It is time to return to fundamentals when it comes to the delivery of social housing, writes Simon Ricketts, as he reviews recent developments in relation to two major projects.
I thought I would start 2020 by trying to establish some common ground, before then mentioning what happened shortly before Christmas in relation to the Elephant & Castle and Old Oak projects, both controversial in different ways. The questions are long but I hope that the answers are short.
Do we all agree that…
- More housing is needed for those who cannot afford homes that are being built by the private sector in their local area, even when these are required to be sold or let at significant discounts to market rates – and that what we call that housing (eg social housing/socially rented) and the nature of the body that delivers and manages it (housing associations or other registered providers, local authorities) are secondary issues?
- The current system of seeking to require developers to deliver that housing (whoever then manages it) is not working and is hugely inefficient, in that: (1) local policy expectations set out in local plans are often not met, due to those expectations being determined not to be viable – leading to prolonged negotiations and local objection; (2) the complexities and multitude of inputs to any negotiated section 106 affordable housing package, often including intricate mechanisms to provide for later reviews of the viability position, are at best a costly distraction for all parties (needing to be tooled up with valuation and QS professionals) and at worst are prone to lead to huge delays and, over time, the prospect of renegotiation where the negotiated outcome is not sufficiently attractive to funders, or where (almost inevitably) circumstances have changed during the long course of the process?
- It is in the public interest for communities within developments to be socially and economically diverse?
- The system worked more easily when much more Government money was available to support affordable housing by way of grant (without grant obviously a requirement to deliver social housing has a huge impact on the viability of a scheme) and that we need to get back to a system that: (1) is simple; (2) delivers housing that is truly affordable for those who need it; (3) is efficient; and (4) does not delay development more generally?
- Government (ie our) money needs to be spent where it can have most beneficial impact and is most needed?
There has been a lot of government tinkering but don’t we have to get back to those fundamentals? I’m not sure that the Government’s promised Social Housing White Paper is going to get us there, given the absence of relevant detail about affordable housing in the Conservatives’ manifesto – talk about owning first homes is a world away from the very different challenges faced by so many.
I am sorry to be a cracked record – see my 28 May 2017 blog post Affordable Housing Tax or 4 November 2017 blog post Viability Assessment Is Not A Loophole, It’s A Noose. We could look at the idea of expanding CIL to include a social housing contribution, so that local authorities can deliver or procure it, with the option of provision on site counting as works in kind? But I have previously been against further rolling out another complex and inefficient regime, ie CIL, and most authorities, hollowed out and stretched as they are, are not currently in any position to deliver or procure social housing at scale. Instead, personally I would simply prefer that we go back to the old way – grants to providers so as to reduce the impact on viability for the developer of providing social housing.
In the meantime, we have to make the current system work. My 8 June 2019 blog post The Bottom Line: Updates On CIL And Viability reported on the RICS professional statement on financial viability in planning, which came into effect on 1 September 2019, and mentioned the revisions made to viability passages of the PPG by the Government on 9 May 2019, reflecting changes to the NPPF that seek to ensure, amongst other things, that detailed viability examination takes place at plan-making stage rather than when applications come forward.
The RICS professional statement sets out the professional responsibilities of the surveyor in the viability appraisal process, to seek to ensure that the surveyor operates with professional independence and integrity throughout. The RICS is now consulting from 13 December 2019 until 9 February 2020 on a draft guidance note Assessing financial viability in planning under the National Planning Policy Framework for England, 1st edition that seeks to set out the methodology to be applied by those professionals, so as to give effect to Government policy.
“We are not seeking comments contrasting the government framework with a market-based appraisal. Comments should focus on whether our draft guidance gives effect to government policy and practice guidance, in an administratively efficient way, in order to deliver the objectives of the NPPF.”
Make your views known.
In the meantime…
Elephant & Castle
Delancey’s proposed redevelopment of the Elephant & Castle shopping centre and London College of Communication has long been controversial. It proposes a large mixed-use development comprising a range of buildings of up to 35 storeys, with a mix of uses including 979 dwellings (proposed to be for rent rather than sale) and accommodation for retail, office, education, assembly and leisure along with a remodelling of the London Underground station. One of the lines of attack for objectors, including the 35% Campaign, has been the perceived lack of “genuinely affordable” housing.
Planning permission was granted by the London Borough of Southwark on 10 January 2019. Just before Christmas, in Flynn v London Borough of Southwark (Dove J, 20 December 2019), the High Court rejected a crowdfunded challenge to the permission brought on behalf of the 35% Campaign. The grounds of challenge all turned on the affordable housing deal that Southwark struck in the section 106 agreement with the developer.
The case doesn’t turn on any particularly interesting legal principles or make any new law. But the facts, set out in careful detail by Dove J, illustrate precisely the concerns that lay behind my attempt just now to establish some common ground:
The policy background is not straightforward, with a changing position both at borough level and at London Plan level.
The Mayor has set out criteria in his 2017 affordable housing and viability SPG for different tenures of affordable housing, including social rent (target rents determined through the national rent regime), affordable rent (rent controls requiring a rent of no more than 80% of the local market rent), intermediate (available for rent or sale at a cost above social rent but below market levels – and eligible only to households whose annual income is within a defined range) and intermediate London Living Rent (only available to households renting with a maximum income of £60,000 without sufficient current savings to purchase a home within the local area).
The adopted London Plan requires boroughs to seek the “maximum reasonable amount of affordable housing…when negotiating on individual private residential and mixed use schemes, having regard to” a number of factors, including “development viability” and the “availability of public subsidy”.
Within the Elephant & Castle area, Southwark’s adopted plan seeks a minimum requirement of 35%, on the basis of a split of 50% social rented and 50% intermediate housing. Its emerging plan seeks, in relation to build to rent developments, a different tenure split for the 35%: social rent equivalent (ie social rent level but not managed by registered provider) 34% minimum, affordable rent (aka discount market rent) capped at London Living Rent equivalent 52% minimum, affordable rent (aka discount market rent) for household incomes between £60,000 and £90,000 per year 14% minimum. The lack of social rent reflects the specific nature of build to rent developments, where it is more efficient for all of the housing to remain under single management rather than for a separate registered provider to be introduced.
At the time Delancey’s application first went to committee on 16 January 2018, its proposal was 36% affordable housing based upon habitable rooms, with the 36% made up as follows: 10% social rent equivalent, 46% London Living Rent, 43% discount market rent. The non policy compliant offer (in terms of tenure split) was based on an agreed viability assessment. Despite a recommendation for approval, members deferred a decision until a meeting scheduled for 30 January 2018 at which they intended to formulate reasons for refusal. The day before the follow-up meeting the developer made further proposals in relation to the affordable housing offer and the application was deferred to a subsequent meeting.
The revised proposal was to replace 33 social rent equivalent units with 74 socially rented units, all to be located on the western part of the development and to be owned and operated either by the borough or by a registered provider. This changed the tenure split (of the 35% affordable housing dwellings) to: social rent 24.9%, London Living Rent 27.9%, discount market rent 47.2%.
In June 2018 the offer was increased again. The developer’s consultants indicated that following “in-principle agreement from the GLA to provide grant funding towards the proposed scheme” the number of social rent units could be increased to 116 homes, or 38.1% of the 35% of the units that were to be affordable.
The application was approved at a committee meeting on 3 July 2018. It was acknowledged in the report that the proposed tenure split was still not policy compliant but was justified by way of the agreed viability appraisal. The report also noted that there would need to be a fallback arrangement in the section 106 agreement to cater for the possibility that the developer might choose after all to develop the western part of the development on a for sale rather than for rent basis (in which case the affordable housing requirement for that part of the site would return to 50% social rented, 50% intermediate).
If all of this does not start to give an idea of the inevitable complexity of negotiations on a scheme such as this, then consider the viability appraisal. As is common with a significant longterm development, where application of the more straightforward benchmark land value plus developer’s profit approach does not reflect accurately the financial modelling of a project over time, viability was judged against a minimum internal rate of return for the developer.
The latest RICS draft guidance defines internal rate of return (or “IRR”) as follows:
“The rate of interest (expressed as a percentage) at which all future project cash flows (positive and negative) will be discounted in order that the net present value (NPV) of those cash flows, including the initial investment, be equal to zero. IRR can be assessed on both gross and net of finance.”
However, unless I have missed it, there is no guidance anywhere as to when an IRR approach is appropriate and how to arrive at and test the inputs and modelling.
The agreed benchmark was 7.15% IRR, with annual growth to 11% over the construction period. Review mechanisms in the section 106 agreement provide that 50% of any excess are to be applied to increasing the affordable housing provision up to a policy compliant level/tenure split.
The claimant had three grounds of challenge. The first turned on an alleged inaccuracy in the way that the GLA’s offer of funding had been reported – it had not been formally confirmed and discussions were at an “in principle stage”. The second alleged that one of the detailed mechanisms in the section 106 agreement departed from the relevant head of term in the committee resolution. The third related to the mechanism in the section 106 agreement for determining the affordable housing to be provided if the western part of the site turned into a “for sale” development, but a deed of variation had been entered into after the challenge was brought, largely correcting the error that had been identified.
Dove J rejected each of the grounds, whilst accepting that each was arguable. (1) The report did not materially mislead members. (2) The section 106 mechanism was not outside the scope of the committee resolution (“True it is that the solutions arrived at are not a literal interpretation of paragraph 364 [of the report to committee], in that they do not include for the provision of land and a substantial cash dowry to construct the social rented units but, in my judgment, that was not required in order to remain within the scope of the delegation granted by the members”). (3) The approach to the fallback (“for sale”) scenario was “entirely rational and appropriate”. Part of the claimant’s criticism of the arrangements turned on whether the additional affordable housing in these circumstances should be social rented units rather than the social rented equivalent units provided for. The judge saw nothing relevant in the distinction:
“In terms of the matters raised by the Claimant the quality of tenure enjoyed by tenants in social rented equivalent properties are, as the nomenclature suggests, equivalent to those in social rented properties. Of course, there may well be nuanced differences between them as a consequence of them being separately defined. Furthermore, they will be managed in different ways as the definition implies. Be all of this as it may, in my view the important point is that the requirement of the officers’ report was a review in terms of affordable housing, and whether the additional habitable rooms were to be provided as social rented or social rented equivalent accommodation was not identified as being in any way a critical point upon which the delegation to the officers of authority to enter into the section 106 obligation turned. Put another way, whatever may be the nuanced differences between social rented equivalent property and social rented units that was not identified as a key requirement in relation to the review mechanism contemplated were the developer to take up the fall-back scenario.”
Will the new guidance make any of this more straight forward? I doubt it. Would proper funding for social rent and social rent equivalent housing? Of course it would.
Old Oak and Park Royal Local Plan
The recent NPPF and PPG changes of course seek to move the viability spotlight to the point at which sites are allocated for development. The Old Oak plan was examined last year under the previous NPPF but viability matters were still centre stage and the inspector’s findings may be an indicator of the detailed scrutiny that is likely to be given to the viability in particular of strategic sites (taken together with proposed policy requirements in terms of infrastructure delivery and affordable housing).
One of the key issues for the inspector was whether the proposed allocation of the 54 acre Cargiant site for residential and associated development was viable. Cargiant had itself attempted development of its site in the past. It had concluded that it would be unviable to contemplate relocating or extinguishing its business and carrying out the development – and took the position that there was no reasonable prospect within the plan period of the Old Oak and Park Royal Development Corporation (“OPDC”) being in a position to carry out such proposals, even by resorting to compulsory purchase and even with the benefit of £250m Housing and Infrastructure Fund monies which had been agreed in principle to be allocated by MHCLG.
My firm acted for Cargiant and so I will restrict myself to pointing out the level of detail to which the inspector went in his interim findings on viability of Cargiant site proposal (10 September 2019) before concluding that the allocation would be unviable and therefore unsound.
The day after the general election, on 13 December 2019, the OPDC announced that it would change its proposals, which will now leave Cargiant in place:
“The Old Oak and Park Royal Development Corporation (OPDC) has today set out a revised approach to deliver tens of thousands of new homes and jobs through collaboration with major public sector landowners.
The regeneration of Old Oak, Park Royal and surrounding areas in west London, has the potential to deliver 25,500 new homes and 65,000 jobs over the next 30 years. OPDC has already approved plans for over 5,000 homes including 1,500 already completed or being built.
The shift in approach has been triggered by recent, rapid increases in industrial land values in west London which mean that it is currently not financially viable to deliver OPDC’s early regeneration plans at Old Oak North. This area, close to the planned new HS2 interchange station, includes the 54-acre site that is owned and operated by Cargiant, which had originally been earmarked for development.
Earlier this year, the Planning Inspector, in his interim report on the OPDC’s draft Local Plan, de-designated the Cargiant site from Strategic Industrial Land, but also concluded that Old Oak North had become commercially unviable for residential-led development at this time.”
Whilst this situation might be taken to be an example of how viability matters can indeed in practice be taken into account at the plan-making stage, I do have concerns:
- There is now a bigger onus on authorities to carry out proper viability work, including work to a sensible level of detail on strategic sites (albeit often with assistance from those promoting those sites for development), and is it actually going to be done?
- Where it is not done, delays will occur in the examination process. At Old Oak, the necessary work had not been done and there was a significant hiatus whilst it was commissioned.
- Development proposals are often not sufficiently worked up, at the stage that the plan is being prepared, so as to enable a sensible viability appraisal to be undertaken. And will developers be prepared always to come clean at the allocation stage as to the challenges they are facing in making the numbers stack up?
- Will there always be participants in the local plan examination process with the motivation and resources to put authorities to proof on the work that has been carried out? If Cargiant hadn’t taken its stance (entailing lawyers and a team of consultants to challenge much of the inputs) I suspect the allocation would have been confirmed without challenge – and then proved over time to be undevelopable.
The next blog post will be shorter, I promise.