Professor Glen Bramley, Institute for Social Policy, Housing and Equalities Research, Heriot-Watt University, Edinburgh – Written evidence (UKH0101)
Background
The Committee will be aware, from previous correspondence, of my background and track record in this field, but in summary I have worked on applied and academic research on issues of planning, housing need and supply since the 1970s and have a strong commitment to evidence-based approaches to public policy.
My recent research for the National Housing Federation and Crisis on ‘Housing Requirements’ [1] has become the widely-accepted basis across the housing sector for a view on the overall scale of new housing provision required in England, including the scale of social housing provision, as clearly acknowledged in the report of the HoC Select Committee for Housing Communities and Local Government in July 2020. I continue to work on updating this evidence base.
Earlier and ongoing work on modelling housing supply and the market [2] highlights multiple problems contributing to the gap between plans and actual completions of housing, while membership of the Board of the former National Housing and Planning Advisory Unit gave further insight into the operation of the system. I have particular expertise and extensive refereed academic outputs relating to measuring planning restrictiveness, defining affordability and housing need, and homelessness, and have collaborated on international comparative literature on housing supply, planning and affordability.
Responses to Questions posed
1. What is the current composition of the UK’s housing sector? How is the sector structured in terms of private ownership, privately rented accommodation and social housing?
The current structure of new provision is skewed to a few large oligopolistic (tending to monopolistic) speculative housebuilders who control most of land the through options and thence the buildout rate as well as quality and mix of supply. These firms are often reporting excessive profits with instances of ridiculously extravagant remuneration packages for senior executives.
More broadly, private home ownership rose rapidly to approaching 70% (middle of range for Europe) in the 1990s but has since fallen back and struggled to recover, while private renting rose rapidly from the late 1990s with the ‘buy to let’ boom. Social renting survives as a moderately large sector by international standards but is currently relatively starved of investment to expand to meet need and demand. Intermediate tenures have a potential role but are squeezed by a range of factors.
2. What social and demographic factors shape housing demand in the UK? What are the expected future trends in housing demand?
Obviously population trends are important, including births (in the longer term), deaths, and migration. Demographic growth since the 1990s has been consistently well above previous historic trends. International migration (pre Covid) had not fallen anywhere near the ‘low 10s 000s’ net level aspired to by successive Conservative governments, despite Brexit. I see this as reflecting ‘Global Britain’, in the sense that the UK is a country with strong international connections and profile, whether in terms of language, culture, trade, finance, education, past migrations, and so forth. While net flows from Europe have fallen, flows associated with family reunion and cultural ties, higher education, and higher skilled job opportunities continued to show buoyancy, while regrettably flows associated with refugee/asylum situations relating to conflicts and tyranny are likely to remain substantial.
Life expectancy has paused, but arguably this is due to poverty (and now Covid) and should be reversed as a matter of public health policy. There is still a big backlog of suppressed household formation (numbers were estimated in my 2018 study [2], and could be updated).
Internal migration is ‘endogenously’ generated by where jobs are, where housing is, and the general environment and opportunities. In the South of England the housing market areas are very large and open, so people move around to find opportunity. There is a post-Covid shift to seeking more housing space indoor and outdoor, implying some shift in demand towards houses with gardens and away from high rise flats. Taken in conjunction with the post Grenfell fiasco in addressing cladding and fire issues, the future market for high rise flats built on urban brownfield near central cities is likely to be weak for the foreseeable future. The idea that London should house all of the need/demand it generates has never made sense and it makes even less sense now.
Taking all of this together, I would argue, as I have consistently argued for nearly three decades, that a traditional extrapolative demographic approach using household projections based on recent trends is inadequate and inappropriate as a basis for planning future housing provision [3]. Demographic information should be an important input to analysis, alongside evidence on affordability and current unmet housing needs, leading to an assessment of potential future outcomes on all of these indicators under a reasonable range of assumptions. This should be informed by appropriate economic models of system behaviour under a reasonable range of assumptions; while my own models are an exemplar of such an approach, I have suggested that there could be role for a semi independent body, analogous to the OBR, to maintain such a set of analyses and models.
3. Does the Government’s target of 300,000 new homes per year accurately reflect housing demand? Is this target achievable?
Target level. This target reflects the combination of potential demand under economic normality plus need that should be met on a basis of reasonable standards and expectations around occupancy and affordability, through a range of tenures, some of which are subsidised. Some recent events might slightly reduce the total – e.g. my previously published 340,000 might be shaded down a little towards 300,000 in the light of demographic events. There are new factors to be considered also in terms of ‘replacement’ of irremediably energy inefficient stock, given decarbonisation commitments, which may increase numerical targets.
Achieveability. I don’t think it will or would be achieved by the current policy mix, which relies excessively on an unsatisfactory private speculative housebuilder model, which has consistently shown sluggishness in building out an ever-growing stock of planning approvals [4]. I think the key steps to bridge the gap are (1) invest significantly in social rented housing supply as well (statistical evidence from my modelling consistently shows that more social rented supply increases private sector output as well); (2) publicly-led development partnership agencies in key growth areas to control release of sites in terms of phasing and parcelling out to different providers to introduce diversity in supply; (3) collaborative regional planning to identify and support major growth in designated areas; (4) effective land value capture for infrastructure and social/affordable housing subsidy.
4. What is the balance of demand for new housing between homes for private ownership, privately rented homes, and social housing? How does this affect the type and tenure required of new homes?
There is a high level of aspiration for home ownership but not all of this can be realistically achieved given prices/affordability relationships and prudential mortgage lending rules/norms, having regard also to contemporary ‘flexible’ labour market conditions. Some of that gap may be bridged by appropriate shared ownership /shared equity tenure provision for which models and providers exist, although government schemes such as Starter Homes and First Homes have been poorly targeted, poor VFM and open to abuse by housebuilders.
Many younger households are forced to rely on private renting, which is a reasonable option for many in short to medium term but of doubtful value as a long term option. Many of these households are paying unaffordable rents and are adversely affected by lack of reasonable tenure security and the effects of that on any bargaining power they have with landlords over repairs etc. There is a need to reform tenure along lines pioneered in Scotland, with perhaps second generation style rent regulation following German models. There is a need for an expanded scale of social rented housing provision in most if not all areas, and some role also for intermediate rental offerings for single earner working households (including key workers) especially in pressured areas like London. To reiterate a point made in the previous section, increasing social rented provision is not something which will ‘crowd out’ or displace private provision – the statistical evidence from my models shows consistently that where and when social housing supply is higher, private completions are also higher.
I have recently refreshed a review of evidence on the backlog of housing need in England, which will be published by NHF shortly. Using fairly conservative definitions and assumptions this shows that immediately prior to Covid a total of 3.22 million households containing 7.86 million people experienced one or more current housing needs, and of these 1.25 m households containing 3.58m people required social rented housing based on reasonable norms of affordability. A further 0.39 million households containing 0.64 million people experienced core or wider homelessness for which social renting is also the most appropriate solution. These figures had been relatively stable over the previous year, although some increases were apparent in overcrowding and concealed households.
A further point should perhaps be made for the longer term: the scale of relative demand in these sectors, and relative costs/affordability, could be affected by future taxation and subsidy arrangements, as there is a growing clamour for the reform of the way housing values and gains are or are not taxed and the way these impact on owner occupiers, landlords, tenants and second home owners.
5. What can be done to ensure there is a good balance of new homes where they are needed across the UK?
This has been a continuing story of frustration and difficulty in England since the 1980s. An attempt at a rational approach was developed in the 2000s following the Barker review, involving a national agency (NHPAU) providing policy and analytical support to a multi-partite regional strategic planning approach (involving LAs, LEPs, GORs) to agree numerical targets for ‘housing market area’ based groups of local authorities subject to a ‘Duty to Cooperate’. This was all summarily scrapped in the name of ‘localism’ in 2010.
From about two years, later the Government realised that was not going to work, so a succession of further interventions were instituted, including the National Planning Framework and then various versions of a simple numerical algorithm which has attracted much criticism, with reserve powers on central government to impose targets on recalcitrant local authorities (which have been shown to not actually affect actual cases much at all so far).
I believe that the right approach would look more like what was attempted in the period 2006-10, but with a clear picture of what the key outcomes to be achieved over 15-25 years were to be and appropriate analytical capacity to show what plans and policies would be likely to achieve those outcomes. By outcomes I mean measures of housing affordability, satisfaction of tenure aspirations, standards of space/occupancy, minimised involuntary sharing and concealment of households, low levels of homelessness etc. This analytical work could be serviced by government analysts and/or commissioned researchers/consultants but I believe some more independent oversight board would be desirable.
Such analysis can then produce target numbers at the levels of sub-regional areas and their constituent local authorities. The MoHCLG has produced numbers for individual LA’s based on a simple algorithm, which has attracted much criticism. The various versions of the algorithm have all suffered from being over-simplistic, and arbitrarily constrained, quite apart from not being integrated with any overarching spatial planning framework. I have produced ‘better’ versions of it which at least take explicit account of starting point, affordability, capacity, different levels of need, and so forth, but it has to engage with an appropriate level of planning (sub-regional, and in some instances, cross-regional).
There is an ongoing discussion about the need for collaborative working at sub-regional scale in planning, particularly where major housing development is mooted, notably in the Greater South East. I believe that local authorities should have a duty to cooperate with neighbouring authorities in sub-regional groupings, but these should ideally come together on a voluntary basis in ways which reflect both housing market geography and historical structures and practices. Within these groupings the local authorities would collectively determine the most appropriate allocation of numbers (i.e. the sum of top down targets for that group of LAs) to the most appropriate major development locations within their combined area.
There will still need to be some strategic steer from central government, informed by the independently-governed analytical work, on the balance between London, the Greater South East, and other regions.
There is in my view a need, as well, to move away from the presumption that most new housing need can be met on brownfield land within existing urban areas. I do not believe that this is consistent with broader evidence on the sustainability (economic, social and environmental) of different urban forms [5], while as already noted there are significant indications of shifting lifestyle preferences and market demand post Covid, away from high density central city living. From a purely pragmatic housing supply point of view, my econometric models of housing supply show a persistent negative effect of brownfield land share on completion rates.
6. Is the construction sector able to deliver the UK’s housing demand? What barriers are facing the sector?
This links partly to comments under 1 above, about the typical UK large speculative housebuilder model. The structure of sector would benefit from more diversity, including a greater role for smaller and medium sized enterprises, providers using modern construction methods, ‘custom build’, housing associations and local authorities (building not just social housing but intermediate tenures and in some instances market sale housing). A key barrier is reliance on initiative of private landowner/developers. The way this leads to slow buildout rates on large sites has been extensively analysed, for example by the Letwin Inquiry [6] and a succession of reports on buildout rates [7]. The very large overhang of existing planning permissions provides further evidence of an unresolved problem.
In my view the currently structured housebuilding industry and its associated landowner interests cannot be expected to change this situation without further intervention from government. Although interactions with the planning system are part of the picture, as discussed below, they are far from the whole story. In my view the current setup gives insufficient role for proactive public policy and management to be brought to bear on major development opportunities. There is a need to reduce risk/uncertainty for particular players through more partnership, including counter cyclical measures when necessary, and more coherence and certainty about content and phasing within large sites, including the provision of infrastructure. I think the best lessons from the past come from the ‘Mark 3’ New Towns and Town Expansion schemes active in the 1970s. In areas where major housing growth is planned, I favour publicly-led development vehicles involving the Local Authorities, Homes England, and potentially major local landowners in partnership. They should be encouraged and empowered to acquire land initially at existing use value with a fixed markup, masterplan the site, ensure the timely provision of infrastructure linked to a clear phasing programme for the development, and appropriate provision of a range of affordable housing including social housing across the site. Sites of different sizes should be disposed of subject to planning briefs and, I would argue, via auctioning of building licenses rather than freehold sale, with conditions on mix and phasing. This way the development vehicle can more or less manage the rate of buildout and achieve the diversity of suppliers and type and tenure mixes which are all considered desirable. .
With regard to the construction industry itself, others will doubtless comment in more detail but at a strategic level there is clearly a need to develop a robust, high quality and sustainable sector using offsite modern construction methods using long term sustainable low carbon materials, and also to ensure a higher level of skills training. In the shorter term there will doubtless be calls for more flexibility on labour migration or temporary worker permits, as in agriculture, as well as concerns about the after-effects of Brexit and Covid on material supplies and logistics.
7. The Government has published its proposals for reform of the planning system. How can the planning system be shaped to meet housing demand?
I do not believe that the case for the key proposal of zoning, whereby zones allocated for development or regeneration would see the principle of development in accordance the plan established at plan approval stage, removing local authority discretion at the individual planning consent stage, is well-made or evidenced.
One specific piece of refereed academic research which examined ‘Permission in principle’: research conducted a couple of years ago, published in the reputable journal (Planning Theory and Practice), suggested that this (as an incremental extension) would not make much difference, while moving towards a more general zoning system was fraught with risk and uncertainty regarding outcomes, and from international experience could not be said to be a panacea.
I think fundamental problem for housebuilding is risk. That explains a lot of developer behaviour e.g. hoarding permissions, drip-feeding supply, seeking very high margins. Therefore, good policies for supply should address alleviating risk, including commitment to counter-cyclical support, diversity of tenure and mix options and allowing switching in appropriate circumstances, underpinning early provision of infrastructure and services, and providing more assurance that other neighbouring sites will be developed in compatible and timely ways, One might add to that list, the government admitting to and paying for its mistakes e.g. in the instance cladding and the weakening of building control and fire safety regulation. I frankly do not see the mooted zoning system as really addressing or assisting with most of these issues. Just allocating or zoning some land for development will not incentivise owners or developers to build it immediately – why won’t they just bank it, like the existing one million outstanding permissions?
There is a major and widely-acknowledged problem of staffing and resources in local planning departments. With the greater emphasis and focus on Local Plans, more planning resources will be needed to keep Local Plans up to date and to service the Local Plan decision/inquiry process
I think it became increasingly clear over the last two decades that the major reason why many/most local authorities were reluctant to agree to major additional housing numbers was that arrangements for the delivery of appropriate infrastructure were not robust. This requires (a) adequate staffing in planning and associated departments, (b) resources for the up-front infrastructure, whether from government/Homes England/utilities/prudential borrowing or whatever; (c) control over coherent phasing of sites; (d) cooperation with neighbouring local authorities. The Zoning approach makes an already difficult challenge more difficult.
I do not support the removal of Local Authority powers to negotiate s106 agreements which are appropriate to specific development proposals. Notwithstanding considerable variations in local practice, S106 has been the most successful overall attempt in UK since 1945 to promote inclusionary mixed tenure development and capture of significant land value for this purpose and for associated local infrastructure. S106 has had a crucial role in the delivery of affordable homes built in most years since the early 200s, as documented in my book [8], in work by other academics (e.g. Crook, Whitehead and colleagues) and the Ministry’s own published statistics. It goes with the grain of local decision making and facilitates development by making it more acceptable to local communities.
My own statistical modelling of housing supply at local authority level over time shows clearly that the local level private housing completions has a strong positive relationship with the level of social/affordable housing completions, when controlling for other factors. In other words, the provision of affordable housing, typically linked through s.106 to private development, helps to motivate local authority support for the principle of development and the approval of proposed schemes.
The need for greatly enhanced levels of social housing is widely acknowledged and covered elsewhere in this evidence. It is highly unlikely that anything approaching these levels would be achieveable without s.106. That partly reflects the inevitable limitations on public expenditure budgets to underpin required subsidy levels, as well as the issue of access to land for social housing providers. It is extremely undesirable that social providers should be engaging in competitive bidding for land in the open market; rather, land values should be driven down by the expectation or norm that ALL housing developments should face an obligation to support a proportion of social /affordable housing as well required supporting infrastructure.
In the full technical version of my report to NHF and Crisis on Housing Requirements [1], in Appendix A, I modelled the subsidy cost of the recommended programme of social and affordable housing provision, assuming a maximised effective use of s.106. This showed that the larger part of the subsidy requirement for affordable housing (£17.3bn pa) could be met from fully effective use of s.106, compared with the cost to the government budget (£14.3bn for all affordable housing, of which £7.5bn was for social rented housing).
The operation of s.106 can doubtless be streamlined to speed processes and reduce uncertainty, through the use of standard clauses and ready reckoners for viability, and this would help to overcome the common objection that this mechanism adds to delays in the approval process.
8. What can be done to improve the quality of new homes? How can the design and aesthetics of new homes be improved?
There is no overall agreement on what constitutes beauty and good design, with tastes changing over time and varying across society. Design codes may be able to avoid some of the worse outcomes and some cost in terms of potential uniformity. Locally based development control is one guard against crassly bad design. Perhaps there could be an option to involve this where a proposal is noncompliant with the operative design code.
On the technical and build quality side, the dominant issue facing us is energy efficiency in use standards, and arguably also embedded energy in construction methods. There are inspirational models to consider, such as the so-called Passive House, but we are a very long way from coming anywhere near such standards in contemporary new build, and central government has repeatedly retreated from imposing building control standards to reflect these aspirations.
9. Is the workforce equipped with the professional, digital and other skills required to meet housing demand, for example in the construction, planning and design sectors? What can be done to overcome skills shortages?
The short answer to this is ‘no’. I have already mentioned specific shortages in relation to planning skills, and I think this could be extended (in the local authority sector) to project management and land acquisition/disposal skills. Basic construction skills shortages are widely publicised in the construction sector.
I believe that a reason for persistent skill shortages in the sector is its strongly cyclical character, creating uncertainty about medium or longer term work and career prospects and discouraging people from entering the sector. A further factor with traditional construction is the character culture of typical building sites, which is not necessarily conducive to the nurturing of skills and consistent quality production. There is a general line of argument to move away from the spec market model, which is characterised by cyclical instability and traditional on-site technology, towards a model which supports a more stable level of output giving more certainty of work, with also a greater role for off-site manufacture. This may be partly through diversity of providers (Letwin, Note [6]) and more social rented and intermediate tenure housing
10. How does the Government interact with Local Authorities to deliver more homes? How can this relationship be improved?
Clearly one area here is (sub-)regional planning.and cooperation, as discussed in section 5 above. There is clearly a missing link in current policy here.
Another area is the role of the planning inspectorate and the associated ministerial role on appeals, call ins, etc. In general the inspectorate do a good job in arbitrating issues arising in the implementation of the principles of the National Planning Policy Framework, and are a factor promoting stability and reassurance within the system. I think in the case of housing numbers disputes I believe they could have been more strongly supported by analytical input of the kind I discuss in section 5. above and elsewhere.
Ministerial decisions overriding inspectorate recommendations are an area for greater concern, which may raise issues around propriety and ‘special interests’ gaining privileged access.
11. What are the main opportunities and areas of innovation for meeting the UK’s housing demand?
Three I would highlight, drawn out from the above, would be:
Notes on sources
[1] The full technical version of my report to NHF and Crisis on Housing Requirements is Bramley, G. (2019) Housing Supply Requirements across Great Britain for low income households and homeless people. Main Technical Report of Research for Crisis and the National Housing Federation. Edinburgh: Heriot-Watt University. HousingSupplyMay2019.pdf (hw.ac.uk)
[2] Evidence on these sub-regional market interactions is developed in Bramley, G. & Watkins, D. (2016) Housebuilding, demographic change and affordability as outcomes of local planning decisions: exploring interactions using a sub-regional model of housing markets in England' Progress in Planning, 35pp, http://dx.doi.org/10.1016/j.progress.2014.10.002; a wider review of housing market models in planning is in Bramley, G. (2013) ‘Housing market models and planning’, Town Planning Review, 84:1, 11-34.
[3] This critique of household projections is developed in conference presentations to the British Society for Population studies, particularly Bramley, G. & Watkins, D. (2014) ‘A sub-regional housing market model for England with endogenous migration and household formation: its role in assessing the adequacy of planned new housing’ BHPS Conference, Winchester, Sept 2014, Special Session on Demographic projections and Forecasts, and also in Bramley & Watkins (2016) [Note 1]. More recently I commented on the 2018-based ONS household projections and the difficulties created for housing target-setting in planning in a Blog entitled ‘Household Projections and Housing Need – Death Spiral or Death Knell?, Household Projections and Housing Need – Death Spiral or Death Knell? – I-SPHERE (hw.ac.uk)
[4] See especially Bramley, G. (2016) ‘Pushing on string: demand and supply’, Built Environment, 41:2, 144-165; and also Bramley, G. & Watkins, D. (2016), Note 1, and Letwin (2018), Note 6
[5] For key findings on ‘social sustainability’ and urban form from the EPSRC ‘CityForm’ research see Dempsey, N., Brown, C. & Bramley, G. ‘The key to sustainable urban development in UK cities? The influence of density on social sustainability’, Progress in Planning, 77:3, 89-141, http://dx.doi.org/10.1016/j.progress.2012.01.001 ; Bramley, G., Dempsey, N., Power, S., Brown, C. & Watkins, D. (2009) ‘Social sustainability and urban form: Evidence from five British cities’, Environment & Planning A, 41:9, 2125-2142, http://dx.doi.org/10.1068/a4184 ; for key findings on density and house prices see Dunse, N., Thanos, S. & Bramley, G. (2013) ‘Planning policy, housing density and consumer preferences’, Journal of Property Research, 30:3, 221-238; http://dx.doi.org/10.1080/09599916.2013.795992
[6] Letwin (2018) Independent Review of Build Out: Final Report. Presented to Parliament by the Secretary of state for Housing, Communities and Local Government. Oct 2018 CM 9720.
[7] See for example Lichfields (2020) Start to Finish: What factors affect the build-out rates of large-scale housing sites. Insight series. 2nd edition. Start to Finish (second edition): What factors affect the build-out rates of large scale housing sites? (lichfields.uk)
[8] Gurran, N. & Bramley, G. 2017, Urban Planning and the Housing Market: international perspectives for policy and practice. Palgrave MacMillan
September 2021