#2026NPPF: 5 bits that aren’t working
Imagine. You’re in the smoky, power-filled corridors of the Ministry on Marsham Street in the heart of Westminster. And it’s policy-drafting time. I hope you’ve had your coffee, because we need results. Your boss, and indeed your boss’s boss (the BIG boss) needs the NPPF to actually work this time. Reforms to national planning policy are “integral” to their mission to “kickstart economic growth”. Heck, the last round of reforms you are trying replace were relied on to justify GDP increases of several billion pounds (billions of pounds which have yet to materialise). So. Don’t muck it up.
Now, we already have a “tilted balance”, but that’s old hat. The Minister wants something punchier - remember, we’ve got 1.5 million homes to not build. So what sounds even more tilty than a “tilted balance”? A “default yes”… mmm. That could work. A “clear, rules-based system” - right, now we’re getting somewhere. How about - hang on, everyone, listen to this - how about… a “universal tilt”. Mic drop moment. Pens away. Ladies and gentlemen, we have a policy.
Don’t we?
2 months or so into the life of our new NPPF (initial quick hits here) we now know - for all the good bits, and there are good bits - cracks are already starting to show. Important cracks. Cracks that are going to make this “tilt” an awful lot less “universal” than the minister appears to think it should be. Time for a re-draft, anyone?
What am I talking about? So soon? Well… let me give you 5 examples:
1. A “substantial” problem
Sometimes, national policy gives us a steer on how much weight different things should attract in the planning balance, e.g. under the old 2024 NPPF:
“Great weight” was given to conserving and enhancing landscape and scenic beauty in National Parks and National Landscapes; conserving designated heritage assets; and the benefits of using suitable windfall sites within existing settlements for homes.
“Substantial weight” was given to the value of using suitable brownfield land within settlements for homes; and to any harm to the green belt brought about by inappropriate development.
“Significant weight” was to be placed on the need to support economic growth and productivity; new public service infrastructure; “outstanding or innovative designs”.
In the end, of course, the policy balance in the 2024 NPPF at §11(d)(ii) talked about harms both “significantly” and “demonstrably” outweighing benefits. Unless you are caught by a “strong” reason to refuse under §11(d)(i). Inappropriate development in the green belt should only be approved if its benefits “clearly” outweigh its harms, including harm to the green belt which has to be given (as above) “substantial” weight.
Oy vey. So many adjectives. So many adverbs. What on earth did any or all of them actually mean? How were they supposed to relate to one another? Genuinely. Of course, in the end, the exercise is not mathematical. The adjectives are being used metaphorically (i.e. we do not literally go around with weighing scales). Nonetheless, an effective national policy would have to be clear about what its priorities are, wouldn’t it? One of the key benefits of using language to describe weightings in a planning balance exercise is to try to achieve a result that is clear, transparent and comparable with outcomes in other cases. That only works if we are all speaking the same language. And we are not.
Planners use their own terms for weighing, and their own matrices of what words denote more or less weight. And sometimes the court weighs in, e.g. telling us that “the word “substantial” does not denote a greater quantum of weight than “significant””. But it should not require a court judgment. The NPPF should make clear what it intends its priorities to be.
So, what’s happening now?
The 2026 NPPF has consciously and expressly taken the decision to address the “too many adjective” problem by (in effect) boiling everything down to a single adjective: substantial.
So now, for instance (to take only a few examples), substantial weight now attaches to:
Housing need (HO7), business land (E2), town centre development (TC2), telecoms (CO1), renewable energy and electricity networks (W3), water infrastructure (W4), minerals (M3), intensification within settlements (L2), community facilities and public service infrastructure (HC4), defence sites (P6), energy efficiency and climate resilience benefits (CC2/CC3), outstanding or innovative design (DP3(5));
Green belt harm (GB6(2)), Golden Rules compliance (GB8(2)), Protected Landscapes (N4), designated heritage assets (HE6), design conflict (DP3), and intentional unauthorised development (DM8).
This approach creates a different problem from the over-abundance of adjectives under the 2024 NPPF: we now have too few adjectives. If everything is a priority, nothing is a priority. We’re left puzzling out under S5(1) whether a “substantial” benefit “substantially outweighs” a “substantial” harm. A notional housing scheme in the green belt affecting the setting of a listed building, attracts substantial weight to the housing benefit, substantial weight to green nelt harm, potentially substantial weight to Golden Rules compliance, and substantial weight to heritage conservation — with no steer on how to resolve the tension. The problem with a “simplified” approach that brandishes “substantial” as the catch-all expression of priority is that the Government gives no steer through the NDMPs on the relative importance of different considerations in the planning balance. And relative importance is precisely the most difficult and decisive thing for decision-makers to assess.
2. Exceptions swallowing rules
This challenge over relativities is expressed most acutely in the carve-outs to the new presumptions, at e.g. S4(2) or S5(2), the latter of which says that:
“In applying this policy, the circumstances in which the benefits of approving development proposals are likely to be substantially outweighed by adverse effects include, but are not restricted to, situations where the development proposal would fail to comply with one of the national decision-making policies which state that development proposals should be refused in specific circumstances.”
The refusal policies to which that carve-out relates are not themselves listed, but there are (at least) 15 of them (I had a go at a list here) – e.g. relating to heritage, designated landscapes, green belt, biodiversity, design, flood risk, transport, town centres etc. In passing, just a little bit amazingly, it is genuinely not clear whether some of the new NPPF’s policies are in this category or not. Take two examples:
Harm to designated heritage assets: under HE6, there’s no doubt about it. If you caused substantial harm and fail the tests at HE6(5)… well, sorry. “Consent should be refused”. Black and white. But hang on. Most schemes do not cause substantial harm. If you cause harm under HE6(4), but it isn’t substantial, then you weigh that harm against public benefits. Okkkaaayyy. But then what? If public benefits are outweighed by the heritage harm… well, so what? Do you press on to the S5(1) default yes? Or is that a policy that states consent should be refused in specific circumstances (even though… it doesn’t really state that at all).
Protected landscapes: under N4, again - no doubt about it. If you’re major development within a protected landscape and you fail the tests around e.g. “exceptional circumstances”, you “should be refused”. Clear. But what if, for instance, you’re in the setting of a national landscape. There we’re told that you “should be sensitively located and designed to avoid or minimise adverse impacts on the Protected Landscape”. OK. So what if you fail that test? Well… the new NPPF doesn’t quite say.
So it’d definitely be nice if someone had actually told us which policies are intended to knock out the “universal” tilt, but it’s clear enough that there are LOADS. Without clear guidance on how priorities in the document are to be comparatively assessed, the question is going to be whether the exception ends up subsuming the rule. In other words, there are so many potential policy areas that act as gateways to the operation of the core presumption policies. The questions are:
(i) Why should those policy considerations generally reverse the outcome of the positive presumptions operating rather than simply featuring as matters to be weighed as part of the presumption; and
(ii) Given their “gateway” role, will they confound the presumptions’ ability to work at all?
2 months in, and of the 80-odd refusals so far on S4(2)/S5(2) refusal reasons, by far the leading refusenik is DP3 - the design policy (responsible for 63 refusals and counting). I have a couple of questions about DP3:
Does it / can it / should it apply to outline planning applications? If not, shouldn’t it say that? Inspectors have already reached different views on this question.
It covers all kinds of things. Almost none of them have anything to do with being “clear” or “rules-based”. It’s a litany of planning judgments on e.g. whether a scheme promotes social interaction, whether its green infrastructure is high quality, whether the pattern of built form optimises the site’s potential… there are literally hundreds of evaluative judgment-calls required in order to get the bottom of every matter comprised in the laundry list of DP3.
Important matters - sure they are - but why are all of those elevated above other things in the NPPF (like e.g. housing delivery or whatever) such that failing the litany of DP3 tests means that your default yes is likely to turn into a default no. Why is that coherent?
I tell you, whatever else these S4 and S5 policies amount to, they are neither “clear” nor “rules-based”.
3. Needing to show “need”
This “universal tilt” may require you e.g. at S5(1)(j) to prove you’re meeting:
“an evidenced unmet need (including, but not limited to, development proposals involving the provision of housing where the local planning authority cannot demonstrate a five year supply of deliverable housing sites or scores below 75% in the most recent Housing Delivery Test)”.
Two problems with this:
First, it’s properly ambiguous. The ambiguity was explained by Simon Ricketts here. Long story short, it isn’t clear whether housing proposals are intended to be able to evidence unmet need even in cases where an authority could demonstrate a 5 year housing land supply and in excess of 75% HDT score. A planning inspector thought that there could be other ways of meeting the trigger (e.g. focussing on unmet needs for affordable housing). The minister suggested the opposite. Maybe we’ll find out soon. What a mess.
Second, it’s internally inconsistent. When we get to green belt policy under GB7(1)(g)(ii) we get the same phrase: applicants need to show
“an evidenced unmet need for the type of development proposed […] Which, in the case of applications involving the provision of housing, means the lack of a five year supply of deliverable housing sites, including the relevant buffer where applicable, or where the Housing Delivery Test result was below 75% of the housing requirement over the previous three years”
But hang on. Is that supposed to be different? Where’s the “including, but not limited to” language from the S5 definition? Why on earth would there be a rational intention to define the same phrase in materially different ways in different bits of the policy document? Well - two options really. Either the government needs to confirm its position prontissimo, or the High Court will have to get into it before long.
4. Are we “well-related” yet?
Your clear, rules-based, default, universal tilt only actually operates for S5(1)(j) sites if the site is “physically well-related to an existing settlement (unless the nature of the development would make this inappropriate)”.
What. On. Earth. Does. That. Mean.
Does it only apply to sites that are physically contiguous to the existing settlement? No - it could’ve said that. Indeed, HO10(2)(a) says “adjoin or be physically well-related to a settlement”. The drafters distinguish the two, so “well-related” clearly isn’t just contiguity
Is it about having good transport/walking routes in and out of the settlement… maybe, I guess? Is it just about that? Or does it mean something more? And what kind of developments would not be “appropriate” to consider under this criterion? No idea.
In one decision, an Inspector (who went on to grant planning permission anyhow) decided that a site wasn’t physically well-connected to a settlement because it caused limited harm to “the character and appearance of the area, it would also harm the setting and context of the village” even though it was “very close to the settlement […] it can be accessed from an existing road within the settlement and is well located in terms of facilities and services”. The difficulty there is that all greenfield development (which is what S5 is targeted at in the first place) will cause at least limited harm to local character and appearance. So. We need guidance. We really do.
5. Getting in the way of “strategic sites”
HO13 is about, among other things, supporting strategic sites. And it’s a powerful piece of policy. The tricky bit comes at (4):
“4. Where an emerging development plan (however advanced) proposes to allocate one or more strategic sites, development proposals should not be inconsistent with the proposed scale, location and phasing of those strategic sites. This applies notwithstanding the application of policy DM4(2) where relevant.”
I have questions. Many questions. For instance:
Again with the ambiguity. Is this just about sites which are part of or next to an emerging wider parcel (i.e. don’t do something speculative that stops the wider scheme from coming forward)? Or is it intended to cover competing strategic sites in the same plan area (i.e. don’t bring forward a new town over here, because the emerging local plan wants a new town over there). Totally unclear.
What happens if you are inconsistent with HO13(4) - is it a refusal under e.g. S5(2)? It doesn’t say that. Urgh. See above. Groan.
Isn’t “however advanced” a pretty awful idea? You could have an incredibly early stage consultation draft plan with some vague blobs on a map. It could be totally inconsistent with national policy. It could be pre-examination. It could be subject to hordes of objections. Those objections could be (who knows) well-founded. The plan may be years away from adoption. It may never be adopted. The planners might agree that the emerging local plan gets next-to-no weight. They might also agree that there’s no DM4 prematurity objection because e.g. the emerging plan is not at an advanced stage. But then… what… notwithstanding all of that we’re supposed to do nothing to disturb its vision for strategic sites. Why? Why. Why.
A certain senior Inspector I know likes to say that planning is, or should be, mainly common sense. That’s a mantra we should be getting behind, isn’t it. Well, I tell you, say what you want about the new NPPF. Commonsensical it ain’t (and I haven’t even gotten into the definitions of reasonable walking distance to well-connected stations yet!!!!). What do we need? A definition of the S5(2) policies. Guidance on what “well-related” means. We need a clearer steer on how these myriad, “substantial” relativities are actually to be resolved. And we need it soon.
Anyhow, enjoy those shortening evenings and the crunchy leaves underfoot #planoraks. We’re firmly into woolly jumper / winter coat season. And get out the popcorn and the JR-o-meter for the brave, hearty authorities refusing to acknowledge the national scheme of delegation that is literally already part of the law and kicking in at the end of this month. That’ll be fun. In the meantime, here’s praying for some PPG to sort out all the above. Until we get it - stay cool, #planoraks. Keep warm. I hope you’re well. And through all the madness, do your level best to #keeponplanning.
