Body-Worn Cameras and Settlements

AbstractBody-worn cameras (BWCs) have been widely adopted in the United States due to their perceived promise of reducing police misconduct through a deterrent or Hawthorne effect. Much of the available evidence indicates that BWCs reduce civilian complaints, use-of-force, and other police misbehavior. However, there is a tendency for existing research to study minor forms of misbehavior rather than serious misconduct. This study focuses on examining how settlements, a proxy of serious police misconduct, changed with BWC adoption and are an important outcome to study as they may correspond with financial and litigation burdens for police departments. This study uses panel data from 19 police departments over 10 years to assess how settlements changed post-BWC adoption. The results indicate BWC adoption is associated with a 40% reduction in police settlements and a 40% decrease in the median cost per settlement, and suggest BWCs may lead to less severe police misconduct over time.Keywords: Body-worn camerascivil litigationlaw enforcementpolice accountabilitypolice misconduct Disclosure statementNo potential conflict of interest was reported by the author(s)Notes1 Four agencies (Fairfax County [VA], Orange Counties [CA, FL], Palm Beach County [FL], and Riverside County [CA]) were excluded from the study because they did not indicate the year of disbursement. New York was not selected for the paper because their number of payments was 23 times higher than the next highest agency (i.e., New York is an influential outlier).2 A few agencies do not report dates for settlements so the counts for some agencies are artificially deflated. The affected agencies include: Detroit (66 settlements); Indianapolis (four settlements); Phoenix (two settlements); Portland (three settlements). Atlanta does not report settlement data for 2010-2013 so those observations are recorded as missing.3 Linear interpolation was accomplished by filling in missing values between pre-deployment (0) years and post-deployment (1) years as a function of time to model partial deployment (e.g., .2, .4, .6, etc.). This strategy was used as many agencies adopted BWCs over time.4 Jacksonville (FL), Indianapolis (IN), and Prince George's County (MD) fully adopted BWC after 2019 so their adoption is modeled as partial when applicable.5 Phoenix (AZ) and Minneapolis (MN) were targeted for investigation in 2021. Detroit (MI), Los Angeles (CA), Prince George's County (MD), and Washington D.C. completed their consent decree before or during the study period. Baltimore (MD), Chicago (IL), Cleveland (OH), Miami (FL), Portland (OR), and Seattle (WA) fell under DOJ oversight during the study period or completed their consent decree outside of the study period).6 The number of officers is used as an exposure term in the total number of suits model. In the median amount models, the officer rate per 1,000 people is used as a control variable.7 For a follow-up analysis, BWC adoption was modeled as a three-point categorical variable to represent pre-adoption (0), partial deployment (1), and full deployment (2). The results from this analysis align with those presented in the paper.Additional informationNotes on contributorsZachary A. PowellZachary A. Powell is an associate professor in the School of Criminology and Criminal Justice at California State University, San Bernardino. Powell’s work focuses on a variety of police accountability issues. His work has been published in Criminology and Public Policy, Policing: A Journal of Policy and Practice, and Social Forces.