Original Article – The New York Times | Does Moving Poor People Work?
Twenty years ago, federal poverty experts, inspired by the forceful arguments in the landmark book “The Truly Disadvantaged,” as well as by definitive research on the harmful effects of segregation, initiated a government experiment that moved 855 low-income predominantly African-American and Hispanic families out of public housing in poverty-stricken urban areas into less impoverished neighborhoods.
The results of the project have provoked an intense debate.
Under the aegis of the “Moving to Opportunity” program, begun during the first administration of Bill Clinton, the Department of Housing and Urban Development randomly selected a large pool of low-income families with children living in public housing in Baltimore, Boston, Chicago, Los Angeles and New York. Ninety-eight percent of the families were headed by women; 63 percent were black, 32 percent Hispanic, and 3 percent white; 26 percent were employed, 76 percent were receiving welfare, and families had an average income of $12,709 in 2009 dollars.
These families, 4604 of them, to be exact, were then divided into three groups. An experimental group of 1,819 families was offered “Section 8 rental assistance certificates or vouchers that they could use only in census tracts with 1990 poverty rates below 10 percent”; 855 accepted the offer and became part of the study. A second group of 1,346 families was offered more traditional “Section 8” rent subsidy vouchers that could be used in anyneighborhood; 848 accepted.
A control group composed of 1,439 families stayed in public housing and became part of the study. The purpose of the relocation initiative, according to Department of Housing and Urban Development, was to test the “long-term effects of access to low-poverty neighborhoods on the housing, employment and educational achievements of the assisted households.” Researchers also studied how relocation affected the health of those who accepted vouchers.
A paper published in the May 2013 issue of the American Economic Review, “Long-Term Neighborhood Effects on Low-Income Families: Evidence From Moving to Opportunity,” found that after 10 to 15 years, moving out of high-poverty public housing through the M.T.O. program showed mixed results.
There were some positive developments, according to the primary author of the paper, Jens Ludwig, a professor of economics at the University of Chicago and the project director for a final assessment of the M.T.O. program. Ludwig and his six co-authors found improvement in “several key adult mental and physical health outcomes.” These included significantly lowered risk of diabetes and obesity, as well an improved level of “subjective well-being.”
But the Ludwig study also found that “changing neighborhoods alone may not be sufficient to improve labor market or schooling outcomes for very disadvantaged families.” Ludwig reported that this particular form of assistance from HUD –a housing voucher that allowed recipients to move into a “low poverty” area – had “no consistent detectable impacts on adult economic self-sufficiency or children’s educational achievement outcomes, even for children who were too young to have enrolled in school at baseline.”
Ludwig reported similar findings in a follow-up essay published this week by Third Way, a Democratic think tank.
Some of the nation’s most prominent poverty researchers, including William Julius Wilson, a professor of sociology at Harvard and the author of “The Truly Disadvantaged,” consider that the design of the M.T.O. project was flawed, leading to unwarranted conclusions about the lack of improvement in employment and schooling.
Wilson pointed out in an email to The Times that the families in the study who left public housing moved into segregated neighborhoods nonetheless, far from employment opportunities and with equally bad schools – often the same schools. Social conditions were only marginally better than those they had left.
In addition, Wilson wrote, the adults in the program “had been exposed all their lives to the effects of severely concentrated disadvantage, and no matter how long they are followed in their new neighborhoods, the effects of those earlier years are not fully erased.”
Robert Sampson, a professor of sociology at Harvard, argued in a 2008 essay published in the American Journal of Sociology that the project should have been called “Moving to Inequality.”
Sampson pointed out in an email that many of the adults in the program had lived in extreme poverty for decades and that the children, who were on average 11 years old when they entered the program, had spent their early years living in adversity. “The result,” he wrote, “is that developmental effects are difficult if not impossible to study in the research design,” which does not reveal the “lagged effects of severe disadvantage.”
While the M.T.O. participants moved to neighborhoods with somewhat less poverty and crime, their new homes were by no means in flourishing sections of the city. Sampson produced a map of Chicago showing that the overwhelming majority of families moved to areas that still qualified as communities of “high concentrated disadvantage” based on a measure combining poverty rates, unemployment, welfare receipt, female-headed households, racial composition and density of children.
In a separate study, Heather Schwartz, a researcher at the RAND Corporation, reached conclusions more in line with Sampson’s and Wilson’s. Schwartz examined the performance of low-income, mostly minority students in Montgomery County, Md., an affluent majority-white suburb of Washington.
The county adopted policies dispersing public housing so that many of the tenants, who were 72 percent black and 16 percent Hispanic, were housed in middle-class, largely white apartment complexes.
This allowed Schwartz to measure the performance of children from public housing who attended schools with large numbers of well-off white students, against the performance of those who attended schools with largely minority populations and much higher poverty rates.he results are striking. The low-income minority children from public housing all started with similar math scores. But after seven years, those who went to schools where fewer than 20 percent of their classmates were poor shot ahead of those who went to schools where 20 to 80 percent of their classmates were poor. This difference in trajectories is shown in Figure 1, in which the green line tracks math scores for poor children (defined as those receiving “free and reduced-priced meals” – a.k.a. FARM recipients) in relatively affluent schools, and the red line tracks math scores for poor children attending schools with much higher percentages of fellow students receiving FARM assistance.
Early childhood stress affects the “developing architecture of the brain” in ways “that create a weak foundation for later learning, behavior and health.”
Looked at this way, the M.T.O. findings — that participants who were given vouchers for housing in low-poverty neighborhoods made no gains in employment and wage equality compared with those left behind in public housing and that their children showed no improvement in school performance — do not seem surprising.
For one thing, participants appear to have been given little or no support other than modest housing counseling. But the issue is deeper than that: Multigenerational poverty is self-evidently more than a question of housing. It is unlikely to yield to even the best-intentioned one-dimensional approach.
Multifactorial approaches may be more productive. Recent papers such as “The Legacy of Disadvantage: Multigenerational Neighborhood Effects on Cognitive Ability” and “Neighborhood Effects in Temporal Perspective: The Impact of Long-Term Exposure to Concentrated Disadvantage on High School Graduation” are part of a continuing research agenda looking more profoundly into the causes of the intergenerational transmission of disadvantage.
The criticism of the M.T.O. study (which is now complete) points to new avenues for exploration. Even though the interpretation of the results of the housing voucher program has become contentious and somewhat politicized, the debate itself has the potential to be constructive.
Significant change is possible, but more resources and more sophisticated research design will be a necessary next step.
Lines of possible future inquiry include evidence-based evaluations of total-immersion school systems like the KIPP program and a better understanding of the effects of poverty on brain development. Perhaps most importantly, in the debate over “neighborhoods or schools,” would be a concentrated focus on reducing racial and ethnic discrepancies in test scores, according to the economists Roland Fryer Jr. of Harvard and Steven Levitt of the University of Chicago. Fryer and Levitt argue that the elimination of “the test score gap that arises by the end of junior high school may be a critical component of reducing racial wage inequality.”
The two authors write, “we demonstrate that in stark contrast to earlier studies, the black-white test score gap among incoming kindergartners disappears when we control for a small number of covariates.” They add, “There is suggestive evidence that differences in school quality may be an important part of the explanation. None of the other hypotheses we test to explain why blacks are losing ground receive any empirical backing.”
We have to figure out a better way to approach intervention, whether it’s education-based or neighborhood-based or both. Otherwise how can we interrupt the intergenerational transmission of disadvantage we are only beginning to understand?