A hypothetical worked example that shows why London’s brownfield pipeline has three sequential problems — and why fixing one without the others changes nothing
Meet David and Sarah. They have run Meridian Autos, a busy car repair workshop in south-east London, for fifteen years. The site is roughly half an acre. It is an eight-minute walk from a Zone 2/3 station. Their Dad bought the site many years before for £200,000. It is now worth around £1 million as a going concern. They are not looking to sell. But they know, in the back of their minds, that one day a developer will knock on the door.
This is the story of what happens when that developer knocks — and why, under London’s current planning and viability framework, the conversation ends before it starts.
The Site
Half an acre. Zone 2/3. Eight minutes to the station. In any rational housing market, this site could be delivering homes. It is brownfield. It is sustainably located. It is exactly the kind of parcel the London Plan says should be prioritised for residential development. The GLA’s own housing trajectory almost certainly includes sites like this in its five-year supply figures.
A developer runs the numbers. Four to six storeys — prevailing height around four, so there is some scope to push it, but not much further without triggering the kind of gateway processes that add years to a programme. Thirty-five units. Average size circa 720 square feet. Build cost at £335 per square foot — inner London, not central, realistic for a mid-quality scheme. Plus £500,000 for abnormals: remediation of the workshop pit, contamination clearance, site-specific infrastructure.
Total construction cost: £8,942,000.
Problem One: The Land Won’t Trade
The GLA’s viability methodology requires the developer to establish a Benchmark Land Value — the minimum price at which a rational landowner will sell. Under EUV+, that benchmark is the Existing Use Value plus a premium to incentivise the landowner to transact. On a commercial site of this type, the GLA might apply a 15% premium above EUV.
EUV: £1,000,000. Premium: £150,000. BLV: £1,150,000.
Now ask David and Sarah what they actually need to walk away.
The site is worth £1,000,000 in existing form. But selling it is not the same as receiving £1,000,000. They bought it for £200,000 many years ago. The gain is £800,000. As an asset sale by the individual shareholders, capital gains tax applies at twenty percent for higher rate taxpayers on a commercial asset disposal — £160,000. Professional fees: agent, solicitor, accountant, approximately £30,000. Business relocation or closure costs — equipment disposal, loss of goodwill, premises for the remaining stock, lost income during transition: perhaps £100,000 to £150,000.
David and Sarah’s minimum walk-away price — the number below which they are worse off selling than staying — is approximately £1.29 million to £1.34 million.
And that is the likely floor, not the ceiling. They might reasonably want to buy another commercial site to continue trading — add £500,000 to £800,000 for a comparable workshop elsewhere. They might want to retire with something meaningful in their pocket after a decade and a half of work — add £300,000 to £500,000. Or both. The GLA’s BLV of £1,150,000 does not get them to the floor, let alone to a realistic outcome. The system treats long standing businesses as if its owners should be indifferent to selling at the minimum theoretical benchmark. They are not. Nobody is.
This is the land strike. It is not ideological. It is arithmetic. The rigidity of the GLA’s approach to EUV+ ignores capital gains tax. It ignores business value. It ignores transaction costs and relocation. It produces a benchmark that looks reasonable on a spreadsheet if you want to try and defend local planning policies and perpetuate a myth that the private sector can deliver the country’s social housing needs.
The sites are in the trajectory. The land doesn’t trade. The homes don’t get built. And nobody asks why. This is not a city that will yield 88,000 homes per annum.
Problem Two: The Scheme Doesn’t Pencil
Now assume the worst happens to David and Sarah. The business fails. They are forced sellers. A developer acquires the site at £1,150,000 — the GLA’s BLV, below what David and Sarah wanted but that’s life.
The developer applies for planning permission. Under the Mayor’s emergency LPG measures — the most generous fast-track route currently available in London — twenty percent of units must be affordable, with sixty percent of that as social rent.
The residual calculation:
GDV:
– 28 market units at £480,000 average: £13,440,000
– 4 social rent units (720 sq ft × £195/sq ft): £561,600
– 3 intermediate units (720 sq ft × £450/sq ft): £972,000
– **Total GDV: £14,973,600**
Costs:
– Construction and abnormals: £8,942,000
– Finance at 8.5%: £759,070
– Professional fees at 12% of build: £1,073,040
– CIL and S106: £500,000
– Sales and marketing at 3% of market GDV: £403,200
– **Total costs: £11,677,310**
Profit at 17.5% of GDV: £2,620,380
Residual land value: £675,910
The scheme falls £474,090 below the GLA’s own BLV of £1,150,000. That’s right, we have hit what I referred to at last week’s Planning Summit as the breach point. Where residential development is so unviable it’s better to keep the site in existing use.
Yes the developer might cut its profit just to keep the lights on – and many do. However, I will come to why that’s increasingly difficult to do in the next section.
This is the emergency LPG rate — twenty percent affordable, the most generous route available. Now apply the standard London Plan H5 fast-track threshold of thirty-five percent affordable. The RLV goes deeply negative. The scheme is structurally unworkable at any meaningful affordable housing level on this site at this density.
The only scheme that pencils is one with zero affordable housing. The affordable housing obligation is not a marginal drag on an otherwise healthy scheme. It is the difference between viability and non-viability.
Problem Three: The Developer Can’t Get Out
Assume the developer proceeds at twenty percent affordable and takes the hit on its margin. The scheme pencils — marginally. Planning is granted. Construction starts. Two years later, thirty-five units are ready to sell. Twenty-eight market units blending at £480,000. £13.4 million of sales to make.
Who is buying?
Help to Buy — the scheme that underwrote approximately forty percent of new build sales at its peak — was wound down in March 2023 and has not been replaced. Stamp duty applies in full on purchases above £250,000 from April 2025, with first-time buyer relief only up to £500,000 in high-value areas. Mortgage rates remain elevated. The buyer pool for new build flats in inner London — particularly flats rather than houses, particularly from developers without established brand recognition, particularly in locations where there is significant second-hand stock— has contracted materially since 2021.
This is the exit risk. It does not appear in any viability appraisal at application stage. It is real. It is the reason developers are repricing London brownfield risk upward — and in Berkeley’s case, exiting the London market entirely. A 17.5% margin is simply not enough when sales risk is this high. Which means the profit assumption that made the residual work is itself uncertain. Which means the residual land value gets lower still. Far below the breach point.
The units could sit. The developer’s equity is trapped. It is a genuinely terrifying prospect for any development director asked to sign off on the land acquisition.
Three Problems, Systemic Solutions
The housing crisis on this half-acre brownfield site is not one problem. It is at least three sequential problems, each blocking a different stage of the development cycle. Fixing one without fixing the others doesn’t move the dial. They all need resolution. This is systemic.
Here is my take on what this means for brownfield land going forward:
First, in a London context, whilst the LPG is welcome, if anyone thinks the planning framework is going to achieve unit numbers in excess of 10,000 let alone 88,000 per annum then they are clearly smoking something very exotic.
Second, the easiest problem to fix is EUV+ with more reflection of the real costs of selling a business. This would in line with what Mr Justice Holgate actually intended during the Parkhurst case. This will link the chain back to the developers which has been severed but it doesn’t resolve the two greater problems.
Third, affordable housing without a full subsidy just doesn’t work. Today’s land economics do not subsidise the costs of delivery. If it doesn’t work in London on brownfield, it doesn’t work in any other major city.
Fourth, if this carries on the built environment supply chain is going to be damaged irreparably. This must now be about understanding and protecting the UK’s economic welfare rather than standing on ceremony over small percentages of affordable housing.
Fifth, demand-side measures can take several forms. However, in every scenario I can think of, they must return. Even if the land is free, no one is going to develop flats at scale if the costs of production are so high and the sales trajectory is so uncertain. That means that whilst public land and partnerships are important, they will still require huge subsidy to deliver homes if they are to be viable even with 100% land write downs.
Sixth, we cannot abandon affordable housing but we do need to reconceptualise delivery. It is evident that affordable housing without full grant subsidy is not possible on brownfield land. It seems to me a temporary suspension of onsite delivery for anything other than strategic sites is now necessary to get the city building again.
There is merit in a more tax led approach at least in the interim with the receipts being used to acquire second hand built stock. The costs of production of developing a social home are now so high that when you factor in the time it takes – up to seven years, there is I would suggest a compelling case to buying second hand stock and just letting developers focus on delivering.
That is in fact what Councils are already doing. But they are doing this off their own balance sheets rather than receiving an income stream from new developers to help finance this.
We end up with four buckets of work to reform the system. The first is to resolve how we bring land into the system. The second is the costs of production, its now far too high and there is a piece of work around what regulations we actually need. The third is of course demand support as nothing will happen meaningfully without it. The fourth and most contentious is affordable housing. In my opinion these all need to be in lockstep to make the system work again.
I will return to this more in another blog post.
