There’s been a brief algal bloom of discussion on urban blogs on the historical roots of anti-urbanism, particularly Ryan Avent and Stephen Smith. It seems as good a time as any to mention the Brisbane experience. Brisbane is a nineteenth century city, and like most Australian cities, a casual attitude to earlier hunter-gatherer settlement meant it suffered from no shortage of land.
Brisbane has a strikingly sparse density for a city of two million – 918 people per square kilometre. That’s about a tenth of the density of New York city or an eighth of Los Angeles. Though it’s worth noting the city limits are drawn to include more suburbs than many other cities, it’s a pretty obvious feature of the city for even the first-time visitor.
The culture of sprawl certainly runs deep in Brisbane, buying a house on land is the conventional wisdom, and new suburbs have been ever unfolding throughout my life and before, while commute times soar ever upwards. It’s a city that demands a car, but where the ubiquitous suburban blocks are often green. I’ve been in forests overseas with less trees than the Brisbane suburbs. It may be one of the few places to deliver on that part of the garden city vision. Perhaps because of this, I had always assumed that the development pattern was driven solely by cheap land and human nature, despite my own frustration with driving for hours to do anything, or the inconvenience of taxi-ing home to the middle of nowhere after a few drinks.
In fact, much like the US examples Stephen cites, the roots are as much regulatory as organic, and they date back to the Undue Subdivision of Land Prevention Act 1885. This set a minimum lot size of 400 m^2 with a ten metre wide block. The population, though going through a boom, was only a few ten thousands at this time, and it had a huge impact on the development about to occur. The house I am writing this in is on a block of exactly the minimum size and shape specified in the act, even though it was rescinded in 1923. The motivation, as wiki notes, was slum prevention. The cost of not letting people choose smaller houses, if they wanted them, was a city that was too expensive to comprehensively sewer until the 1970s. It was also much hard to keep services like trams economic when cars emerged. Residents were trapped by sprawl, in a well-ventilated cage.
Conflated Automatons 
Phoenix is a sprawl city. There a couple of isolated islands of sprawl, like Maricopa City, that got caught in the financial crisis. North Scottsdale is one homogeneous sprawl similar to Castle Hill in Sydney. Only one major road runs past its southern extreme and like Castle Hill in the 90s there is no major arterial servicing it. Phoenix appears to be collapsing to centers again though, Tempe and Downtown appear to be winning and ASU seems to be driving that. ASU opened up a downtown campus recently as well. Supports Richard Florida’s thesis. See what happens when development money starts flowing again. Phoenix relies on construction a lot for growth.
Yeah. I didn’t really cover it here but actually Brisbane has taken a much more evenhanded approach to density in recent years, allowing highrise apartments in the centre of town, and also a lot more townhouses about the place (semi-detached in UK terms). Like Phoenix, I guess, one of the building challenges is the heat – unlike temperate climates heat causes diseconomies of cooling scale when you increase density. Since the 90s a lot of infrastructure investement is going on too, eg upgrading train lines and services, bringing ferries onto the cleaned up river, etc. Trains still only run half hourly through much of Brisbane, and even in commute hours it’s only a bit more frequent. The more I learn about it, Brisbane seemed to follow a model of development which paralleled US cities, right down to the private transit system which was taken over by the city, and the destruction of the tram lines post-war.