NPF4 call for ideas: analysis of responses

Housing Technical Discussion Paper

The Housing Technical Discussion Paper^[40] (the Discussion Paper) sets out the Scottish 0Government's current thinking on the methodology that could be used for setting 'targets for the use of land for housing in different areas of Scotland' as required in NPF4 by the Planning (Scotland) Act 2019. It has been prepared to inform early discussions with stakeholders on implementation of the 2019 Act and NPF4.

Around 25 respondents, including a number of Local Authority respondents, made a direct response to the Discussion Paper. Other respondents made relevant comments under the housing-related themes or questions and their views are also included in the analysis presented below.

A small number of highly detailed responses raised issues, including relating to calculating the allowances necessary to satisfy paragraph 100 of SPP (which advises that development plans should identify a generous supply of housing land sufficient to maintain a minimum five-year effective land supply at all times). Please note that the relevant Scottish Government policy team has access to all original responses.

Broader themes

In additional to answering the specific questions posed, some respondents made broader comments on housing land and delivery or the Discussion Paper. Some of the issues raised are covered in greater depth under the Housing - general and Housing - affordable themes.

It was seen as unclear how this Call for Ideas process relates to the recent Housing to 2040 consultation, with the hope expressed that the outcome of both consultations will result in a more joined up approach. It was suggested that the ambitions of Housing to 2040 and the potential shift in models of housing provision and delivery, and potential funding mechanisms are all relevant to this exercise. 'Homes' also need to be recognised as an interconnected part of a 'local system' of place, affordability, quality of life, and health and wellbeing.

The importance of making connections and aligning with other strategic processes was also highlighted, with national and regional transport plans and strategic transport projects cited as examples. The need to deliver housing in the most sustainable way possible was identified as an increasing requirement and it was suggested that the overall approach to land for housing must take account of wider objectives of addressing climate change, making the best use of land as a resource, and maximising existing infrastructure investment.

There was also a suggestion that, particularly given the need to agree housing land figures, the preparation of NPF4 requires LDPs and LHSs to become better aligned with the NPF4 timetable and a mechanism will be required to allow this to happen. It was suggested that the current outline suggests that all LDPs will need to be prepared in a window after the adoption of NPF but that, while there could be clear benefits to this, there are massive resource implications for all the key agencies, public sector bodies and other groups involved.

A concern was raised that the Discussion Paper focuses on housing 'numbers' and does not recognise the relationship between the planning system and the role of HMPs in addressing wider issues impacting on housing delivery.

There were differing perspectives on the key drivers of completion rates in Scotland over recent years. On the one hand, it was suggested that there is no evidence establishing a link between planning policy and completion rates. Rather, developer activity tends to be cyclical and is closely aligned to the performance of the economy and to wider social and housing policy changes.

An alternative perspective was that the housing market in Scotland is not functioning properly due to years of constrained supply reflecting a failure to plan for the right homes in the right places to meet housing needs and requirements. It was suggested that this creates a lack of choice and is one of the main reasons the market fails to function.

Other issues raised often reflected those covered in greater detail under the Housing - general and Housing - affordable themes but in summary included that:

• Certainty of funding and resources is needed over a longer period to provide a stable delivery pipeline and consistent standards. This would provide the ability to set deliverable targets over which the public sector has direct influence and control, and which could usefully be included in NPF4.
• The economic, environmental, social and geographic impacts of COVID-19 will have implications for delivery of homes in Scotland for at least the short to medium term, but most likely into the long term when taking account of the impact on local businesses, supply chain, available finance, availability of workforce etc. These impacts should be factored into the methodology where possible.
• Attention needs to be paid to the terminology being used: a number of terms - figures, requirements, housing land targets, flexibility - are used without clarity. Estimates, requirements, generosity and targets are some of the terms in current usage and practitioners have established a better understanding of how and when these terms are used.

Finally, it was also suggested that, whilst the comments in the Discussion Paper regarding the importance of local input are welcome, the proposals are being interpreted by many local authorities as a centralisation of an activity over which the authorities would wish to retain direct influence.

What is your view on the guiding principles set out?

The Discussion Paper sets out a number of guiding principles that could form part of the process for setting targets for the use of land for housing.

Some respondents offered a degree of broad support for the guiding principles, for example as providing a useful frame of reference. Other respondents made a general statement that they did not support the guiding principles or, for example, that most of the guiding principles are reasonable in theory but may have some problems working in practice.

There was also a suggestion that another guiding principle should be added. This was:

Engagement with communities, key agencies, stakeholders and developers will be important nationally and locally. Infrastructure providers and the development industry could positively contribute to this work and in particular bring their expertise to developing delivery programmes to achieve the ambitions of NPF.

The approach to setting national housing land figures is intended to provide early clarity and to reduce conflict and complexity in the local development plan process.

There was support for moving in principle towards an approach which will reduce debate around housing land calculations and reduce conflict and complexity. However, some respondents were concerned that the proposed approach will not deliver against that aim.

One issue raised (discussed further below) was the potential for double examination of housing numbers, and there was a request for clarity on what will be examined at gatecheck and examination phases. There were concerns that it will introduce more opportunities for conflict in the process: it was suggested that at present there may be conflict at the examination of the LDP or SDP, but that this could be extended to the Draft and Final NPF stages, the gatecheck stage and the LDP examination stage.

It was also suggested that there should be a consistent, universal method endorsed by the Scottish Government, on how housing targets are to be set and the evidence on which the final figure is to be based. There was a view that applying an entirely consistent approach across every planning authority, and for this to be transparent and open to comment at an early stage, would instil confidence in the process and reduce the potential for conflict at later stages. To this end, it was seen as important for the process to be accessible and fully open to scrutiny while in a format that enables ease of comparison, assessment and comment across areas.

The purpose of the figures is to ensure that local development plans allocate sufficient land for housing. We propose that the figures in NPF4 are expressed as minimum figures for housing land to accommodate an agreed projected number of homes.

Some respondents stated their disagreement with this principle, or that they agreed only in principle, with reasons given including that:

• It is impossible to state the quantum of land required without first assessing the number of homes required.
• The process should be outcome focused and any land identified as a result of a housing land requirement figure must be presented in the context of being necessary to meet a housing supply target.
• A minimum target will be open to interpretation and will increase opportunities for debate and conflict.

There was a query, as at other questions, as to whether the reference to housing land is to area or in unit numbers.

In relation to minimum figures for housing it was suggested that:

• These are inappropriate in areas of high demand where some planning restraint is required to avoid adverse effects on infrastructure and environmental issues.
• Setting minimum figures in places where demand is low or even non-existent is irrelevant.

It was also suggested that there would need to be a mechanism which establishes what a maximum figure is since the absence of an upper limit could lead to other pressures, such as on infrastructure provision, environmental limits and capacity of public services, which also need to be factored in.

An agreed proportion of this land should be 'deliverable' whilst the remainder should reflect the longer-term spatial strategy for the area. The policy and / or guidance could provide clarity on what is expected to be deliverable land, as well as addressing longer term strategy.

A number of comments focused on what is meant by this guiding principle, including that:

• It is not clear what 'an agreed proportion' means in practical terms, and how this would be measured and established.
• Nor is it clear with whom agreement will be required.
• Nor what is meant by 'deliverable' in this context. The issue of deliverability is discussed further below at the questions on the approach.

Other comments, both under this guiding principle and at the next, considered the most effective or meaningful unit or measure for land. Approaches with which respondents did not agree included those based on:

• Completion or delivery rates.
• The programmed output. It was suggested that this has caused some of the problems with the current system with the five-year programme of completions being used as the measure of supply of land.

In terms of how respondents thought land should be 'described' or measured, suggestions included:

• As a proportion of the land being free of development constraints with a programme in place to remove any constraints on the remaining land by a specific time.
• There should be minimum housing delivery figures and not just minimum figures for identifying land on paper.

While there was support for the principle of focusing on deliverable rather than effective land, some respondents stated their disagreement with this principle, with reasons given including that identifying deliverable land should not be used as a reason to reject other sustainable and deliverable home building opportunities if it cannot be demonstrated that the identified land is free of barriers (market or otherwise) to the delivery of homes through the preferred method.

In terms of approaches going forward, suggestions included:

• There should be a clear plan of action within each LDP to show that every site is, or will go through defined steps to become, deliverable.
• Ensuring there is no contradiction with the guiding principles of "Housing to 2040", which clearly explains the Scottish Government's position on moving away from banking land and property.

There was also reference to longer-term plans that are in place in other parts of the UK and to the lessons learned and areas of best practice which have been established elsewhere. It was reported that the National Planning Policy Framework in England contains wording in relation to the identification of a supply of land for new homes which could provide a useful basis upon which to plan for five years' effective supply with flexibility thereafter. Based on this, suggested wording for NPF4 terminology was:

• Specific, effective sites for years one to five of the plan period.
• Specific, established sites or broad locations for growth, for years 6-10 and, where possible, for years 11-15 of the plan.

In terms of geography and scale, we propose that the minimum figures should be set for all local authority areas in Scotland. This would maximise consistency and simplicity in the planning system as a whole and remove the uncertainty that could arise if, for example, figures at a regional scale required further calculations to set out local contributions to a shared figure. Within this, however, we recognise that there is a need to allow for local flexibility that reflects local circumstances. This could still be achieved as the figures are taken forward in each local development plan.

While some respondents agreed with the principle of setting minimum figures for all local authority areas in Scotland, others did not agree with the local authority level approach, including because it could result in unsustainable development and the undermining of spatial and regeneration priorities.

It was also suggested that setting a minimum housing land figure is inappropriate in areas where:

• High market demand means measures to restrain new housing need to be introduced to respect infrastructure constraints and environmental issues.
• It is low demand that is preventing development and therefore completion rates.

These concerns led to a call for local authorities to be involved directly in the determination of any levels of flexibility to establish the minimum housing land figures.

Other comments addressed the geographies for a minimum housing land figure. They included that the areas should correspond with the operational area of each LDP. It was suggested that, in some areas at least, there is no basis or rationale for the setting of a figure at a regional level. Further, it was reported that for the first round of indicative RSSs and timelines to feed into NPF4 it may be impractical for those outwith former SDP areas to work together to alter HMAs. However, as above, there was also a question as to whether local authorities represent the best geography or whether an RSS-wide target would be more relevant.

Alternative geographies proposed included at the city region/functional HMA to better reflect the way that housing demand occurs and the role of new housing in delivering spatial priorities such as brownfield land regeneration.

The impact of tenure was also noted, including that while functional HMAs reflect wider areas within which private market buyers may wish to purchase a new home, local authority geographies better reflect the social sector, evidently so in the case of council home provision. It was suggested that these distinctions should be retained within the process. It was also suggested that targets for both private and affordable housing tenures should be set at a national level with input from the relevant stakeholder, including local authorities at that point.

A national approach needs to be informed by regional and local knowledge, analysis and input. We therefore propose to define the requirements for land in each area by working with local and planning authorities individually and / or in regional collaborations to ensure national analysis is informed by local objectives.

Comments included that the opportunity for the input of regional and local knowledge and analysis is welcomed and will be crucial in ensuring that the targets in the NPF are realistic and result in the delivery of high-quality homes in the right locations. Other comments included that:

• It is unclear how this could be achieved and at what point regional and local knowledge would play a part.
• While a national approach requires support at the regional and local level, this should not be to the detriment of meeting Scotland's housing need and demand.
• This is a missed opportunity for true collaboration, across the current public-private divide.

There was also a concern that it will be difficult for local authorities and other stakeholders to conduct the required research and provide robust evidence in time to influence NPF 4 land requirements.

With reference to stakeholders, it was noted that the requirement for stakeholders, such as Homes for Scotland, to agree to any changes could mean that this stage will require significant time and reaching agreement may be difficult. It was reported that there are no other areas of planning that involve a primarily private sector industry body such as Homes for Scotland to such a degree, and a number of local authority or city region respondents reported that they would be concerned if Homes for Scotland were to be granted a potential veto role in agreeing the scenarios and assumptions and any potential approach to functional HMAs. There was also a query as to whether authorities would be able to progress with a disputed position should an agreement not be reached with all parties or whether the Scottish Government would put mediation measures in place to allow targets to be set.

In terms of how the approach should be taken forward, other suggestions or comments included:

• As this matter has previously been the preserve of the local authorities, it may require to be considered by local authority representative bodies such as SOLACE and COSLA.
• Collaboration should not be limited to local and planning authorities but should involve other key cross-sector stakeholders.
• Locally based HNDAs should be prepared jointly by local authorities and the CHMA. This would require funding and concise programming.
• It will be important that sufficient time is given to local authorities to feed into this process, as any alternative scenarios and assumptions will require the local authority to develop a robust evidence base. Time requirements for going through the governance and approval processes within local authorities were also highlighted.

Should NPF contain housing land figures for all areas in Scotland or focus on certain areas?

Respondents who addressed this question directly generally thought that NPF should contain figures for all areas, including if NPF is to live up to its title as a national framework. As above, there was also a view that NPF4 must set an over-arching ambitious housing target for Scotland.

A concern raised was that the Discussion Paper appears to suggest that more land may be required in certain high demand areas because of previous undersupply and that this may come at the expense of areas of low demand. It was suggested that areas of low demand are equally important, if not more so, as a means of evening out spatially the pressures of development and encouraging new life in places able to accommodate additional homes. It was also noted that it has not been specified whether 'areas' refers to geographic areas or different sectors.

However, it was also suggested that, as the requirement for housing land varies from area to area, there is a need for NPF4 to focus on certain areas and have the flexibility to apply evidence to support requirements at a local level. It was also argued that locational factors must be considered, particularly between areas of high and low demand, and that minimum figures for housing are inappropriate in areas of high market and commercial demand where some planning constraints are required.

A further comment was that, if the NPF were to only focus on certain areas, then this could most usefully be done for the former city regions in determining cross-boundary housing market areas and then apportioning the target between the different authorities.

Are there areas in Scotland where an alternative approach may be more appropriate?

Comments at this question tended to be brief, with most of those who answered suggesting that an alternative approach is not likely to be required or would not be required for their area. Further comments included that there is no obvious reason why any part of mainland Scotland should need to vary a standard national approach.

However, not all respondents agreed, including because of a view that local knowledge and experience of the housing market should be a key principle of determining housing need and its location.

It was also suggested that, to avoid urban sprawl, a separate approach should be considered for rural areas and the edges of cities, particularly in areas where infrastructure is not suitable for large scale expansion.

What is your view on the proposed approach to setting out requirements for housing land?

A number of respondents stated that, overall, they could not support or did not agree with the proposed approach. This included because of an understanding that if the NPF4 HNDA is run as it is currently anticipated, it will produce a Scottish annual all tenure housing target of approximately 13,434 homes. In contrast, it was suggested that the evidence points to 25,000 homes a year being an appropriate minimum target for Scotland.

Other wider concerns raised included that:

• Given the intention to remove from local authorities the possibility of preparing their own, albeit flawed HNDA based on local knowledge and allowing for flexibility, the proposed approach would result in more confusion and disagreement than the current method.
• It removes the requirement from current SPP to take account of wider economic factors such as capacity, resource and deliverability.
• It assumes that local authorities should aim to provide land for enough new homes to account for the projected increase in new households and for households currently in need.

Timescales

Other comments raised issues related to timescales, including that:

• The Discussion Paper does not address base dates for the calculation of targets in LDPs. It is not clear whether the base date would be the date of publication of NPF4, the adoption date of the LDP, or the base date of the HNDA/household projections.
• The issue of timescales and synchronising NPF4, the LHS and the LDP cycles is not properly addressed. Synchronising LDP and LHS locally has been problematic, and this is going to get even more difficult with NPF4 becoming part of the process.
• It is unclear over what period the housing land target would be set, and for what period detailed allocations would be made.

There was also a query about the timescales associated with the Scottish Government undertaking steps 1 and 2 of the HNDA tool (discussed further below). The timescale-associated concern was whether the CHMA will have the capacity to undertake this modelling for the whole of Scotland in the time available and that, depending on the dates of a local authority's LDP, there could also be a significant timing lag between the running of steps 1 and 2 of the HNDA tool by the CHMA and the running of the remaining steps of the tool by the local authority. It was suggested that this could then bring the reliability of the data and HNDA results into question.

On the updating of data between draft and revised drafts of NPF, experience was reported to suggest this could have significant implications for the end results. There was a query as to what would happen if planning authorities have already started working on their 'local' HNDA to drill down to HMA level for LDPs, or whether this work would have to wait until NPF4 is finalised?

As under other themes, it was suggested that a means of aligning the LDP, LHS and NPF4 will need to be developed in order to avoid conflict at the gatecheck and examination stages, particularly if any of these stages should take place a great time after the publication of NPF.

How 'land' and outputs should be defined

The question was raised as to whether consideration has been given to moving away from a targets-based system, looking at a more informed and rational approach to planning for housing and communities? In this context it was seen as vital that planning retains a focus on delivery, infrastructure, design and the creation of great places for people.

Other comments addressed whether NPF4 should be focused on delivery rather than amount of land and should set out targets rather than (downgrading to) figures. In terms of focusing on delivery rather than land, it was suggested that land availability is not the main/only barrier to increasing housing delivery; there are barriers to the delivery of many sites allocated in plans and/or identified in audits.

Supporting arguments made in relation to setting targets included that this is a clear requirement under the 2019 Act. It was also suggested that the amount of undeliverable land serves as a barrier to the release of alternative, deliverable development opportunities and this will perpetuate if NPF4 housing targets are not set for delivery of new homes.

A number of respondents commented on, or raised queries about, how the 'targets for the use of land for housing' would be defined and presented. One suggestion was that the housing land target should be expressed in housing units but that the Discussion Paper gives the impression that it will be expressed as a land area. Reflecting wider calls for clarity around use of terms, there was a question as to what is meant by 'housing land figures' i.e. whether it refers to a housing land requirement, the housing supply target, or the assessment of housing need and demand from the HNDA tool? Similar issues around meaning and use of language were also raised around LDPs moving to a ten-year timeframe (covered at the final question below).

Deliverability, resources and infrastructure

Issues raised included that while the Discussion Paper states that an agreed proportion of housing to meet the target should be deliverable, it is unclear what 'deliverable' means, to what period deliverability would relate, who 'agrees' the proportion, and when this would be decided. In terms of a planning authority's role, it was reported as being to assist in facilitating delivery and, where possible, allocating sites that are deliverable. However, it was also noted that influencing timing and rate of building is limited within the LDP and Development Management process. Other comments included that:

• All land is deliverable given the right circumstances.
• Experience has shown that in some locations almost no land is deliverable due to market circumstances.
• Delivery of public sector housing is dependent almost exclusively on the availability of public investment.
• Planning authorities have relatively little control over delivery of private sector homes which is the responsibility of housebuilders and developers in the main.
• At no point in the proposed method for setting housing land figures are available and future infrastructure requirements considered.

The issues raised sometimes culminated in a call for a careful definition of the term 'deliverable' and for:

• Identification of barriers to development and possible means of supporting delivery to be a key outcome.
• The link between deliverability and infrastructure provision to be recognised.

In relation to how the performance of planning authorities in maintaining an adequate land supply should be judged, it was suggested that:

• It should not be based on past delivery or future programming which are dependent on the housebuilding industry, or on wider economic factors.
• Where genuine shortfalls do occur, a more plan-led approach is needed to enable the right sites to come forward, rather than the plan being considered out of date.
• Authorities could specify broad areas or settlements where there is available infrastructure or identify key criteria to be met. It must also be clear than any additional sites will actually be capable of increasing build rates and addressing any shortfall in a reasonable timescale.

Tenure

Views were mixed on whether NPF should set an all tenure figure. Those supporting this approach commented that it is important that HMPs take on a clear role in planning for housing and that would include on tenure-based requirements. An alternative perspective was that whilst a tenure split is needed for LHS purposes, it is questionable whether there is any point defining this within the LDP as the split does not influence actual planning decisions.

However, others were looking for NPF to set out a tenure-based breakdown, sometimes described as a breakdown between private and affordable housing. Reasons given included that predicting land required does not consider sufficiently the potential that the level of affordable housing need may be higher than the market could support. There was an associated concern that this could lead to an over-supply of land in areas where there is insufficient resource or capacity to deliver all of the identified need for affordable housing. It was also noted that the COVID-19 pandemic means that there are now many unknowns in terms of capacity, resources and deliverability and that the need for affordable housing is only likely to rise.

Role of the HNDA Tool

Concerns sometimes focused on the appropriateness and efficacy of the HNDA process (discussed further below). In summary, they included that using the HNDA Tool could result in an annual housing target of fewer than 15,000 homes, significantly below both current delivery levels and the number of news homes that are shown to be required by the best available independent evidence.

Scenarios and assumptions

It was considered essential that factors beyond demographic projections are taken into account. but it was suggested that it is not clear how this is to be factored into the process.

Further comments included that:

• The phrase 'invitation to local authorities' to contribute to scenario and assumption choices' is of concern. Local authorities and the wider HMP should not be invited but seen as key partners in the setting of any housing land target/requirements that affect their area.
• Robust and credible evidence would be required to justify alternative scenarios and the proposal is clear that agreement with stakeholders would be required. Such a process would be similar to the current situation of local authorities conducting their own HNDA which is a high resource and time-consuming undertaking.
• It would be difficult to get agreement for alternative targets from a range of stakeholders with conflicting priorities for housing targets.
• It would be helpful not to call the consideration of alternative projections 'scenarios' as this has a distinct meaning in a later stage of the HNDA model.
• Housing land completions, based on local authority HLAs, should inform the minimum housing land figure. It is important to consider that for 'flexibility' to be informed by recent completions, the average completions over a particular period will reflect prevailing market conditions, the development plan strategy in force at that time, and the availability or otherwise of an effective five year land supply (under current provisions). The implications of whether completions have come from allocated or windfall sites should also be considered, and the share indicated.

It was also suggested that in areas where functional housing markets cross local authority boundaries, and the submission of alternative assumptions and scenarios would be necessary, the proposed mechanism and timescale for input to the preparation of the NPF may not be achievable. Given this, it was requested that specific consideration to be given to the opportunity to submit local area alternatives at any time including between the draft and revised draft NPF4.

Finally, a concern was raised with respect to local authorities' ability to consider the requirement and propose alternative scenarios and assumptions and it was reported that there are no details provided of how this will be independently scrutinised.

Role of/impact on communities

Respondents also raised issues around local communities, including that the environmental, social and economic policy factors and the spatial implications for communities of targets must be taken into account. This should include allowing for adequate and comprehensive processes for engagement with communities and their representatives to ensure an understanding of NPF and to build trust in the planning system.

Also from a communities' perspective, the recommendations of the Scottish Government's 2019 HLA research^[41] were noted, including that the process for preparing HLAs, and their form and content, should be standardised across Scotland, with clear guidance provided. It was suggested that communities who feel they are impacted by the development of new homes on sites allocated in the LDP will want to understand how numbers for new homes have been assessed and that an easy-read guide to both the HLA and HNDA processes would be beneficial.

Rural and island communities

It was suggested that the proposed approach does not adequately reflect the rural dimension, including the needs of island communities. Specific concerns included that:

• There is no reference to housing land to enable rural repopulation, despite this being a high-level outcome of NPF4, and that an alternative approach to setting targets for land will be required, particularly to address the lack of affordable housing opportunities.
• It does not consider whether the proposed approach will have a different effect within island communities.

Should NPF provide a single housing land figure or a range?

Views were mixed, with some respondents noting both pros and cons to providing a single figure or a range. Advantages associated with a single figure, including those identified by respondents who supported this approach, included that it would result in a clearly defined housing target that would be subject to minimal debate at the development plan gatecheck and examination, thus easing the development plan preparation process.

While preferring a single figure, support was sometimes caveated by the requirement for that single figure to be based on local knowledge and expertise. One local authority respondent noted that they would only support this approach if the relevant environmental, social and economic factors, including infrastructure constraints, have been taken into account and there has been significant input by local authorities and local communities. Should this not be the case, they felt that a well-informed and realistic housing land target range would be more appropriate, and that this would allow the local authority to consider these matters.

Advantages associated with providing a range, including those identified by respondents who supported this approach, included that it offers a greater level of flexibility for changing circumstances.

Disadvantages or possible tensions identified in relation to providing a range included that:

• Where ranges are set there is a tendency for the bottom (or easier) reaches of the range to be selected. It was seen as representing a risk that local authorities would plan to meet their minimum target.
• The development industry is likely to push for the land allocations to meet the higher end of the range and communities are more likely to want the lower end with the planning authority having to justify its position either way.

Finally, it was noted that this question is assumed to refer to a single figure or range being provided for a single local authority area.

Is the HNDA Tool an appropriate mechanism to base housing land figures on?

There was some support for using the HNDA Tool to provide a baseline to start the process of setting housing land supply targets, albeit the importance of considering any outputs against local environmental, social and economic factors was again stressed. It was suggested that the HNDA outputs are a useful starting point only.

Factors seen as supporting the use of the HNDA Tool included that:

• The LHS process is not currently subject to change and housing targets within LHSs will continue to be underpinned by HNDAs. SPP requires LDPs and LHSs to be aligned as much as possible.
• The HNDA process has brought added value to the collective understanding of housing systems within HMPs and this should not be lost within any new system of planning for housing.

It was also reported that the HNDA process has brought added value to the collective understanding of housing systems between housing and planning professionals working within HMPs and seen as important that this is not lost within any new system of planning for housing.

Those who offered some support for the Tool having a role, along with those who did not or only raised concerns, highlighted a range of issues they saw as affecting its usefulness and/or the validity of its outputs.

There was a view that running the initial numbers using the HNDA Tool is a relatively simple part of the process, and whilst it can be done at the national level, it does not particularly save time or effort to do so.

Population projections

Comments about the population projection figures used to populate the HNDA Tool included that:

• They are limited by the use of self-perpetuating, trend-based projections.
• The default in the HNDA Tool does not align with the household projection change as set out in NRS projections. Instead, it allows the user to capture household change from the base year of the model's run. It therefore potentially can lose the household projection change from the base date of the NRS household projections. The HNDA Practitioners Guide provides essential clarity that future household formation from NRS projections has to be met through the provision of additional housing units. The HNDA Tool therefore will require modification to ensure that that the base date of the NRS household projections, be it 2016, 2018 or 2020, is adopted for the housing land figures.
• The current COVID-19 crisis largely invalidates the 2016 trend-based projections as a reasonable estimate of future demographic need when the need for affordable housing against market housing is taken into account.

It was suggested that HNDAs rely heavily on historic, secondary data showing household formation trends, meaning that where households have not been forming quickly in the past (perhaps due to affordability issues) will be reflected in the HNDA outputs.

There was also a view that an objective assessment of housing need based on sustainable economic growth ambitions, rather than just past trends of household projections, is a more robust method for setting housing needs, particularly over a ten-year period. It was suggested that a more rounded approach than is afforded by a standardised methodology is needed to avoid setting housing numbers that exacerbate existing negative trends rather than focusing on the potential for sustainable growth.

HoTOC^[42]

There was a concern that the HoTOC method may significantly underestimate those households who are homeless as well as those that are overcrowded and concealed.

With reference to the use of HoTOC for the calculation of existing need across Scotland, it was reported that the CHMA has advised in the past that the HNDA toolkit data and scenarios are only relevant for a maximum of five years. One concern raised was that the existing need count is likely to under-estimate housing need using the proposed indicators of homeless households, overcrowded households and concealed households.

It was also suggested that HoTOC works to a strict definition of what a household is, which means that those who might wish to form a household, such as young single adults living with parents but who want their own home, are excluded. Those living in homes that are physically unfit or households in homes that are not affordable to them were also identified as being excluded.

A range of additional data sources that could be used to establish the existing need count (as identified in a 2018 report for Crisis)^[43] were highlighted, including:

• Concealed/sharing households as a percentage of all households.
• Backlog housing needs - percentage of all households with one or more of affordability problems, overcrowding, concealed, sharing or unsuitable housing.
• Core homelessness - the number of households estimated to be rough sleeping or in similar situations, or in hostels, unsuitable temporary accommodation or sofa-surfing.
• Wider homelessness - households who are statutorily homeless but not in the 'core' group above, plus other households at significant risk of falling into homelessness in the coming year.
• Annual net new need for affordable housing - the number of households newly forming, plus half of net migrant households to the area, who are unable to afford market rent, less the annual flow of relets of social housing (excluding transfers).
• Chances of rehousing - the annual flow of lettings of social housing to new tenants as a percentage of the backlog housing need.

The potential impact of the Rapid Rehousing work going on across Scotland was also highlighted, with one local authority respondent reporting that they are looking into innovative alternative models to tackling homelessness. They noted that this work will affect the backlog of existing need on which scenarios are based, and that the toolkit scenarios need to consider the impact of Rapid Rehousing and Housing First.

Other data range limitations

In terms of other limitations of the HNDA Tool and the figures that it would produce, comments included that:

• The range of data used in the HNDA Tool is limited as the needs for specialist /adapted housing will not be counted.
• Local authorities would want to reflect housing needs arising from size/type/ locational pressures which would not be reflected in a count of need at local authority level through the HNDA tool.

It was reported that local authorities have good quality local data to triangulate with national datasets.

Local applicability and adaptability

It was noted that, given that there will be limited scope for adjusting figures locally in the development plan, the figures need to accurately reflect local circumstances. However, one local authority reported their past experience that the HNDA is an imperfect tool which produces results that are unreliable, and which do not reflect local knowledge or actual local conditions. This respondent considered that a new approach to housing supply should be taken but that, in the interim, the Council-based process of preparing HNDAs should be examined independently and that the CHMA should be commissioned to explore whether, on the basis of that independent assessment, the HNDA tool and process can be improved or if it should be replaced by a more effective mechanism.

It was also suggested that the Tool is not dynamic and provides no opportunity for updating mid cycle. It was reported that under current arrangements for assessing the effectiveness of the land supply, the supply side is updated annually through HLA processes but the demand side is only updated every five years through the incorporation of the updated HNDA within the development plan. This means that supply is never being compared to an up-to-date assessment of demand and it was suggested that both elements of the equation should be updated on the same cycle.

Rural communities

In line with the general concerns expressed about the appropriateness of the approach to rural and island communities, it was suggested that current methodologies for HNDAs do not operate at sufficient detail to identify pockets of deprivation and housing need within rural areas, including in relation to homelessness and concealed households. It was also reported that HNDAs entrench past trends reinforcing depopulation.

Going forward

Some respondents highlighted issues to be borne in mind if the HNDA Tool is to be used. These included that:

• It is important that the most up-to-date household projections are used for this stage of the process. (New household projections are due in July 2020).
• It may be more helpful for NPF4 to express its target for housing land as an estimate based on steps 1 and 2 of the HNDA which local authorities then use to produce detailed local area-based targets for housing delivery.

In terms of suggested alternative approaches, proposals included carrying out primary research in the form of household surveys, with the information gathered used alongside other information sources to estimate the number of households and aspirant households.

As above, there was also reference to a range of additional data sources that could be used to establish the existing need count, as set out in the 2018 Crisis report.

Should there be scope for local and planning authorities working together to reflect functional housing market areas that cross local authority boundaries? What approaches could be used to achieve this?

One perspective was that local and planning authorities working together makes sense, or that it is imperative that local and planning authorities are able to continue working together to determine housing needs at a cross-boundary level. The Glasgow City Region was cited as an area within which demand for housing is extremely integrated across the area, with local authority boundaries largely arbitrary.

On a similar theme, it was noted that the former city regions already have mechanisms in place for working together although, given that these regional groupings will presumably be voluntary, there was a query as to what would happen in the event that a functional HMA boundary is identified which crosses local authority boundaries but no agreement is reached between those authorities to work together.

Further comments included that NPF must be able to consider functional housing markets and submarkets not only in terms of establishing demand but also in determining suitable locations for housing development in line with regional spatial priorities. It was suggested that such an approach would require the establishment of a regional body with a robust decision-making process, associated oversight and with the collaboration of all relevant local authorities. However, there were also concerns about whether the timescales for NPF4 processes will make it difficult to establish targets at functional market area level.

It was suggested that, where local authorities cannot meet their full minimum housing land requirement within the Council boundary, there must be a duty to co-operate with neighbouring authorities to redistribute the housing requirement.

Other comments on how any approach should be framed included that:

• It would need to be accompanied by clear information on how each authority would retain its own responsibility for achieving the delivery of sufficient new homes to meet the requirements.
• NPF processes must allow sufficient time to allow cross-boundary work and analysis to be conducted and agreed by each authority through its governance procedures.
• Measures should be put in place to avoid an instance where proper planning for housing land in a single HMA straddling multiple authorities becomes a race to the finish, whereby the last authority to publish its LDP is left holding a requirement to meet the remainder of the housing land requirement.
• The role of functional market area considerations in decision making within LDP examinations and planning decisions needs to be clarified. The Scottish Government should provide relevant guidance.
• Clarity is required around how delivery and supply are monitored and then how such information is utilised within decision making. A national HLA on a digital platform would assist in this regard.
• Any approach must not result in a single local or planning authority area having to provide a burdensome level of resources to support the process.

Finally, it was seen as important that sufficient infrastructure is available or capable of being made available to support housing development and its delivery. It was suggested that a joined-up approach is required to infrastructure delivery and meeting expectations and a concern was raised that the reference in the 2019 Act to an infrastructure levy is considerably weakened by the inclusion of a 'sunset' clause which removes the regulation making power if no regulations are made within seven years of the Act.

Should NPF apply a level of flexibility to the HNDA tool results to ensure a proactive approach to managing the supply of land for housing in a positive way? Should the level of flexibility be informed by recent housing completions?

While some respondents agreed that NPF should apply a level of flexibility to the HNDA Tool results, as under the guiding principles, others commented that they were not clear as to what was meant by 'flexibility' in this context and how it relates to the currently used 'generosity'.

Clarification was sought as to whether 'flexibility' is effectively the same as the generosity allowance which is currently set at 10-20% in SPP, and that the 'minimum housing land figure' is the equivalent of the current housing land requirement. It was also reported that as section 75 regulations only stipulate a 25% contribution to social housing it is difficult to see how the backlog can be cleared through NPF forecasts alone.

There was also a query as to whether flexibility would be around a standard figure across all authorities or varied according to circumstance, and if varied, on what evidence would the rate be based.

One perspective was that the addition of a flexible percentage to the strategic housing target represents an essential safeguard. It was suggested that this should ensure that the development plan could maintain the minimum five-year effective land supply at all times. It was also suggested that the current range of between 10-20% of the strategic housing target seems to be arbitrarily fixed and unrelated to the potential performance of the LDP.

There was a call for a form of ambition bias, with the ability to flex targets upwards only, ensuring delivery targets exceed the base need and demand uncovered through evidence gathering. An alternative perspective was that an approach that allows for flexibility would defeat the purpose of having prepared an HNDA, introduce the prospect of challenge and remove certainty for communities.

Other concerns raised included that the proposed approach introduces the potential for flexibility at the wrong stage - the view was that HNDA outputs should inform an LDP-related policy decision on the number of new homes the planning authority reasonably expects could be delivered. Another alternative proposed was that the level of flexibility should not be set within NPF4 but following a full HNDA and consideration of local factors including past completions, market conditions and the growth aspirations of the local authority. The level of flexibility could be considered through the gatecheck process in the preparation of the LDP.

It was also suggested that the inclusion of two separate targets (housing supply and housing land requirement) leads to confusion and that, should a target be required, then a single figure produced with local knowledge and subject to less dispute would provide greater clarity for both developers and planning authorities. It was also suggested that some flexibility could be included to allow for extra capacity in addition to that single number, but that it should be represented as an accepted percentage point departure from the same figure and not a separate, defined figure in its own right.

It was also seen as important that NPF should not seek to artificially inflate the supply of land for housing other than to incorporate flexibility to allow for sites failing to come forward or delivering more slowly than programmed, as is currently the case between housing supply targets and housing land requirements. It was reported that many parts of the country are still struggling to see a return to pre-recession completions and that requiring planning authorities to allocate more land to ensure additional flexibility will not necessarily mean that more homes are delivered. Further, delivery on the ground is not something which planning authorities can control on the majority of sites and simply allocating more land could potentially have a detrimental impact on the delivery of other priorities, such as urban regeneration and the reuse of brownfield sites, as developers will understandably opt to concentrate on the less complex greenfield sites first.

Also in relation to how flexibility can or should be exercised, it was suggested that there should be:

• Clarity around the factors that will be considered when any adjustments are made.
• A robust methodology and consistency across Scotland and that this could equate to a flat rate of generosity, making housing land calculations easier to defend during the proposed gatecheck process.
• Recent housing completions.

In relation to whether flexibility should be informed by recent housing completions, comments included that:

• An approach based on past completions will be able to reflect variations in delivery at market area geography and by tenure. In arriving at a locally agreed view, consideration could be given to completions using measures including for example an average of a five- or ten-year period.
• They can be a useful indicator of what can be delivered but should be treated in the same way as the HNDA Tool outputs and must be subject to considerations on environmental, social and economic factors.
• Any flexibility could be informed by the track record of the planning authority in delivering new homes against its (current system) housing supply target and/or against identified need and demand.

However, concerns were also raised about the use of recent house completions data, including that in an area with a constrained housing market, recent completions may be low, masking unmet demand. Peaks and troughs in development activity may occur for a variety of reasons.

An alternative perspective was that such an approach would effectively reward failure since a planning authority that under-delivered in the past would 'enjoy' lower requirements in the future and associated short-term political gain, regardless of implications in terms of actual housing need.

Should NPF housing land figures be met in LDPs as a minimum?

Comments from respondents who thought that NPF housing land figures should be a minimum included that this approach would continue the approach set out in the current SPP at paragraph 116, that plans should indicate the number of new homes to be built over the plan period, and that this should be increased by a margin of 10-20% to establish the housing land requirement, in order to ensure that a generous supply of land for housing is provided.

In terms of taking forward an approach based on a minimum figure, it was suggested that each LDP should have clear approaches in place to effectively and quickly address any housing delivery shortfall that may arise if some of the preferred sites do not deliver homes as anticipated.

Issues raised by those who did not support the approach included that a key tenet set out at the start of the reforms to the planning system was to improve the participation of communities and subsequently increase public trust in the system. The local authority respondent raising this issue went on to suggest that the imposition within the NPF of minimum targets without accessible and rigorous consultation with communities is likely to further degrade their trust in the planning process.

It was also suggested that setting a minimum figure would put too many restrictions on authorities as there should be room for recognising local knowledge and understanding. There were additional concerns that if the minimum figures set out in NPF4 are unattainable (particularly in areas of low demand) then this will leave authorities exposed to development in unsuitable locations/areas.

One view was that, in order to attain the simplicity sought, the Scottish Government should remove the scope for minimums, flexibility and debate whilst taking a simpler approach based on locally agreed knowledge and average completions. An associated comment was that if LDPs are required to meet housing figures, there should be a specific requirement on the housing industry to meet these targets over the lifespan of the LDP and any under-delivery of these houses should be ignored at LDP gatechecks and examinations.

LDPs are moving to a ten-year timeframe. Housing land audits generally programme land supply for a five-year period. For LDPs to have a ten-year land supply available upon adoption what mechanisms could be used to ensure land is brought forward in accordance with the LDPs spatial strategy?

Broader issues raised included that there is currently no legislation that ties housing to planning, and the Scottish Government therefore needs to examine how legislation can be aligned. It was also suggested that to ensure land is brought forward in accordance with the LDP's spatial strategy, there is potential for much closer alignment between the HLA process and delivery programmes to help confirm site effectiveness and programme investments.

There was also a call for modifications to the HLA procedure to be introduced as a matter of urgency. It was suggested that annual audits have a tendency to overestimate rather than underestimate the available effective land supply, including because of the inclusion of anticipated annual completions from sites that have no planning consent. As an example, it was reported that one 2013 HLA contained 20 sites which were 'allocated' but did not have planning consent and that the total contribution from these sites to the forward five year effective land supply was equivalent to 30% of the overall effective land supply identified in the audit. Further, the 2018 HLA revealed that these 20 sites had only delivered 1,103 houses in the previous five years, as opposed to the 3,327 anticipated.

Other comments included that:

• The requirement to maintain a five-year effective housing land supply, and how it is calculated, needs to be reviewed as it continues to be a source of confusion in the system, and the way it is applied is undermining the plan-led system.
• The five-year effective period is an important consideration as maintaining a five-year supply is currently a trigger for paragraphs 32 and 33 of SPP.

Clarity was sought around whether there will be a requirement going forward to maintain a five-year effective supply and on the approach to housing shortfall.

It was also suggested that it is unrealistic to allocate housing land and provide effective supply for a decade-long period. Further, that whilst LDPs can identify sites for delivery in years one to five and safeguarded for deliver in years five to ten, it will be extremely challenging to have an effective ten-year supply identified in an LDP at the outset, as there is so much uncertainty with delivery. Reasons given included that, with the validity of forecasting over the five-year period or longer reported to be low, the usefulness of creating a ten-year supply is questionable.

In addition to the suggestion that significant new allocations of deliverable land will need to be added to LDPs, ideas concerning how a ten year-related supply could operate included that:

• NPF4 should define how land supply might be managed over a ten-year period within an LDP and should include circumstances that might trigger a review of land supply and alignment with LHSs, for example, to determine whether more land may require to be released.
• At least 80% of allocated sites should be effective at the point of being allocated in an LDP. The remaining 20% must be able to address six of the seven tests of effectiveness within Planning Advice Note 2/2010: Affordable Housing and Housing Land Supply.
• There would need to be an acceptance that sites that are not currently effective but would expect to be effective within the ten-year period would be included.
• The non-effective supply should be capable of becoming effective within the ten-year plan period but where there are constraints to be overcome these should be set out, monitored and updated in the LDP Action Programme.
• Local authorities should be encouraged to allocate a buffer of safeguarded land above their NPF requirement, to account for slippage and ensure the minimum requirement is met.
• Where a planning authority proposed to rely on windfall development to make up a proportion of its housing land supply, there should be a reasonable prospect of windfall sites coming forward. Any assumptions must be supported by compelling evidence that they are justifiable and will lead to the necessary delivery of homes.

Other comments also addressed the relationship between five- and ten-year cycles. For example, it was reported that even though LDPs are reviewed every five years currently, they should allocate sufficient housing land for a ten-year period. It was suggested that when the preparation of LDPs moves to a ten-year cycle, authorities will be able to produce an update of the LDP, and in many cases this would likely be to update the housing land supply. It was also suggested that:

• The Scottish Government should provide clear guidance on how the five-year housing land supply should be calculated. It should include that if a shortfall develops, immediate action should be taken, and additional land released.
• Failure to have met the housing supply target of the preceding five-year period should always result in the shortfall being carried forward for the subsequent five-year period to ensure the housing requirement is met.
• NPF should set out policy that would allow planning permission to be granted for sustainable development of non-allocated sites. Where a planning authority cannot demonstrate a five-year supply there must be a strong presumption in favour of granting planning permission for sustainable development on unallocated sites to address the shortfall.

Suggestions as to other mechanisms to ensure land is brought forward included that:

• A range of agencies should be required to co-operate and invest time and funds for a minimum period of time. Such an approach should be focused on bringing forward vacant or derelict or underused urban brownfield sites.
• In areas of high demand, any approach needs to ensure there is a gradual improvement of local services to accommodate an increased population. A mechanism that prioritises the release of land not previously identified would, however, be counterproductive and would negate the plan-led approach.
• A single template for HLAs should be provided for all planning authorities to use and this should feed into an online database that can be readily accessed and used to monitor completions and deliverable land in each planning authority, in each region, and across Scotland as a whole.
• Sites that are currently thought to be deliverable should be programmed out, at a locally achievable rate that the market is capable of absorbing, from the year the home builder attached to the site (or another developer) demonstrates the completion of new homes will begin.
• Annual delivery must be robustly monitored and reviewed through the annual HLA. Verification should be provided by the Scottish Government, to ensure the housing supply target is verified and the housing land supply remains effective.

Should the Scottish Government play a role in the housing land audit process?

One view was that the HLA process should be standardised and prescribed by the Scottish Government, to include a fixed template and possibly a sign off procedure to ensure absolute consistency across Scotland.

An alternative view was that the HLA should remain a matter for local authorities as they have direct knowledge and understanding of how the local housing markets operate. However, it was reported that the failure to agree the annual audit within an appropriate timescale with Homes for Scotland can be a drawback and that consideration should be given to setting defined timescales for all parties involved in this process.

Other comments included that, in order to ensure consistency and compatibility in assumptions and methodology for data collection across Scottish local authorities, the Scottish Government should be involved to some extent. Comments sometimes addressed when or how the Scottish Government should be involved with variations proposed including that they:

• Should take forward Recommendation 13 of the report of the Independent Review of the Planning System in 2016 and establish a clearer definition and policy context for how housing land and housing delivery are measured and dealt with.
• Should play an important role in setting out the remit and requirements of the HLA and additional guidance on the HLA process would be welcomed. One suggestion was this should equate to ensuring a reliable, annual, genuinely agreed audit is carried out for each planning authority and then acting to ensure the content of an audit can be properly interpreted and, where necessary, acted upon.
• Could be represented in HLA meetings to ensure that they are aware of emerging issues.
• Would have greatest impact by playing an enhanced role at gatecheck and examination stages.
• Could consider becoming involved where there are disputes in the audit between the local authority and the development industry.
• Should use new powers to ensure LDPs with inadequate housing land supplies cannot be adopted.
• Should have a monitoring role.

Finally, there was support for presenting housing land on a digital national platform, and it was suggested that the Digital Taskforce provides an opportunity to develop a digital platform that could provide clarity and reduce unnecessary debate.