The Long-Term Benefit of Gentrifying Cities

One popular strain of urban-policy thinking opposes gentrification—the arrival of affluent people into poor neighborhoods—and argues that poverty should be rectified by ever greater expenditure on public housing. The opposite might be true: Government spending can help, but it can also hurt, as badly designed public-housing projects have done. So long as gentrification brings rich and poor together, and offers the latter greater opportunity to take part in a healthy economy, it looks less like a villainous process and more like a heroic one.

Few places illustrate the aspirations and failures of American housing policy as well as the Techwood Homes in downtown Atlanta—one of the first federal housing projects. Its completion, in 1935, even drew the attendance of Franklin D. Roosevelt, who switched on its electricity. To make way for the development, the old slums—in which roughly a quarter of residents were Black—had been cleared away. But the 604 new units were for white tenants only until 1968, when civil-rights laws forced integration. Like the ramshackle shacks it replaced, Techwood fell into disrepair. By the 1990s, Techwood had resegregated, becoming almost exclusively Black, and turned into a byword in Atlanta for urban decline. Gates and windows lay shattered; residents complained of squalid living conditions; drug trafficking and gang violence were out of control.

[Judd Kessler: The hidden cost of ‘affordable housing’]

In 1993, Atlanta received one of the first grants awarded by the federal HOPE VI program—which aimed to knock down the most decrepit public-housing projects in America and replace them with better housing—to demolish and rebuild the Techwood Homes. The demolitions took place just before the city hosted the 1996 Summer Olympics. If you went to Techwood’s site today—sandwiched between Georgia Tech and the Coca-Cola museum—you would see a commemorative plaque but almost none of the original brick buildings that fell into dilapidation. Instead, you would find yourself in Centennial Place, a mixed-income community with subsidized apartments alongside private, market-rate housing. It was intentionally designed to reduce the isolation of the urban poor—and it’s succeeding.

A new, rigorous study of 200 HOPE VI sites, including Techwood, shows that the redevelopment significantly improved the lives of children. The reasons reveal a crucial fact about economic opportunity: To have social mobility, you need social integration. “Just giving people cash, just giving people education, doesn’t do as much as if you pair it with connections that then help them,” Raj Chetty, a Harvard economist and one of the study’s authors, told me.

From 1993 to 2010, HOPE VI spent $17 billion to knock down and remake projects. The program was controversial from the outset. The original residents did not have a right to return to the townhomes and other smaller structures that replaced the demolished larger complexes, and ultimately only 28 percent of residents came back. There were fewer units in the lower-density replacements. An influential 2002 report by the National Housing Law Project and other groups titled “False HOPE” argued that the new mixed-income model was “a social engineering scheme built on a number of inaccurate, irrelevant, and harmful assumptions about low income families and their neighborhoods.” These critiques were made during a time of growing revulsion against slum clearance and heavy-handed urban-renewal attempts; the reentry of the creative classes to city centers was only beginning to gain notice. Activists and some academic critics derided HOPE VI as a state-sponsored gentrification program, doomed to harm the people it was intended to help.

But the new study, “Creating High-Opportunity Neighborhoods: Evidence from the HOPE VI Program,” found nothing of the sort. Chetty and six colleagues at Harvard, Cornell University, and the Census Bureau used tax-return data to track outcomes for residents decades after they lived in neighborhoods changed by the HOPE VI experiment. The benefits to children living in the new low-density housing project are considerable: Their earnings in adulthood increased by 2.8 percent for every year they lived in the new developments instead of the old ones, the researchers calculated. This holds even when the researchers compare siblings within the same family. Overall, children whose families moved into revitalized units earn 16 percent more than they otherwise would have earned, they are 17 percent more likely to attend college, and boys are 20 percent less likely to go to jail in adulthood. The future-income increases alone greatly exceed the up-front cost of rebuilding, the authors argue.

The paper contends that the HOPE VI program delivered such significant benefits to children because the demolitions and reconstructions “increased friendships between children from low- and high-income families in high schools near public housing sites.” (In this case, cross-class friendship is measured empirically through proprietary data from Facebook.) “Distressed public housing projects were essentially islands that had limited social interaction with nearby communities,” the paper argues. “The HOPE VI program built a bridge to surrounding communities, allowing public housing residents to benefit from interacting with those residents.” The exact reason is still being worked out: It could be straightforward—that high-achieving peers boost education for others—or more subtle (for example, that closer contact with surrounding areas yields more introductions to parents who can provide job referrals). But integration does seem to matter a great deal. The researchers’ results show that simply living in the newly constructed projects was insufficient. Children who moved into these revitalized neighborhoods but experienced no increase in cross-class friendships saw essentially no benefit.

The study does not focus on children displaced by the demolitions. But Matthew Staiger, one of the paper’s co-authors, has separate research showing that they too went on to have markedly higher earnings. For adults, however, the findings are more mixed. Although neighborhoods became considerably richer after the HOPE VI revitalizations—household incomes increased by 45 percent, and poverty rates dropped by 12 percentage points—this is due entirely to richer adults moving in. The residents of the original projects, most of whom are scattered to other neighborhoods, are no better off in terms of income even years later. One recurring finding in social-mobility research is that interventions targeted to poor children yield significant results; for adults who are already in poverty, improvement is harder to attain even with expensive interventions.

Our understanding of what kinds of Americans experience upward mobility—and which ones don’t—has improved immensely over the past 10 years. Much of that progress is due to the work of Chetty, whose use of administrative data collected by government agencies has provided granular answers to questions sociologists and economists have debated for decades. These papers, often co-written with prominent economists such as Nathaniel Hendren, John Friedman, and Lawrence Katz, point toward a cohesive set of findings: Whatever their parents’ circumstances, the kind of neighborhood children grow up in substantially affects their life outcome, for better or for worse. Economists are used to explaining life outcomes as a result of financial or human capital, but Chetty says the cumulative research shows that social capital is just as important.

This lesson is especially stark when applied to childhood poverty. It is unsurprising that poor material circumstances at birth predict poverty in adulthood. Less obvious is the fact that poor children living in concentrated poverty—like those in infamous superblock towers—face worse outcomes than poor children who live near wealthier peers. We know this from studies following the life trajectory of children living in poor, segregated neighborhoods after they were moved by government programs—as in the Gautreaux Assisted Housing Program from the 1970s to the 1990s and the Moving to Opportunity (MTO) experiment in the 1990s. The MTO results, written by Chetty, Hendren, and Katz and published in 2016, stunned the economics field after finding that young children whose families left high-poverty census tracts after receiving housing vouchers went on to have 31 percent higher incomes in adulthood (alongside improvements in college attendance and reduced rates of single parenthood). The HOPE VI results show the converse of the MTO experiment to be true too: Kids don’t necessarily have to change locations to improve their life outcome—neighborhoods can be made to improve around them.

These results in economics ultimately vindicate foundational ideas in sociology—developed by scholars such as William Julius Wilson and Robert Sampson—that the concentration of disadvantage and social isolation worsen the effects of material disadvantage. The findings also raise an immediate question: If the primary reason that HOPE VI improved outcomes was that it boosted social capital for disadvantaged children, how can that positive intervention be replicated elsewhere? Laura Tach, a Cornell sociologist and co-author of the paper, told me that the HOPE VI program could have boosted social integration through several mechanisms: The demolition of towers (which in many cases were packed together in huge superblocks) made the outside world physically easier to access, reductions in violence and crime made outside connection psychologically easier, and complimentary community-support services for job training and after-school programs may have directly brought people of different social backgrounds together.

[Jerusalem Demsas: The real villain in the gentrification story]

There is another process that improves neighborhoods around poor children, both by bringing higher-income peers nearer to them and by reducing the violence they are exposed to. This process often occurs without explicit governmental intervention or cost. The problem is that it is regularly dismissed as gentrification, a phenomenon that is not usually cheered. The most common objection to gentrification is that it results in displacement of incumbent residents. The empirical evidence for this is weaker than conventionally assumed. One paper examining children who received Medicaid benefits in New York City from 2009 to 2015 found no elevated rates of moving for those in gentrifying neighborhoods. The HOPE VI study suggests that gentrification should improve outcomes for kids, so long as it actually improves social integration.

This is not guaranteed, certainly, but is perhaps more likely to happen than further governmental action. HOPE VI remade some of America’s worst public-housing projects for the better. But it also cost the government $170,000 per unit (in 2022 dollars). The Trump administration has called for a 43 percent reduction in public-housing spending. It is especially hostile to the idea of using federal funds as an explicit tool to break up concentrations of poverty. The lessons of HOPE VI provide a tantalizing clue about how social mobility can be engineered in America by building bridges between rich and poor. How unfortunate it is that the current administration is unlikely to even try.