Does Melbourne's CBD need density controls?

Planners and urban designers have mostly sought to increase densities in Australia’s cities. Society’s low-density love affair has brought us unending sprawl, with all its social, environmental and economic ills. Attempts to rein in our metropolises’ spreading girth and create more sustainable, liveable and economically-efficient cities have focused on strategic increases in density to support public transport use and reach viability thresholds for local amenities within walking distance.

Now, courtesy of Leanne Hodyl’s elegantly presented Churchill Fellowship report, we are being asked to consider whether densities have got too high in central Melbourne. After the raging success of the Postcode 3000 program in attracting people to live in the city centre, are we witnessing the creation of an unliveable CBD?

The problems identified in the report include poor internal amenity, poor public realm amenity, unfair constraints on the future development of neighbouring properties (‘inequitable development’) and a lack of commensurate open space and community infrastructure provision. There is no doubt that there are examples of all of these undesirable outcomes (and, of course, many examples of tall buildings that don’t cause them). The question is whether this warrants density controls.

In one of Melbourne’s many efforts to distinguish itself from Sydney, we don’t (in the main) seek to control development density, instead focusing on its form. This is because it is the siting, scale and shape of a building that most influences its amenity and that of the surroundings, rather than the floor area or the number of dwellings or people it accommodates. Insufficient internal and external daylight and sunlight, disrespect of heritage and character, visual bulk and wind are all a result of form, not density.

It is true that designing a building to avoid adverse impacts (typically through boundary setbacks) reduces its density compared with a vertical extrusion of the site to the same height. But the reverse is not necessarily true: reducing density does not automatically reduce the adverse impacts of a building. And the density of an acceptable building on one site may be quite different to the density of an acceptable building on the next, due to differing interfaces. In other words, density controls are a very blunt tool for managing the ill effects of tall buildings.

Density does have impacts independent of form — notably, demands on infrastructure and congestion of transport networks. However, this is largely a question of funding. At face value, there is merit in the notion of a density-based infrastructure contribution scheme. You want more floorspace? You provide or pay for infrastructure upgrades.

To a point, street congestion is merely an influence on character. Busy streets offer a different experience to quiet streets. Not better or worse, just different.

Ultimately, however, we need to ensure people can get around safely and efficiently. We have accepted the principle that roads will be congested at times in big cities, and the only way to provide for growth is to get more people onto alternative modes of transport (including cycling and walking). We are now beginning to see trams and cycle lanes reach critical levels of congestion too. But, like community and utility services infrastructure, this is just a funding problem. Density-based contributions can fund upgrades in public transport and cycle infrastructure.

Of course, there is an absolute limit on the space available within the street network for moving people about. And increased provision for public transport and cyclists may require more space. But we are far from wringing the last unit of movement capacity out of our wide streets. The trend of reallocating space from private cars to tram stops and cycle lanes has a long way to go before we reach their absolute limit.

So, density controls aren’t the answer. How, then, should we control tall buildings to avoid the kinds of problems beginning to afflict the CBD?

Setting the question of infrastructure aside, a leading cause of the other problems is insufficient building separation. Until now, most taller development has occurred on larger lots, separated from other taller buildings by narrower lots or through the ability to provide generous setbacks. However, changes in technology and market preference are now allowing tall buildings on small sites. The potential for such buildings to be constructed ‘cheek by jowl’ with other taller buildings and the inability of small sites to accommodate generous setbacks has been the cause of many of the issues raised in Ms Hodyl’s report.

Since most of the larger sites in the CBD have been developed, and further growth relies on the development of smaller properties, how can the ill-effects of taller buildings be avoided?

The answer is that a more sophisticated but enforceable building separation control is needed. The 24m separation sought by Clause 22.01 quite simply rules out the development of any property less than about 40m wide. This has led to it being ignored (even in situations where the neighbouring property could, in future, be developed, in which case the ‘first come first served’ principle has been adopted) because to do otherwise would stymie growth in the CBD.

A new building separation control should take into account that:

• Towers on narrow sites may be able to obtain their amenity primarily or wholly from the front and rear, avoiding the need for side setbacks for internal amenity or equitable development reasons.
• Towers on narrow sites have less impact on the public realm due to their narrower profile. Indeed, many of the larger towers in the CBD are the equivalent of three or more narrow towers in footprint.
• The external impacts of a building relate at least partly to its height, so the scale of setback requirements should be linked to height.
• Where neighbouring properties are unlikely to be redeveloped for towers (due, for example, to significant heritage values or strata titling), they can provide the necessary tower separation avoiding the need for a setback from that boundary.

We need to address the problems being caused by some taller buildings in the CBD. But, attractive as it might sound for reasons of administrative simplicity, density controls are not the answer. Instead, it lies in more nuanced and, therefore, implementable side and rear setback controls.

Mark Sheppard is a Principal at David Lock Associates and author of ‘Essentials of Urban Design.’ This article recently appeared on the David Lock Associates website.