The basics #7 - signed, sealed, delivered
What makes housing sites “deliverable” - a brief (non-technical) history.
These are treacherous times for #planoraks, with all this talk of taking axes to some things and overhauling others (if you’d like to talk about zoning, then allow me to welcome you to Euclid).
But let’s bank on having a town and country planning system for at least another week or two. Which may make it worth settling in for just 5 minutes if you have them for a fresh round of #planorak basics.
Let’s talk about housing delivery.
We’re almost 15 years into the requirment for LPAs to show a 5 year housing land supply (hats off to PPS3). Of course, falling below a 5 year supply of deliverable sites triggers our much-discussed tilted balance. Which may make this important stuff. But what makes a housing site “deliverable”?
Under PPS3, deliverability had 3 ingredients. Sites had to be:
Available now.
Suitably located for development which would contribute to sustainable, mixed communities.
Achievable, i.e. with a reasonable prospect that housing will be delivered within five years.
The 2012 NPPF added two extras: sites needed to be viable + sites with planning permission were presumed deliverable until the permission expires absent clear evidence that the scheme wouldn’t be implemented within 5 years.
How high a bar was that intended to set? How sure did an LPA have to be that a site would come forward before it could insert that site into its housing trajectory? That takes us to the Court of Appeal in the 2017 St Modwen case where Lord Justice Linbdlom said that:
Sites can be “deliverable” without ever being delivered, or even likely to be delivered.
That’s because “deliverable” under the 2012 NPPF didn’t require certainty or even any probability that homes would actually be delivered.
All the test needed was a “realistic prospect” of delivery - a low bar.
Roll onto 2018 - another NPPF, another layer added to the definition of “deliverable”. All of the above elements, more or less, remain but now with a further limb:
“Sites with outline planning permission, permission in principle, allocated in the development plan or identified on a brownfield register should only be considered deliverable where there is clear evidence that housing completions will begin on site within five years.”
Add to that, PPG which talks about “firm progress” and “clear relevant information” and you might well think that the Government was trying to respond to (/reverse) the Court of Appeal in St Modwen by raising the bar to justify including sites in housing trajectories.
That was certainly the impression planning inspectors received in cases like the Woolpit or Woolmer Green appeals which applied a considerably higher bar to the 2012 NPPF approach. In particular, there is no longer a presumption that sites without detailed planning permission are deliverable, and the onus falls on the LPA to demonstrate (with “clear evidence”) that sites without detailed planning permission will be delivered within 5 years.
This shifting definition has led to some surprising results - particularly when we find “transitional” local plans examined under the 2012 NPPF vs. appeals being decided at almost the same time under the 2018 or 2019 NPPFs. For example, in this Hanslope appeal decision from 2019 (full disclosure: I acted for the appellant) Milton Keynes lost its 5 year supply of housing land only a few months after it had adopted Plan:MK because of - in a nutshell - the change in definition between the 2012 and 2018 NPPFs. See also the more recent decision in Harrold (again full disclosure: I acted for the appellant) which the Secretary of State has agreed should be quashed because (among other things) that Inspector failed to give reasons to explain the differences between the 2012 and 2019 definitions of “deliverable”.
So overall - until recently - the pendulum appeared to be swinging against LPAs. But not so fast. Here are 2 bang-up-to-date pieces to add to this puzzle which might suggest it’s swinging the other way again:
The Secretary of State has now accepted that the list of categories of “deliverable” site in the NPPF isn’t closed. Other kinds of site can be deliverable if they’re available / suitable / achievable. Albeit, as above, we’re still after “clear evidence” for major sites without detailed planning permission: Harry Bennett of Lichfields gave an excellent summary of the implications of that case here.
The latest SoS Woburn Sands decision has lots of controverisal material for housing land supply #planoraks, including on the circumstances when you can (apparently) include post-base date evidence to support deliverability, and what is said to be the ongoing relevance of St Modwen approach to the “realistic prospect” test even under the 2019 NPPF (albeit the Inspector made this point by relying on East Bergholt which was a case about the 2012 NPPF 🧐). I promised at the top that this peice would be non-techincal so I won’t dwell on base dates any further (thank heavens for that).
In the end, a familiar refrain for regular readers: assuming you’ve defined it properly - deciding whether or not a site is actually “deliverable” requires the fact-sensitive exercise of planning judgment with which the courts won’t normally interfere.
So much for what policy says now. What should it say next? Of course, housing delivery will be under strain this year and next like never before. Some say the answer is to ditch the deliverability test altogether. I asked Philip Barnes about that idea, and he said (memorably, I thought!):
“The Government hasn’t advised that less babies should be conceived and born during the pandemic so I’m not expecting them to say less homes need to be built as a result of it.”
Still, times are a-changing and who knows what is next for “deliverability”. Here’s a head-scratcher for you (answers on a post-card to our new planning task-force): how could a “deliverability” test fit into our fancy new zoning system? That is, thank goodness, a topic for another day.
Stay well, #planoraks.
