but forecastable.
So the mayoral lever the urbanist would inherit is already a flat, transparent, London-wide development charge that he alone sets. Which raises the question his platform has to answer: if that instrument exists and he would control it, why has it not solved the problem he describes?
The answer sits in two regulatory restrictions, and they are the article’s finding. First, CIL in general cannot be spent on affordable housing; Section 106 remains the only route. Second, Mayoral CIL is narrower still, restricted by regulation to transport, which is why its proceeds fund Crossrail rather than homes.
In 2018 to 2019, the last year with a full published breakdown, affordable housing accounted for £4.675 billion of the £7.0 billion of contributions agreed, against £830 million for local CIL and £200 million for the Mayor’s. The fixed charge is roughly a seventh of the negotiated one. The uncertainty a lender cannot price is concentrated almost entirely in the portion a fixed charge is legally forbidden to carry.
The current system is not accidentally arranged that way. When the Mayor set the 2019 schedule, he declined to offer relief where Section 106 sums exceed Mayoral CIL, on the reasoning that viability problems arising from the combined demands “of Mayoral CIL and section 106 agreements” are better addressed “by making any necessary adjustments to the latter.” In other words: the fixed charge stays fixed, and the negotiated obligation absorbs the shock. Certainty for the levy is purchased with uncertainty in the obligation. Some part of the system has to flex, and the design decision was that it would be the affordable-housing side.
Which means the real proposal is not “introduce a fixed charge.” It is “fold the affordable-housing obligation into the fixed charge.” That policy has a name, statutory authority and a death certificate. The Infrastructure Levy was legislated in Part 4 of the Levelling-up and Regeneration Act 2023 to replace CIL and largely absorb Section 106, and the Ministry’s stated benefit for it was, almost word for word, the urbanist’s argument: preventing developers from negotiating down their contributions by setting the levy at a fixed level. No regulations were ever made. In July 2024 the incoming government confirmed it would not proceed, after stakeholders raised concerns about the levy’s design and its transitional arrangements.
That does not make the idea wrong. It does mean any candidate proposing it is picking up an instrument dropped on grounds of design, and owes an account of what he would do differently.
Would a fixed charge raise less money?
His own claim is that it would not.
“What I’m suggesting is that actually there is a simple taxation system which the developer pays exactly the same amount of money, but he knows how much it is at the start.”
Revenue neutrality is the hard part, and the arithmetic is unforgiving in three places.
- A flat rate applied to a variable-margin market either sets the rate low enough for marginal sites, which loses revenue on strong ones, or high enough to capture value on strong sites, which stalls the marginal ones. Negotiation exists to price that variance. Removing it does not remove the variance.
- Affordable housing and infrastructure would compete for one pot. Section 106 currently delivers 44% of new affordable homes, and 27,700 units is not a number a levy can quietly underdeliver.
- Collection is not the binding constraint. Local authorities were already sitting on more than £8 billion of unspent contributions, on a Home Builders Federation estimate cited by the National Audit Office, while 86 of 308 planning authorities had an up-to-date local plan as at February 2025, down from 149 in 2019. A charge that is easy to calculate still needs a plan to spend it against.
The objection to folding affordable housing into a single charge is not hypothetical, and it has been made in these terms before. Hugh Ellis, director of policy at the Town and Country Planning Association, said when an earlier government floated replacing Section 106 that “Section 106 is all we have got”, and that to “rip it up” would mean “you will deliver no affordable housing”. It would be, he said, “very reckless to rip up the Section 106 regime, which is delivering the majority of the country’s affordable homes, unless you’ve got a very detailed alternative ready to go.” He said that in August 2020, before the Infrastructure Levy was drafted, legislated and dropped. The five years since have run the experiment he asked for, and the detailed alternative was never ready.
There is a counter-argument, and it is the strongest thing in the urbanist’s case. Revenue neutrality measured against today’s output is the wrong test, because today’s output is the problem. If a knowable charge restores lending to the tier of builder that has fallen from 39% of supply to about 10%, the base grows. A slightly lower take on considerably more homes is not a loss. That proposition is testable, and nobody has tested it, because the only instrument designed to try was withdrawn before a single regulation was written.
He is candid that the design is unfinished.
“Well, the election is two years away, and I’m still working on the process for larger projects and mixed use projects.”
The certainty argument is sound and the small-builder evidence supports it. The instrument is not new, the Mayor already holds a version of it, and the version he holds is barred by regulation from the one function that would matter. Between now and May 2028, the platform needs three things it does not yet have: a rate that survives contact with viability across 33 boroughs, a lawful route to make a fixed charge fund affordable homes, and an explanation for why the last attempt at exactly this failed. The first two are technical work, arithmetic and statute. The third is not going away on its own: the Infrastructure Levy collapsed only twenty-five months ago, and no candidate has yet said why this attempt would fare differently.
What this article could not establish
The correction to his own figures holds up, and the regulatory bar is documented: a fixed, published, per-square-metre charge already exists in London, and it is barred by law from touching affordable housing. The causal step his platform depends on is a different matter, that a bank which will not lend against Section 106 uncertainty would in fact lend once the number were fixed instead. No study tests it. The instrument built to test it was withdrawn before a single scheme used it, and the nearest research, the RTPI work above, points the other way, reporting that lenders wait for permission rather than pricing the uncertainty. Restoring lending to the tier of builder that fell from 39% to about 10% is the claim the proposal rests on, and it is the claim with the least evidence behind it.
We also went looking for an independent planning economist willing to say on the record whether a single flat rate can be viable on marginal London sites without being value-destructive on strong ones. We did not obtain one in time for publication. The closest published answer is a 2023 study for the housing ministry by LSE London and the University of Liverpool, co-authored by Professor Christine Whitehead, modelling the Infrastructure Levy. It concluded that “rates that might be appropriate for one type of development may crowd out others”, and that the trade-off turns on whether the goal is to maximise proceeds or to shape what gets built. That is the same tension, measured, and left unresolved.
These arguments do not settle in print. RealTimes convenes b2bv2v26 London on Tuesday 8 September, doors open at 14:00, across three canvases: Vision, opened by Peter, then SkyScraper Dialogue and Value. Tickets at RealTimes.co.
How this article was sourced
Every quotation above is verbatim from the conversation recorded on 14 August 2026. He states an intention to stand in 2028; he is not yet a candidate, and RealTimes does not characterise him as one.
RealTimes convenes the event he opens on 8 September, which is a conflict a reader is entitled to weigh. This piece disputes the central plank of what he proposes. It was written by the editorial desk, was not submitted to him for approval, and any correction or response he wishes to make will be published here in full.
Frequently asked questions
- What is Section 106? A planning obligation under section 106 of the Town and Country Planning Act 1990, negotiated case by case between a developer and the local planning authority as a condition of permission. It is the only route through which affordable housing is secured, and it delivered 44% of England’s new affordable homes in 2023 to 2024.
- How does CIL differ? The Community Infrastructure Levy is a fixed rate per square metre published in advance, not negotiated. It is discretionary for authorities, adopted by 52% of them as at November 2024, and it cannot be spent on affordable housing.
- Why did small housebuilders disappear? Their share of new homes fell from 39% in 1988 to around 10% by 2020. The negotiated contributions system rewards scale, because larger developers can fund the consultants and lawyers who negotiate obligations down, while the timing of the obligation makes small schemes difficult to finance.
- What would a fixed development charge look like? In practice it already exists. Mayoral CIL charges £25 to £80 per square metre by borough band and up to £185 for central London offices, indexed annually. The proposal at issue is extending that model to cover the affordable-housing obligation, which no current instrument does.
- Has a fixed charge been tried anywhere? In England, yes and recently. The Infrastructure Levy was legislated in 2023 to replace CIL and absorb most of Section 106, precisely to stop contributions being negotiated down. It was never commenced and was abandoned in July 2024.