Seminar Rising Temperatures and Domestic Violence in Peru: Evidence and Mechanisms

Abstract

Climate change is increasing the frequency of extreme heat events, yet consequences for household violence remain poorly understood. This paper provides causal evidence on the relationship between daily temperature and domestic violence (DV) in Peru, where over half of ever-partnered women report having experienced intimate partner violence.

We combine three administrative datasets — police reports, Emergency Center visits, and a national helpline — covering 1.8 million incidents across 1,874 districts over 2017–2024, matched to high-resolution daily weather data. Days above 32°C increase police-reported DV by 19%, and cumulative lag models confirm these are genuinely new incidents rather than rescheduled violence.

We investigate three mechanisms. Temperature increases male alcohol consumption, with proportionally larger heat-driven increases in DV involving intoxicated perpetrators. Temperature also reduces hours and labour income among informally employed women — but not among men or formally employed women — pointing to the absence of contractual income protection as the binding constraint, though we acknowledge the direction of causality between labour outcomes and violence remains an open question. We find no evidence for a depression pathway using the PHQ-9 as a screening tool.

The findings suggest that income-buffering policies targeting informal workers — such as heat-triggered extensions of conditional cash transfers — could reduce climate-driven domestic violence.