This study examines how COVID-19 mobility restrictions affected crime patterns using detailed spatiotemporal police-recorded crime data from São Paulo State, Brazil. In particular, the analysis captures changes in crime across both spatial dimensions---ranging from the state to municipalities and census blocks---and temporal dimensions, including years, weeks, and times of day, providing a comprehensive understanding of how crime evolved during the pandemic.
The results show that property crimes, such as theft and robbery, temporarily declined following the introduction of mobility restrictions. However, the impacts varied substantially across regions and time periods, indicating that the effects were not uniform. For example, differences were observed across neighborhoods, with distinct patterns in urban slums (favelas). In addition, changes were observed in crime hotspots and time-of-day patterns, revealing more nuanced shifts in crime.
These findings highlight the importance of examining not only overall trends but also when and where crime occurs. The study offers a new perspective and analytical framework for understanding more comprehensively how major societal disruptions, such as pandemic-related restrictions, influence crime, and provides insights that may inform future crime prevention and urban policy.
[Article Information]
Title: A Multi-Scalar Causal Approach across Space and Time: An Application to the COVID-19 Pandemic in São Paulo State, Brazil
Authors: Katsuo Kogure, Yoshito Takasaki
Journal: Journal of Quantitative Criminology (an international peer-reviewed journal in criminology)
Publication date: April 15, 2026