"You remove the abandoned automobiles, all the abandoned trash where you can hide firearms," said lead author Charles Branas, an associate professor of epidemiology at Penn's Perelman School of Medicine. "All of a sudden, it makes it more challenging for illegal guns to move around the neighborhood."
The authors said the decrease in crime might also be partly explained by the "broken-windows" theory, which proposes that boarded-up windows, vandalism, and other signs of decay can serve as a magnet for serious crime, because they suggest that no one is in control.
Told of the study findings, Police Commissioner Charles H. Ramsey said both explanations made sense - the elimination of hiding places for guns, and the broader perception that an environment is safe.
"If you bring something special to a neighborhood, it helps to make people feel a little bit better about where they are and where they live," Ramsey said, adding that he expects the city's Mural Arts Program to have a similar impact.
Other studies have linked crime rates with elements of the surrounding environment, including a Penn study on take-out liquor stores. But the new research appeared to be the first to look at the effects of transforming vacant lots.
The greening program, a major effort begun under Mayor John F. Street that has received national attention as a weapon against urban blight, involves trash removal and the planting of grass and a few trees. A low wooden fence is typically erected around the border.
Deborah McColloch, director of the city Office of Housing and Community Development, which oversees the program, said she initially saw it as an economic-revitalization tool, not as a means for reducing crime. She said she learned of the Penn research while it was still under way, and was intrigued.
"We were excited because it added this other layer of interest to something that we were already working on," McColloch said.
The greening effort is paid for with city and federal funds, and the amount devoted to clearing new lots varies from year to year. Once cleared, they receive regular maintenance from crews hired by the Horticultural Society, at a total cost of about $600,000 to $700,000 per year, McColloch said.
The research on the crime rates was funded by the Centers for Disease Control and Prevention and the National Institutes of Health.
Daniel Webster, a prominent gun-violence researcher at Johns Hopkins University who was not involved in the Penn study, said he was so impressed by it that he planned to use it in his classes.
The hiding of firearms on vacant lots is a significant phenomenon, said Webster, who has studied how young men first become involved in gun-related activity.
"Often the first guns that teenagers possess are ones that they find," said Webster, codirector of the Center for Gun Policy and Research at Hopkins.
While there was a net reduction in gun-related crimes near the greened lots, the study authors saw a net increase in reports of disorderly conduct in these neighborhoods. This could be because the beautified lots had instilled a sense of collective ownership, and thus a tendency to report minor offenses that might otherwise been met with resignation, said John MacDonald, one of the Penn researchers.
"It creates an environment where people feel like there's a stake in reducing illicit behavior on these properties," said MacDonald, an associate professor of criminology.
The researchers analyzed crime data from 1999 to 2008, both in the areas surrounding the 4,436 greened lots and in the areas surrounding 13,308 lots that had yet to be improved - matching three vacant "control" lots with each treated space.
They carved up the numbers on three levels: By counting the number of crimes in the same census tract as a given lot; by counting crimes in smaller units called block groups (which measure roughly four blocks in Philadelphia); and by using a "point-based" approach - assigning different statistical weights to crimes depending on how far they were from the lot in question.
Crime numbers were adjusted to account for any socio-economic differences among neighborhoods.
While the greened lots appeared to play a role in keeping certain crimes in check, the raw numbers didn't necessarily go down in those neighborhoods. The citywide rate of gun assaults, for example, stayed about the same before and after greening - roughly 20 assaults per square mile per year, at a glance, near the improved lots. But near the vacant lots, there was a significant increase in gun assaults - in unadjusted numbers, a jump of roughly four crimes per square mile per year.
In other words, by studying both kinds of neighborhoods before and after, the authors showed a net effect of greening for certain offenses: either the number of crimes went up by less near the greened neighborhoods than near the vacant lots, or it decreased by more.
The researchers didn't study murders and shootings because there weren't nearly enough to demonstrate statistical associations with so many thousands of lots. The study also did not evaluate lots in the Northeast, as the greening program has taken place predominantly in other parts of the city.
Separately, the researchers also studied whether the greened lots had an impact on health, using data from a biannual survey. Residents who lived near greened lots reported that they felt less stress and engaged in more exercise than those near blighted lots, though they were also more likely to report high cholesterol.
The authors said these health findings were less reliable, in part because the survey suffered from a low response rate.
Nevertheless, one community leader said she was sure that greened lots were good for residents' health.
"It's a walkable community," said Rose Gray, vice president of community and economic development for the nonprofit group Asociación Puertorriqueños en Marcha.
Bob Grossmann, who oversees the lot-greening program for the Horticultural Society, said none of these benefits comes as a surprise. But to have them outlined in the pages of an academic journal is a welcome development, he said: "It's been great to see it verified through data."
Contact staff writer Tom Avril at 215-854-2430 or email@example.com.