The revised NPPF landed yesterday and the initial reaction across the planning world has been warm to positive. There’s some good news on strategic sites, grey belt viability, and — buried on page 52 of the consultation response — a quiet retreat on medium sites that deserves scrutiny.
Overall they have pushed things as far as they can within the current orthodoxy although I agree with Zach Simmon’s Planoraks blog. They could have been much more radical by putting national policy on a statutory footing. But alas, this is a Government that is pro-development but not pro-developers. The result is a heck of a lot of compromises.
The standardisation fight the sector won
Back in March I wrote into this consultation warning that the draft’s proposed standardised viability inputs — fixed developer profit at 17.5% for private and 6% for affordable, a x10 EUV-to-greenfield benchmark land value formula, mandatory growth testing, IRR/ROCE baked into local plan viability assessments, and tighter landowner premiums — would produce “unrealistic viability outcomes” that ignore how much development risk varies site to site.
Annex B has been abandoned. The existing PPG methodology stays in place rather than being replaced with the rigid national assumptions the draft proposed. We are instead promised a new PPG on viability as soon as possible.
DM5 itself has also moved in the right direction. It’s been amended to let costs that weren’t accounted for at local plan stage form part of a viability assessment at application stage — a direct answer to the argument that markets move faster than the planning system can revise its assumptions. The catch is footnote 22, which now defines the threshold for “changed significantly” as a recession or similarly severe economic shock. That’s a materially higher bar than the ordinary-cycle cost inflation and borrowing-cost volatility that actually erodes scheme viability most years. Flexibility was restored in principle; the trigger for using it was set high.
Plan making viability and trouble for the Draft London Plan
I read the NPPF with the Draft London Plan’s viability troubles in mind. The new NPPF underpins the importance of robust viability evidence supporting Local Plans. Page 20 :
“Where a viability assessment is submitted with a development proposal, this should be based upon and refer back to the viability assessment(s) that informed the relevant development plan policies. It should fully evidence all inputs and assumptions used in the assessment, and explain any differences from those used for viability assessment that informed the relevant plan policies.”
If the basis of the Plan’s viability is flawed, then what are the implications of this reference loop? Read together with DM5(1) — proposals according with up-to-date plan policy “should be assumed to be viable” — the whole architecture rests on one assumption: that the plan-stage viability testing behind the policy was done on defensible terms in the first place. Take that assumption away and DM5(4) stops being a tidy transparency requirement and becomes a mechanism for re-litigating the plan itself, application by application.
That’s precisely the question mark hanging over the draft London Plan. In Viability: Heads or Tails, I set out that City Hall’s own testing found only 23% of 989 development scenarios viable, before applying market-realistic cost and revenue assumptions. The plan’s targets (35% affordable in Inner London, 25-20% on the periphery) are explicitly premised on “the public sector taking a more interventionist role” through land assembly and de-risking gap-funding of a kind Philip Barnes described as “without precedent during my 40 years” of watching this system. DM5(1)’s presumption of viability is being extended to a draft plan whose own evidence base says the numbers don’t work absent a Gov sponsored intervention nobody has actually committed to deliver — and DM5(4) then obliges every applicant to “refer back to” and explain any departure from that evidence. It’s like reading Kafka isn’t it?
A Framework built around the idea that plan-compliant schemes shouldn’t need a viability fight is only as sound as the plans underneath it. London may turn out to be the test case for what happens when the plan itself was the bet.
Strategic sites: a category the Framework finally admits exists
The bigger structural change is the new Glossary definition of a “Strategic site” — broadly, 1,500+ dwellings, multi-phase, infrastructure-heavy, requiring a masterplan — threaded consistently through. This closes a gap that’s sat in planning policy for years: a 2,000-home new community and a 200-home infill site have never been formally distinguished, despite facing entirely different delivery economics, hold periods and capital structures.
The practical payoff is GB8(3)(b): a strategic site is now, on its own, one of three grounds on which a viability assessment against the Golden Rules can be submitted — alongside demonstrable unaccounted PDL costs and schemes of a wholly different type to what the plan tested. Strategic grey belt sites also no longer need to separately demonstrate they’re “multi-phase” to access that route. For sites currently rendered non-viable by c.50% affordable housing requirements under the Golden Rules, that’s not a technical tidy-up — it’s a genuine unlock, and probably the single biggest viability story in the document.
Data centres get their own plan-led framework too — E1 allocation duties, E2 power and grid connection criteria, and essential-infrastructure status for flood risk purposes — a sensible, overdue recognition that power availability is now a locational constraint as real as land itself.
One thing that should worry the strategic-sites crowd
Set against all of that: minimum Local Plan periods have been cut from 15 years to 10. On the surface this looks unrelated to viability, but it sits awkwardly next to the Framework’s new enthusiasm for strategic sites, which by definition run delivery periods well beyond ten years. A shorter statutory planning horizon creates less incentive for authorities to think big, allocate strategically, and commit to the infrastructure that large-scale development depends on — precisely when the Framework has just built an entire policy architecture to reward exactly that kind of ambition. Spatial development strategies might plug the gap eventually, but in the meantime, don’t be surprised if some emerging plans get scaled back rather than scaled up.
The retreat on helping SMEs
Which brings me to the bit that’s actually a shame. Government has confirmed it will not proceed with giving applicants discretion to meet affordable housing requirements on medium sites through cash payments in lieu of on-site delivery. The consultation response is explicit about why: allowing that flexibility “could have a detrimental impact on the manifesto commitment to deliver the biggest increase in social and affordable housebuilding in a generation.”
You won’t find that decision written into a standalone medium-sites policy, because there isn’t one. Annex B now defines “Medium development” as a sub-set of major development — 10-49 homes, up to 2.5 hectares — and is explicit that “policies in this Framework which relate to major development involving the provision of housing also apply to medium development,” with the only carve-outs (footnote 71) being policies N4 and N5, neither of which touches affordable housing. So this wasn’t a proposal that got drafted and then rejected. It was a decision not to write a medium-sites exception into that cascade at all, which means HO5 and PM12’s full major-development affordable housing regime — on-site by default — now applies to a 12-home scheme on a tight urban site exactly as it does to a 1,500-home strategic allocation. A government that is pro-development, not pro-developers.
