A planning permission is often reported as though it were a building. It is not. It is a removal of one specific legal obstacle, granted by one body, under one set of tests. Everything else that has to be true before a crane arrives remains untrue on the day consent is issued, and in London a great deal of it stays untrue for years. Understanding that gap matters, because most of the public argument about housing delivery is conducted as though the binding constraint were the decision notice, when frequently it is not.
What a permission actually settles
Consent establishes that a particular development, in a particular form, is acceptable in land use terms subject to conditions. It does not establish that the site is buildable, financeable, insurable, serviceable or owned outright by the applicant. It does not establish that the ground is what the site investigation assumed. It does not settle the party wall position, the rights of light exposure, the utility capacity, the crane oversail, or the highway agreements needed to get material in and out. Each of those is a separate negotiation with a separate counterparty who has no obligation to be quick.
This is why the phrase implementable consent is doing more work than it appears. A consent can be perfectly lawful and still sit in a drawer because one of these parallel tracks has not landed. When schemes are described as stalled, the interesting question is almost never why did the council take so long. It is which of the tracks broke.
The pre-commencement stack
Most permissions of any size arrive with conditions that must be discharged before work starts. Contamination, archaeology, drainage strategy, construction logistics, materials, ecology mitigation, energy strategy, sometimes each in sequence rather than in parallel. Individually these are reasonable. Collectively they form a queue, and the queue is served by a discharge function that is usually the least resourced part of a planning department because it produces no headline output.
Two practical points follow. First, the drafting of conditions at committee stage has more effect on delivery timing than almost anything else in the report, and it is the part least scrutinised. A condition that requires approval of a strategy before any development commences is very different from one that permits demolition and enabling works to proceed. Applicants who negotiate that distinction at the right moment save quarters, not weeks.
Second, discharge submissions are frequently made before the design is developed enough to answer them. A drainage strategy submitted on the basis of an outline layout gets returned, and the clock restarts. The teams that move fastest are the ones that treat condition discharge as a design programme item from day one rather than as post-consent administration.
Section 106 and the timing of obligations
The agreement that accompanies a major permission is not simply a list of payments. It is a schedule of triggers, and the triggers are where delivery risk lives. An obligation tied to commencement behaves differently from one tied to first occupation, and a scheme can be viable under one arrangement and unfundable under the other because of what it does to the cash flow curve.
Negotiating those triggers is skilled work on both sides. A local authority has a legitimate interest in securing benefits early, because a scheme that stops halfway leaves it with a hole and no infrastructure. A developer has a legitimate interest in matching outgoings to revenue, because paying for everything before the first sale is what turns a marginal scheme into an unbuilt one. Where the two sides fail to engage seriously with the sequencing, the usual result is an agreement that is signed and then immediately becomes the reason nothing happens.
The finance problem is a design problem
Development finance is priced against a set of assumptions made at a moment in time: build cost, sales values or rents, exit yield, programme length. Consent takes long enough that at least one of those assumptions is usually stale by the time it lands. If build cost inflation has outpaced value growth in the interim, the scheme that was appraised as viable at submission is not viable at consent, and the consent becomes a piece of paper whose main function is to protect the land value.
What is less often acknowledged is how much of that exposure is designed in. Deep basements, complex structural transfers over retained facades, unusual cladding build-ups, single-stair configurations that later require rework, plant strategies that depend on a grid connection nobody has confirmed: each of these converts a market movement into a stop. A scheme with a simple structural grid and a conventional envelope absorbs a cost shock. A scheme carrying three separate technical bets does not. Design decisions taken early, often for good architectural reasons, determine how much bad news the project can survive.
The things nobody controls
Utility capacity deserves particular attention because it is the constraint most likely to surprise people who have not built recently. Electrical connection in parts of London is not a formality and cannot be assumed from the presence of a substation nearby. The lead times are set by an organisation with its own programme, its own queue and no interest in the applicant’s committee date. The same is true of certain highway and rail interfaces, where possession windows are scarce and dictated years ahead.
Vacant possession is another. A site with sitting occupiers, whether commercial tenants with security of tenure or operational uses that must be relocated, has a delivery date governed by lease events rather than by planning. It is entirely normal for a consent to be granted years before the site can be cleared, and entirely normal for that fact to be invisible in any published statistic about permissions granted versus homes started.
What this means for how we count
The consequence of all this is that permissions granted is a poor proxy for supply, and everyone in the industry knows it while continuing to use it because it is the number that exists. A more honest picture would separate consents that are ready to build from those that are held for value, those that are unimplementable pending an infrastructure event, and those that were never intended to be built by the party that obtained them.
That distinction has policy consequences. If the binding constraint on a particular site is grid capacity or vacant possession, then reforming the decision-making process will not move it. If the constraint is a viability position that only works at a certain interest rate, then the lever is economic rather than procedural. Treating every stalled scheme as evidence of the same failure produces interventions aimed at the wrong thing.
A practical checklist
- Before submission, list every non-planning consent and connection the scheme needs, with an owner and a lead time. If the list has gaps, the programme is fiction.
- Draft conditions actively. Ask which ones genuinely need to precede commencement and which can attach to a later trigger, and make that argument in writing while the officer is still writing the report.
- Sequence obligations against the cash flow, not against the calendar.
- Identify the two or three technical bets in the design and ask what happens to the appraisal if any one of them fails. Reduce the count if you can.
- Treat vacant possession and utility capacity as design constraints, not as legal housekeeping.
None of this is exotic. It is simply the recognition that consent is the start of the difficult part, not the end of it, and that the professionals who deliver most reliably are the ones who have internalised that before they draw anything.
