Researcher Roland Sturm of the Rand Corporation throws cold water on the notion that so-called food deserts have a role in the obesity crisis.

Reported by Ezra Klein on his Washington Post Wonkblog, Sturm concludes that everyone drives everywhere anyway, so what difference does it make what's nearest by? Especially given that his study is of 13,000 adolescents in California, there's some merit to the notion. But I always understood the notion of food deserts to be in poor urban areas, where fewer residents have access to private transportation and are therefore more dependent on what's available in walking distance, or bus distance. Also, that the walking alternatives are more likely to be small stores, with less — and less appealing — produce for sale, making less healthy processed foods a greater proportion of such residents' food universes.

The money graf, for me, is Klein's next to last: "Sturm doesn’t disagree with the idea that supermarkets can benefit a neighborhood: They provide more food variety and more options for eating healthily. But he cautions against turning to a supermarket expansion as a way to address American obesity."

OK, got it. Having a supermarket nearby IS better after all. But supermarket expansion isn't a way to address American obesity. 

Well, I can't say I'm convinced by the argument "everyone drives, anyway" (my paraphrase), but I never thought supermarket expansion was a top-10 or -20 measure anyway. There's plenty we have to do before we underwrite the construction of supermarkets in areas that haven't otherwise shown the ability to sustain them financially, on the belief that then, people who've had a lifetime of poor nutritional options will start buying fruit and vegetables.