The release of an important study from the Equality of Opportunity Project at Harvard University showing that moving the urban poor to areas with jobs and other opportunities increases upward mobility probably received less coverage than it deserved.
For those of us in public policy, this new study is something like finding the Higgs-Boson particle in physics. Many believe a link exists between moving from a high-poverty community to a low-poverty place can break the dynamics of concentrated poverty. But we had no analysis to back that up.
From 1994 to1998, the Department of Housing and Urban Development conducted a demonstration called Moving to Opportunity for Fair Housing (MTO). Housing authorities in five cities — Baltimore, Boston, Chicago, Los Angeles, and New York — recruited 4,600 low-income families to participate in the program, which tested if moving the poor to high-quality neighborhoods makes a difference.
Using a random-assignment model, program participants were placed in one group offering housing counseling and housing vouchers that they could only use to move away from high-poverty neighborhoods to low-poverty ones. A second group of participants was offered Section 8 vouchers but with no stipulation that they move to a low-poverty community, nor did this group receive housing counseling. A control group was not given vouchers and continued to live in public housing.
Using a battery of outcome measures for children and adults, researchers initially found little impact on educational attainment or earnings more than 10 years after the demonstration concluded. As is the case in scholarly study, there were meaningful debates about the methodological framework for assessing impact, but no clear evidence that the demonstration had a significant impact.
It turns out that methodology was key in disguising the impact on children’s educational attainment and future earnings. Studies using the MTO data after it was first available lumped all children together. Using the same MTO demonstration data, Raj Chetty (an economist at Harvard University) and colleagues [link:http://www.equality-of-opportunity.org/images/mto_paper.pdf|
reanalyzed the data] and found that younger children (12 years or under) reaped great and significant benefits from moving to better neighborhoods. Younger children in this category had an increased probability of attending college and earning significantly more money than the control group.
The pattern was not found in older children (13 years or over). As one policy analyst writing in The New York Times said: “All told, this re-analysis transforms what was previously seen as influential evidence that neighborhoods are unimportant into the more nuanced finding that moving while young can be tremendously beneficial. Indeed, the net present value of the extra earnings that will eventually accrue to a child who moved at age eight is $99,000, meaning that for a family with two children, the program yields $198,000 in extra earnings. The extra tax that these two children will eventually pay is probably $22,000, more than enough to offset the extra cost of the voucher program, relative to the alternative of public housing.”
Wow! Here in New Jersey, this new analysis of the MTO data should give comfort to those who hoped that the Mt. Laurel court cases would lead to growing suburban communities in the state taking on a share of the urban poor located in their region. The map should remind us that segregation by race and income still persists in our state.
Communities such as East Orange, Irvington, and parts of Newark struggle with concentrated poverty. The shaded blue places represent concentrations of African-Americans who earn below 50 percent of Essex County’s median-family income. This is in stark contrast to communities as you move outward in Essex County, which are predominantly white and wealthy.
I feature the new re-analysis by Chetty for two reasons. For one thing, this is a great example of carefully done policy research directed toward an important question. The research allows us to revisit, with more than anecdotes or a hunch, the question of public policy encouraging the regional dispersal of the poor. Too often in policy research we cannot address big questions due to limited resources and time. Here, the process through which important research findings emerge from colleagues challenging each other and looking at the data in many different ways produced important knowledge. Random-assignment studies in policy questions are not applicable for every challenge, but we ought not dismiss the method as much as we do in state-level research on difficult social and economic challenges.
The other reason this study is timely is the continuing problem of concentrated poverty in many places, including New Jersey. The MTO demonstration was a response to the 1991 civil disturbances in Los Angeles. It is with some irony that Baltimore was one of the demonstration sites. We have learned much about the dynamics of poverty, and there is much more to learn. My hope is that we can actively use the reanalysis done by Chetty and the Equality of Opportunity Project to shift the policy dialogue from poverty to opportunity. We face continuing eruptions in places and among people that lack opportunity if we do not find the will to rethink the parameters of opportunity in the United States.
Let me be clear. Mobility strategies are not the only answer. Place-based strategies including education reform, community economic development, and workforce development must bear the largest share and focus when we consider antipoverty policies. But those policies face an unyielding barrier to success if urban places are segregated by wealth and race. As an advanced democracy, we should not deny that we are facing profound questions as to who gets to live the American Dream. It is only when we understand and learn that can we renew.