4. Calculating The Extent Of The 5 Year Effective Housing Land Supply
- Question 3 in the consultation paper focused on changes to paragraph 125 of the Scottish Planning Policy.
- Different views on the best calculation to use have caused considerable debate. We wanted the approach to this calculation to be clearer and more consistent.
- We proposed that the 'housing supply target' is averaged out over the life of the plan to provide a broad guide on how many homes should be expected over 5 years. The extent of land available for housing at any given time would then be compared with that number.
- Our view was that this would be more appropriate, as well as easier to understand and apply consistently, than more complex and detailed calculations that take into account any shortfall in the number of homes built.
As set out in our consultation analysis report there was both support, and opposition to, these proposals. In summary, the analysis of responses showed that there are strong and contrasting views on a number of issues:
- There was wide, but not unanimous, agreement that a clearer approach to calculating the extent of the 5 year land supply would be beneficial for all.
- Many respondents supported the proposal because:
- An agreed approach would help to reduce disputes and debates, if robust and straightforward.
- It would be helpful to planning authorities in defending their local development plan and making balanced planning judgements.
- It was viewed as consistent with the approach taken to Housing Land Audits and the planning performance frameworks.
- It would be helpful if decisions were based on the most recently adopted Housing Land Audits, rather than ad-hoc assessments on a case by case basis.
- The proposed approach is flexible where a shortfall arises as it allows for programming and delivery to vary over the life of the plan.
- The residual approach was not supported given its reliance on inaccurate figures on completions.
- They support an approach for proposals where a shortfall in the supply emerges, but different views on the steps that should be taken to address this including that it should not be 'determinative'.
- There were broader concerns about the role of this calculation.
- Detailed further amendments were suggested by those who supported the proposed changes to the policy including: clarification of the roles of the housing land requirement and housing supply target; a clearer definition of the effective housing land supply; clarification of the role of the programme as opposed to land supply; the timescale for the land supply (whether it should be for 5 years or the plan period); clarification of the base date for the calculation; a tenure-based approach; sub-market area calculations; and prioritisation of brownfield development over greenfield.
- Those who supported the approach also raised concerns that where build out rates are achieved early in a plan's lifetime there may still be a need to identify further land.
- Many respondents opposed the proposal because:
- The proposed methodology allows authorities to 'write-off' shortfalls in housing delivery. This is critical and should be a material consideration.
- The proposals do not reflect the approach taken by Reporters and the Court.
- Plans should over-provide for housing in the first place, and it is important that there is flexibility to respond to emerging shortfalls.
- The pandemic is not a reasonable basis for using the average methodology, given that shortfalls were arising before this.
- The problem is of supply rather than demand and there is a significant backlog in affordable housing.
- The residual approach is needed to establish whether the land requirement will be met.
- The average approach will lead to thousands of new homes not being delivered.
- The proposals remove accountability from the system and does not encourage local authorities to work with others to deliver homes.
- The changes would cause difficulties in infrastructure provision.
- The proposed approach to addressing the shortfall would in fact be a type of tilted balance.
- Alternatives were suggested including; a universal, plain-English methodology for planning authorities (prepared and proposed by Homes for Scotland); limiting changes to clarify that the housing supply target forms the basis of calculations; requiring a shortfall to be recovered by the end of the plan period, rather than within 5 years.
The following evidence has been taken into account in reaching our final policy position.
Current approaches to calculating the 5 year effective housing land supply
Some respondents suggested that decision-makers, including Reporters, routinely apply the residual approach, and that there is no need for the proposed clarification. However, our analysis of a sample of appeal cases shows that whilst the residual approach has been recognised in many cases, other methods have also been taken into account. Whilst a reliable conclusion can generally be made that a shortfall exists, Reporters have often come to a view that it is not possible to quantify its scale whilst there is no single methodology to refer to.
Examples referred to in responses showed that where the pace of delivery is being factored into housing land calculations, in some areas it would be extremely difficult for a planning authority to sustain a 5 year effective land supply, particularly in the latter years of a plan period where delivery has been lower than originally programmed. This may be the case despite the fact that there is arguably sufficient land and consents available which are not being built out quickly enough for a range of reasons.
Heads of Planning Scotland has produced evidence on the availability of housing land across all local authorities in Scotland, from planning performance reporting.[xiii] Whilst the data may be approximate and is not complete, this shows that:
- There was an established housing land supply in 2018-19 for more than 390,000 homes to be built across Scotland.
- The approximate / estimated requirement for the same period is for 110,000 homes – 22,000 homes per year over 5 years.
- Planning permission was granted for around 36,000 homes in 2018-19.
- In 2018-19 around 160,000 homes were programmed to be built over the next 5 years – around 32,000 per year.
- Previous completions (based on analysis of historic information) have been around 86,000 over the last five years or 17,200 per year.
- The highest annual completions in recent years for Scotland was approximately 26,000 in 2007-8.
Reasons for a shortfall in homes delivered emerging
Whilst we acknowledge that the figures provided by planning authorities are likely to be approximate, we suggest that it shows that lower levels of completions than programmed (i.e. an emerging shortfall) are unlikely to have arisen as a result of the availability of land or in securing planning permission for proposals.
This is supported by evidence we received outlining the reasons for a shortfall arising including:
- the capacity of the industry;
- the market absorption rate;
- access to finance by purchasers;
- delays in the planning system / protracted consenting processes;
- refusal of applications contrary to officer recommendations;
- land controlled by non-developers;
- plans not including sufficient or effective housing land for housing.
It is clear that there are multiple reasons perhaps including, but not limited to, the availability of land for a shortfall in the number of homes being built. Housing delivery is a complex issue that we are seeking to address in the fuller review of policy in NPF4. In the meantime, the evidence suggests the need for a cautious approach when considering whether or not releasing additional land or development is justified in specific cases. A shortfall in homes delivered is unlikely to be substantially resolved by exceptional releases of housing land, potentially on sites that are not sustainable and with no guarantee that they will progress to construction as programmed.
Strengths and weaknesses of the average and residual methodologies
We have undertaken analysis of land supply assessments applying the residual and average approaches, focusing on the local authorities which experience the greatest number of planning appeals for the exceptional release of housing land. This shows that:
- During times in which completions are low (for a range of reasons, as set out above) the residual method often leads to a substantial uplift in the number of homes to be achieved in meeting the land supply – for example, in one area (West Lothian) the annual requirement would be more than three times average completion rates since 2011.
- This issue can become very acute towards the end of a plan period when more homes have to be delivered in fewer years.
- This methodology can lead to an adjusted 5 year effective housing supply target which appears to be unrealistic when taking into account data on completions to date.
- A shortfall demonstrated by the residual approach does not necessarily arise as a result of a lack of available land but can be due to the programming assumptions of developers which are subject to change.
- Therefore, this approach may not lead to an accurate conclusion about the sufficiency of the land supply.
In comparison the average method:
- Tends to require a lower number of units to be met to achieve the 5 year effective land supply. This is considered to be realistic for the period between now and NPF4 being adopted, given that it is broadly similar to past levels of completions.
- Is not re-adjusted to account for actual delivery rates and is therefore less responsive and flexible.
- If applied, is likely to require planning authorities to maintain a supply of land which is similar to a rolling average of completions.
- For the plan areas where the presumption is triggered most often, would still require previous completion rates to be exceeded and therefore remains ambitious.
We accept that neither approach is perfect, but our view is that the arguments against the average method, and for the residual method, may be overstated. This, together with the inaccuracy of data involved in audits, the subjectivity associated with the effectiveness of housing land, and the externalities affecting programming suggests that calculations on the land supply should not be the determining factor in planning decisions to the extent that it outweighs other factors required to make a development sustainable.
We have therefore concluded that:
- There is a wide range of views on the approach to calculating the 5 year effective land supply. In our view, there is no perfect methodology and data on land and programming should be interpreted with caution.
- Planning authority statistics suggest that there is, overall, more than enough land allocated than is required.
- The benefits of the residual method are, in our view, overstated. It conflates land availability with the rate at which homes are being built-out. The latter is influenced by many different factors including, but not only, the availability of effective land. Sites may be suitable for development, and land may be available, but a technical 'shortfall' could still arise based on the residual calculation. This can lead to undeliverable requirements that cannot be recovered by planning authorities when the delivery rate is shaped by many other factors.
- It is not in the public interest to recommend a methodology which leads to conclusions that there is insufficient land available when completions are influenced by a variety of different factors.
- We recognise that the average method also has weaknesses and may be viewed as less precise and inflexible as it does not factor in emerging shortfalls in the number of homes being delivered. However, we are reassured that it is reasonable and realistic as an interim approach, as it is broadly similar to levels of completion that have been achieved in recent years. As a result, we believe that it has a valid role as a consistent benchmark to inform decision making.
- The housing land requirement is applicable when the development plan is being prepared to ensure that it includes a generous supply of land. We initially proposed that the housing supply target, which does not include an allowance for generosity, should form the basis of this calculation. However, we also recognise the importance of a positive approach to planning which aspires to deliver a generous supply of land for homes to meet our future needs. We have therefore decided that the housing land requirement, which includes an additional percentage for generosity (generally between 10 and 20 per cent), should form the basis of this calculation.
- The disputed nature of the evidence and unreliability of data in housing land audits suggests that the calculated 5 year land supply should only ever be used as an indicative number. Whether or not further land should be released for development should be a matter for planning judgement taking into account the facts and circumstances of each case.
- There should still be scope to support applications where it is clear that there is unmet demand for housing and this could be achieved by retaining a simpler and clearer version of the presumption. A clearer policy on this has been set out in the final policy amendments in Section 7.
There is a problem
Thanks for your feedback