Massachusetts Institute of Technology: A Simple Intervention that Empowers Low-Income Families to Overcome the Poverty Trap
Many low-income families might desire to move into different neighborhoods — places that are safer, quieter, or have more resources in their schools. In fact, not many do relocate. But it turns out they are far more likely to move when someone is on hand to help them do it.
That’s the outcome of a high-profile experiment by a research team including MIT economists, which shows that a modest amount of logistical assistance dramatically increases the likelihood that low-income families will move into neighborhoods providing better economic opportunity.
The randomized field experiment, set in the Seattle area, showed the number of families using vouchers for new housing jumped from 15 percent to 53 percent when they had more information, some financial support, and, most of all, a “navigator” who helped them address logistical challenges.
“The question we were after is really what drives residential segregation,” says Nathaniel Hendren, an MIT economist and co-author of the paper detailing the results. “Is it due to preferences people have, due to having family or jobs close by? Or are there constraints on the search process that make it difficult to move?” As the study clearly shows, he says, “Just pairing people with [navigators] broke down search barriers and created dramatic changes in where they chose to live. This was really just a very deep need in the search process.”
The study’s results have prompted U.S. Congress to twice allocate $25 million in funds allowing eight other U.S. cities to run their own versions of the experiment and measure the impact.
That is partly because the result “represented a bigger treatment effect than any of us had really ever seen,” says Christopher Palmer, an MIT economist and a co-author of the paper. “We spend a little bit of money to help people take down the barriers to moving to these places, and they are happy to do it.”
Having attracted attention when the top-line numbers were first aired in 2019, the study is now in its final form as a peer-reviewed paper, “Creating Moves to Opportunity: Experimental Evidence on Barriers to Neighborhood Choice,” published in this month’s issue of the American Economic Review.
The authors are Peter Bergman, an associate professor at the University of Texas at Austin; Raj Chetty, a professor at Harvard University; Stefanie DeLuca, a professor at Johns Hopkins University; Hendren, a professor in MIT’s Department of Economics; Lawrence F. Katz, a professor at Harvard University; and Palmer, an associate professor in the MIT Sloan School of Management.
New research renews an idea
The study follows other prominent work about the geography of economic mobility. In 2018, Chetty and Hendren released an “Opportunity Atlas” of the U.S., a comprehensive national study showing that, other things being equal, some areas provide greater long-term economic mobility for people who grow up there. The project brought renewed attention to the influence of place on economic outcomes.
The Seattle experiment also follows a 1990s federal government program called Moving to Opportunity, a test in five U.S. cities helping families seek new neighborhoods. That intervention had mixed results: Participants who moved reported better mental health, but there was no apparent change in income levels.
Still, in light of the Opportunity Atlas data, the scholars decided revisit the concept, with a program they call Creating Moves to Opportunity (CMTO). This provides housing vouchers along with a bundle of other things: Short-term financial assistance of about $1,000 on average, more information, and the assistance of a “navigator,” a caseworker who would help troubleshoot issues that families encountered.
The experiment was implemented by the Seattle and King County Housing Authorities, along with MDRC, a nonprofit policy research organization, and J-PAL North America. The latter is one of the arms of the MIT-based Abdul Latif Jameel Poverty Action Lab (J-PAL), a leading center promoting randomized, controlled trials in the social sciences.
The experiment had 712 families in it, and two phases. In the first, all participants were issued housing vouchers worth a little more than $1,500 per month on average, and divided into treatment and control groups. Families in the treatment group also received the CMTO bundle of services, including the navigator.
In this phase, lasting from 2018 to 2019, 53 percent of families in the treatment group used the housing vouchers, while only 15 percent of those in the control group used the vouchers. Families who moved dispersed to 46 different neighborhoods, defined by U.S. Census Bureau tracts, meaning they were not just shifting en masse from one location to one other.
Families who moved were very likely to want to renew their leases, and expressed satisfaction with their new neighborhoods. All told, the program cost about $2,670 per family. Additional research scholars in the group have conducted about changes in income suggest the program’s direct benefits are 2.5 times greater than its costs.
“Our sense is that’s a pretty reasonable return for the money compared to other strategies we have to combat intergenerational poverty,” Hendren says.
Logistical and emotional support
In the second phase of the experiment, lasting from 2019 to 2020, families in a treatment group received individual components of the CMTO support, while the control group again only received the housing vouchers. This way, the researchers could see which parts of the program made the biggest difference. The vast majority of the impact, it turned out, came from receiving the full set of services, especially the “customized” help of navigators.
“What came out of the phase two results was that the customized search assistance was just invaluable to people,” Palmer says. “The barriers are so heterogenous across families.” Some people might have trouble understanding lease terms; others might want guidance about schools; still others might have no experience renting a moving truck.
The research turned up a related phenomenon: In 251 follow-up interviews, families often emphasized that the navigators mattered partly because moving is so stressful.
“When we interviewed people and asked them what was so valuable about that, they said things like, ‘Emotional support,’” Palmer observes. He notes that many families participating in the program are “in distress,” facing serious problems such as the potential for homelessness.
Moving the experiment to other cities
The researchers say they welcome the opportunity to see how the Creating Moves to Opportunity program, or at least localized replications of it, might fare in other places. Congress allocated $25 million in 2019, and then again in 2022, so the program could be tried out in eight metro areas: Cleveland, Los Angeles, Minneapolis, Nashville, New Orleans, New York City, Pittsburgh, and Rochester. With the Covid-19 pandemic having slowed the process, officials in those places are still examing the outcomes.
“It’s thrilling to us that Congress has appropriated money to try this program in different cities, so we can verify it wasn’t just that we had really magical and dedicated family navigators in Seattle,” Palmer says. “That would be really useful to test and know.”
Seattle might feature a few particularities that helped the program succeed. As a newer city than many metro areas, it may contain fewer social roadblocks to moving across neighborhoods, for instance.
“It’s conceivable that in Seattle, the barriers for moving to opportunity are more solvable than they might be somewhere else.” Palmer says. “That’s [one reason] to test it in other places.”
Still, the Seattle experiment might translate well even in cities considered to have entrenched neighborhood boundaries and racial divisions. Some of the project’s elements extend earlier work applied in the Baltimore Housing Mobility Program, a voucher plan run by the Baltimore Regional Housing Partnership. In Seattle, though, the researchers were able to rigorously test the program as a field experiment, one reason it has seemed viable to try replicate it elsewhere.
“The generalizable lesson is there’s not a deep-seated preference for staying put that’s driving residential segregation,” Hendren says. “I think that’s important to take away from this. Is this the right policy to fight residential segregation? That’s an open question, and we’ll see if this kind of approach generalizes to other cities.”
The research was supported by the Bill and Melinda Gates Foundation, the Chan-Zuckerberg Initiative, the Surgo Foundation, the William T. Grant Foundation, and Harvard University.