Buildin’ in the Rain: flood risk in the courts
Wellies on. Brollies up. You may’ve noticed this already, friends, but… it’s getting wet out there. July to December 2023 was the UK’s wettest period on record. A series of creatively named storms – Babet, Ciarán, Debi, Elin, Fergus, Gerrit and Henk – turned Britain into what one hydrologist described as “a sopping wet sponge”. Of course, with the storms came the floods. And floods on a pretty biblical scale which led to untold property damage, evacuations, injuries and even fatalities. On account of our warming climate, this position is only going to get worse.
So. How does the planning system manage the competing considerations of (a) finding land to meet desperate needs for more development, with (b) a country that is wet and is getting wetter.
A key policy for trying to do that is the “sequential test” for flood risk. The requirement to test alternative sites in sequence with the aim of steering development where possible toward lower areas of flood risk has been a feature of national planning policy for decades – before the advent of the NPPF in 2012 it was found in PPS25, and before that (and long before the days of hyperlinks!) PPG25.
Now. There have been two big court decisions on the sequential test in the last few weeks:
In mid-January, the Substation Action case in the Court of Appeal - here; and then
Only yesterday, the Redrow / Mead case in the High Court - here (full disclosure - I acted for Redrow).
What do they have to tell us?
A lot. They are - without wanting to sound dramatic - essential reading for anyone involved in these issues in practice, and they will both require (in this humble blogger’s opinion, anyway) updates to the PPG at the very least.
NB this blog post is not for the faint of heart! It is, I’m sorry to say, a bit too involved for my “basics” series. Long-time readers will know I do my very best to keep things short and sweet. But these matters can get a little technical. This post skims the surface. But, it goes without saying, if you need to get deeper into the weeds, this is one to take some specialist advice on. Annnnnyway….
First thing’s first - before we get into flood risk - a point of much more general application…
Can the PPG change national policy?
Yes.
In yesterday’s Redrow/Mead judgment to which I will return, Mr Justice Holgate in the Planning Court held that the PPG has the same legal status as the NPPF.
So, legally at any rate, the PPG is not “only” guidance. The PPG can, like the NPPF, make planning policy. It can, and it does. The PPG doesn’t simply have to provide guidance on that policy. The court recognised that, as a matter of policy, the PPG is intended to support the NPPF. But that’s policy, not law. There is, the Court said, no legal reason that the PPG cannot change or be inconsistent with the NPPF.
This point may come as a surprise to you if you’d been following a series other High Court judgments which had held the PPG to be “guidance not policy”. Or if you recall the Minister who introduced the PPG talking about “a simplified set of clear, concise guidance”.
We’re moving on from that now. In a nutshell, the PPG is a tool which can, and now regularly is (see e.g. even the other day the updates on 4 year housing land supply) not just explaining national policy but making it.
Next, let’s get into the flood risk…
When do you have to do a sequential test?
Let’s start with the key bit of policy. It’s §168 of the NPPF which is essentially unchanged since 2012 - it says that:
When exactly you’re supposed to do a sequential test has been a topic of massssssive confusion. Let’s take a step back:
There are different sources of flooding. Including from seas and rivers, from rain, from rising groundwater. Areas at risk of sea flooding and river (or “fluvial”) flooding are within Flood Zones 2 and 3 mapped by the Environment Agency. For areas at risk of flooding when it rains (i.e. “surface water flooding”) there are no such standardised “zones”.
In the original PPG, it was clear that the sequential test was only interested in locating development, so far as possible, out of Flood Zones 2 and 3. It was not concerned with other kinds of flood risk, e.g. surface water flood risk. Those kinds of risks could be managed on-site e.g. through well designed drainage schemes.
This was important because doing a sequential test can be a laborious process; and surface water flood risk arises on almost every larger site – got a depression in the middle of your site where it puddles in heavy rainfall? That’s some surface water flood risk. But if you can design your flood risk and drainage scheme to manage the puddles, do you really still need a sequential test? Under the old PPG, no.
And - we now know - same is true under the NPPF. In the Substation Action case, the Court of Appeal said that “the provisions of the Framework do not, however, require an applicant for development consent to demonstrate that there are no other sites reasonably available if any part of the development is to be located in an area where there is a risk of flooding from surface water” (as distinct from a risk of fluvial flooding).
The Substation case was considering the current language of the NPPF. But it was not considering the current PPG, which was updated in August 2022.
The new PPG says that “the Sequential Test ensures that a sequential, risk-based approach is followed to steer new development to areas with the lowest risk of flooding, taking all sources of flood risk and climate change into account.” That reference to all sources of flood risk has led lots of authorities and indeed inspectors to assume that a sequential test may be required simply on account of surface water flood risk. But is it really so different from the reference at §168 NPPF (“the aim of the sequential test is to steer new development to areas with the lowest risk of flooding from any source”) which we know from the Court of Appeal in the the Substation Action case did not have that effect?
Individual planning applications won’t normally need a sequential test if your site is already allocated (and the allocation was itself sequentially tested), or your site is at a low risk of flooding - on which see above! - without reference to any drainage schemes or flood mitigation proposals which are part of your scheme. So you take the site as it is. Even if - with your development in place - there wouldn’t be any real world flood risk. Common sense? Shmommon shmense! [Really? Ed.]. That sequential testing for you.
Here’s a common issue: what if you have only part of your red-line area within an area of flood risk, then if e.g. there is no built development proposed in that area, is a sequential test still required? Here’s an unhelpful answer: views differ. Planning Inspectors have taken mutually irreconcilable approaches on the issue. The NPPG and PPG aren’t determinative on it. The Planning Inspector’s Training Manual sets out a clear view on the matter, but (i) PINS has been very clear that isn’t policy or guidance, and (ii) the clear view isn’t based on anything. So long as views continue to differ, think carefully about including unnecessary areas of land in your red line which are at risk of flooding.
Of course, if you pass the sequential test, you may then need to pass the “exception” test too. But that’s for another blog post. I think we all will need a break after this one.
What sites have to be sequentially tested?
Again - the starting point is the NPPF above at §168 - “Development should not be allocated or permitted if there are reasonably available sites appropriate for the proposed development in areas with a lower risk of flooding.” But hang on:
What kind of sites are “reasonably available sites appropriate for the proposed development”?
Within what catchment area?
Can they be smaller than your site? Bigger?
Available now? Or some time in the future?
What happens if after you’ve looked high and low, there are some sites which are sequentially preferable to yours in flood risk terms – must the application be refused?
These questions (and more) have been analysed by the High Court in a bumper judgment handed down yesterday by Mr Justice Holgate: here.
The judgment dealt with (and dismissed) two claims – one by Redrow Homes, a national housebuilder, and another by Mead Realisations, a land promoter based in Somerset. The claims were heard together in January 2024. Again, full disclosure – I acted for Redrow. To give you the key facts:
The claims challenged two inspectors appeal decisions:
The Redrow decision (dismissing an appeal for up to 310 homes, land for a primary school and some community facilities in Bushey in Hertmsere, Hertfordshire): https://acp.planninginspectorate.gov.uk/ViewCase.aspx?caseid=3314268
The Mead decision (dismissing an appeal for up to 75 homes in Wick St Lawrence, Weston-super-Mare in North Somerset): https://acp.planninginspectorate.gov.uk/ViewCase.aspx?caseid=3313624
The Redrow site was subject to some surface water flood risk, and parts of the site (albeit not parts intended to be built on) were within flood zones 2 and 3. The Mead site was within flood zone 3. In both cases, it was decided – eventually – that a sequential test was required in relation to flood risk. In both cases, permission was refused on appeal in part because of findings of conflict in relation to the sequential test.
The key questions for the Court were on how the sequential test works. In particular:
What kind of sites are “reasonably available sites appropriate for the proposed development”?
What relevance do needs for the development have – i.e. what happens if your site isn’t the top of the tree in terms of flood risk, but it is accepted that development will be needed on many, many more sites including those which are less sequentially preferable.
The updated PPG says:
That reference to a “series of smaller sites” has proven… problematic. To take the Redrow case, even though the “proposed development” included 310 homes and other things, the Inspector had found 9 sites to be sequentially preferable in flood risk terms which were much, much smaller so could only accommodate a fraction of the proposed development. When there had been evidence put to the Inspector which explained why the different elements of the scheme relied on one another.
The Judge decided this approach was wrong because:
The PPG refers to a “series of smaller sites.”
The word “series” connotes a relationship between sites appropriate for accommodating the type of development which the decision-maker judges should form the basis for the sequential assessment.
So a proposal should not automatically fail the sequential test because of the availability of multiple, disconnected sites across a local authority’s area. The issue is whether they have a relationship which makes them suitable in combination to accommodate any need or demand to which the decision-maker decides to attach weight.
On the facts of the Redrow case, the Inspector was wrong to find 9 smaller sites to be sequentially preferable to Redrow’s site because “there was no indication of any connection or relationship between those sites”.
Do alternative sites have to be available for development within precisely the same timescale as your scheme?
No.
But timing is obviously relevant - the PPG tells us so (“… available to be developed at the point in time envisaged for the development”).
How relevant? That’ll depend on the facts of your case, the nature of your development and the needs you’re seeking to meet.
Mr Justice Holgate leaves that issue, and a raft of other issues, up to the planning judgment of the decision-maker.
Is need relevant to the sequential test?
Yes. It can be. If e.g. the developer makes a case that a particular need for a particular kind of development has the effect of (i) limiting the area of search your sequential test needs to cover, and (ii) building parameters into the search for e.g. the size of the site needed, the scale of the development, it density etc. A specific need can be relevant then to how the sequential test works.
But what about general need, i.e. a 5 year housing land supply shortfall in an area? The PPG suggests it’s totally irrelevant. Wellllll… maybe not quite.
Let me give you an example:
The Redrow site is in Bushey. The relevant authority is Hertsmere. We will, inevitably, need thousands - many thousands - of homes in Hertsmere to make for many years of shortfalls in delivery. They will, inevitably, need to be spread out over tens or even hundreds of sites.
But the Inspector held that the Redrow site failed the sequential test because (after you knock out the 9 small sites above) there were… 3 sites. 3. Which had not been excluded as unavailable. On the planning authority’s own evidence, avoiding areas altogether which contain a lower flood risk than the application site… was and is not possible. Not even close.
When circumstances like that arise, the Court decided that this could be an “obviously material” factor which could justify reducing the weight to be given the sequential test in the overall balance, i.e. you can get your planning permission anyway even if the sequential test is failed.
Again, that could be a surprise to many of you. If you’ve assumed that failing a “footnote 7” policy in the framework adds up to - in the language of §11(d)(i) - “a clear reason for refusing the development proposed”… you may be wrong. In the end, the Court is reminding us, the NPPF is just another material consideration in the balance. You can fail the sequential test, but if land at risk of flooding will inevitably be needed all over the area, there may be good reasons to grant planning permission anyway.
Are you feeling exhausted yet? As I said above, there’s lot to digest in these decisions, and the above is just a primer. But I hope you find it a useful start. Good luck out there, #planoraks. I hope you stay well. Or at least, stay dry. Keep refreshing those PPG pages - you never know when change is afoot. And in the meantime, even though pesky lawyers keep getting in the way, do your level best to #keeponplanning.
