A vast new national study showing how income mobility dramatically varies by city.
We know that a child's chances of escaping poverty are highly dependent on factors beyond his control – his parents' income, their educational background, the school he attends, the demographics of his classmates, the characteristics of the neighborhood outside his door. But it turns out that a child's prospects for upward mobility are also highly influenced by the most basic question of geography: Depending on which city his parents raised him in, he has dramatically different odds of growing up to be someone who isn't poor.
Academics from Harvard and UC Berkeley recently set out to study if local tax expenditures (and particularly tax credits for the poor) in fact help lift children out of poverty, promoting intergenerational mobility. They figured they could better understand the benefits of tax policy by leveraging the variety of ways that it differs across the country. Their massive study – "the most detailed portrait yet of income mobility in the United States," as The New York Times put it this morning – spanned 740 "commuting zones" across the U.S., roughly analogous to metropolitan areas but covering the entire country. And they examined an entire cohort of American children born in 1980 and 1981, whose parents' income and later adult income was calculated using millions of anonymous earnings records.
The researchers found that larger tax credits for the poor do correlate with higher income mobility. But their research more strikingly turned up "substantial variation" across the country in a child's prospects of escaping poverty, and for reasons that go likely well beyond local tax policy. This map from the paper illustrates inter-generational mobility, comparing a child's adult income rank with his or parents' income rank (defined by income centiles). The darker the areas, the lower the intergenerational mobility:
The researchers, Raj Chetty and Nathaniel Hendren of Harvard, and Patrick Kline and Emmanuel Saez of UC Berkeley, have also released their full dataset online, and the Times has built a powerful set of interactive tools with it. Regionally, the research shows that lower-income children living in cities in the Northeast, Upper Midwest and West Coast (New York, Minneapolis, San Francisco) stand a much better chance of reaching higher income brackets in adulthood than children living in Southeastern cities and the industrial Midwest (Atlanta, Memphis, Indianapolis).
More specifically, this means that a poor child in Seattle, raised by a family in the bottom fifth of national family income, has about a 10.4 percent chance of becoming an earner in the top fifth of national income by the time he's 30. For a child in Atlanta, that figure is only 4 percent.
In Charlotte, North Carolina, a child whose parents make just $16,000 a year, or in the 10th percentile nationally, grows up on average to earn in the 30th percentile. For a child growing up with similar means in Salt Lake City, that adult income is in the 43rd percentile. And 11 percent of those low-income Salt Lake City children will go on to become top earners (in the top fifth of the U.S. population). For Charlotte children, the comparable figure is just 4 percent.
What separates all of these places is more than just the strength of the local economy or the generosity of the tax benefits. As the authors write in the executive summary:
We do find higher rates of upward income mobility in areas with high rates of economic growth over the past decade, but the vast majority of the difference in mobility across areas is unrelated to economic growth.
So what is it related to? The researchers identify several factors with strong correlations: school quality, social capital and family structure (metros with more two-parent households seemed to have higher income mobility). They also found that low inequality and economic and racial segregation across metro areas were tied to higher income mobility, an idea we've written about before. Segregation hinders literal mobility – the ability of poor people to cross town to reach jobs and other opportunities.
This implies that the geography of cities themselves – how far out the jobs are, where people live relative to them, whether there's good transportation connecting the two – may have much to do with these nationwide patterns in the geography of upward mobility. You may have long suspected this to be true. But now here is some strong but depressing data to back this up: The kind of city a child grows up in may well determine the kind of life he goes on to live.