Source: Chetty, Hendren, Kline and Saez, "The Economic Impacts of Tax Expenditures: Evidence from Spatial Variation Across the U.S."
Source: Chetty, Hendren, Kline and Saez, "The Economic Impacts of Tax Expenditures: Evidence from Spatial Variation Across the U.S."

In Atlanta, children born into the bottom fifth of the national-income distribution have just a 4 percent chance of rising to the top fifth. That's a fraction of the shot they would have if they lived in another city: The odds would be one in 10 if they lived in New York, Los Angeles or Boston.

This comes from a terrific piece in yesterday's New York Times by David Leonhardt, which showed how some areas of the U.S. have strikingly low income mobility. The new research Leonhardt draws on also has interesting answer -- and one that contradicts much prior research -- to a different question: Does moving to a better neighborhood make a difference for poor children?

A number of earlier studies that came out of a public-policy experiment -- the "Moving to Opportunity" program run by the Department of Housing and Urban Development -- had said no. The new study says yes, moving might be able to help.

Here's the key graph:

The graph shows that the older a child is at the time of moving to a better neighborhood, the smaller the effect on his or her future success. By age 25, moving doesn't make a significant difference. But at age 10, a move has 60 percent of the impact of being born in a better neighborhood.

But wait: What if the families that moved out of bad neighborhoods earlier were just better at raising children? Isn't moving itself a sign of good parenting? The researchers anticipated this sort of issue by looking at families with multiple children. And, indeed, length of exposure trumps family: If one sibling gets 10 more years in a good neighborhood, that sibling does much better. (See page 33 of the study for related graphs.)

If a child's chance at having a better life depends on where he or she is raised, then that reinvigorates the case for public policy to promote income mobility and equality of opportunity across generations. It makes us wonder whether the government should be helping poor families physically move toward opportunity.

That's been tried before: From 1996 to 2006, the federal Moving to Opportunity program provided rent subsidies for several thousand poor families to move from poor, inner-city neighborhoods into areas with less poverty. It was a trial program, and the results were underwhelming.

Moving didn't improve the success of poor children in school. It didn't improve their physical health and had mixed impacts on mental health. It seemed to encourage crime from boys, though generally the program seemed to help girls. (Girls appeared to acclimate to the new environment, while boys rebelled.) For parents, moving to a better neighborhood didn't make them any more likely to get a job or get off welfare.

These findings led researchers to believe that it wasn't the place, it was the family -- a dispiriting conclusion that undermined the case for housing policies and challenged the idea that government can help give poor kids a real chance at success.

The Moving to Opportunity research isn't in direct contradiction with the new work. With many of the moved children only in their 20s, we don't yet know whether they'll have higher-income careers. The earlier findings had suggested they would not. Now we have a reason to think such programs can make a difference in the lives of poor families. That's worth celebrating.

(Evan Soltas is a contributor to the Ticker. Follow him on Twitter.)