A change of government usually means a policy reset, including for planning.
The arrival of New Labour kicked off a journey that led initially to PPG3 in 2000 and then a shift in direction following the 2003 Barker Review [2], which included the National Housing and Planning Advice Unit (NHPAU) and stronger ‘top down’ targets via Regional Spatial Strategies (RSS).
In 2010, the coalition government began unwinding that approach and took forward the localism agenda from 'Open Source Planning' [3]. However, it still took two years to formally revoke RSS [4] and formulate a replacement national planning policy [5]. Subsequent reform followed a twisty road that passed through countless milestones including White Papers; a 2018 NPPF; an abandoned mutant algorithm; on-off support for stalinist targets; a 2022 consultation on a backbench-facing policy adjustment; before latterly culminating in the December 2023 NPPF and a long-term plan for housing.
14 years later, Sir Keir Starmer becomes Prime Minister with a very clear platform to "get Britain building again" [6]. There are manifesto proposals to drive forward national infrastructure (via NSIP) and to update the NPPF with the aim of “making it easier to build laboratories, digital infrastructure, and gigafactories” [7] which will need to provide a much-needed reboot to how we plan for economic growth.
But it is on housing that Labour’s manifesto is most strikingly bold: 1.5m homes over the next parliament, with some punchy messaging about its willingness to take on policy shibboleths like Green Belt [8], "tough action" to ensure up-to-date Local Plans, a new generation of New Towns, and new mechanisms for strategic (not regional) planning. We can expect a draft NPPF before parliamentary recess [9] suggesting lessons have been learned from how long it took previous new Governments to cascade a national goal for planning into actual policy that guides plan making and deciding on applications.
The headline target bears repeating: 1.5 million homes in five years. This takes the last Government's 300,000 per annum ambition, but makes it accumulating, so any under-delivery in one year is added to what is needed for the rest of the period. It’s 50% more than was achieved in the last parliament [10].
Self-evidently, this raises all sorts of practical questions: around the housing market, access to skilled workers, the capacity of Registered Providers, and so on.
But this blog focuses on the practical planning challenge. A good starting point is a stock take of what the new Government is inheriting as it takes office, and what this might mean for how it approaches its reform agenda in the short term.
The planning system is currently targeting annual delivery well below 300,000
The Housing Delivery Test (HDT) results [11] show the combined annual monitoring benchmarks are well below 300,000 (in 2022 they totalled 259,000) and targets (requirements) in Local Plans currently add up to just 230,000.
More than half of LPAs (168) built fewer homes in the past three years than needed whilst 37 built twice as many as their target. 60 LPAs failed the HDT to the extent they are subject to the presumption in favour of sustainable development, but two thirds of these are insulated from the consequences of this by Green Belt.
Of the 305,000 homes in the current Standard Method [12], just under a third is in London, which has a Plan with a target of only slightly more than half, and no mechanism for redistributing what it cannot provide outside the M25. There is little evidence that the other areas subject to the 35% urban uplift are realistically capable of delivering on their elevated need figures (see Figure 1). Over 75,000 of housing need is in locations where Green Belt is a significant constraint and many of these areas surround tightly bound urban areas themselves with unmet need.
Figure 1 Summary of current standard method compared with current plan requirements and recent delivery – London and rest of England
Source: Lichfields analysis of current plans, DLUHC and standard method (PPG)
Most areas have plans that are – or soon will be – ‘out of date’
There is local plan coverage across most areas, but - even now - a number still pre-date the 2012 NPPF, and many others were adopted five or more years ago (see Figure 2, left). Preparation of plans has been slow, and very few were submitted for examination in 2023 (Figure 2, right).
Figure 2 Local Plan Age at April 2024 (left) | Number of local plans submitted for examination (right)
Source: PINS / LPAs / Lichfields analysis
Our modelling for the LPDF in 2023 [13] estimated that, by 2025, only around 20-25% of LPAs will have an up-to-date local plan (see Figure 3 below).
Figure 3 Age of Strategic Local Plans since adoption
Source: PINS / Lichfields analysis
Although there are some new local plans being prepared, and the LURA proposes to introduce a new system (originally due to commence in November 2024) in which plans are prepared (in waves or as and when plan makers choose) in a 30+4 month period, it is unlikely that the proportion of local planning authorities with an up-to-date local plan will increase significantly until late in this parliament, and any extra site allocations proposed therein are unlikely in of themselves to deliver housing before 2029 except in a small number of cases.
The number of residential planning permissions is below what is needed to deliver 300,000 net additions per annum and falling
There have been insufficient number of planning permissions granted to meet the 300,000 per annum target. The Competition and Markets Authority (CMA) [14] found that:
“The planning system is exerting a significant downward pressure on the overall number of planning permissions being granted. Over the long-term, the number of permissions being given has been insufficient to support housebuilding at the level required to meet government targets and measures of assessed need.” (para 34 of Summary)
The CMA cited Lichfields research which observed that delivering 300,000 annually would require a stock of approximately 1.4m homes with permission at any one time, and approving circa 350-375,000 dwellings a year [15]. Other modelling taking into account differential lapse rates by region (and presence of drop-ins and duplicates) indicates some 520,000 units might need to be granted permission in a single year [16]. Whatever the precise number, we know at minimum 1.8m homes with permission would be required.
Analysis carried out in 2021 compared the stock of live planning permissions against the number of homes needed (based on the Standard Method) or adopted local plan target (where relevant) and showed that there are certain parts of the country that have an acute shortage of consented sites (see Figure 4). In a number of locations, the stock and flow of consented sites has been suppressed by Green Belt and restrictions imposed by water and nutrient neutrality problems.
Figure 4 Homes with Permission relative to housing need or planned requirement (2021)
Source: Glenigan / ONS / Lichfields analysis
The number of homes granted planning permissions has been falling (see Figure 5) and is now below even the number of recent net additions, indicating that there is likely to be a strong drop off in output. Data released on 13th June [17] shows only 236,000 were approved (year to Q1 2024) supporting an implied delivery rate of 190,000 per annum (or 944,000 over five years) and the number is likely to fall, because the number of planning applications for residential development submitted last year was down 12% (for 'major development'). The number of major development applications has almost halved in seven years (down to 56% of 2016-17 level) and the number of minor development applications is down a third (down to 65% of 2016-17 level).
Figure 5 Number of homes granted planning permission by year
Source: DLUHC based on Glenigan / Lichfields analysis
September 2023 analysis by Planning Magazine found 39% of LPAs were unable to meet their five-year land supply obligations [18]. This had fallen to 26% in April 2024 [19]; but this was due to the lower expectations of the soon to be reversed December 2023 changes to the NPPF [20].
Savills forecast a reduction in housing delivery in England to 170,000 per annum, citing lack of support from Help to Buy and a poorly performing planning system. They suggest need for “a greater volume and range of sites, supporting a diversification of …delivery” [21].
Most projects take at least 2-3 years to pass through planning and deliver a first housing completion and on average then build out at rates no higher than 200 per annum
The third edition of Start to Finish by Lichfields [22] finds that lead-in times for new residential developments are counted in years (see extract of analysis at Figure 6), and although it can vary, the clear implication is that for homes to be completed within this parliamentary term (ending 2029) they will in the main need to emerge from the stock of existing permissions or new approvals granted in the first 2-3 years of the parliament. The subsequent the rate of build out is largely determined by the number of outlets and affordable housing (where viable); larger sites have higher build rates, but even so, the LQ to UQ range for sites of 2,000+ homes is 100 – 188 per annum.
Boosting housing completions in the first couple of years of the coming parliament will therefore largely be a product of what can be done to immediately unlock stalled applications already in the system (such as those held back by nutrients), general economic confidence, or through other financial measures. Beyond that, given the time it takes to prepare a planning application (typically 6-12 months), any new policy designed to achieve the increase in output would need to be in place almost immediately.
Figure 6 Median timeframes from validation of the first application to completion of the first dwelling (left) | Average build-out rate by size of site in dwelling (right)
Source: Lichfields analysis
Decisions on planning applications for housing are unpredictable
There has been a presumption in favour of sustainable development since 2012 and – coupled with an absence of up-to-date local plans in that period – that policy helped progressively drive an increase in the number of homes granted permission from around 180,000 per annum in 2012 up to the rates of >300,000 experienced from 2016. Many were granted at appeal, and in combination with emerging Local Plans it has driven the annual output over recent years to consistently above 200,000. But there is a lack of predictability. PINS data suggests for major dwellings appeals, under 45% are now approved, down from 59% in 2012/13. There has seemingly been an increase in refusal of applications by LPAs on allocated sites, and research by Lichfields in 2018 [23] showed that even when an LPA officer had recommended approval and it was subsequently refused by elected Councillors, Inspectors went with the decision of politicians (to refuse) 35% of the time, with more subjective planning judgements – such as on landscape - being the explanation in most cases. There is a perception in the development industry that the revisions to the NPPF in 2018, PINS custom and practice, and a succession of legal judgments have eroded the 'force' of the presumption compared to its original impact in 2012 [24].
What does all this mean for achieving 1.5m homes?
This analysis shows that, even before one considers other potential barriers to delivery (construction industry capacity, planning resources in LPAs and PINs, funding for affordable housing and infrastructure and so on), the planning context is challenging:
-
Net additions (completions) are unlikely to exceed 200,000 in the short term and will need to ramp up. Addressing the backlog by 2029 means a residual annual target of >300,000 in years 4-5 [25], many more than has been achieved in the recent past.
-
The increase in supply needed to approach the 1.5m will realistically not emerge to any great extent from new strategic or local plans prepared pursuant to any new policy (or New Towns), but in large part by encouraging the immediate submission and approval of planning applications ahead of - or alongside - Local Plans [26]. This includes on unallocated sites, particularly in areas where there has been historic housing under-delivery, and including Green Belt (hitherto seen as ‘no go’ area for speculative applications in all but the rarest situations).
-
In view of the low starting point (of local plans, land supply and permissions), the policy support for housing delivery required to encourage applications achieve this step-change would necessarily need to go beyond the 2012 or 2018 iterations of the NPPF and be rapid indeed in its effect.
At its 2023 Conference, Labour stated it would advance a "housing recovery plan; a blitz of planning reform to quickly boost housebuilding to buy and rent" [27]. What might this mean for the soon-to-be-revised NPPF? There are five planning policy elements likely to be necessary if the Government is serious about aiming for its 1.5m target:
- Reform and strengthen the presumption: The circumstances in which the presumption in favour of sustainable development applies to decision taking would need to be broadened/clarified, including through out-of-date plans, lack of five-year housing land supply and past under-delivery. The 'tilted balance' part of the presumption would need to be strengthened ('extra tilt') with some prescription on weights applied in the balance of benefits and harm to improve consistency of approach. This might include clarifying that the impact on 'everyday' non-valued/non-protected landscapes is not a reason to refuse applications for otherwise acceptable proposals in sustainable locations.
- Update local housing need: the standard method needs updating (probably with a simple, stock-based approach with affordability adjustment for prices and rents) to achieve an assessment of need that better reflects market signals and distributes it more evenly, to open up the number of local authority areas where land is released for development [28].
- Unlock poor quality Green Belt: The Labour Manifesto is clear that the general extent and purposes of Green Belt will not change, which means it can retain protection via up-to-date plans. But there is no realistic route to 1.5m that does not pass through Green Belt areas. Housing delivery in the short term (and incentivising Local Plan production in Green Belt areas) means LPAs without up-to-date local plans should no longer be able to apply current levels of resistance to development behind the protective shield of ‘very special circumstances’ (VSC). This could be achieved by reducing the current automatic 'substantial' weight afforded to harm to the Green Belt in specific circumstances, particularly for previously-developed land or 'grey belt', and by clarifying the factors that can represent VSC justifying development. For new local plans, there should be a requirement for strategic policies to review Green Belt to meet needs, with a 'brown-grey-green' sequential approach to finding sites to release [29] so that new local plans can then provide current levels of protection to the Green Belt retained.
- Remove technical blockages: NPPF reform can address some practical issues holding up housing delivery including on the flood risk sequential test, reinforce/clarify the need to recognise the legal responsibility of statutory undertakers/regulation, ensure a proportionate approach to mitigation on nutrients [30], and provide alternative means of supporting affordable housing delivery where there is a lack of demand from Registered Providers for on-site s.106 affordable homes in some locations/cases [31].
- Lay the ground: for the rolling out of strategic plans, local plans and nationally-significant New Towns that will provide the platform for effective, plan-led delivery of housing and economic growth in the medium to long term. This will be necessary to show that, even if housing delivery in the short term is boosted by more planning applications running ahead of local plans (with all the political challenge this may cause), there is a credible plan-led route to housing delivery within reach for most areas.
None of this is a silver bullet that diminishes the need for complementary measures to overcome other barriers to housing delivery. It would bring obvious political challenge. But it would be a policy platform that puts the planning system on a 'recovery' footing, consistent with the manifesto on which Labour has been elected. It appears to be the minimum needed - for the next five years - to reduce the risk that the Government's new dawn for housing delivery finds itself perceived as a false one.
[1] That date was also memorable, for me if nobody else, as being the day I received a job offer from Lichfields as a planning assistant. 27 years on, I’m still here.
[2] The Barker Review of Housing Supply – available here
[3] The Conservative Party Green Paper on planning – available here
[4] After the CALA litigation
[5] Published in March 2012 and available here
[6] See the Manifesto here
[7] See the Manifesto section on Economic Infrastructure here
[8] See this piece in The Sunday Times (£)
[9] See this analysis by Planning Magazine (£)
[10] See this almost valedictory pre-election (24th May 2024) analysis from DLUHC here.
[11] HDT 2022 available here
[12] Seen by many as outdated due to reliance on old, 2014-based projections and volatile due to annual changes in affordability ratios.
[13] Source: Timed Out? A projection of future local plan coverage in 2025 under prevailing policy conditions
[14] CMA Market Study Final Report into housebuilding – available here
[15] CMA Market Study into Housebuilding: Planning Working Paper – available here
[16] Source: Taking Stock. This reflects that, with a third of housing need in London and a much higher lapse rate in the capital, the stock of permissions nationally would need to be much greater.
[17] DLUHC Planning Applications January to March 2024 – available here
[18] Planning Magazine analysis September 2023 here (£)
[19] Planning Magazine analysis April 2024 here (£)
[20] The December 2023 NPPF exempted some LPAs from having to demonstrate a five year land supply, changed buffers, and meant others (with a draft plan) only needing a four year supply.
[21] Savills Housing Completions Forecast for England, May 2024 available here
[22] Start to Finish is available here
[23] Source: Lichfields: Refused for Good Reason August 2018 available here
[24] See for example this report in Planning Magazine (£) with Gladman Developments quoted as saying: “Many applicants, including us here at Gladman, have begun to avoid the appeals system as it is now extremely difficult to assess your chances of success even where the presumption in favour of development is engaged… Until this situation is resolved by government, and clarity is provided, there will be even fewer applicants who will decide to take the significant financial risk of conducting an appeal when the chances of success are so clearly diminished."
[25] This was explicitly recognised by Keir Starmer in his Times interview when he said: “It’s more likely, I think we’ll ramp up over the parliament. Therefore, towards the end of that, we’ll be doing more than the 300,000 [a year].”
[26] Including by way of some SoS interventions on stalled proposals or current emerging Local Plans that are pushed through the system
[27] See Labour’s 11th October 2023 press release here
[28] Currently, many housebuilders are not active in otherwise strong local housing markets because deliverable housing land is not released by the planning system.
[29] The definition of ‘grey belt’ is stated to be be broader than brownfield, covering areas without permanent structures such as car parks and some green spaces with “little intrinsic beauty or character” – see The Times here (£)
[30] Taking forward recommendations such as those outlined in this Lichfields work for the HBF here
[31] Addressing the ‘cliff edge’ issues highlighted by Inside Housing here