Over a million people in the United States are employed in private security and law enforcement, yet very little is known about the effects of private police on crime. The current study examines the relationship between a privately-funded university police force and crime in a large U.S. city. Following an expansion of the jurisdictional boundary of the private police force, we see no short-term change in crime. However, using a geographic regression discontinuity approach, we find large impacts of private police on public safety, with violent crime in particular decreasing. These contradictory results appear to be a consequence of delayed effect of private police on crime.