Lawrence Katz, from the Harvard economics department, was here yesterday to give the Institute for Research on Poverty’s annual Lampman lecture. Katz’s talk, “Neighbohoods vs. Schools,” wasn’t what I’d guessed it’d be about,1 but still posed an interesting question about economic geography.2 Using data from the Moving to Opportunity project, Katz was trying to isolate the effects of moving to a higher-quality neighborhood on economic security as well as personal well-being, without including the benefits of moving to a better school district.
The MTO project, launched in 1992 in several large cities, offered a lottery-based housing voucher to families living in the highest-poverty areas of the cities in question, with the stipulation that they must use the vouchers to move into a neighborhood with less than 40% poverty. In most cases, this meant moving from a neighborhood absolutely mired in poverty to a slightly more stable lower-middle-class neighborhood. Because the school districts in these new neighborhoods were not much better than the original school districts, this program provided a natural experiment in which to measure the raw effects of living in new environmental conditions.
What Katz and his colleagues found was that economic security didn’t change at all: the movers were just as likely to have found employment as those who stayed in the high-poverty neighborhoods. What did show a significant change, however, were measures of health and well-being. Girls in particular scored much higher in several quality-of-life indices once they had moved to a new neighborhood.
It’s a fascinating finding in urban geography, but I was surprised to find that Katz’s only statistical information about the new neighborhoods was demographic. That is, the only data he used about the moved-to neighborhoods versus the moved-from neighborhoods was their raw poverty rate, crime rate, and the test scores of the school district. This, I think, misses what might be the primary causal mechanism: the “public environment” of impoverished versus stable neighborhoods.
What I would have wanted to include in the data would have been geographical data about the moved-to versus the moved-from neighborhoods—things like size and quality of public parks, number of playgrounds, age and quality of housing stock, accessibility to transit, and environmental indicators such as air pollution. These sorts of measures, after all, represent the geographic “trickle-down” mechanism by which the presence of economically stable families is made material in a way that benefits the entire neighborhood. If you live next door to a rich person who leaves their locked home every day in a chauffeured car to his job ten miles away but invests nothing in the community, then it is hard to imagine what benefit you receive from living in the “same” neighborhood as him. If, by contrast, the upper- and middle-class residents of a neighborhood invest in (and demand from the government) the kinds of high-quality public services which are available to all, then the poor co-residents of that neighborhood really will benefit.
Using this kind of data, Katz’s study then could have tried to determine how much better the moved-to neighborhoods were in their provision of these kinds of public amenities than the moved-from neighborhoods. If this turned out to be an explanatory variable, it could be suggestive as to the mechanisms through which “better” neighborhoods lift the fortunes of everybody living in their borders.