The three proposed apartment towers next to Flemington Racecourse railway station, on land in Ascot Vale.
The three proposed apartment towers next to Flemington Racecourse railway station, on land in Ascot Vale, currently under examination by Planning Panels Victoria.

The idea that every new building, neighbourhood and suburb must be “diverse” is a truism in urban development.

It might be imagined that most attention in new developments would therefore be focussed on fostering diversity on crucial variables like income and cultural background.

The reality is that politicians and planners almost always find that too hard so they mostly focus their attention on easier objectives like creating a mix of household types e.g. singles, couples and families with children.

The standard approach is to require new developments to provide a range of dwelling sizes, from one bedroom through to three or four bedrooms. The theory is singles and couples will self-sort into the smaller dwellings and families with children into the larger ones.

So inner city developers who respond to market demand by providing mostly one and two bedroom dwellings often face stiff opposition.

For example, here’s a report in Monday’s paper on what Melbourne City Council’s Barrister, Juliet Forsyth, put to Planning Panels Victoria in opposition to a proposed development on Flemington Racecourse:

This development appears to be aimed at the investor market and will deliver more of the same: small one and two bedroom apartments with few measurable controls to ensure reasonable levels of amenity and [an] appropriate mix of dwelling types.

I discussed another instance yesterday; the City of Yarra’s rejection of a proposed development in Cremorne on the grounds that 80% of apartments would have one bedroom (see Is 20% two and three bedroom apartments diverse enough?).

This idea that there has to be diversity of household types at the building, neighbourhood, or even suburb level is flawed thinking. It misunderstands what cities are about.

As I’ve discussed before, what proponents of this view don’t get is that while big cities provide greater diversity (and are rightly lauded for it), they also provide more opportunities for specialisation (see Are bigger cities less diverse?).

This is the key insight:

  • You are more likely to find people who fit closely with your opinions, tastes and idiosyncrasies in a big city like Melbourne than in a smaller place like (say) Geelong…
  • But the consequence is that once having found like-minded people, you’re also more likely to spend most of your time with them and ignore the lesser matches you’d tolerate in a smaller place.

Wellesley College researcher Angela Bahns and colleagues found students at smaller universities have a greater diversity of friends – in terms of lifestyles and opinions on issues such as abortion – than students at larger institutions. They subsequently compared big cities like New York with small ones like Iowa City and came up with similar results.

This is the “similarity-attraction effect”:

It influences everything from whom we date and hire to where we choose to live. The bigger the pond, the more likely we are—consciously or not—to swim around until we find a group of like and like-minded people.

Geographical concentrations of newly-arrived migrants are one manifestation of this effect at the aggregate level. So are families with children; they generally like to be near other families because there are benefits e.g. sharing child care, coordinating social activities around children. Singles like to co-locate too e.g. more opportunities to meet-up, fewer families objecting to noise.

In most cases this isn’t something to fear; it’s how the diversity dividend of big cities works – more diverse at the bigger scale but more specialised at the smaller scale.

The idea that all buildings, neighbourhoods or suburbs should have much the same demographic profile – that any significant deviation is a deficiency – misunderstands what humans want and how cities work to deliver benefits to their residents.

Cities are and always have been “spiky”; barring the invention of teleportation they always will be. That’s obvious in the lumpy way jobs are distributed spatially but it’s also true of other activities that benefit from co-location.

More singles and couples want to live in an apartment complex like the proposed Nylex development I discussed yesterday because they’re the ones who place the highest value on the benefits of the location and are more tolerant of the restrictions it necessarily imposes. Families with children mostly see better value elsewhere, perhaps a terrace or townhouse.

This clustering of like-minded people for mutual benefit (agglomeration economies!) is generally a good thing, but there’s one very important caveat; geographical concentrations of socio-economically disadvantaged residents are problematic because agglomeration can also amplify negative forces (see Does place matter for the life prospects of children?).

However there’s no compelling underlying logic or morality that legitimises planners trying to design in diversity of household types at the micro-scale in places like the inner city. And it’s of doubtful utility anyway; larger apartments in desirable locations are more expensive and mostly get taken by well-heeled singles and couples rather than families with children.

Planners would be better advised focussing on increasing opportunities for low to moderate income households of all types and sizes to gain access to affordable dwellings. One way to do that is to increase the supply of housing across the metropolitan area so that it matches demand.