Injuries, Collisions Declined Significantly in First Seven Months; Speeding Violations Subsequently Fell, Suggesting Influence on Drivers’ Behavior

In the United States, automobile crashes are responsible for about 40,000 deaths and 2 million injuries a year, resulting in an estimated $340 billion in economic costs. In a new study, researchers examined the effectiveness of automated speed cameras, which detect and fine speeding motorists, over a 10-year period. Within the first months, the cameras reduced collisions and injuries, and speeding violations subsequently declined, suggesting influence on drivers’ behavior. Use of the cameras also addresses concerns about race-based harms associated with conventional approaches to traffic safety.

The study, by researchers at Rutgers University, the University of Chicago, and Columbia University, is published in PNAS.

“Our findings demonstrate that automated speeding enforcement can reduce dangerous driving and prevent crashes, while diminishing concerns about the unequal policing associated with traditional traffic enforcement,” explains Aaron Stagoff-Belfort, a PhD student in sociology at the University of Chicago, who led the study.

Traffic enforcement has historically been the responsibility of local law enforcement agencies. Traffic stops are now the most common form of contact between police and the public, and these encounters have expanded the scope of racially biased policing, as officers conduct so-called pretextual stops for minor violations that are often unrelated to road safety or crash prevention, and that can turn fatal.

In their study, researchers examined the effectiveness of nearly 1,900 automated speed cameras set up in New York City between 2014 and 2023. The study is among the first to offer causal evidence of the safety impacts of automated speed enforcement, and did so in the largest program in the United States and one of the most extensive automated enforcement programs ever implemented.

In the first seven months following activation, speed cameras reduced total collisions by around 30% and collision-related injuries by around 16% on average within 900 feet of intersections, the study found. The largest effects were observed in Manhattan and Brooklyn, and in areas with higher baseline traffic volume and greater intensity of camera enforcement.

Beginning in the second month of use, speeding violations dropped sharply, indicating the rapid influence of automated enforcement on drivers’ behavior.

“Many U.S. cities are turning to automated speed cameras as a promising alternative to traditional traffic enforcement by police, aiming to reduce collisions without the risks of racial bias or harmful police encounters,” notes Jonathan Ben-Menachem, a PhD student in sociology at Columbia University, who coauthored the study. “Our findings provide compelling evidence for an intervention that equitably addresses one of the nation’s leading public safety challenges.”

Among the study’s limitations, the authors note that they studied just one city, limiting the generalizability of their findings to jurisdictions with different types of programs. Also, their data on collisions are based on incidents reported to the police, which may undercount the actual number of crashes. And because New York City’s camera program is limited to school zones, the use of cameras in these areas may differ from use in other areas.

“Automated speeding enforcement can serve as a central pillar of effective traffic safety programs, reducing vehicle crashes, promoting changes to drivers’ behavior, and mitigating the negative consequences of relying solely on police to manage road safety,” adds Brenden Beck, associate professor at Rutgers-Newark School of Criminal Justice, who coauthored the study.

###

Summarized from an article in the PNAS, Can Speed Cameras Make Streets Safer? Quasi-Experimental Evidence from New York City, by Stagoff-Belfort, A (University of Chicago), Ben-Menachem, J (Columbia University), and Beck, B (Rutgers University). Copyright 2025 National Academy of Sciences. All rights reserved.