Reducing Crime Through Environmental Design: Evidence from a Randomized Experiment of Street Lighting in New York City

The field experiment described in this paper was conducted in the Spring and Summer of 2016 in NYC. Through a unique partnership between NYPD and the NYC Mayor's Office of Crime Justice, we randomized the provision of temporary street lights to the city's public housing developments, allowing us to avoid the potential challenges that could result due to spurious time trends as well as selection bias. Managed by NYCHA, NYC's public housing developments are officially home to more than 400,000 New Yorkers (and perhaps an additional 100,000 non-official residents), making NYCHA the second largest landlord in the United States after the U.S. military. NYC's official public housing population is indeed large enough to place it among the forty largest cities in the United States, making it an ideal setting to study the effect of street lighting in urban areas.

Given the cost of providing enhanced street lighting at scale, the city wanted to launch a pilot study to investigate the extent to which increased lighting would be effective in reducing serious crime. We worked closely with the City for nearly two years to develop the field experiment described in this paper. The intervention deployed temporary lighting towers to housing developments across NYC. These towers emit approximately 600,000 lumens---a measure of brightness---making them extraordinarily luminous. Towers were equipped with an automatic timer set to switch on at sunset and off upon sunrise.

2016-02-29

2016-08-31

Index crimes occurring outdoors at during the nighttime hours

Index crimes occurring outdoors at during the daytime hours

In order to select developments for the study, the New York Police Department provided a list of 80 high-priority developments based upon their elevated crime rates and perceived need for additional lighting from among the 340 NYCHA developments in NYC. From this list, we randomized 40 developments into a treatment condition that would receive new lights and 40 developments into a control condition via paired random sampling, stratifying on each development's outdoor nighttime index crime rate and size in the two years prior to the intervention; treatment developments were then randomly assigned a lighting dosage.

In order to maximize statistical power, we also randomly assigned the dosage of lighting among the treatment group. Three hundred and ninety-seven lighting towers were available to be randomly assigned amongst the treatment group. For operational reasons, the City decided that two light towers would be allocated to each campus, regardless of square footage. The remaining 319 lighting towers were assigned to the 39 developments according to a random number drawn from a uniform distribution linked to the square footage of the developments. This created exogenous variation in the the number of lights per square feet across the developments.

Light towers were deployed in the field between February 29, 2016 and March 7, 2016; the lights remained illuminated during all nighttime hours for the following six months. Control group developments received no additional outdoor lighting ("business-as-usual").

Randomization was completed using the statistical software package, Stata.

The unit of randomization is the public housing community.

Yes

40

80 (for treatment vs. control analysis) 40 (for analysis of the effect of lighting dosage)

2 (for treatment vs. control analysis)