Understanding interpersonal violence: the impact of temperatures in Mexico

Produced as part of the Evaluating the performance of climate policies CCCEP research programme theme

This research investigates how and why high temperature has an effect on criminality, using data from Mexico. The mainstream theory is that human psychology could be affected by heat, leading people to act violently. Other theories suggest that committing a crime may be easier, or that social interactions are more frequent, during a hot day.

The study uses daily crime records for Mexico on accusations and processed individuals from the ‘First Instance Courts’ from 1997–2012 and daily temperature and rainfall data from the Mexican National Weather Service. The research shows that an increase of 1°C linearly increases the accusation rate of all types of crimes by 1.3%. The crime accusation rate is expected to be one third higher during a hot day (>32°C) than during a cold day (<10°C).

These results are not fully compatible with the idea that a psychological or physiological effect would cause increases in criminality. This is because psychological effects should only appear outside the comfort zone of the human body (20–25°C). Additional analyses show that a significant share of weather-related crimes can be explained by higher alcohol consumption (9%) and changes in time use during weekends (17%). Also, 28% of weather-related crimes are committed at night, when temperatures are mild.

This research therefore suggests that the additional crimes are committed because of changes in time use and habits by both criminals and their victims during hot days. This implies that increased surveillance and prevention policies during hot days could be more effective in managing heat-related criminality than physiological moderators such as air conditioning. However, more research is needed to assess the effectiveness of different measures in reducing heat-related criminality.

Key points for decision-makers:

  • An increase in 1°C increases the rate of accusations of all types of crime by 1.3%.
  • The relationship between temperature and crime is found to be linear. Physiological changes cannot alone explain the results because they are controlled by thermoregulation and not linked to temperature linearly.
  • A significant share (9%) of weather-related crimes can be explained by higher alcohol consumption.
  • Another share (17%) can be explained by changes in time use during weekends.
  • Also, 28% of weather-related crimes are committed at night, when temperatures are mild.
  • The researchers suggest that surveillance and prevention policies may be more effective in helping to manage crime on hot days compared with physiological moderators (for instance air conditioning).
  • The researchers find that Mexico would experience an increase in accusation rates by 3.3% under the RCP4.5 climate change scenario by the end of the century.

ISSN  2515-5709 (Online) – CCCEP Working Paper series

ISSN 2515-5717 (Online) – Grantham Research Institute Working Paper series