I want to get my hands on the original paper, but this write-up in the New York Times looks extremely promising. The idea that where you are born–not just where within a city, but in what city–matters to your life chances. While the article spends a lot of time trying to make what is going on about urban form in their discussion of Atlanta, the actual data as displayed do not appear to follow from that: New York city is a place that appears to have a great deal of mobility, but so does Houston, if I understand the graphic properly, which I am not sure that I do. Miami is one of those regions that flips around in sprawl measures, so…how do we deal with that in these figures?

That said, please do not think that I think it’s just fine that people in Atlanta live in segregated communities and have three- and four-hour commutes. I don’t. But there is probably more going on here than a physical mobility. I suspect it is a symptom rather than a cause of low social mobility, though that doesn’t make those commutes less burdensome or destructive to family life or economic security to those who face them.

Here’s one answer I can get behind; schools can offer ways to leverage opportunities within place as well opening richer opportunities than suggested by three-hour commutes:

Lawrence Katz, a labor economist who did not work on the project, said he was struck by the fact that areas with high levels of income mobility were also those that established high school earliest and have long had strong school systems. Mr. Katz, a Harvard economist and former Clinton administration official, called the work “certainly the most comprehensive analysis of intergenerational mobility in the contemporary U.S.”

Social mobility can also go the other direction as well. I wonder what is going on with that?

Alter.net summarized the story and highlighted the five worst cities, all of which are in Alabama, Mississippi, and Tennessee–which brings us full circle, I think. While we all want to talk about about cities matter, and they do, I think we also have a clear case that states matter a great deal in the setting the urban social context in which schools and urban form and the other factors develop. For all those people out there who want to shriek at me that “we’ve never tried” small government…that’s baloney. We have a big variation in the social safety set and in social inclusion programs  in state and provincial governments around the world, and we have strong differences in outcomes. Whether those are causal we can debate, but we can’t claim no variance or experimentation.  While I don’t have the data and this is conjecture, what we see here in these data are likely one product of that variation.