One of the most politically charged and ideologically loaded question in the field of metropolitan studies is that which concerns people’s motivations to live where they live within a given urban region – i.e., why some people choose the suburbs over the city, or the countryside over the suburbs, etc. In the US, this question has strong racial overtones because of the raging debate (outside of academia, mostly) as to whether black residential segregation in central cities is the result of blacks choosing to self-segregate.
(It’s not, a least according to Harvard Economist Edward Glaeser, but the debate still rages. Just type “blacks choose to live” in Google and you’ll see).
Whatever the case may be, liberal (and neoliberal) economists believe that individuals are, by and large, autonomous, rational utility-maximizers. In other words, they assume that people make decisions – including their choice of residential location – based on a clear set of preferences (about such things as “peace and quiet”, “commuting time”, “privacy”, “access to recreational amenities”, etc.) which are themselves more or less idiosyncratic. Some people prefer the “bundle” of characteristics they find in the suburbs, and that is all there is to it.
As an aspiring social engineer (I’m only half joking), I – as other urban planners and aspiring social engineers – am interested less in the “revealed preferences” of suburban dwellers as in how these preferences are formed. Are people’s residential locational preferences really idiosyncratic (i.e., somehow inherent to each person)? Or are they somehow influenced by the environment in which we are brought up, as suggested recently by a group of researchers at Laval University? To paraphrase the article from Le Soleil which reports on the findings of these researchers, is there a “suburban gene”?
Seeing as the article didn’t really answer the question – or attempt to answer it – I decided to do a bit of research myself. I remember distinctly that in planning school we were told that the best predictor of where one ends up when one has kids is the kind of (sub)urban environment where one grew up. Childhood memories supposedly trump all other preferences and “acquired tastes”. However, I couldn’t find the exact reference so I did a bit of research online and found this very interesting Australin “twin study” which looks at the locational decisions of identical vs. non-identical twins to see 1) if twins’ choices are correlated and 2) if identical twins are more or less likely to make the same choice as non-identical twins.
Notwithstanding the very real definitional difficulties involved in trying to determine what is urban, suburban and rural (an important subject which deserves its own separate post), the research design proposed by Whitfield and colleagues is an interesting one – and perhaps the only allowing us to really get at this question of “nature or nurture” in the formation of residential locational preferences. And the results are truly fascinating: both “additive genetic” and “shared environmental” influences are important, both the former becomes more important with age whereas the latter seems to become less influential (see Figure 1 below, reproduced from the article).
So what does it all mean? Well first of fall, it is interesting to note that both nature and nurture matter. But more interestingly: nature and nurture do not matter to the same degree at different stages of life (assuming that older cohorts are different simply because they are older and not because of some unobserved variation between cohorts). Note that subsequently a Dutch study and a Swedish study replicated the same research design and that the first one did not find evidence of “additive genetic influences”; however, both studies did find that environmental influences from childhood were important determinants of residential location choice – especially for younger people.
Some of you might ask (quite rightly): why does this matter from a policy, planning or social science perspective? One thing to observe, it seems to me, is that growing up in the absence of density replicates the absence of density. It follows, then, that we need to continue densifying the suburbs because that is the only way we can make density more acceptable to the next generation. Indeed, if “shared environmental influences” (i.e., environmental influences from childhood) explain 50% of the variance in residential location for people between 35 and 40 (i.e., people with young children or about to have them), then we need to influence the “shared environmental influences” of their children… so that they will find a wider range of options acceptable when they are making locational decisions 35 years from now.
Just to be clear: I do not imply here that density is aesthetically better and that we need to impose these aesthetic views upon those who have “objectionable” sensibilities. People’s sensibilities are what they are and I am not judging them. I am simply observing that low density living is self-perpetrating – and that since we can’t afford (economically or ecologically) to live like the inhabitant of Aurora in Azimov’s The Robots of Dawn, we will need to tame density or way or another.
Great article, Joel. Just as we learn that Montreal lost 1.20% of its population last year (mostly to Laval and the south shore), I found myself wondering how big of a proportion the baby boomer’s exodus has to be taken into account. Most of our parents (speaking of francophones at least) grew up into a somewhat rural environment and, as they get close to retirement, will try to replicate some elements from their childhood in their lifestyle. That is, at least, what most of my francophone friend’s middle class parents (and my own) are doing these days: going back to “sources”.
On the other hand, an ever going larger proportion of kids in Quebec now come from an urban environment and younger generations would logically seem to be more tolerant to issues related with high density living spaces. Coupled with the higher nativity rates (12% in Montreal vs 11.1% provincial average), could we expect the suburban exodus to notably slow down on the long run?
Who on earth suggested that racial segregation in US cities is “the result of blacks choosing to self-segregate”? I was under the impression that inner-city black ghettoization in the US was primarily the result of “white flight” – i.e., white people migrating to the suburbs and leaving blacks behind in the inner city. If anything, it was the whites who were self-segretating. Weren’t black Americans’ generally poorer economic circumstances the primary factor that prevented them from buying into the car-oriented suburban lifestyle, rather some kind of desire to live apart from white people?
Some further reading of interest:
– An academic paper arguing that job location and intolerance to commuting may a partial factor explaining black urbanization vs white suburbanization in the US (http://www-siepr.stanford.edu/workp/swp00007.pdf).
– Suburb apologist Joel Kotkin, armed with some stats from fellow suburb lover Wendel Cox, shows that millennials or “boomer echoes” in the US aren’t settling in inner cities but rather living in them transitionally (http://www.forbes.com/sites/joelkotkin/2011/07/20/why-americas-young-and-restless-will-abandon-cities-for-suburbs/). The data is interesting but I disagree with the conclusion, which is basically that we should keep building suburbs because that’s what people want.
– Smart Growth America challenges Kotkin’s conclusion and argues that the city-suburb dichotomy is false and obscures American’s desire for denser, more walkable living environments (which of course supports the Smart Growth agenda…) (http://www.smartgrowthamerica.org/2011/07/28/cities-versus-suburbs-is-the-wrong-debate/)