Redefining suburbia

Redefining suburbia

This study focuses on the ‘mainstream’ segment of the housing market – those for whom suburban ideals remain important

Study by MJP Architects

Download the study as a pdf

• Existing research into consumer choice suggests that these homebuyers choose suburban neighbourhoods because they are relatively affordable and they represent a desirable lifestyle - the 'suburban dream'. We postulate that while this lifestyle is often associated with lower densities, it can be achieved at much higher densities than is currently the norm.

• The various benefits of increased density collectively improve the sustainability of a development.

• Five hypothetical design studies of housing and block configurations demonstrate this, achieving densities of 57 – 120 net (1) dph (dwellings per hectare) or 285 – 528 net bph (bedspaces per hectare)

• A sixth study, of a hypothetical community of 5,000 people, shows that significant positive trade-offs for the community as a whole begin to occur if the average net density is above 50 dph.

• The study establishes a range of density at which many of the benefits of higher density living can be achieved without sacrificing what remains the aspiration of most British homebuyers – their own house and garden with its own front door in a safe, quiet, and leafy street.

• These proposals depend on the efficiency of road systems sustaining density across a community and suggest how land use and infrastructure costs for housing can be significantly reduced.

Los Angeles 15 dph
Milton Keynes 17 dph

London 42 dph

Suburbia and density

Of the UK’s population of 49 million, 30million live in suburbia. Due to changing demographics and culture, our cities have seen a revival in recent years. However, a significant proportion of homebuyers, especially families with children and first time buyers continue to choose a suburban lifestyle. Government policy and legislation such as PPG32 (2), aim at promoting higher densities, is thus often seen to be running counter to the wishes of the majority of the population. This study directly addresses this issue.

The benefits to society as a whole of raising residential densities are too important to be ignored:

1. Transport - walkable neighbourhoods dense enough to support public transport can offer residents the choice of owning or not owning a car, supporting a more efficient use of resources through public transport, as well as enhancing the independence of those unable to drive.

2. Economics - infrastructure costs per household tend to be lower in higher density developments.

3. Environment (3) - more compact developments leave more land available for recreation, agriculture and wildlife.

4. Society – both higher densities and the factors often associated with them (more walkable communities with a diversity of housing/land use types in closer proximity) have been reported to be associated with a better ‘sense of community’ and an increased likelihood of knowing one’s neighbours (4).

This paper explores ways in which these collective benefits of higher density living can be reconciled with the needs of individuals who aspire to a typically lower density, ‘suburban’ lifestyle. The importance of this sector of the housing market should not be underestimated - while there has been a significant renaissance in urban living in recent years, this has involved mainly childless households.

Lyon 93 dph
Kowloon 1250 dph

However, families with young children still make up 20% of the housing market - for them and others the benefits of a suburban lifestyle are very real (5). Furthermore this is the market segment that makes ‘aspirational’ choices in home-ownership and may therefore have a disproportionate influence on the preferences of other groups. New housing in England continues to be built on average at 25 dph but more than half of this is built at less than 20 dph – the impact of this sector on the land taken for new housing makes it very significant indeed.

(2) Planning Policy Guidance 3: ‘Housing’ – 2000, Office of the Deputy Prime Minister(3) See the (BRE) British Research Establishment ‘Ecohomes’ environmental ratings for homes which includes an encouragement for higher density by awarding points for‘making effective use of the building footprint’ on the proportion of building over two storeys.(4) Capital Gains Housing Federation and Urban Design Compendium, English Partnerships(5) See BMRB, ‘Kerb Appeal’, Forest, Kennett and Leather, ‘Home Owners on New Estates in the 1990s’, Policy Press, Mulholland Associates, ‘Towns or LeafierEnvironments’

Density and the suburban lifestyle

US research into why people choose suburbia reveals that it is the aspiration for a type of lifestyle rather than the attractions of a particular environment that is driving the process (6).

While this lifestyle is perceived to be associated with lower densities, UK studies (7) have shown that there is often little correlation between residents’ description of a neighbourhood as ‘urban’ or ‘suburban’ and its actual density. Densities are perceived as low by residents if they include open spaces, low height to space ratios, low artificial light levels, low traffic levels, private gardens and entrances, etc. These criteria do tend to set an upper limit on density.

However, we argue that substantially higher densities can be achieved than the current norm while retaining these characteristics.

A review of recent research on housing choice (8) produced the following list of qualities that are consistently seen as positive by a majority of homebuyers. These qualities have been ordered to form a spectrum, from those pertaining to the design and layout of individual homes, to those relating to the neighbourhood as a whole.

The ones nearer the top of the list are those traditionally associated with suburbia and with lower densities. The ones nearer the bottom of the list are generally unachievable without a fairly high average density. An individual’s decision about where to live or buy can be seen as a process of prioritising and making trade-offs between these qualities based on the options available.

Those who choose a home in a low-density neighbourhood are prioritising the qualities near the top of the list. However, in an ‘ideal world’, they may also appreciate the qualities at the bottom of the list. We suggest that those qualities at the top of the list, while traditionally associated with low densities, can actually be achieved at far higher density than is currently the norm. If so, then a ‘best of both worlds’ may be possible, which would offer homebuyers a balance of all the qualities on the list.

(6) Shlay A.B., ‘Castles in the Sky: Measuring Housing And Neighbourhood Ideology’(7) Llewelyn-Davies and Bartlett School of Planning, ‘The Use of Density in Urban Planning’, DETR, 1998(8) Capital Gains, ‘But Would You Live There?’

Two neighbourhoods compared

This suggestion is perhaps best illustrated with a comparison of two existing neighbourhoods. The diagrams analyse two different neighbourhoods in the same suburban district, to the northwest of Milton Keynes - Greenleys and Wolverton.

Because of their proximity, they form a very useful comparison, as many of the qualities at the bottom of the above list (i.e. those collective or neighbourhood qualities that are normally beyond the control of an individual developer) can be controlled in this context – general location, the quality of the local school and services, access to public transport, etc. Also, both developments contain 3 bedroom houses of approx.120 sq.m. with 1 off-street parking space. The type of construction is also comparable generally – two storey brick buildings with pitched roofs.

Greenleys, Milton Keynes

• The first, Greenleys, was built in the 1970s, and has a net density of 25 dph - approximately 125 bedspaces per hectare – the current UK average for new residential developments. A majority of the houses are semidetached, though some are detached, and all have garages.

• The cul-de-sac layout is designed for cars, not people – although it helps keep out through traffic, the wide roads encourage higher speeds, and pedestrian access to local shops, etc., is difficult.

• The traffic is channelled onto wide, high speed arterial roads, and a large proportion of the land is ‘SLOAP’9 – e.g. buffer zones to protect the residential areas from this higher speed traffic, unusable as public amenity space. Greenleys: 1970s car suburbs

• As a consequence of the road system a gross density of as little as 7 dph is typical of many similar areas of Milton Keynes.

• The layout of the houses makes for social discontinuity and isolation – the streets are not well-defined or overlooked, leading to a sense of insecurity for pedestrians.

• The houses have generous private space, but much of it is hard to use, either in narrow strips separating houses to the sides or front ‘aprons’ overlooked by adjacent houses.

Roads : 20%

SLOAP (space left over after parking): 40%

Storeys : 2

FAR (Floor Area Ratio) : 0.4

Off-street parking : 100%

Low density: defined perimeter
High density: quality public space

Greenleys: 1970s car suburb
Density: 25 dph (125 bph)

Wolverton, Milton Keynes

• The second, Wolverton, was built in the 1880s and has a net density of 52 dph. It consists of terraced houses. Many residents take advantage of off-street parking accessed from a rear alley.

• The roads are narrower, and the grid form helps distribute traffic, reducing the need for larger, faster arterial roads. It is more permeable to pedestrians, and the grid intersections help slow traffic.

• The road system sustains the density of the block across the whole neighbourhood.

• The layout of the houses is such that the street feels well-overlooked, and thus more secure.

• The houses have much less private open space, but all of it is usable and in a comfortable relation to the houses. In spite of the much smaller plots and higher densities of Wolverton, the dwellings are seen as more desirable than in the more traditionally suburban Greenleys.

The design and layout of Wolverton allows it to provide most of the qualities from the top of the housing choice list nearly as well as Greenleys, in spite of being over twice as dense. The increased density means that it can also provide many of the benefits at a ‘neighbourhood’ level from the bottom of the list. Because of its density and the way it is designed, Wolverton is a ‘best of both worlds’ scenario. The most enduringly popular suburbs in fact are often developed at densities similar or higher than Wolverton - e.g. Clifton in Bristol, Whalley Range in Manchester, Worcester Park in London, Trumpington in Cambridge, and many others.

Recent research in the capital reports that: 'While it is difficult to generalise, thriving suburbs tend to be developed more intensely (10).

Roads : 20%

SLOAP : 0%

Storeys : 2

FAR : 0.4

Off-street parking : 100%

10 'A City of Villages' p 38

Five studies: housing and block layout

The following studies take a three-bedroom family house as a ‘building block’ to explore the maximum densities that can be achieved given the requirement to maintain the main ingredients of the suburban ‘lifestyle’. This focuses on issues of design at the scale that is usually within the control of the individual developer – house, street and block layout. Design approaches and ‘tricks’ are identified that help to achieve the higher densities.

The first three studies satisfy the following criteria/assumptions:

• They provide the qualities associated with traditional suburbia.

• Houses are 125 sq.m. with 3 bedrooms, 1 off-street parking space and a private garden or patio.

• They involve simple two/three story dwellings, with construction costs comparable to those for a typical detached suburban house built of similar materials.

• A grid block layout is assumed for this hypothetical study. In a real situation the grid may not be rectilinear, but might shift in shape to account for topography, landscape features, preexisting buildings, etc.

• Net density includes local access roads – assuming a grid or similar diversified access system, density is sustained across the settlement

• The urban structure is not considered here – a real design would be likely to include a mixture of all three different housing types, plus others not shown here, within an urban structure that offers a variety of different densities.

• These are designed to ‘Lifetime Homes’ (11) standards except where otherwise noted.

1. Courtyard Housing

Eight houses are grouped around a landscaped parking court, which is open to the public with a mid-block pedestrian cut-through. The houses overlook a protected parking area. These little communities of 8 houses create something of the sense of exclusivity of a cul-de-sac, avoiding what some perceive as the anonymity of the long Victorian terrace. However, from a design point of view, care must be taken to ensure that the façades facing the public street remain animated, to avoid creating inward looking communities that make the public street feel desolate and unsafe.

Density : 57 dph net (285 bph)

Housing mix 100% 3b 5p houses

Storeys 2

Offstreet parking 100%

Proportion of road 36%

11 See ‘Designing Lifetime Homes’ – Joseph Rowntree Foundation (Recommendations with implications for site planning are: if a car parking space is provided it should have a 900mm zone, which can be shared, in addition to the standard 2.4m, level access to entrance, located at entrance level a WC and a living room capable of accommodating a single bed

12 Where b=bedroom, p=person

2. Terraced Court Housing

This develops the principle of the traditional Victorian terrace, as in the Wolverton example, but courtyards are introduced to encourage smaller communities. As all houses have front doors facing the public street, they have the advantage of providing a lively public realm.13 The density achieved is significantly higher than for the small courtyard study.

Density: 77dph, 383 bph

Design 'tricks' include:

• A sense of enclosure is created for this space with the addition of ‘carriage houses’ entered at first floor level, above garage parking, at each end. These also help to provide frontage facing onto and animating the cross streets.

• Some allowance is made for variation at the ends of the terrace rows. Depending on the situation, a ‘reallife’ design might take this variation further, and incorporate existing trees etc. to provide punctuation to the rows of houses, until an appropriate tradeoff between density and amenity is achieved.

Note that the maisonette in the 'carriage house' would need a lift to achieve lifetime homes standard.

3. Mews Housing

This example achieves a higher density by eliminating the parking areas, substituting garage parking beneath three-storey houses.

The terraces of houses are arranged around ‘mews’ or mid-block alleys; a special dwelling type on either side of the entrance to these alleys provides two maisonettes, a lower one with a ground level patio and an upper one with a large balcony at second-floor level forming an arch over the entrance to the mews street. (The upper maisonette would not have level access to the front door, a 'Lifetime Home' requirement, unless a lift is provided).

This type and density begins to depart from a suburban feel; it does however have the bonus of offering flexibility and choice in the use of car space – the garage might be converted into an office, playroom, or spare room. It might lend itself to infill developments in existing suburban areas, in particular if it is anticipated that the area will densify and become more ‘urban’ in character over time, thus perhaps eliminating the need for a car in the future.

As in the Courtyard study, the grouping of the houses into smaller communities helps generate a sense of identity. Again the design must ensure that the dwellings do not present a blank wall to the public street; for example, the units backing onto the public cross streets may have their front doors on the opposite side, in order to enliven the street.

The Courtyard, Terraced Court and Mews housing studies demonstrate that densities between 57-87 dph can be achieved with family housing with the addition of some maisonettes in Mews 'gateway' blocks.

Lifetime Community

The study is taken further to produce hypothetical 'Lifetime Communities'. The dwellings shown in block studies 4 and 5 produce higher density and a variety of dwelling mix, within a short walk of public transport nodes. By incorporating maisonettes and flats, more dwelling choice is offered for different stages of life. The inclusion of higher density

blocks along key routes could provide space for retail, office or recreation-based employment or alternatively allow lower density around the periphery. Thus the neighbourhood starts to obtain a structure,which further reinforces a sense of community and distinctiveness.

4. Mews housing with terraced maisonettes

By locating maisonette over flat or maisonette along a public transport route parallel to the courtyards, a net density of 111 dph, with bed-spaces per hectare rising to 485 or 528, depending upon the mix and height of the block. The ratio of car space to dwelling reduces to 94% as the corner ‘gateway’ blocks do not have a off–road dedicated parking space. To provide level access to upper level front doors, a 'Lifetime Home' requirement, lifts are provided in the corner blocks.

Density: 111 dph net (528 bph)

5. Mews housing with flats and maisonettes

The illustrations show flats and maisonettes located within 4 storey ‘gateway’ blocks, with maisonettes over the adjacent mews houses, increasing net density to 120 dph and 482 bph and a car-to-dwelling ratio of 76%. These figures can be adjusted by varying the mix of dwellings, the height of blocks and the required car parking ratio, depending upon the character of the neighbourhood and the local transport services.

Lifts in the 'gateway' blocks serve the upper-level flats and maisonettes to allow level access to front doors – a ‘Lifetime Home’ requirement. House type mix can be further varied by the inclusion of larger homes, such as a taller ‘mews’ house type, providing for extended families and aspirational homes.

Density : up to 120 dph net (482 bph)

A community of 5,000 dwellings

This exploration of the design of street/ block layout and individual houses, flats and maisonettes tells us that many of the qualities at the top of the 'housing choice' list can be achieved at densities significantly higher than 50dph. However, it does not tell us anything about achieving the qualities at the bottom of the list.

These depend on wider issues such as the urban structure of the neighbourhood as a whole, location in relation to employment/amenities, quality of public services and infrastructure. If the density is high enough, a walkable community can be created – one that supports a full public transport infrastructure, making a private car a luxury rather than a necessity.

Another study – this time at the scale of a community of 5,000 dwellings, puts the examples above in context and establishes the ‘average’ density necessary for a walkable community. Research by the DETR and others suggests that a viable regional link needs to have 5,000 dwellings within a 10 minute walk.

This is equivalent to 800m, which, assuming a grid block layout, might give a 600m radius as the crow flies. This creates a circle of 113 hectares. Local infrastructure for 5,000 homes is taken from an LDA study (14) and assumptions as to the land take are based on typical existing examples, give a total of 13.0 hectares, leaving 100 hectares to residential. A net density of 50 dph (15) would be required to accommodate 5,000 dwellings; this is the minimum density for a walkable suburb.

The viability of public transport is a key to achieving some of the ‘wider community’ advantages of higher densities. Public transport can generate a ‘community feel’ as it allows a diversity of people to live in a neighbourhood, including those who cannot or do not wish to drive. A correlation has been shown between not driving and knowing one’s neighbours.16 Densities higher than 50 dph will generate a greater range of local services and stores.

Conclusion

Most studies of housing densityhave focused on urban housing and urban lifestyle, taking advantage of demographic trends which have driven inner-city revival. However, a significant proportion of homebuyers, especially families with children and first-time buyers, choose suburbia because of it's relative affordability and the criteria associated with individual houses, private gardens and entrances, green space, privacy and off-street parking.

What these combined studies demonstrate is that these criteria can be achieved at substantially higher densities than anticipated in current practice. In a community of over 5000 dwellings at a density above 50 dph or densities varying around an average of 50 dph we demonstrate that it is possible to achieve a pedestrian community which brings a viable public transport system and many facilities, usually associated with an urban environment, within ten minutes’ walk.

A mixture of dwelling types including one and two person flats can increase densities further, extending the range of choice of facilities and local employment and opening up the potential of a more diverse demography of young and old, families and single people. This could be a suburbia with a real lifetime community.

A fundamental characteristic of our proposal is that gross densities across a community sustain the net densities of housing groups. This is the key to efficient land use and sustainability and requires a new approach to highway engineering which rejects distributor roads which dilute density, in favour of permeable road networks which sustain density, reduce traffic speed and create a more pedestrian-friendly and neighbourly environment.

Such an approach can significantly reduce land use and infrastructure costs. The high-density suburb of the future should be tranquil, private, and green; it should feel like home, as well as being dense enough to be walkable, and to support a richer diverse community.

Further research

The studies so far are a quantitative demonstration of relationships between house types, layouts and density, and the potential for walkable suburban communities. Further research would engage with the following issues:

Roads To sustain high gross densities, current highway engineering criteria will need to be reconsidered. A repertoire of types of road, street and mews will need to be developed

Open Space and Variety Differences between net densities of housing groups and gross densities across the wider settlement will allow patterns of open space and greenery to be developed to create new leafy kinds of suburbia.

New and Existing Settlements The studies need to be tested in different contexts including the densification of existing settlements. These should be real case studies.

The Housing Market The studies are based on assumptions about suburban values which can be realised at high densities. This is a generalization which needs to be tested against various market conditions.

Bibliography

Appleyard, D., ‘Liveable streets’, University of California Press, 1981

BMRB International, ‘Kerb appeal,’ Popular Housing Forum, 1998

BRE Checklist and website www.bre. co.uk/ecohomes ‘Ecohomes – Rating Prediction Checklist’

Julie Brewerton and David Darton, ‘Designing Lifetime Homes’, Joseph Rowntree Foundation, 1997

Caitriona Carroll, Julie Cowans and David Darton, ‘Meeting Part M and Designing Lifetime Homes’ Joseph Rowntree Foundation 1999

CABE, ’The Value of housing design and layout’, Telford, 2003

CABE, ‘Making Higher Densities Work’, 2005

Castells M., ‘The Rise of the network society’, Oxford: Blackwell 1996

Congress for the New Urbanism : www. cnu.org

Cope, H., ’Capital gains : making highdensity housing work in London’, Housing Federation, 2002

Davis, M., ‘City of quartz : excavating the future of Los Angeles’, Verso, 1990

Ellis, J., ‘Explaining residential density’, Places, vol. 16, no. 2, Spring 2004, p. 34-43

English Partnerships, ‘Urban design compendium’, Llewelyn-Davies, 2000

English Partnerships, ‘Sustainable urban extensions : Planned through design’, English Partnerships, 2000

Fishman, R., ‘Bourgeois utopia: The Rise and fall of suburbia’, Basic Books, 1987

Forest, Kennett and Leather, ‘Home owners on new estates in the 1990s’, Policy Press, 1991

Garreau, J., ‘Edge city : Life on the new frontier’, Anchor Books, 1992

GLA, ‘A City of villages : Promoting a sustainable future for London’s suburbs’,

GLA, 2002 GLA Architecture and Urbanism Unit, ‘Housing for a compact city’,

GLA, 2003 Independent Transport Commission, ‘Suburban future’, July 2004

Jacobs, A., ‘Great streets’, MIT Press, 1993

Jacobs, Jane, ‘Life and Death of Great American Cities’ King, A (ed.), ‘Re-presenting the city : Ethnicity, capital and culture in the 21st century metropolis’, Macmillan, 1996

Llewelyn-Davies and Bartlett School of Planning, ‘The Use of density in urban planning’, DETR, 1998

Local Government Management Board, ‘Sustainable settlements’, UNE & LGMB, 1995 (recommends average net density of 40-50dph to support a good bus service)

London Development Agency – www. lda.gov.uk

Mace and Gallent, ‘Sustainable suburban areas’, Bartlett School of Planning, 2000

Marsh, M., ‘(Ms)reading the suburbs’, American Quarterly, vol. 46, 1994, p. 40-48

Mulholland Associates, ‘Towns or leafier environments : A survey of family home buying choices’, House Builders Federation, 1995

Of•ce of the Deputy Prime Minister, ‘Better places to live by design: A companion guide to PPG3’, Telford, 2001

Platt, S., ‘Housing futures : informed public opinion’, Joseph Rowntree Foundation, 2004

Rudlin, D. & Falk, N., ‘Building the 21st century home: The sustainable urban neighbourhood’ Architectural Press, 1999

Shlay A.B., ‘Castles in the sky : Measuring housing and neighbourhood ideology’, Environment and Behaviour, Vol.17 No. 5, Sept 1985, p. 593-626

Sorkin, M., ‘Variations on a theme park : the new American city and the end of public space’, Noonday Press, 1992

Town and Country Planning Association, ‘Home truths : Setting out the evidence of the need for new homes’, TCPA, 2003

Urban Task Force, ’But would you live there? : Shaping attitudes to urban living’ DETR, 1999

Acknowledgements

Many thanks to John Ellis whose studies of housing in the U.S. form the basis of this work and without whose guidance this would not have been possible.