68. A Housing Design Audit for England

With the drive to deliver more homes across the country has come a loud call for those developments to be of a high standard of design in order to deliver high quality, liveable and sustainable environments for residents.  Research has consistently shown that high quality design makes new residential developments more acceptable to local communities and delivers huge value to all – socially, economically, environmentally and as regards long-term health outcomes.

Housing design audits were a tool developed and used by CABE in the mid-2000s as a systematic approach to assess the design quality of the external residential environment.  Over the summer of 2019 the Place Alliance conduced A Housing Design Audit for England, which evaluated the design of 142 large-scale housing-led development projects across England against seventeen design considerations.  The work was jointly funded by the CPRE and harnessed the support of a diverse alliance of organisations including: the Home Builders Federation, Civic Voice, Academy of Urbanism, Design Council, Urban Design Group, UK Green Building Council, and the Institute for Highways and Transportation.  It was underpinned by professional input from Arup, JTP, Spawforths and URBED and a network of specially trained volunteers across the country.

The audit elicited enough data for comparisons to be made regionally and against the results of previous housing design audits conducted over a decade ago.  It establishes a new baseline from which to measure progress on housing design quality in the future and to begin to ask some of the questions around why housing design in the volume housebuilding sector is like it is.  The intention was never to name and shame or to showcase the small number of already well known exemplars, but instead to rigorously demonstrate where we are as a country on the design of volume housing, and to identify what we are doing well and where there is room for improvement.

The background to the study and the approach used is included in Blog 64: Housing design quality, auditing England.  Here the focus is on the findings and the recommendations that flow from the analysis.

What are we delivering and where?

The audit showed that there has been a small overall improvement in housing design quality nationally since the last audits that were conducted between 2004 and 2007.  However, because this improvement is from a low base – what CABE at the time called “An uncompromising and unflattering picture” – the large majority of new housing developments are still ‘mediocre’ or ‘poor’.  Three quarters of the audited projects fell into these categories.

If advice in the National Planning Policy Framework was followed – that “Permission should be refused for development of poor design that fails to take the opportunities available for improving the character and quality of an area and the way it functions” – then one in five of the audited schemes should have been refused planning permission outright.  The design of many others should have been improved before relevant permissions were granted.

When examined regionally, the picture is patchy.  The most improved region was the West Midlands, with the South East and Greater London (the best performing region) also showing very significant improvements.  The East Midlands and South West scored least well, and significantly lower than the English average.  Positively, a wide distribution of ‘good’ and ‘poor’ scores across all regions shows that it is clearly possible to deliver high quality housing environments (and substandard ones) right across the country.  At the same time, if the improvements of the three best performing regions are stripped out of the findings, then the small national gain in housing design quality largely disappears.

What are we getting right and what are we getting wrong?

The audit examined seventeen design considerations, and in two of these we seem to be doing well.  Designing for safety and security faired best, suggesting that the Secured by design parameters of recent decades (e.g. well overlooked street spaces) have been successfully mainstreamed across much of the country.  Schemes were also typically successful at integrating a variety of housing types, demonstrating that local needs, at least as regards a variety of house types and sizes and the integration of different tenures in a sensitive manner, are generally being successfully balanced with market imperatives.

At the other extreme, it is no surprise that the least successful design elements nationally related to overly engineered highways infrastructure and the poor integration of storage, bins and car parking.  These problems have been leading to unattractive and unfriendly environments dominated by large areas of hard surfaces (tarmac or brick paviours), parked cars and bins.  Low-scoring schemes also performed especially poorly in the categories relating to the architectural response to the context and establishing a positive new character for development.  Development often had little distinguishing personality or ‘sense of place’, with public, open and play spaces poorly designed and located to encourage social interaction.  Housing units are frequently of an obviously standard type with little attempt to create something distinctive.

Some design considerations were marked by a broad variation in practice nationally.  These include how well streets are defined by houses and the designed landscape, and whether streets connect up together and with their surroundings.  Also whether developments are pedestrian, cycle and public transport friendly and conveniently served by local facilities and amenities.  The combination of many of these preceding factors influence how ‘walkable’ or car-dependent developments are likely to be.  Many developments are failing in this regard with likely negative health, social and environmental implications.

Finally, whilst the majority of schemes are achieving the basic minimum energy efficiency requirements set out in legislation, the audit identified that significant numbers are still falling short of even this minimum standard.  This, combined with the known and persistent performance gap between the ‘designed’ and ‘as built’ energy performance in new homes and the failure to deliver a green and bio-diverse landscape in many projects, amounts to a sub-standard response to the environmental challenges we face.

Why is there such variation in practice?

In sum, whilst some limited progress has been made in some regions, overwhelmingly the message is that the design of new housing environments in England are not good enough.  So why is this?  The research correlated the findings on design quality with a range of other factors in an attempt to understand.

When correlated with the nature of the site being developed it was clear that audited design outcomes were progressively worse as projects moved away from the urban core and reduced in density, and if they were built on greenfield, as opposed to brownfield, sites.  The additional constraints imposed by a stronger pre-existing urban context – often with existing infrastructure, heritage and natural assets, and a street network to plug into – seemed to encourage a more sensitive design response.  Generally we are not good at building at lower densities and on greenfields.

One notable finding is that better designed schemes achieve higher sales values and vice versa in every region, amounting to a 75% uplift nationally.  At the same time there is a continued trend (by a factor of ten) towards delivering sub-standard design outcomes for less affluent communities.  In other words, less-affluent communities get worse design.  Standard housebuilding development models undoubtedly make it easier to invest in better design when development values are higher.  But just because values are low, does not mean that good design cannot be afforded.  The cost factors separating ‘good’ from ‘poor’ design are likely to be a relatively small proportion of total development value (across all markets) and low value locations may anyway show a higher return on investment and be more profitable to develop because land values are low compared to the gross development value that can be achieved.  Indeed, a minority of schemes with low market value buck the trend and achieve ‘good’ and ‘very good’ design outcomes whilst high value schemes sometimes deliver only ‘mediocre’ or ‘poor’ design results.  Better design can be afforded, but we just don’t do it.

When the design practices of our largest developers were examined, it was also apparent that individual developers often had widely varying audit scores, sometimes right across the audit scale from ‘very good’ to ‘very poor’ design.  The practices, priorities and inconsistencies within and between housebuilders are clearly very significant as regards the quality of housing developments that are being realized.

The practices of local authorities are also very inconsistent, notably around the use of proactive site-specific design governance.  The audit confirmed that to achieve ‘good’ or ‘very good’ outcomes requires more than a passive check against a generic checklist of design principles, it requires a proactive and site-specific process of guidance and accompanying peer review.  The most effective design governance tools were therefore shown to be design codes and design review, although these tools are used far less than other more generic policies and principles.

A series of more detailed case studies accompanying the audit demonstrated a repeated story that, in order to meet housing numbers, poor design is either getting through on appeal or at the stage of reserved matters when design is being repeatedly dumbed down.  This fatally undermines the Government’s own policy on design in the NPPF and sends a message that design quality does not really matter.

A final element of the audit involved asking residents what they thought about the environments that they now lived in.  This was compared with the views expressed in the Place Alliance’s recent England-wide survey of Councillor’s attitudes towards residential design.  Whilst new residents are generally happy with the environments around their new homes, this contrasts strongly with the perspective of local councillors whose views can be taken as a surrogate for those of the wider community.  They regret what they see as the imposition of poor design and the loss of local character in their areas.  Both residents and councillors see a negative impact from unduly car and roads dominated environments, but the lack of choice (to choose good design) in many areas and the emotional investment made in a new home, may go some way to explain the divergence in views between these two groups.

What can we do?

What was clear from the research was that the responsibility for poor housing design is a shared concern, across housebuilders, local planning and highways authorities, and Government.  Based on the findings A Housing Design Audit for England makes recommendations for each.

Recommendations for Government

1. Be more prescriptive on density: The clear benefits of designing at higher (not high) densities is apparent. The best schemes averaged 56 dwellings / hectare, approaching double the current national average of 31 dwellings / hectare.  Instead the current national average for density is almost exactly the average density of schemes scoring ‘poor’ in the audit (32 dwellings / hectare).  Government should be more prescriptive in seeking more urban densities (compatible with other contextual factors) in the NPPF, densities of at least 50dph that are able to support public transport, and a mix of uses and local facilities.
2. Seek to spread learning from the best practice and publicise it: The Government should work with the industry to seek out, and proactively showcase good design by volume housebuilders, for example through online case studies and dedicated national housing design guidance covering matters from detail design (e.g. the design of bins and storage) to strategic urban design concerns relating to the location and connectivity of housing.
3. Understand design in less affluent contexts: Commission research into delivering design quality in less affluent areas, including into how standard housing units can be used in more creative ways to deliver distinctive places and how local authorities may become more engaged in delivery through public / private partnerships or other means.
4. Issue guidance on the design of parking: How parking is handled can make or break the design of residential environments. National research on the successful integration of parking across different densities should be commissioned as the basis for guidance to be adopted on the subject nationally and locally.
5. Publicise the rejection at appeal of poor quality schemes: More forcefully advise the Planning Inspectorate to reject schemes that do not live up to the Government’s own design aspirations as set out in the NPPF – regardless of whether local housing targets have been met or not – and publicise these decisions.
6. Require a place-first approach to highways design: Highways authorities should be required to take a ‘place first’ approach when dealing with the design and adoption of highways. This could begin by requiring highways authorities to adopt Manual for Streets or an equivalent place-focussed guidance on highways design and by issuing national guidance on what it is reasonable to charge for adopting trees and other landscape elements.
7. Continue to audit progress: A Housing Design Audit for England provides a new baseline from which to measure progress on the design of housing, but the Place Alliance will struggle to repeat the exercise given the resource implications and the reliance on voluntary input. The Government has a duty to monitor the design quality of the residential sector and should fund its own repeat audit no later than 2024.
8. Audit small housebuilders and social housebuilders: A Housing Design Audit for England has focussed on the products of the large volume housebuilders. The work of other key sectors have been omitted and could valuably provide the basis for other follow up audits.

Recommendations for housebuilders:

1. A big leap needs to be made: Whilst a small overall improvement in the design of external residential environments was detected, this comes from a low base and is very patchy: geographically, between different housebuilders, and even across the regional operations of individual housebuilders. The vital importance of good design to the industry as a whole as regards building a positive reputation, encouraging acceptance of new housing locally, and easing the path towards regulatory permissions, has still not been accepted.  From small scale easy wins, like dealing adequately with bin storage, to more complex challenges, such as injecting character and into streets and making them walkable, a big leap needs to be made on design by the industry as a whole.
2. A new ethical approach to design: The largest housebuilders should set the ethical standards for the industry at large. They are building developments which will have profound impacts for many decades: on the places and communities they help to shape, on the social wellbeing and health outcomes of their customers and future occupants, and on the environment at large.  The negative impacts of how we have been building are well known and understood.  A responsible and ethical approach for housebuilders to take is to seek net gains on all these fronts.
3. Invest in an internal design infrastructure: Housebuilders should invest in their own internal design governance teams and processes in order to set a high ambition on design, whichever the sub-market they are building for. Large housebuilders should consider mechanisms for internal learning and coordination on design, notably for bringing all their regional operations and subsidiaries up to the standards of the best.
4. Resident satisfaction should not be taken as a sign that all is OK: The emotional investment in a new home masks the downsides of poor design for residents. In particular lower density suburban and rural schemes are failing to exploit the space and landscape advantages of their contexts.  These reflect potential qualities that new residents strongly value and reveal the need for a greater focus on how to design well in less dense locations.
5. Examine the economics of housing design: The factors impacting on design and their economic implications for viability are poorly understood and opaque. The industry, in partnership with others, should commission and publish research into the economics of housing design in order that design decision-making can be better understood.

 Recommendations for local authorities (planning and highways):

1. Set very clear aspirations for sites (in advance): All design governance tools help to deliver better design outcomes – it is far better to use them than not – but the use of proactive site-specific tools, notably design codes, is the most effective means to influence better design. Such tools give greater certainty (for housebuilders and communities), but their use and the design ambitions they espouse should be made crystal clear in policy well in advance of sites coming forward for development.
2. Design review for all major housing schemes: Local authorities should themselves establish or externally commission a design review panel as a chargeable service and all major housing projects should be subject to a programme of design review. Advice on how to do this can be found in Reviewing Design Review
3. Deal once and for all with the highways / planning disconnect: Highways authorities should take responsibility for their part in making streets and places, not simply roads and infrastructure. Highways design and adoption functions should work in a wholly integrated manner with planning (development management), perhaps through the establishment of multi-disciplinary urban design teams (across authorities in two tier areas), and by involving highways authorities in the commissioning of design review.
4. Refuse sub-standard schemes on design grounds: The NPPF is very clear in its advice that “good design is a key aspect of sustainable development”. Consequently ‘poor’ and even ‘mediocre’ design is not sustainable and the ‘Presumption in favour of sustainable development’ does not apply.  Local planning authorities need to have the courage of their convictions and set clear local aspirations by refusing schemes that do not meet their published design standards.
5. Consider the parts and the whole when delivering quality: Some well designed large schemes are being undermined by a failure to give reserved matters applications adequate scrutiny or through poor phasing strategies resulting in the delivery of disconnected parcels of housing only development. Delivery of design quality requires both the whole and the parts to be properly scrutinised at all stages during the design and delivery process.

For some housebuilders, as long as there is a ready market for poor quality design, and they can continue to get this product through the planning system, there is little incentive to improve.  For some planning authorities, the short-term imperative to deliver new homes may continue to trump the long-term negative impact of the environments that are being built.  For some highways authorities, the very notion of good place-making is, as yet, simply not on the radar.  Collectively, as the results of the audit show, we need to significantly raise our game if we are to create the sorts of places that future generations will feel proud to call home.

Matthew Carmona

Professor of Planning & Urban Design

Bartlett School of Planning, UCL

@ProfMCarmona

January 2020