Your planning risk shorthand broke on 17th August
Martin Alderson, director of GapSense, explains why the planning risk shorthand used by development lenders changed on 17th August, with settlement boundaries, housing delivery figures and Green Belt policy now creating materially different credit questions for land deals.
Here is a file most development lenders will recognise. A paddock on the edge of a village, four or five units, borrower wants twelve months to get consent and then refinance onto a build facility. No permission yet. The land is worth one number if it gets consent and a much smaller number if it does not, and your security sits somewhere between the two.
Historically there was a quick way to get comfortable. You looked up the council’s Housing Delivery Test result. Below 75% and the presumption in favour of development was engaged across the district, which told you the wind was behind the application. Above it, less so. Crude, but it took about ninety seconds and it was roughly right.
That stopped working on 17 August.
Since then the presumption applies inside a settlement whatever the council has delivered. Unconditionally. The authority could be at 23% like Basildon or 381% like Blackburn with Darwen and it makes no difference to whether the presumption is engaged. Step outside the settlement and you are into Policy S5, where only certain forms of development should normally be approved, and there the delivery figures still matter a great deal.
Which means the paddock is now a completely different credit question depending on where the red line sits. Not which council. Where the fence is.
I would like to tell you there is a national map of settlement boundaries you can check. There is not.
We went looking, which is how we ended up drawing 7,147 of them ourselves using Office for National Statistics built-up areas. Those are a proxy. They are not settlement boundaries in the planning sense and I am not going to pretend otherwise. Plenty of local plans define their own, but the coverage is uneven and some of them were drawn a long time ago for a plan that had nothing like this hanging off it. Nobody built a national layer because until recently nobody needed one.
Our 7,147 built-up areas cover about 11% of England. Treat that as an indication rather than a measurement. But it tells you roughly how much of the country now carries the presumption automatically, and it tells you that a fair amount of money is going to turn on exactly where the edge of it falls.
Most of what has been written since 17 August has been about the other number, the rise from 59 to 91 districts below the 75% threshold. Birmingham falling from 79% to 57% is a genuine story. But if you are underwriting land, reading that number the way you read it in July will get you into trouble, because it has quietly changed job. It used to be the answer. It is now one input, and for anything inside a settlement it is not even the relevant one.
There is a flip side that I suspect is being missed. Forty districts just fell below 75%, and for a countryside site that is help rather than harm. Under S5(1)(j) a result below 75% means unmet housing need is deemed to exist, which is one of the routes to consent outside a settlement. So a borrower with a paddock in a district that just failed is arguably in a stronger position this month than last. Not a good position. A stronger one.
The place I would expect people to get genuinely caught out is the Green Belt. S5 does not apply there at all, so the instinct is that housing figures have stopped mattering. They have not. The same triggers come back in through footnote 41 on the grey belt route. If your credit paper has dropped that argument because someone saw a Green Belt notation on a constraints plan, it has dropped one of the better arguments on the file.
Two practical suggestions, for what they are worth.
If you are packaging land deals, put a view on the settlement position in the pack, explicitly, rather than leaving the valuer to infer it. And do not be surprised if valuations take time to converge. Two valuers may quite reasonably take different views on the same field.
One more thing. The 91 only tells you about one of the two housing triggers. The delivery test is published nationally, five-year housing land supply is not. Councils evidence their own position and there is no complete national dataset, so there will be districts sitting outside that 91 where the supply trigger is engaged instead.
How many? I cannot give you a defensible national figure. At the moment nobody publishes one, which is its own kind of answer about where this system currently stands.
Related
Grey belt is changing Green Belt appeals, but not in the way many expected
AI is not the due diligence risk. Unverified AI is
The short‑term cost of planning reform
The rise of the “forever renter”
The development finance problem nobody has quite solved
Unprepared landlords are facing a compliance crisis under the Renters’ Rights Act
The mortgage market is losing tomorrow’s borrowers today
Private credit should fund a transition, not postpone a problem
Speed in specialist finance starts before the deal reaches a lender
Rental fraud shows it’s time for mandatory digital identify checks
Section 21 is making renting harder, not easier
