Suburbs and climate change: a home-grown brawl

For as long as anyone can remember, green activists and academic town planners have been damning low-density suburbs as environmental Hiroshimas. So imagine the shock and confusion when one of the country’s leading green lobbies publishes research suggesting that dense inner-city zones unleash more greenhouse emissions than car-loving fringe suburbs. Talk about an inconvenient truth.

Assuming it is the truth, of course. Stakeholders, from here and overseas, have been brawling about that ever since. The fight was sparked by comments attached to the Australian Conservation Foundation’s 2007 Consumption Atlas. Essentially, the atlas is an interactive online tool enabling users to find out a locality’s “carbon pollution” emissions, water use and “eco-footprint” by entering the relevant postcode. According to the ACF, its atlas “maps patterns of consumption and its environmental impact across Australia”. The explosive content is found in a linked “main findings” report.

The ACF’s central finding is that environmental conditions relate to levels of general human consumption. The higher the consumption, the worse it is for the planet. This is standard green stuff. The problem came when it was given a spatial twist. Here's the quote in full:

Yet despite the lower environmental impacts associated with less car use, inner city households outstrip the rest of Australia in every other category of consumption. Even in the area of housing, the opportunities for relatively efficient, compact living appear to be overwhelmed by the energy and water demands of modern urban living, such as air conditioning, spa baths, down lighting and luxury electronics and appliances, as well as by a higher proportion of individuals living alone or in small households.

In each state and territory, the centre of the capital city is the area with the highest environmental impacts, followed by the inner suburban areas. Rural and regional areas tend to have noticeably lower levels of consumption.

These trends in are closely correlated with wealth. Higher incomes in the inner cities are associated with higher levels of consumption across the board.

These words would return to haunt the ACF. As far as the report goes, no attempt is made to qualify the comments in deference to green dogma about the evils of low-density suburbia. This is surprising and, in a sense, brave. By pointing the finger toward inner city professionals, the ACF risked offending people it relies on for support. The taint of green guilt is usually applied to other demographics.

Hanging like a juicy fig, the comments were ripe for the picking.

American consultant Wendell Cox, perhaps the world’s most vocal advocate of suburban development, wasted no time in putting them to use. Cox has cited them in various articles and blog posts. Particular ire, though, has been directed at a report his firm prepared for the Australian Property Council’s residential arm in September 2007. Called Housing Form in Australia and its Impact on Greenhouse Gas Emissions, the report (PDF 3.29KB) describes itself as an “analysis of data” from the ACF atlas.

Put simply, Cox classified the atlas’s environmental data according to distance from city centres. He did a similar thing with 2006 census data relating to housing type, automobile use and population density. Then he matched them up. This method confirmed the ACF’s finding. Lower GHG emissions were associated with longer distance from the (urban) core, detached housing, more automobile use and lower population density. Cox drew the logical conclusion that squeezing suburban growth is no way to combat climate change.

Naturally, the ACF wasn't so happy to see its perfunctory comments lit with neon lights.

But the ultimate swipe at Cox came from a different quarter, Queensland’s Griffith University Urban Research Program. Under the directorship of Brendan Gleeson, Griffith is emerging as a prominent link in the chain of progressive, deep-green urban studies programs winding through the Australian university system. Last November, Gleeson, along with colleagues Rowan Gray and Matthew Burke, weighed in with a paper of their own, Urban Consolidation and Household Greenhouse Emissions: Towards a Full Consumption Impacts Approach.

The trio start off with a general tour around the state of research into the effects of urban consolidation on GHG emissions. Surprisingly, there isn’t much out there. “ … consolidation has been partially justified by alleged environmental benefits”, they say, “including the claim that it promotes household energy and greenhouse efficiency … Our review finds that evidence supporting this claim is lacking.” This is news to anyone sick of being lectured about the urgent need to rein in our cities.

Eventually, they get around to tackling Cox, who, they say, “uses a flawed methodology, and so draws conclusions that are highly questionable”. The gist of their argument is that correlation does not establish causality. Correspondence between distance from the core, detached housing, more automobile use, lower population density and lower GHG emissions doesn’t establish a causal relationship between these factors. Yet in a response to the Griffith paper, Cox agrees and points out that there are no references to causation in his report. The term he uses is “association”.

Cox goes on to hint at the most critical point, one that the Griffith trio conveniently dodge. The argument that correlation isn’t causality cuts both ways. If it isn’t valid to argue for causality in the direction of distance from the core, detached housing, more car use and lower population density to lower emissions, it’s just as invalid to do so in the opposite direction of proximity to the core, multi-unit housing, less car use and higher population density to lower emissions.

Green activists and planning dons are waging a campaign to impose their compact city vision on the rest of us, despite homebuyer preferences and the interests of a vital economic sector. Surely they’re obliged to prove their case. After the ACF atlas, Cox’s analysis and the failure to turn up usable evidence, that’s a challenge. The Griffith trio acknowledge this problem, but add “we read it differently”. The difference, though, is just a resort to obfuscation: “[i]n our opinion, current empirical evidence suggests influences on household greenhouse emissions and energy demands are complex and context dependent …”

Moving on, the Griffith boys fire a secondary shot at Cox’s method. While he treats household GHG emissions as a single factor (or variable), the “built environment” affects emissions from each set of household activities (travel, housing, recreation, food, clothing and so on) in different ways. Without specific research, however, they can’t show how Cox’s findings should have been otherwise. Their discussion amounts to little more than supposition.

Let’s finish where we started. Why was the ACF so blasé about the spatial dimension of its findings? Because GHG emissions are a function of overall consumption. Consumption shaped by low-density housing, whether it concerns transport, energy or infrastructure, doesn’t figure prominently in the composition of aggregate consumption. According to the atlas, it’s swamped by the profusion of consumption patterns across the urban landscape. This leads to a conclusion the ACF won’t like: if the object is to cut GHG emissions, singling out low-density suburbia represents a poor ordering of priorities.