Abstract
This paper studies the effect of weather and air pollution on violent crimes and homicide. We use all crimes committed in New York City between 2006 and 2020 and match each individual crime to the measures of the nearest weather, pollution station respectively. Our results show that neither weather nor pollutants have a contemporaneous effect of on homicide. In contrast, higher concentrations of carbon monoxide increase violent crimes in a U-shaped relation, that more rainfall linearly reduces violent crimes, and that sulphur dioxide has an inverted-U shaped relationship with violent crimes. Our results have implications for policing and environmental regulation.