Is Urban Design Evidence-Based?

Evidence-based plans. That is the call these days. An interesting issue is how urban design gets tangled up in this call for "evidence” and comes out the other end not very well off when in fact it is evidence-based.

The Independent Hearings Panel that looked at the Auckland Unitary Plan took up the call on evidence-based plans, and good on them. In their overview report they state (1):

“constraints on choices should be based on evidentially-validated thresholds and not on the consent authority’s preferences”.

To make the point clear they repeat it again a bit later on (2) :

“The reasons for objectives, policies and rules must be based on objective evidence rather than anecdotes or subjective considerations”.

But what is objective and what is subjective? One of the curious aspects of their decision relates to urban design and its place in resource management. You get the feeling that the Panel don’t like urban design. For example (3):

Good design is based on principles rather than rules. Mere reference to good design or the listing of preferred design principles is ill-suited to a regulatory framework which imposes binary ‘grant/decline’ outcomes. Discretionary decision-making must be exercised on the basis of relevant and clear objectives, policies and assessment criteria rather than on subjective preferences.

The Panel are not the only ones to hold this view. The Productivity Commission in their draft report on urban planning said something similar (4):

The role and skillset of urban planners needs to be more tightly defined. A more tightly defined role for urban planners would require some elements of the planning profession (notably urban designers and those promoting New Urbanist ideals) to develop:

• a greater understanding of the scope of influence of the planning profession;
• a clearer focus on the evidential basis for decisions; and
• a greater understanding of the responsibilities that come with the use of the coercive powers of the state.

What I find odd about these comments is that my understanding of urban design is that it is grounded in empirical observations of town and cities and how people interact with the spaces and places created. It is far more grounded in observation and reality (evidence if you wish) than much discourse about what is and isn’t good urban planning. The whole debate around zoning, land supply and housing affordability is driven more by theories about what should happen than evidence about what is happening, for example.

The empirical nature of urban design is what attracted me to the topic. You only need to look at the classic texts on urban design like the Death and Life of Great American Cities, Responsive Environments and the works of Jan Gehl to see how important observation is to the basic building blocks of urban design, while much urban planning is normative.

Urban design is grounded in people’s behaviour when in different urban environments – why do some city environments attract people and activity, while others appear to repel people and activity?  It is clear to me that people are very sensitive to the urban environment and small physical changes can lead to big changes in behaviour. This should not be taken as environmental determinism; rather it is an acknowledgement that we live in a complex world where people and businesses have options and choices, and those choices get shaped by the places around them. As Winston Churchill said - "man makes the town, then the town makes the man".

Urban design identifies and creates conditions that help to make places safer, more inclusive and more user-friendly and therefore successful in economic, social and environmental terms, over time. In so doing, it recognises that there are many different user groups that need to be considered, and that their needs and wants do not completely overlap. So for example in the city centre there are workers, business visitors, shoppers, local and out-of-town tourists, residents, educational users, recreational users (walkers, joggers, cyclists), as well as general visitors. Urban design therefore has a strong element of equity running through it - places that enable all people to use them. From equity comes efficiency.

In urban efficiency terms, good urban design assists with the generation of positive externalities that arise when people and businesses are congregated together (agglomeration benefits). It also helps to avoid negative externalities - in a dense urban environment, a poorly designed building that does not relate to adjacent buildings or public spaces can create a deadening effect on these neighbouring places and buildings, fostering crime and reducing amenity, for example. This is a negative cost on others that the poorly designed building does not carry. Urban design address these externalities by applying a design-based approach, rather than rigid standards - surely a good thing.

I often wonder if the dislike of urban design arises from:

1. observations of built environment outcomes suggests that individual actions in relation to how a site is developed (particularly the public-private interface) don’t necessarily add up to a wider benefit - contrary to the cornerstone of economic orthodoxy that pursuit of individual welfare adds up to community welfare, so long as environmental externalities are addressed;
2. You could say that ‘equity’ of access to and use of urban environments is as important to urban design as urban efficiency, so again its swimming against the tide;
3. Urban environments and their inhabitants are incredibly sensitive to small changes. Sometimes, micro management is needed. But micro management is not the flavour of the month.  

Getting back to urban design, I accept that it can be its own worst enemy sometimes. There is no universally accepted definition of urban design, and as a result urban design evidence can often be wide ranging and not comparable. Principles rather than standards and rules are used due to the complexity of the urban environment and our lack of the precise understanding as to the myriad of interactions involved.

Urban design can therefore be perceived to be “subjective”. But that is a easy stone to throw at urban design if you don't like the message.

Often the only recourse to objectivity is to compare and contrast different urban environments; that is to compare the proposed urban environment with an existing environment with similar qualities and to see if the claimed benefits and costs (positive and adverse effects)  are the same or different. This is, of course, not without difficulty, because of the different contexts involved. Principles need to be applied to a place in a considered way.

In terms of the degree of prescription involved in urban design, this flows on from the above point about small changes having ramifications that flow through an urban area (in complex, interconnected systems, small changes can have amplified effects). Cities are complex systems; to manage complex systems, sometimes a few simple rules are needed, in other cases detail is the deciding factor. When you are in a car at 80km per hour, details of the urban environment don't really matter and a few simple rules probably work OK. When you are at the scale of a person walking down a street at 5km per hour detail matters; detail is everything.

Until we are in a position to fully understand the links between the built environment, human behaviour and wider social, economic and environmental outcomes then there will remain a degree of judgement in urban design as to what may or may not make a successful area or precinct. But that doesn't mean we should down play the basic messages of urban design and their empirical foundations.

Footnotes:

1: Page 36, IHP Panel report to AC Overview of recommendations 2016-07-22

2: Page 37, IHP Overview

3: Page 37, IHP Overview
4: Page 328, Productivity Commission Draft Report: Better urban planning

Auckland Unitary Plan, intensification and open space

In this blog I want to look at the spatial allocation of commercially feasible intensive development and access to open space. When planning for apartments (intensification), rather than worry too much about access to public transport and proximity to services in town centres, should planning look more closely at access to open space in terms of determining where it may be appropriate to locate apartment developments?

In previous blogs I've discussed the policy shift implicit in the latest version of the Proposed Auckland Unitary Plan (PAUP) away from intensification associated with centres and transport routes to intensification based around amenity. This shift is being driven partly from considerations of commercial feasibility. But it is also partly an attempt to widen the base demand for intensification.  Proximity to open space has to be a key part of this.

The PAUP as notified included the concept of intensification around open spaces. The following is policy 2.1.2 in the notified version. I've highlighted the relevant bit:

Enable higher residential densities and the efficient use of land in neighbourhoods:

a.within and around centres and within moderate walking distances from the city, metropolitan, town and local centres

b.in areas close to the frequent public transport routes and facilities

c.in close proximity to existing or proposed large open spaces, community facilities, education and healthcare facilities

d.adequately serviced by existing physical infrastructure or where infrastructure can be efficiently upgraded.

It could reasonably be argued that in terms of actual zonings,  the PAUP as notified did not take forward (c) above in any meaningful way. Most intensification seemed to be driven by (a) and (b), although later moves by council to free up zonings started to bring in more (c) areas. 

The revised version of the policy following the hearing of submission dropped (d) - don't worry about infrastructure constraints, they can be fixed; and added in 'corridors' and expanded out (c) to include employment. Here it is:

(5) Enable higher residential intensification:

(a) in and around centres;

(b) along identified corridors; and

(c) close to public transport, social facilities (including open space) and employment opportunities.

You might say that not much of the urban area is left out of the policy by the time you add (a) + (b) +(c). Of course what the policy doesn't tell you is how much intensification might occur in these different areas.  That comes later on in the Plan.  

 As previously identified, most commercially feasible redevelopment in the existing urban area is in the form of apartment developments. An interesting relationship exists between apartments and open space. Apartments near open spaces sell for more than apartments away from open space. This relationship does not hold for stand alone houses. The relationship for apartments reflects the trade offs involved in apartment living. People accept smaller internal living spaces and less private outdoor space if close to a park. Parks and open spaces also provide secure outlook from apartments. When you are beside a park, there are no concerns  that views may be built out, or sunlight and daylight reduced, by next door development. 

An Auckland Council report[1] has confirmed this relationship:

We find evidence that proximity to parks and public open spaces has a positive impact on apartment prices.2 That is, if we compare two apartments that were equivalent in all respects except distance to the nearest park, we would expect the apartment that was closer to the park to command a higher price. Higher prices in turn reflect the higher level of amenity that apartment owners (or occupants) receive from proximity to a park. Proximity to both regional parks and to local / neighbourhood parks has a positive impact, which suggests that parks of varying significance and size are valued by apartment-buyers.

No surprises there. The higher prices for apartments near open spaces suggest higher demand, which is a good thing. 

The report did not assess access to beaches and coastal areas. My pick would that the price gradient (if that is the right word) for apartments would be even stronger for coastal access and views. 

We can't add much more coastline to the region (unless we go much further north and east), but we can add more open spaces. So to help support and enable more apartment developments:

1. Do we need more opportunities for apartments close to open spaces and
2. Do we need more open spaces?

Equally, will the lack of open space actually hold back apartment development from occurring in some areas? The analysis of commercially feasible development is just a mechanical exercise of costs and returns. It doesn't take into account the many factors that will influence buyers.

I have obtained the following figures on the area zoned open space under the PAUP (decisions version).

Open Space Type	Area (ha)
Open Space - Conservation	34,655.54
Open Space - Informal Recreation	8,410.70
Open Space - Sport and Active Recreation	3,084.03
Open Space - Community	90.96
Open Space - Civic Spaces	3.31

The largest category of open space - Conservation - includes the Hunua and Waitakere Ranges, as well as the large regional parks in outer lying areas of the region.

It is useful to organise the above data by Local Board area. If we drop out the rural Local Boards of Waitakere, Rodney and Franklin, then a lot of that Conservation land is taken out of the picture. 

When open space zoned land is organised by the remaining 'urban' Local Boards we get the following data (see Figure 1).

Figure 1: Open Space zoned land (ha), by urban Local Board

If we then relate that open space area to the number of dwellings in the Local Boards, as of 2013, then we get the following figures on open space zoned area per dwelling (Figure 2). This data is converted to square metres of open space per dwelling. The urban area 'average' is identified. The urban area average is about 230m² of open space per dwelling.  This sounds quite generous.

Figure 2: Square metres of open space zoned land per dwelling, 2013, by Local Board

There is no set standard as such in terms of how much open space should be available per dwelling (or per 1000 people). Access to open space (e.g. walking distance) may be more important than gross area. 

But lets set aside those details. What I'm looking at here is the broad picture.

Notice how some of the inner city Board areas have relatively low levels of open space. The quality of that space may be quite high (think of the Domain), but in quantity terms, there is below average provision.

If we then arrange the Local Boards by the number of commercially feasible apartments provided for by the PAUP and compare that to the open space available, we get the following figures.

Local Board	Open space (ha)	Number of possible apartments	M2 of open space per apartment
Maungakeikei Tamaki	512.2	31,192	164
Howick	818.0	22,615	362
Waitemata	250.9	19,318	130
Devonport Takapuna	231.7	18,801	123
Upper Harbour	885.8	17,720	500
Whau	285.3	13,745	208
Hibiscus&Bays	1720.4	11,300	1,522
Albert Eden	268.9	10,170	264
Kaipatiki	644.8	7,793	827
Orakei	533.9	7,674	696
Puketapapa	329.0	6,691	492
Henderson Massey	583.6	2,270	25,71
Mangere Otahuhu	604.8	1,649	36,68
Papakura	265.1	1,339	19,80
Otara Papatoetoe	585.0	666	87,83
Maurewa	779.7	12	649,767

Of course each new apartment does not get exclusive access to this area of open space land. New apartment dwellers will have to share that space with others - existing residents in the area and visitors.

The number of possible apartments could be added to the number of existing dwellings to get a feel for total demand, but total demand also needs to take into account other use of space (like sportfields used by a variety of codes who draw players from across the region).

To keep it simple, I've just looked at the growth side of the equation.  

The data suggests some areas where more open space may help, or where more intensification could be accommodated, given the stock of open space land.

Devonport - Takapuna Local Board has a low level of existing provision of open space and substantial growth in apartments, which doesn't sound good. But it is a Board that enjoys access to good beaches, so probably the lack of open space will not hold back apartment development. On the other hand Orakei Local Board is coastal and has a relatively high rate of open space provision. Could it take more apartment development?  Howick and Hibiscus and Bays might be in the same category.

Away from the East Coast beaches, Mungakeikei-Tamaki, Albert Eden and Whau Local Boards all have relatively high rates of possible apartment development. In these areas, access to open space is likely to be important. But they have relatively low levels of provision of open space.  Will this be a break on achieving the amount of intensification proposed?

At the other end of the scale, places like Manurewa probably will not benefit from more open space in terms of enabling apartment development.  

In my blog of the 12 October 2016, I talked about the role of the middle ring suburbs - how they are expected to take a substantial proportion of growth, but are areas where often there is the largest gap between plan enabled growth and commercially feasible development. Closing that gap could perhaps involve a much more targeted approach to apartments and open spaces in these areas. 

In short: more and better quality open spaces in middle ring suburbs may be more likely to drive apartment development than other forms of intervention. 

---

[1] How Do Aucklanders Value Their Parks? A hedonic analysis of the impact of proximity to open space on residential property values  August 2016 Technical Report 2016/031