In Part One I put a price on delay.
What I could not tell you then was whether any of it is about to change.
Ministers now treat the big structural reforms as delivered, and attention has moved to the application process itself, with the stated ambition of making that journey “as fast and easy as possible.”
That is the right target. Whether the remedy matches the ambition is a fair question, and research published by Lichfields for the Land Promoters and Developers Federation goes some way towards answering it.
The proposals deal with validation requirements, consultee referrals, legal drafting and the sheer volume of consultation responses. What they do not touch is the technical work. On a strategic site, that is where a great deal of the time goes.
So here is the honest answer before I get into the detail:
The system we end up with may be more predictable without being much faster.
For a landowner, those are not the same thing. But they are not nothing either.
Why did the system slow down in the first place?
Councils are handling fewer applications than they used to and taking longer over them. I covered the numbers in Part One. The question is why.
Part of it is straightforward. Authorities can only ask for information where it is proportionate and relevant to the application. That is the rule. In practice, the incentives pull the other way.
Ask an applicant for too much, and the cost of producing it falls on the applicant.
Ask for too little, and the consequences can land on the officer.
Add judicial review and the growing politicisation of planning to that, and deferring a decision or referring it on can start to feel a good deal safer than making it.
I do not say that as a criticism of planning officers. Most of the ones I deal with are doing a difficult job with too few colleagues. But it does explain how delay becomes built into the system rather than being an occasional failure of it.
It also tells you something about reform. If the delay is a product of incentives, changing the forms will not be enough on its own.
Getting to the starting line
Validation is the point where your application formally enters the system and the clock starts.
As I said in Part One, a medium-sized site can need around 30 separate assessments before a council will accept it. Some authorities publish local validation guidance running to more than a hundred pages.
The Government’s answer is a national validation list.
There is a catch, and it is a technical one worth understanding. Validation lists are set through legislation. Planning policy sits above that statutory framework rather than replacing it. Lichfields’ view is that if the aim is a genuinely consistent national system, a regulatory approach will probably be needed to get there.
The evidence behind the change is also thinner than you might expect. The Government’s most recent study of validation delays draws on research from 2019-20, using cases from 2016-18. It excludes applications of more than 1,000 dwellings, outline applications with reserved matters, and appeals.
Read that list of exclusions again if you own strategic land, because it describes your site.
None of which makes the reform pointless. A consistent national list would be a real improvement. But the time spent assembling information and getting an application validated still burns fees and management time, and the statutory clock has not started yet.
Who else gets a say, and for how long?
Once you are validated, the next source of delay is the statutory consultees.
Ecology, flood risk, highways, heritage. All of it needs specialist input, and quite properly so. The issue is how often that input is sought when it is not needed, and how long it takes to arrive.
Lichfields found that 30 to 40% of referrals to Natural England and the Environment Agency were considered unnecessary, and that around 35% of National Highways referrals offered scope for efficiency.
The research also collects some examples that will raise a wry smile from anyone who has been through it. My favourite is a ball strike assessment requested at outline stage.
The proposed reforms would narrow the circumstances in which consultees are brought in, separate genuinely legal or technical responses from advisory comment, and, most importantly, introduce fixed response windows with consequences for missing them.
This is the change most likely to show up in the headline figures, because consultee delay currently sits inside the statutory determination period. Shorten it and the published determination times move.
Whether it shortens your application is a different matter. On a complex site, the consultees you need to hear from are usually the ones with something substantive to say.
The part that happens after you have won
Section 106 agreements are a different animal.
By the time an application has a resolution to grant, the principle of development has essentially been settled. Everybody agrees it is going to happen. And then the lawyers start drafting.
Lichfields puts the average time to complete a Section 106 agreement at one year and five months, and that figure has grown by 20% over the previous two years.
As I showed in Part One, that is not a neutral delay. On the worked example of a 100-home site, six months of additional delay added nearly £10,000 in interest. A Section 106 period of a year and five months can add somewhere in the region of £25,000 to £30,000.
There are some genuinely odd contributors to this. Some authorities charge their own planning department internally for the council’s legal assistance, which does not obviously speed anything up.
Town Legal LLP has produced template Section 106 wording for sites below 50 units, and the research recommends extending that principle of standardised drafting to larger applications.
The Government’s current Section 106 roadmap is aimed mainly at the capacity of Registered Providers to take on affordable housing units. That is a real problem. It is not the same problem as the negotiation taking a year and a half.
A newer problem: objections at scale
There is one pressure on the system that barely existed when I started in this industry.
A survey of local authorities found that around seven in ten had seen representations increase over the previous two years, and four in five of those attributed at least part of the rise to AI.
When an objection can be generated in seconds at no cost, the volume of material an officer has to read goes up without the amount of substance in it going up to match.
I want to be careful here. People have every right to object to development, and a well-argued objection from a neighbour deserves proper consideration. That is how the system is supposed to work. The problem is a planning department reading two hundred versions of the same letter.
The Government’s proposed response is a Written Ministerial Statement restating something that was already true: objections do not carry more weight simply because there are more of them. The research also points to pilots in West Oxfordshire and Milton Keynes using summarisation tools to recover officer time.
This strand is different from the others. It is not removing an old cause of delay. It is trying to stop a new one becoming entrenched.
What does it all add up to?
Every one of these reforms is aimed at something real.
Validation can be made more consistent. Unnecessary referrals can be cut. Section 106 drafting can be standardised. Consultation responses can be made manageable.
If you own a site and all four land as intended, the process you go through should be tidier and more predictable than the one I described in Part One.
But notice what all four have in common.
They are aimed at the process around your application. None of them touches the work inside it.
Your ecology surveys still run to the seasons they run to. Your highways modelling still takes as long as it takes. Your viability assessment still has to be done, argued over and done again.
That is the distinction I want to pick up in Part Three, along with the more awkward question of how long reform itself takes to arrive, and what a landowner can genuinely control while everybody waits.
