clock menu more-arrow no yes mobile

Filed under:

Want to help poor kids? Help their parents move to a better neighborhood.

Spencer Platt/Getty Images

You rarely see new middle-class parents respond to the expenses related to the birth of a child by moving to a high-poverty neighborhood where houses are cheaper. This is because, intuitively, people tend to believe that growing up in a "bad" neighborhood is bad for children. Based on this idea, the federal government even briefly ran a program that specifically gave poor families money to move to nicer neighborhoods.

Academic studies of this voucher program initially gave it a failing grade, finding that very minor benefits for children could be detected. Poverty, it seemed, was either baked into the fundamental structure of capitalism or else baked into the fundamental character of low-income families. Nothing so banal as a move to a lower-poverty neighborhood could change things.

Except now it seems this old conclusion was most likely mistaken.

That's according to a new study by Raj Chetty, Nathaniel Hendren, and Lawrence Katz and a companion study by Chetty and Hendren. These studies are the basis of an amazing interactive feature from the New York Times that interested parties can (and should!) while away hours playing with. But the basic finding of the research — moving poor families with young children to less-poor neighborhoods would produce huge gains for the children when they grow up — shouldn't get lost in the fog.

The old CW was that intuition about neighborhoods was wrong

Both normal people and academic sociologists drawn to a qualitative approach have long believed that growing up in a high poverty neighborhood full of struggling adults has deleterious impacts on the life outcomes of children. But among economists, the conventional wisdom has recently been that this is wrong. That's because in the mid-1990s the federal government was inspired to create the Moving to Opportunity Program which gave 4,600 randomly selected families living in public housing money to move elsewhere.

The lottery design made it possible to do very high-quality research, and the research showed that the program didn't work. Lottery winners had incomes no higher than lottery losers, and while winners had somewhat better mental health, their children ended up about the same as the children of lottery losers.

A more careful study shows a big boost to young kids

The Chetty, Hendren, and Katz study looks back at this old research and finds that the conclusion was too hasty.

If you disaggregate younger kids from older kids, it turns out that "moving to a lower-poverty neighborhood significantly improves college attendance rates and earnings for children who were young (below age 13) when their families moved." The reason the old study found no positive impact is that "perhaps because of disruption effects," kids who moved as teenagers seem to have ended up worse off.

In part due to the higher educational attainment, young lottery winners from the 1990s now have incomes that are 31 percent higher than those of young lottery losers. Lottery winners are less likely to live in high-poverty neighborhoods as adults, and less likely to be single parents. These findings suggest that moving to opportunity, if targeted at appropriately young families, can seriously disrupt the cycle of intergenerational poverty.

A bigger study reveals the best and worst counties for poor people

The great thing about the Moving to Opportunity study is that its lottery-based nature allows for extremely high-quality experimental evidence. The bad thing about it is that since there were only 4,600 lottery winners, the actual amount of number-crunching you can do with it is limited.

That's where the second study by Chetty and Hendrin comes in.

They used a giant set of tax data to take a "quasi-experimental" look at the impact of moving from one county to another, based on tax returns from 5 million children whose families moved between counties from 1996 to 2012.

This kind of survey isn't quite as ironclad as a lottery, but it's pretty strong evidence and backs the basic conclusions of the randomized study. Even better, the data set is so large and rich that it lets them assign values to every county in America. For example, a poor child who grows up in Brooklyn will earn $1,500 less at age 26 than the average poor child. By contrast, a poor child who grows up in King County, Washington (the Seattle area), will earn $2,960 more than the average American.

We don't know what makes the "bad" neighborhoods bad

Of course, even better than moving people from one neighborhood to another would be to actually fix whatever it is that's wrong with the neighborhoods they are already in. Unfortunately, these studies don't really tell us what that is.

Could it be bad schools? The stress of living in high-crime areas? Lack of local job opportunities or positive role models? The presence of lead and atmospheric toxins?

We can't quite tell, though we separately have evidence to suggest that all play a role.