The Philadelphia Foot Patrol Experiment: Randomized Evidence That Foot Patrol Can Reduce Violent Crime

AI Summary: Research question: Can placing police officers on foot patrol in violent crime hotspots actually reduce violent crime?

Bottom line: Yes. In the first large-scale randomized controlled trial of foot patrol in violent crime hotspots, the Philadelphia Police Department assigned more than 200 rookie officers to 60 violent crime hotspots during the summer of 2009. Compared with matched control locations, violent crime fell by 23 percent, preventing an estimated 53 violent crimes after accounting for crime displacement. Importantly, the benefits were concentrated in the city's highest-crime hotspots, suggesting that foot patrol works best when deployed in places with a sufficiently high baseline level of violence.

For decades, criminologists generally believed that foot patrols did not reduce crime.

That conclusion was largely based on the famous Newark Foot Patrol Experiment conducted in the late 1970s. While foot patrol improved public confidence in the police and reduced fear of crime, it appeared to have little measurable impact on actual crime rates. As a result, many police agencies continued using foot patrol primarily as a community policing strategy rather than as a crime reduction tool.

We believed it was time to revisit that question. This is an AI-generated summary of our research article.

Advances in crime mapping, geographic information systems, and hotspot policing had fundamentally changed how police could deploy officers. Instead of assigning foot patrols across large beats, officers could now concentrate on very small areas where violence consistently clustered.

The Philadelphia Foot Patrol Experiment was designed to determine whether that difference mattered.

A Randomized Controlled Trial

Working with the Philadelphia Police Department, we identified 120 violent crime hotspots across the city using several years of geocoded violent crime data. The locations were paired according to their crime levels and then randomly assigned so that 60 hotspots received foot patrol officers while 60 comparable hotspots served as controls.

More than 200 newly graduated police officers were deployed in pairs to patrol these small geographic areas for up to 16 hours per day, five days each week, over a twelve-week summer period. The patrol areas averaged only about 1.3 miles of streets, allowing officers to maintain a highly visible and continuous presence.

Because the experiment used random assignment, it represents one of the strongest research designs available for evaluating policing strategies.

The Results

The findings challenged decades of conventional wisdom.

Compared with the matched control hotspots, violent crime in the foot patrol areas declined by 23 percent. After accounting for both crime displacement and diffusion effects, the intervention prevented an estimated 53 violent crimes during the twelve-week experiment.

This was not simply a matter of officers generating more activity reports.

The reductions occurred in serious violent offenses including homicide, robbery, and aggravated assault occurring in public spaces. The results demonstrated that carefully targeted foot patrols can produce measurable reductions in violence.

Not Every Hotspot Benefited Equally

One of the most important findings emerged only after examining the data more closely.

Foot patrol did not reduce violence equally across all hotspot locations.

Instead, the intervention was most effective in the highest-violence hotspots. Once locations reached a threshold of approximately six violent crimes during the three-month pre-intervention period, the treatment areas consistently outperformed comparable control locations. In lower-crime hotspots, the differences were much smaller and generally not statistically significant.

This finding has major implications for police deployment.

Rather than distributing officers evenly across many moderately active locations, agencies may achieve greater crime reduction by concentrating foot patrol resources in the relatively small number of places experiencing the highest levels of violence.

Why Did It Work?

The study was not designed to isolate the precise mechanism responsible for the crime reduction, but several explanations are plausible.

Highly visible officers increase the perceived certainty of detection. Potential offenders are more likely to believe they will be interrupted, questioned, or arrested when uniformed officers are consistently present in small geographic areas. In deterrence theory, increasing the perceived certainty of punishment is generally more influential than increasing its severity.

The officers also engaged in substantial proactive policing.

Compared with control locations, treatment areas experienced large increases in pedestrian stops, vehicle stops, disorder interventions, narcotics enforcement, and arrests. Interestingly, these proactive activities were concentrated most heavily in the highest-crime hotspots—the same places where crime reductions were greatest.

Whether the reductions resulted primarily from officer visibility, proactive enforcement, community engagement, or some combination of these factors remains an important question for future research.

Did Crime Simply Move Somewhere Else?

Whenever police concentrate resources in one place, an obvious concern is displacement.

We specifically examined whether violent crime simply shifted into nearby streets.

Some displacement did occur around the treatment areas, but it was considerably smaller than the crime reduction achieved inside the hotspots themselves. After accounting for those surrounding areas, the experiment still produced a substantial net reduction in violent crime across the city.

The overall public safety benefit therefore remained positive.

Why This Research Matters

The Philadelphia Foot Patrol Experiment helped reshape thinking about one of policing's oldest tactics.

Earlier research had suggested that foot patrol mainly improved public confidence without reducing crime. Our findings demonstrated that this conclusion was incomplete.

Foot patrol can reduce violent crime—but where officers are deployed appears to matter enormously.

The study also reinforced a broader lesson emerging from evidence-based policing. Police strategies are often most effective when they focus on micro places where crime concentrates, rather than spreading resources thinly across entire neighborhoods or districts.

Targeting the right places may matter as much as choosing the right strategy.

Citation

Ratcliffe, J. H., Taniguchi, T., Groff, E. R., & Wood, J. D. (2011). The Philadelphia Foot Patrol Experiment: A randomized controlled trial of police patrol effectiveness in violent crime hotspots. Criminology, 49(3), 795–831.