Across the United States, about 15 percent of city land is vacant or abandoned. These areas can foster criminal activity and make urban residents, particularly in low-income neighborhoods, feel unsafe. Findings published in the Proceedings of the National Academy of Sciences from Penn and other institutions revealed that remediating these spaces can dramatically affect both perceptions of crime and vandalism and the acts themselves.
Neighborhoods where vacant lots were cleaned up experienced a 29 percent reduction in gun violence, 22 percent decrease in burglaries, and 30 percent drop in nuisances like noise complaints and illegal dumping.
Residents living near those same spaces also reported feeling much safer post-remediation, with 58 percent having fewer security concerns when leaving their homes and more than three-quarters saying they significantly increased use of their outside spaces for relaxing and socializing.
“Our analyses showed a substantial reduction in crime,” says Penn criminologist John MacDonald, the senior paper author, “particularly, a large reduction in gun assaults in neighborhoods that were in the lower 50 percent of income distribution in Philadelphia.”
The study, led by Charles Branas, an epidemiologist at Columbia University, is believed to be the first randomized controlled trial of its type.
To analyze the relationship between restoring vacant plots and crime in the city, the researchers randomly assigned 541 lots to receive extensive, minimal, or no restoration (the control). At the experiment’s end, every plot received trash and debris removal, land grading, planting of new grass, and lot maintenance.
Through police reports filed 18 months leading up to and following the restoration, the researchers then looked at applicable crime data. To understand resident experiences and remediation effect on the land, the team conducted repeated interviews with 445 randomly sampled residents living near the lots and also placed anthropologists in two neighborhoods.
“There is a lot of descriptive and theoretical research in criminology suggesting you can substantially affect crime by changing conditions of the built environment and making places less attractive to crime,” MacDonald says. “But we have very few scientific studies that have experiments on a scale like this.”
The results hold promise that quick, inexpensive tactics can substantially improve neighborhoods. The strategy, chosen specifically to improve local conditions block by block, yields a high return on investment and doesn’t displace long-term residents, like some other costlier methods unintentionally do. It also has the potential to improve people’s health and well-being, says Eugenia South, an emergency-medicine physician at Penn Medicine.
“Feeling unsafe when you leave your home contributes to the experience of chronic stress in low-resource neighborhoods, which ultimately contributes to poor health and persistent health disparities,” she says. “The fact that participants felt safer after cleaning and greening of vacant lots is an important and exciting result.”
The intervention, MacDonald adds, could be scalable across the United States. “What we’re seeing in Philadelphia is just a microcosm of what you see in legacy cities from Milwaukee to Youngstown, Ohio, to Scranton,” he says. “This suggests that one additional benefit of improving land could be to reduce crime.”
Future work includes conducting a longer term data analysis of the vacant lot cleanup and running a similar trial that focuses on bringing vacant houses to code.
The study was funded by the National Institutes of Health (Grant R01DA037820) and the Centers for Disease Control and Prevention (Grant R49CE002474). Researchers include Penn’s John MacDonald, professor of criminology and sociology in the School of Arts and Sciences; Eugenia South, assistant professor of emergency medicine in the Perelman School of Medicine; Douglas Wiebe, associate professor of biostatistics and epidemiology in Perelman; Charles Branas, chair and professor of epidemiology, Columbia University’s Mailman School of Public Health; Philippe Bourgois, professor-in-residence, University of California, Los Angeles; Michelle Kondo, research social scientist, U.S. Forest Service; and Bernadette Hohl, assistant professor, Rutgers University School of Public Health.
This piece originally appeared on
