Big Benefits from Moving to a Better Neighborhood
The importance of neighborhoods in shaping opportunity and life outcomes is well-documented. As such, it’s not surprising that studies have found that moving to a better neighborhood is generally beneficial for children. However, a new paper by Eric Chyn of the University of Michigan suggests we may have been vastly underestimating just how impactful relocation can be. Spotlight recently talked with Chyn to understand what’s distinctive about his research and what it means for public policy. The conversation has been lightly edited for length and clarity.
Before we get into your work, can you talk a little about the existing research on the impact of low-income families relocating to better neighborhoods?
There is a broad literature looking at the impact of growing up in a disadvantaged area with several studies looking at how using housing vouchers to relocate families influences life prospects. In particular, there are two important relocation programs that have been studied.
A first set of studies looks at Chicago’s Gautreaux mobility program which provided housing vouchers to several thousand low-income households. Some families used these vouchers to move to suburban areas while others relocated to more disadvantaged urban areas. Research showed that those who moved to the lower-poverty suburbs had significantly better outcomes than those who moved to urban areas. Policymakers and academics took notice of this and thought, “Ok, maybe vouchers are a good way to change neighborhood conditions and create better outcomes.”
Another set of studies analyzed the Moving to Opportunity (MTO) program, which, unlike Gautreaux, was a randomized experiment. Beginning in the 1990s, the program recruited families living in public housing to join the experiment and have a chance to win a voucher. Long-run analysis by Raj Chetty and two coauthors was published last year and shows that children who were young when they moved through MTO benefited from relocation in terms of adult labor market outcomes. Additional analyses shows there were not similar benefits for older children and adults.
Your paper finds even larger benefits for this relocation than the existing literature. Talk me through your work and how it differs?
One thing that struck me was that 25 percent of families eligible for the experiment participated. I also read an MTO report that interviewed participants and revealed that these parents were particularly motivated to protect their kids from neighborhood crime. This implied that the control group parents who volunteered for the lottery but didn’t end up moving may have been especially proactive in trying to ameliorate the harmful effects of a high-poverty environment. Overall, this caused me to wonder whether the effect of relocation might be different for a broader population of public housing residents.
To understand the effects of vouchers and moving for a more general population, I studied households forced to relocate due to public housing demolitions that occurred in Chicago. These demolitions were unique because the housing authority wanted to remove all high-rise public housing in the city, but they couldn’t tear everything down because there were too many buildings. Instead, they chose some buildings to destroy and some to leave intact, often for idiosyncratic reasons such as a pipe breaking in the middle of winter that caused the entire building heating system to shut down.
This context sets up a perfect “natural experiment” to study the impact of vouchers. CHA provided displaced residents with housing vouchers to replace their project-based housing, while residents in nearby buildings that remained intact did not receive these vouchers. Hence, I compare the outcomes for displaced individuals with the outcomes for those who were not displaced to estimate the impact of moving. There is no selective participation because individuals do not volunteer to move – the housing authority chooses who moves when it decides building destruction. This lets me examine whether the average public housing child – including those whose household might not be motivated to move – would benefit.
And what did you find?
The big headline findings are that the children who were displaced by the demolitions went on to have notably better adult labor market outcomes than those not forced to move. Specifically, they are 9 percent more likely to be working and have 16 percent higher annual earnings as adults.
These results differ from previous work because I find that all children, regardless of their age when their familes moved, benefited. As I mentioned, the most recent long-run analysis of MTO finds that only children who moved when they were young had better outcomes.
So what about this change in neighborhoods is having such a large effect on children?
The two ideas that are often brought up are schools and crime.
If you look at the Moving to Opportunity study and previous work on public housing demolitions by Brian Jacob, you see that these public housing families aren’t moving that far and the children aren’t attending high-quality schools. So, you don’t see a lot of effects on measures of schooling outcomes like test scores.
However, the other way that neighborhood relocation might affect outcomes for disadvantaged children is by limiting exposure to neighbors and peers engaged in crime and other risky behavior. So, in addition to labor outcomes, I analyzed whether there were effects on criminal behavior, finding that displaced kids have fewer arrests for violent crime in the long run.
There seems to be pretty clear benefits to relocation? Are there also costs associated with the demolition of housing projects?
As an economist, you always want to know the total cost and benefit. I am talking about just one part of the whole: the benefits for those who are relocated. In terms of costs, we have limited research on the effects of relocated public housing residents on their new neighbors. Also, the demolition could have imposed costs on residents of buildings that were not selected for demolition. I tried to understand this by looking at crime rates in the areas where demolitions occurred, finding no statistically significant effects in my sample of demolitions during the mid-1990s. If anything, the data provides suggestive evidence that crime decreased. So, things may have been getting better for residents in intact buildings, and at the very least crime didn’t seem to have gotten worse.
What are the policy implications of your findings?
My research supports the argument that it’s better to provide housing assistance through vouchers rather than project-based public housing. I wouldn’t say that my results apply to all public housing throughout the U.S., but to the extent that some public housing is located in high-crime, low-income environments, my research suggests providing vouchers is a good way to generate better outcomes for children. It’s important to note that this conclusion is consistent with the recent work by Raj Chetty and coauthors showing that MTO vouchers generated long-run benefits for young children.
Eric Chyn is a PhD Candidate in the Department of Economics at the University of Michigan. He will be joining the Department of Economics at the University of Virginia next year.
The views expressed in this commentary are those of the author or authors alone, and not those of Spotlight. Spotlight is a non-partisan initiative, and Spotlight’s commentary section includes diverse perspectives on poverty. If you have a question about a commentary, please don’t hesitate to contact us at firstname.lastname@example.org.
You can also sign up to receive our weekly newsletter and other updates here.