Everyone has theories for why well-educated, higher-income professionals are moving back into parts of cities shunned by their parents’ generation.
“There are all sorts of potential other amenities, whether it’s cafes, restaurants, bars, nicer parks, better schools,” said Ingrid Gould Ellen, a professor of urban policy and planning at New York University.
“But a huge piece of it,” she said, “I think is crime.”
New research that she has conducted alongside Keren Mertens Horn, an economist at the University of Massachusetts in Boston, and Davin Reed, a doctoral student at N.Y.U., finds that when violent crime falls sharply, wealthier and educated people are more likely to move into lower-income and predominantly minority urban neighborhoods.
Their working paper suggests that just as rising crime can drive people out of cities, falling crime has a comparable effect, spurring gentrification. And it highlights how, even if many Americans—including, by his own words, President-elect Donald Trump—inaccurately believe urban violence is soaring, the opposite long-term trend has brought wide-ranging change to cities.
The new research looked at confidential geocoded data from the 1990 and 2000 censuses, and more recent American Community Surveys, to identify the neighborhoods where more than four million households moved. Using citywide violent crime data from the F.B.I., the scholars tracked the changing probability of different demographic groups moving into central cities, as opposed to suburbs, as crime fell.
Higher-income and college-educated movers—and to a lesser degree, whites—appeared significantly more sensitive to changing crime levels in their housing decisions than other groups. Lower-income and minority households, for instance, didn’t become more likely to move to cities as they grew safer.
It’s entirely likely that the arrival of more affluent residents affected crime, too—either by increasing opportunities for property crime in the short term, or by adding eyes on the street and pressure on the police in the long run. Because this research looked at moves that occurred after crime was already falling, the authors believe the movers were reacting to changes in crime and not simply causing it themselves.