Abstract: I use two sources of spatially disaggregated data, crime records and calls to a helpline, to document a positive and linear association between ambient temperature and domestic violence in Mexico City. My findings rule out that this association is entirely driven by changes in the reporting behavior of victims. I find that the association between temperature and domestic violence is stronger for domestic violence than for other types of crime: Higher temperatures, by increasing irritability and victimization risk, lead to more domestic violence incidents. My results show a differential impact of temperature even over small geographies within the city. I document a large heterogeneity across neighborhoods, and show that the impact of temperature is strongly correlated to neighborhood characteristics such as poverty and housing attributes.
Presented at: Maastricht Workshop on Applied Economics of the Environment (2024), 6th QMUL Economics and Finance Workshop (2024), Tor Vergata Ph.D. Conference (2024)